Jump to content

David Broadland

Moderators
  • Content count

    50
  • Joined

  • Last visited

About David Broadland

  • Rank
    Advanced Member

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. David Broadland

    Landslide Lisa's record as mayor of Victoria

    Stephen, Thank you for your comment. Just to clarify, we learned you were running for a council seat the day this went to press. In any case, I wouldn't have contacted you for comment had I known earlier. Your 2014 interview with Lisa Helps speaks for itself. This story is about Mayor Helps' record, not yours. But I believe your interview of Helps could be useful to voters in terms of comparing what she said in 2014 compared with what she did. I would recommend to voters that they read your interview. That's why you were included. My apologies if there's any implication in my story that you interviewed Helps with the idea in mind that you might run for mayor, too, in 2014. That wasn't my intention. At the time, I knew you as a dedicated journalist and believed that would be your path into the future. When you announced you were running for mayor, I was as surprised as anybody. You fail to mention above that Focus' Judith Lavoie interviewed you and that your story was on an equal footing with our stories about the other mayoralty challengers. I agree with you that you raised questions about Gene Miller's writing following the 2014 election. We listened to you. To the extent that we agreed with you, we have taken a different course this time around. Thanks again for your comments, and I wish you the best at the polls. David Broadland
  2. For this reporter, three key moments defined Mayor Lisa Helps’ controversial first term. BACK IN OUR January 2014 edition, in a story titled “Tough questions for Lisa Helps,” freelance journalist Stephen Andrew played cat-and-mouse with then City of Victoria Councillor Lisa Helps over whether she was running for mayor. We had asked Andrew to interview Helps because we’d heard she was ready to declare. Andrew couldn’t get Helps to admit the obvious—and he gave no hint whatsoever that he was considering running too. Helps told Andrew that, as mayor, she would never go to China (on a trade mission). In the next four years she went twice. She told Andrew “things are very positive” that the new Johnson Street Bridge would “come in on time and on budget.” The bridge, then scheduled to be open to traffic in the fall of 2015, is still unfinished, its final price still unknown but likely to be over $115 million (voters approved a $77-million project). Helps told Andrew that Victoria couldn’t afford a new Crystal Pool; now she favours building a pool at a cost that’s pushing $80 million. In the election later that year, Helps did run for mayor. So did Stephen Andrew. Helps defeated incumbent Mayor Dean Fortin by 89 votes, earning her the moniker “Landslide Lisa.” Andrew came in fourth, trailing Ida Chong. (Andrew has announced he’s seeking a seat on council in the upcoming election.) Helps seems to have a clearer path to victory this time around. None of her announced competitors have the kind of name recognition she does, and that may be all it takes for her to hold on to her $104,000-per-year job for another four years. Below, I will take you through three issues Mayor Helps faced in her first term. The mayor agreed to answer questions submitted to her by email (A link to our questions and the mayor's full responses can be found at the end of this article.). Then-councillor Lisa Helps in 2014, before Sir John A. Macdonald's statue was an issue for her. (Photo by Tony Bounsall) LET'S START WITH THE HURRIED REMOVAL of the statue of Sir John A. Macdonald from the grounds of City Hall in August 2018. Helps’ public description of how this came about links the action back to the City’s commitment in 2015 to undertake the Calls to Action identified by the Truth and Reconciliation Commission (TRC) that were within municipal government authority. Yet, following Victoria’s removal of the statue, former Truth and Reconciliation Commissioner Murray Sinclair told The Canadian Press, “The problem I have with the overall approach to tearing down statues and buildings is that it is counterproductive to…reconciliation because it almost smacks of revenge or smacks of acts of anger, but in reality, what we are trying to do, is we are trying to create more balance in the relationship.” Revenge? Anger? How did Victoria suddenly become Canada’s poster child for how not to conduct reconciliation with First Nations? First, let’s place the issue in its proper context. The stated goals of the TRC were to “raise awareness of the history and impacts of the residential school system,” and to “enable a process of healing and reconciliation between those affected and non-Aboriginal governments, communities and individuals.” In 2012, as many as 3,000 people attended a TRC event in Victoria, many of them non-Aboriginal. The TRC set a high standard for openness and inclusion in order to create what Sinclair describes as “more balance in the relationship.” Sadly, the process that led to the removal of the Macdonald statue appears to have involved just three members of the non-Aboriginal community and had zero transparency. Mayor Helps played the lead role in the process that led to this “counterproductive” action. The extraordinary committee behind the removal of the statue was established at an in camera meeting of City council in June 2017. Creation of that committee was based on a seven-page recommendation authored by Helps and Councillor Marianne Alto. Part of their recommendation was that the activities of the committee, which they called “the City Family,” would only be reported to the public “at the discretion” of Helps. The document is striking for its lack of clearly defined objectives. None of the five recommendations of the TRC that were specific to governments were addressed. Instead, Helps and Alto requested a looser arrangement in which the Family would “take responsibility for doing that work with integrity, an open heart, and a willingness to work in diverse ways and take the time needed.” They declared: “Reconciliation is the way forward; it is the process, not the outcome. Reconciliation is how, not what.” Indeed, the only concrete action Helps and Alto foresaw, aside from paying First Nations members of the City Family for their participation, was their idea to “document the program on film, as a record of the work and for observation and use by other municipalities and/or organizations interested in a Reconciliation program.” Notably absent from the document was the signature of City Manager Jason Johnson, who, along with Helps and Alto, had met with local First Nations leaders as the process was developed. (Johnson was fired by City council soon afterwards. No explanation for his termination has ever been provided.) I asked Mayor Helps which of the five TRC “Calls to Action for Government” justified, in her view, removal of the Macdonald statue. “Not every act of reconciliation is specifically dictated by a single TRC Action,” Helps responded. “The entirety of the TRC Report suggests the complexities that will challenge government at every level as they strive to take actions that are meaningful and make a real difference in relationships between Indigenous and non-Indigenous communities.” Helps also quoted some of the opening lines of the TRC Report, which linked Macdonald to Canada’s residential school system for Aboriginal children. She also listed three of the principles of reconciliation developed by the TRC, but made no attempt to show how those principles were interpreted by the City Family to support removal of the statue—a conclusion Sinclair does not support. There is no record of the monthly City Family meetings, held in the mayor’s office, that’s available to the public, so we are unable to examine its decision-making process using the methods by which the actions of municipal governments are normally monitored. But one would think that if the City Family was intent on building on the foundation created by the TRC, there would be a record of the City moving to implement the specific recommendations of the TRC—the ones that don’t need any interpretation. The most specific of these was this: “We call upon federal, provincial, territorial, and municipal governments to provide education to public servants on the history of Aboriginal peoples, including the history and legacy of residential schools, the United Nations Declaration on the Rights of Indigenous Peoples, Treaties and Aboriginal rights, Indigenous law, and Aboriginal-Crown relations.” I asked Helps if the City of Victoria had acted on that specific call to action. “Staff are considering how best to provide this education in a formal programmatic way,” Helps said. So, three years after the City committed to adopt the TRC recommendations, it has done little more than “consider” the most specific of the TRC’s recommendations to government. In the City’s defence, Helps pointed to two events it held related to reconciliation: “Five employees from the Mayor/City Manager’s office attended a seven-hour learning event ‘Reconciliation—Journey of our Generation’ in April 2018 presented by Dialogue and Resolution Services. Many staff also attended a City Hall Lunch Time Lecture Series on December 4, 2017: Speaking our Truth: A Journey of Reconciliation with Monique Gray Smith.” Helps also listed a number events involving local First Nations and the City of Victoria, but it’s clear from what’s missing in her response that the City has made little effort to create even the educational program for City employees called for by the TRC. Why, then, was a largely unaccountable committee that included three City councillors allowed to free-range around First Nations’ issues? As I mentioned above, Helps and Alto recommended to City council that any public reporting of the City Family process should be at Helps’ discretion. Had the mayor ever released any information about what the Family was considering? The only evidence Helps offered was an op-ed penned by her in the Times Colonist in September 2017. While her op-ed mentioned a debate on social media about Macdonald, Helps implied there would be no removal of the statue without first engaging with the larger community. Her op-ed noted: “It’s in this deeper context that we’ll be able to have a conversation about Macdonald’s future at the doors of city hall.” That conversation, of course, never happened. Helps’ Times Colonist op-ed raises questions about her own understanding of the role municipal governments have in First Nations issues. In it she conflated “reconciliation” with non-Aboriginal support for First Nations’ land claims. Helps wrote, “As part of the [reconciliation] process, we need to understand as a council and as a community what role city hall and local settlers played in removing the Songhees and Esquimalt Nations from their lands upon which the city was built… Once we understand the role of the city and local settlers in dispossession and decolonization [sic], we can acknowledge our wrongdoing, provide appropriate restitution and make an apology.” Helps no doubt meant “colonization,” not “decolonization.” But her stated intention of extending “reconciliation” to include consideration by Victoria City council of First Nations’ traditional territories and “restitution” is striking. Helps seems to be venturing well outside her area of responsibility as mayor of Victoria and intruding into the legal domain of the provincial and federal governments, which are responsible for making final treaties with First Nations. Ironically, after committing to the TRC’s principles of reconciliation in 2015, the first gesture the City made to the Esquimalt and Songhees people was the removal of the derelict Checkers Pavillion on top of Beacon Hill. That place has historic, cultural and sacred significance to First Nations. Chief Andy Thomas of Esquimalt Nation had openly requested that the City give back the small area occupied by the pavillion so the Songhees and Esquimalt Nations could build a longhouse on the site. The City agreed to the longhouse; however, it did not agree to transfer title of any land. The process of deciding to remove the pavillion was conducted in full public view as municipal governments are required to do. There was no uproar. When Helps was running for mayor in 2014, she told Focus one of the three main planks in her platform was greater public engagement. “I think there’s a real disconnection to City Hall right now because of the lack of meaningful public participation on a whole raft of issues, big and small,” she said then. The exclusion of the public from her consideration of the statue’s fate—which turned out to be a really “big” issue—brings into question the sincerity of Helps’ commitment to public engagement. What has the experience taught her? “I understand that this process didn’t feel good to the public and I wish we had known or thought to keep the public and council more in the loop about the work of the City Family, in a way that also feels true to the Family’s process,” she said. “That is the big lesson learned from this—the hunger the community has to participate in reconciliation and the feeling that this process somehow took that away from them.” But Helps’ “I wish we had known...” is at odds with a statement made to local media by Janice Simcoe, one of the First Nations members of the City Family. Simcoe told the Times Colonist: “We expected there would be opposition, uproar…I think it was the only process that could have taken place without all this coming to a screeching halt…I don’t have any regrets.” So Helps must have known, too, that she was short-circuiting public process. Yet that didn’t keep her from acting. The mayor’s timing on the removal—a little more than two months before the October 20 civic election—and her inability to explain how the City Family was building on the TRC’s recommendations warrant some skepticism about her leadership and motives. It’s noteworthy that Helps’ re-election campaign played up the predictable backlash on social media and sought donations for her campaign to fight against the “hatred” that was directed at her. According to an email circulated by Helps’ campaign, that backlash helped raise “thousands of dollars.” Let’s move on to another difficult moment in Helps’ first term. Then-Victoria Police Chief Frank Elsner in 2015 ON DECEMBER 4, 2015, Mayor Helps was asked by a Global TV journalist whether Victoria Police Chief Frank Elsner was being investigated. Helps responded: “No. The [Police] Board has full confidence in our chief. He’s the best thing that’s happened to this town and Esquimalt in a long time.” But news reports over the following days quickly proved Helps (and Esquimalt Mayor Barb Desjardins) had misled journalists. An investigation of Elsner had been conducted under their disciplinary authority as co-chairs of the Victoria Police Board. Two weeks later, Police Complaint Commissioner Stan Lowe released a report on the mayors’ investigation of Elsner’s conduct. Lowe’s report confirmed that Elsner had been accused of exchanging “inappropriate” tweets with the wife of another member of the Victoria Police Department (that person is referred to below as “the Member”). But the real substance of his report was its examination of the conduct of Helps and Desjardins. Lowe’s report provided a step-by-step account of how an internal investigation into Elsner’s conduct was initiated, the ways in which the investigation was flawed, his rationale for stripping Helps and Desjardins of their authority to discipline Elsner, and an order for an external, public-trust investigation of Elsner’s conduct. Below, I’m going to focus on just one aspect of Lowe’s report, which was Helps’ and Desjardins’ failure to abide by two important preconditions that Lowe had stipulated before allowing them to do an internal investigation. An internal investigation meant Helps and Desjardins would make the determination as to whether or not Elsner would be disciplined. In order not to completely overtax the reader’s patience, I’ll consider the mayors’ failure to abide by just one of Lowe’s preconditions. Lowe had stipulated: “There had to be disclosure of the allegations to the Member serving under the command of Chief Constable Elsner, and the Co-Chairs should obtain the Member’s informed views as to whether he wished to initiate a complaint or request a public-trust investigation under the Police Act.” By that Lowe meant that the mayors had to accurately explain to the Member exactly what Elsner had done. Once they were sure he understood what had transpired, he was to be asked whether he wanted the mayors to proceed with an internal process or if he wanted the process turned over to Lowe’s office as a public-trust investigation. Lowe’s report states that Helps and Desjardins agreed to his preconditions in late August 2015. Lowe’s report picked up the story a few weeks later: “…our office was advised by counsel for the Co-Chairs… that the affected Member did not wish an investigation. On the understanding that my two conditions had been satisfied, I supported the decision to proceed with this matter as an internal discipline matter. It was my expectation that if the investigation revealed evidence of conduct that could constitute a disciplinary breach of public trust, the Co-Chairs would raise the matter with our office.” Lowe’s oversight of the Elsner investigation might have ended at that point had it not been for the aforementioned media reporting in early December 2015 in which Desjardins and Helps claimed there had been no investigation of Elsner. Prompted by Helps’ and Desjardins’ false claims, Lowe requested the records of the mayors’ investigation. In his report, Lowe made many highly critical observations. Here we will consider just one. About the mayors’ report, Lowe stated: “[T]he Member is described as advising the Co-Chairs of his meeting with Chief Constable Elsner and the information the Chief provided to the Member. It appears that the Co-Chairs did nothing to correct the Member’s misguided appreciation of the circumstances, despite the Co-Chairs knowing the information provided to the Member was false and misleading. In advising our office that the informational pre-condition had been met, no mention had been made that the Member had received false and misleading information from Chief Constable Elsner. Given the circumstances as contained in the report, it is clear that the Member’s decision was influenced by misleading information; therefore, the pre-condition had not been fulfilled.” Note in particular Lowe’s statement “…despite the Co-Chairs knowing the information provided to the Member was false and misleading…” Lowe is saying, in effect, Helps and Desjardins consciously omitted telling the Member the truth. That is, they lied. I asked Helps for her version of what took place. “We were instructed by our lawyer that the Office of the Police Complaint Commissioner [OPCC] required us to do four things: meet with the Member whose wife was involved in the matter and advise him generally of the investigation, meet with the Chief and advise him that there was going to be an investigation, advise the Board of the matter and hire an independent investigator. Although we were most uncomfortable meeting with the Member who was impacted, we reluctantly agreed to meet with him after the OPCC insisted that we do so. After we had done each of the four things we were directed to do, the investigation proceeded without our interference.” Helps agreed that Elsner had added to his problems by not being forthright with the Member: “It has now been determined by another discipline authority in an external discipline process that former Chief Elsner provided false information to the Member whose wife was involved in the matter. That is very serious misconduct and worthy of significant discipline. The apparent providing of false information to the Member by former Chief Elsner was something completely beyond our control,” Helps asserted. “Further, it was not part of what the OPCC had authorized Mayor Desjardins and I to pursue through the internal discipline process. Our mandate, as authorized by the OPCC, was very limited. We were authorized to deal only with the issues of whether Elsner had engaged in an inappropriate relationship with the wife of a VicPD member and whether Elsner had improperly used police social media accounts.” Helps’ version of events differs with Lowe’s in one significant way. Lowe concluded that Helps and Desjardins knew Elsner had provided the Member with false and misleading information and they did nothing about it. Who is a voter to believe? Let me return to where we started, on December 4, 2015, when Helps was asked by a journalist whether an investigation had taken place. Even though one had, Helps said: “No. The [Police] Board has full confidence in our chief. He’s the best thing that’s happened to this town and Esquimalt in a long time.” In trying to decide whether to believe Lowe or to believe Helps, it’s hard to get past the fact that the mayor lied in public to a journalist about the Elsner situation. It’s evident that Helps’ and Desjardins’ intention was to protect Elsner. If we believe Lowe, then they protected Elsner even though they knew he had provided false and misleading information to the Member. Helps now describes that as “very serious misconduct and worthy of significant discipline.” She was in a position to make that judgment back in 2015. Yet she participated in a cover-up of Elsner’s misconduct anyway (other allegations about Elsner surfaced latter). Helps allowed her enthusiasm for Elsner’s fresh approach to community policing to overwhelm her responsibility as an elected official to side with the truth. The OPCC’s public-trust investigation was completed in 2017. Legal maneuvers by Elsner (at City taxpayers’ expense) have prevented release of the report, but in April 2018 the BC Court of Appeal ruled that Lowe’s action to remove the mayors as the disciplinary authority and conduct a public-trust investigation had been a reasonable interpretation of the Police Act. Now it’s time for voters to make their ruling on the mayors’ conduct. One of two surprising repairs needed on the new $115M Johnson Street Bridge before it even opened. LASTLY, LET’S REVISIT HELPS’ RECORD on what the City has called “the largest infrastructure project in Victoria’s history.” You knew I would get there, didn’t you? In January I wrote about the surprising appearance of two six-foot by six-foot steel plates that had appeared on the steel work of the new bridge. There was one plate bolted onto the underside of each of the bridge’s signature rings, and they amounted to a physical defacement of what had been promised to be an architecturally significant structure. They seemed to me—and many others—to be an obvious insult to the design integrity of the bridge that the City promised would be “world class” and “iconic.” Naturally, I had questions that only the City could answer. The City’s official spokesperson, on everything, is the mayor. That policy is contained in the City’s written guidelines on how employees, including the mayor, should respond to media. So I emailed Mayor Helps a few questions—the first questions I had posed to her in the first three years of her term. Was she aware of the plates? When did she find out? Was the City given any options to consider? These were not difficult questions to answer, but the mayor didn’t respond to five emails. After my story was published, Helps issued a statement through her Facebook page. In that statement she claimed the article contained “a number of serious factual errors and inaccuracies,” but didn’t specify what those were. I sent the mayor more questions, including a request to make public what those “serious factual errors and inaccuracies were.” Normally, a public official that makes such a claim would have proactively provided that information without being asked. That’s the process: If we make a mistake, the subject tells us about the mistake we made, and if they are correct, we acknowledge our error and publish the correct information. So I requested that the mayor make those mistakes clear. Then something peculiar happened. Mayor Helps’ inadvertently copied me on a “proposed response” to my questions that she had meant to send only to acting City Manager Jocelyn Jenkins and private engineering consultant Jonathan Huggett. “Do you see any downfalls in this approach?” the mayor asked Huggett and Jenkins. Later, realizing what she had done, Helps emailed me: “David there you have my response. Sent before my morning meditation and copied to you inadvertently. But truth may walk through the world unarmed. So please feel free to use what I have said.” She had written: “I trust all of the reporters at the Times Colonist. I trust all of the reporters at Vic News. I trust all of the reporters at CBC and CFAX. I trust all of the reporters at CTV, CHEK, and GLOBAL. This trust has come through hard conversations, good reporting and relationship building. I do not trust you. As such I feel that however I answer your questions you will use the answers to suit your own needs, not to serve the public good.” Mayor Helps made no attempt to identify any errors or inaccuracies in my story. In almost 30 years of community reporting, I had never experienced such an evasive response from an elected official. It was very Trumpian: First, accuse the messenger of spreading fake news. Then lock him up in solitary confinement. Mayor Helps has still not provided those alleged “factual errors and inaccuracies.” And no wonder. Six months after I wrote about the plates, the City, in response to an FOI, handed over to Focus the communications between the various engineers involved in adding those steel slabs to the new bridge. The communications showed they had been judged necessary to beef up a structural weakness that had been built into the bridge. The engineers had decided the juxtaposition of four “weld-access holes” in each of the rings could lead to fatigue cracks forming in the fracture-critical steel of the rings. The shape and nature of the holes should have been controlled during fabrication by properly executed fabrication drawings, but detailing in the drawings was inadequate. Ensuring that those details were adequate was ultimately the responsibility of the Engineer of Record on the project. The problematic holes had been created over a year before any effort was made by the project engineers to find a solution to the potential fatigue problem. The records obtained by Focus showed engineers had misled the City about when the problem was discovered, who was involved in considering options, and who was ultimately responsible for creating the problem in the first place. The engineers’ explanation that “schedule” had figured into their decision to add the plates neglected to include the fact that they had done nothing about the problem for over a year, and during that year, additional work on the rings had made it more difficult and costly to properly address the problem. Those communications also answered all the questions I posed to Helps. Nobody at City Hall had been informed. Helps first learned about the plates when I sent her my first set of questions. Moreover, the records obtained from the City showed Helps had been misled by the very engineers who had told her our story was inaccurate. For this article, the one you are reading now, I pointed out to Helps that the engineers she had trusted had actually misled her on important facts about the story. Would the City lodge a complaint with the engineers’ professional association? Her response was confusing. She misremembered how one of the engineers had come to be employed by the City and concluded that the only thing that mattered was that the Engineer of Record had signed off on the project. The Engineer of Record was one of the engineers that had misled the City. What does this story say about Mayor Helps’ record? It shows that she learns from experience. When confronted by a journalist about whether or not Elsner’s conduct was the subject of an investigation, Helps chose to lie in order to protect Elsner, whom she clearly admired for his enlightened approach to community policing. But when Helps was presented with my questions about the bridge, there was no denying the existence of those six-foot by six-foot, 1500-pound plates. She wisely clammed up and let someone else do the lying for her. And her trusted engineers—whom she no doubt admired for their project management expertise and powers of persuasion— obliged. By publicly accepting their misrepresentations, she protected them, too. We might also conclude from her record on the bridge, on Elsner, and on the Macdonald statue that Mayor Helps doesn’t trust ordinary people in the community, and what they have to say about these issues, unless they share her own point of view or those of her advisers. Genuine dialogue might have brought movement in the direction she wanted to go—in each case—to a screeching halt. THIS IS FAR FROM A COMPREHENSIVE examination of Mayor Helps’ first-term record. She played a prominent role in many other issues for both the City of Victoria and the CRD. But as thin a slice as this is, what should be evident to readers is the immense complexity and pressures facing any would-be mayor of Victoria these days. Helps has taken on that complexity with enthusiasm and energy, and the three examples above barely capture the extent to which she has had to spread her efforts. Had she tried to do less, it’s possible she wouldn’t have made the misjudgements I’ve outlined here. But her enthusiasm for participating in government, and her definite position on contentious issues, has driven her to be fully engaged. That has had both costs and benefits for the City. David Broadland is the publisher of Focus Magazine. Questions Focus asked and Mayor Helps' full responses: Questions Focus asked and Mayor Helps' full responses.pdf
  3. The distance travelled in autos each day by CRD residents continues to grow, but there is a surge in the uptake of all-electric cars. THE REGION'S LATEST EXAMINATION of how people here currently choose to get from point A to point B as they go about their daily lives has been released by the CRD. And, no surprise, the Origin Destination Travel Survey, based on the participation of 7,159 households last fall, shows that about 84 percent of the total distance travelled by CRD residents takes place in private automobiles. And, in 2017, we collectively drove—or were driven—an additional 50,000 kilometres each day in private autos, compared with 2011. Victorians’ overwhelming preference to travel by automobile isn’t much different from, say, residents of London, Ontario, the Canadian city closest to ours in population. There, 88 percent of the total distance travelled each day is done in private automobiles. Victoria’s slightly lower level of auto reliance is made up for by a higher rate of transit usage. One of the most interesting findings in the CRD’s $330,000 survey was a 19-fold increase—since the last survey in 2011—in the number of battery-electric vehicles (BEVs) in our region. You read that correctly: a 19-fold increase. My neighbour, who now drives a Nissan Leaf, is part of a growing trend. My neighbour’s Nissan Leaf. The 2018 version can travel about 150 miles on a full charge. Perhaps the most intriguing aspect of this trend is that it wasn’t highlighted in any way by the CRD. Why is that intriguing? Think about it. The dominant preference for travel in our region is by private automobile, and that preference is essentially stable. At the same time, there is an accelerating uptake of all-electric private automobiles. Our strong preference to get from A to B in our cars, along with the growth in electric vehicle ownership, suggests an obvious pathway the CRD could encourage to reduce transportation-related carbon emissions. Yet the CRD chose not to highlight this emerging pathway in its own analysis of the survey’s results. Why? I’ll address that question later. First let me outline why lack of leadership from the CRD on the electrification of regional transportation could slow down the uptake of electric vehicles—the most obvious pathway for reducing transportation emissions. It’s well known that the shift from gasoline-fuelled automobiles to BEVs is being driven by public policies that assist that shift. Financial incentives to purchase BEVs have made a significant difference in those jurisdictions where the shift to BEVs is advancing—like Norway. In BC, the Province has financial incentives in place—currently up to $5,000 against the purchase price of a BEV. But other incentives make a difference, too, like allowing BEVs to travel in HOV lanes, and free parking. Public charging infrastructure will be required, and unless local governments move quickly and visibly to provide that, they are telegraphing hesitation and doubt to potential early adopters. That infrastructure needs to be out on the streets, funded with public money. If the 19-fold increase in BEVs measured by the CRD’s latest survey holds up for the next five years, there could be as many as 30,000 BEVs in use by then. That would knock a significant hole in the region’s daily emissions. If the CRD’s analysis of its survey didn’t promote the surge in electric vehicle use, what did it emphasize? That’s best illustrated in a June 26 article in the Times-Colonist headlined “More people in capital travelling by bus, bike and on foot.” The paper quotes Oak Bay Mayor Nils Jensen, chair of the CRD’s Transportation Committee: “As we see the trend—slowly we’re moving toward more people walking, biking and busing…So I think it’s very positive.” The most attention-worthy word in Jensen’s statement is “slowly.” After all, the CRD’s survey shows that the distance travelled in autos each day by CRD residents has actually increased since the last survey. In other words, the single largest source of emissions produced locally—travel in fossil-fuel-powered automobiles—is getting larger. Why, when there is an urgent need to reduce emissions, would any politician describe the trend of increasing emissions as “very positive”? Perhaps the reason is that the CRD’s official transportation plan focuses on increasing the level of “active” modes of transportation, like cycling and walking—and running to catch the bus. As long as some increase in those modes can be detected—no matter how small their contribution is to overall travel—politicians like Jensen can claim that the CRD is meeting its objective. Don’t worry, we’re on it. Let me push your be-very-worried button for a moment. Last year, climate scientists at Scripps Institute of Oceanography published a study in the peer-reviewed journal Proceedings of the National Academy of Sciences. The study found a five-percent probability that there will be a three-degree-celsius increase in average global temperature (above the pre-industrial era average) by 2050. That level of increase would lead to what the scientists termed a “high impact” on human and natural systems. They reckoned such a scenario could lead to “catastrophic” effects on human civilization. Five percent? Why worry? On that question, one of the study’s authors, Dr Veerabhadran Ramanathanb, noted, “When we say a five-percent-probability high-impact event, people may dismiss it as small but it is equivalent to a 1-in-20 chance the plane you are about to board will crash. We would never get on that plane with a 1-in-20 chance of it coming down but we are willing to send our children and grandchildren on that plane.” Humanity has, sort of, collectively decided not to get on that plane. Canada gave up its boarding pass when it signed on to the Paris climate agreement. As a consequence, the federal government has set a target of 150 megatonnes for total emissions by 2050—just over 30 years from now. But Canada’s transportation sector alone accounts for 173 megatonnes right now. Here’s how hard that’s going to be to meet: Victoria drivers of fossil-fuelled cars would have needed to reduce the total distance they travelled each day by 15 percent since the last CRD travel survey in 2011. Instead, the distance we travelled went up. Moreover, the CRD’s travel survey misses a significant portion of vehicle use by residents of the region. Commercial truck use and trips made for commercial or business purposes are not included in its survey. Given these undercounts, it’s prudent to assume that the effort and money spent on moving more people toward walking, biking and busing has not significantly impacted regional transportation emissions. By the way, if you are hoping that increases in fuel efficiency for internal combustion engines will allow us to keep buying gasoline-powered vehicles, consider this: A 2015 study by the University of Michigan showed that between 1991 and 2015, US vehicle fleet fuel economy increased by only 6 percent. Let’s go back to the question of why the CRD didn’t bother to emphasize that 19-fold increase in the use of BEVs. I asked John Hicks, the CRD’s transportation planner in charge of the survey, why that shift hadn’t been highlighted. Hicks said, “the number of BEVs themselves are still very small and the 19-fold increase came off of a very low base of only 100 in 2011. It is therefore not possible to identify any concrete trends based on the limited data that we have available.” Hicks is right about the danger of using small numbers to predict trends, but shouldn’t that apply to changes in the use of, say, bicycles, too. In 2011, CRD residents used bicycles for 1.7 percent of the total distance they travelled each day. By 2017 that had risen to 3.0 percent. That 1.3 percent increase over 6 years is a small number too, about 0.2 percent per year. But that hasn’t prevented Jensen and others from assessing it as a “very positive” trend. Besides, the CRD needs to pay attention to what’s happening in other progressive places—like Norway—as the number of manufacturers and the driving range of electric vehicles grows, and lower-cost models become available. In the last month for which data is available (April 2018), BEVs accounted for 37 percent of new car sales in Norway. Why is that happening? Norway’s federal transportation plan calls for all new passenger cars and vans sold in 2025 to be zero-emission vehicles. That’s where we are headed. The CRD can either facilitate that, or dither over whether it’s really happening. As it turns out, the CRD is waking up. Last June it quietly (no local media covered it) sought public input for its “brand new Electric Vehicle and Electric Bicycle Infrastructure Planning Project.” The survey sought “to better understand regional interest in—and opportunities for—new charging infrastructure across the region.” The CRD’s Nikki Elliott told Focus that nearly 700 responses were received. Results were not available at the time of our inquiry. David Broadland is the publisher of Focus Magazine.
  4. FOIed emails show engineers forgot about a serious flaw until it was too late to fix properly. They then forgot that they had forgotten. AFTER STALLING FOR SIX MONTHS, the City of Victoria finally released documents that provide details about why the lifting section of the new Johnson Street Bridge had to be repaired before it was even installed. “Repair,” by the way, is the word used by the engineers involved. The records the City released include emails from Ryan Andrews, PCL’s project manager, to private consultant Jonathan Huggett, whom the City has been paying about $300,000 a year to watch over Victoria taxpayers’ interests on the project. PCL is the Edmonton-headquartered company the City contracted to build the bridge in 2013. The released records show Huggett and another engineer misled public officials about the circumstances that led to the need for the premature repair. In turn, those public officials, including Mayor Lisa Helps, misled the public. As you may recall, our January story pointed out that a one-metre by one-metre steel plate had been bolted to the underside of each of the new bridge’s 50-foot-diameter rings at the 12 o’clock position. The plates were not part of the intended design, and appeared to be a last-minute addition needed to reinforce some structural weakness in the rings. For our story we contacted Huggett and Helps for information about what had necessitated the plates. A close-up of the large bolted-on plates that made a last-minute and unexpected appearance on Victoria's brand-new $115-million bridge Huggett’s response to our questions at the time produced little factual information. Mayor Helps ignored our questions about whether she and council had been advised about the plates. Following publication of our story, Helps issued a statement through facebook claiming our story “contained serious factual errors and inaccuracies.” Her statement had apparently been created by then acting City Manager Jocelyn Jenkins, with Huggett’s guidance. Keith Griesing, a professional engineer with Hardesty & Hanover, the company that engineered the lifting section of the bridge, wrote a letter to the City explaining how the problem that led to the repair had been discovered. Griesing denied any responsibility for the circumstances that led to the bolted-on plates and blamed the Chinese fabricator. In the six months since then, Helps and the City were unable to provide any examples of “factual errors and inaccuracies” in our original story. However, after receiving the records we requested by FOI, I am able to report that I did make a factual error. The size of the plates are actually six feet by six feet—almost four times as large as we reported. But the emails from Andrews to Huggett paint a rather different story than that told by Huggett, Helps, and Griesing. In two emails from Andrews to Huggett, Andrews appealed to Huggett to intervene with Hardesty & Hanover to hasten resolution of a serious issue that became known as “NCR 155.” That referred to four weld access holes in each ring described in a non-conformance report (NCR) by Atema dated January 4, 2017. Atema was the company performing PCL’s quality-control inspections at the steel fabrication facility in China where the lifting section was built. Atema’s report recorded that the weld access holes were “discovered” on December 9, 2016. The concern about the small holes, not openly stated in the Atema report, was that their location and physical nature would concentrate stress and could eventually lead to the formation of fatigue cracks in the rings. Such fatigue cracks could threaten the structural integrity of the bridge’s superstructure. In Andrews’ first email to Huggett about the issue, which he sent on March 31, 2017 following a meeting with Hardesty & Hanover that day, Andrews wrote: “For a brief recap, you heard H&H mention this has been a known issue for over a year, yet it wasn’t communicated to PCL…until December 2016 at which point the NCR was initiated…” I’ve added the italicization in that sentence to draw your attention to that point. It shows that Andrews told Huggett on March 31, 2017 that the problem created by the weld access holes “had been a known issue for over a year.” That would mean the problem had been evident to Hardesty & Hanover before March 31, 2016. Both Huggett and Griesing have, in their separate public explanations of the issue, implied that the problem was discovered in December 2016 as a result of Atema’s diligent inspection and quality control. So what’s the big deal about the eight months missing from both Huggett’s and Griesing’s explanations? During those eight months, as well as during the four months between rediscovery of the problem and the date when Andrews emailed Huggett, work had continued on the rings. That year of work made it more difficult to fix the problem properly. Neither Huggett nor Griesing responded to requests for an explanation of why those eight months had disappeared from their public accounts of what happened. Perhaps there is no reasonable explanation for why the problem wasn’t addressed as soon as it was first identified. It appears the problem was simply forgotten. Did someone then wake up in the middle of a dark, December night and have an OMG! moment? How embarrassing this must be for those involved: Engineers forgetting that they forgot. But that’s not the full extent to which Huggett and Griesing misled the City and the public. Andrews’ March 31, 2017 email to Huggett continued: “Now it also needs to be noted that this so called ‘non-conforming’ condition has no design details and no comments were made on the shop drawings. We have not raised issue of this prior due to our interest in just getting it repaired and over with, but being that it continues to linger it now has significant impacts for both PCL and the City—those being time and money.” Andrews, then, disputed that the weld access holes were the fault of the fabricator. He believed they were a result of design details and comments not being provided by the designer—Hardesty & Hanover. Indeed, Atema’s NCR 155 included notations that pointed out the lack of clarity in the shop drawings for the weld access holes. Ultimate responsibility for approving those shop drawings lay with Griesing. Martin Bache, a 40-year veteran of the heavy steel fabrication industry in Canada, described the proper procedure required for approval of weld access holes: “Weld access holes in fracture-critical members must be designed by the Engineer of Record [Griesing]. Competent detail draftspersons would be expected, during preparation of the shop drawings, to identify closed chambers where the Engineer of Record may have forgotten to show on his plans weld access holes without which the required welding cannot be performed. They would then issue an RFI [request for information] pointing this out, and asking the Engineer of Record how they should proceed.” According to Bache, then, Griesing would ultimately be responsible for the completeness of the design of every weld access hole that was required, since every steel member in the rings was designated “fracture-critical.” But in January 2018, after publication of our original story, Griesing wrote the City of Victoria and stated, “There was no ‘design flaw’ by Hardesty & Hanover nor any other of the City consultants involved; it was assembly by the fabricator that did not conform to the design plan requirements nor to the applicable detailing and fabrication standards required in the specifications.” Andrews’ email to Huggett calls that claim into question. And now we come to the nub of why Victoria got a new bridge delivered with such highly visible repairs. Andrews’ email to Huggett continued: “I’m requesting your support by having a separate conversation with H&H/MMM on this subject and resolution thereof. Having to remove the cover plate to do these repairs should be the item challenged as this is what will create the delay expressed above. Or in other words, give me a repair that requires no removal of the cover plate.” There were three general approaches that could have been taken to address the potential for fatigue caused by the weld access holes. PCL’s fabricator proposed a welded repair limited to the weld access holes themselves, but Hardesty & Hanover rejected that approach. A second approach was to add steel to the outside of the rings, and that’s the approach that was eventually taken. The third approach would have been to remove steel plate (“the cover plate”) from the rings in the area around the access holes and rework the problematic area. That would have preserved the intended design. But Andrews lobbied Huggett to steer away from this latter course. Removing the cover plate at that point in the project could have created a problem that he outlined to Huggett in a second email a few days later: “[T]he whole reason the [cover] plate was installed was [because] it was continually discussed between all parties that all welding needed to be complete before proceeding [in March 2017] to vertical assembly due to concerns of the heat induced from welding causing the shape of the rings to change.” The implication was that if the choice was made to remove plating and rebuild the problematic area of the ring, the fit between the rings and the other major bridge components—obtained during the month-long trial vertical assembly that had just been completed—couldn’t be counted on. Andrews estimated that taking the route of removal of plate and doing an internal fix would add a “week or more” to the schedule. Compare that “week or more” to the year that had passed between the time the issue first became known and the point when trial vertical assembly had been completed. In that context, Griesing’s explanation to City council that “impact to schedule” was an important factor in the choice of bolted-on plates is…laughable. Remarkably, the record of Huggett’s communications released by the City doesn’t include any communications coming from Huggett. He appears not to have responded to Andrews’ emails and not to have consulted with Hardesty & Hanover or with MMM as per Andrews’ request. Neither did he inform anyone at the City of Victoria about the issue or the options. It’s clear that an alternative to bolted-on plates was possible and would, by Andrews’ estimate, have added only a “week or more” to the schedule. But, according to the records released by the City, Huggett didn’t advocate for any outcome with anyone. Let’s reflect on the absence of any emails from Huggett in response to Andrews’ appeals. Huggett is not a City of Victoria employee. He operates what appears to be a one-man project management business in the Vancouver area. His contract with the City does not require him to use the City’s email server, which automatically backs up all City employee emails and preserves them for the purpose, among others, of being available for access-to-information requests. When Focus requested Huggett’s communications on this issue, the City had to ask him to go through his emails and find relevant records. In other words, the City created a situation where a private contractor could, in effect, decide what emails to provide. The problem here should be obvious. The City has no way of knowing what interactions took place between Huggett and employees of Hardesty and Hanover, PCL or MMM. The product these companies delivered was seriously defective, yet Huggett, when questioned about the bolted-on plates, provided little more than an airbrushed account of why that had occurred. From what I understand, the City has now entered into a similar arrangement with Huggett regarding the City’s plan to build a new $75 million swimming pool. What is that definition of “insanity” again? Does the City have any legal recourse to seek compensation as a result of the delivery of a defective bridge? As I noted in my story in our May/June edition, in April 2016 the City agreed to “release and forever discharge” PCL, MMM and H&H “from all debts, claims, demands, damages, expenses and costs (including without limitation, legal costs) of any nature or kind that are in any way related to the Project and either known or which ought to be known by the [City] as of [April 23, 2016].” According to PCL’s Ryan Andrews, the defects that necessitated the bolted-on plates were a “known issue” at the time the City signed away its right to seek legal recourse for anything related to known issues. David Broadland is the publisher of Focus. The City of Victoria's full release of documents related to the bolted-on plates can be downloaded here. (9.1 MB file size) Related stories: Bridge design flaw hidden for a year, then given quick-and-dirty repair Victoria City Hall continues cover-up of bridge design flaw Why are Victoria City councillors accepting a world-class bodge? The Wounded White Elephant
  5. Is Fisheries & Oceans Canada ignoring Washington State research on chemical contamination from sewage treatment plants? ARE THREE LARGE SEWAGE TREATMENT PLANTS located on the Fraser River estuary contributing to the decline of the Southern Resident Killer Whale population? Between them they discharge 1.1 billion litres of effluent every day of the year into the estuary and nearshore marine waters. The largest, Iona Island, provides only primary treatment and has been permitted by Fisheries and Oceans Canada to continue at that level until 2030. We now know that the reproductive health of the orca population depends heavily on the availability of Fraser River chinook salmon, but, according to fisheries scientists, chinook runs on the Fraser are now only 25 percent of historic numbers. Recent research in Washington has found a strong link between the survival rate of juvenile chinook salmon and chemical contamination of their natal estuary. Is the survival rate of Fraser River juvenile chinook being similarly impacted by contamination from the Annacis Island, Lulu Island and Iona Island wastewater treatment plants? Currently, these three plants provide treatment for over 1.8 million people, and that population is not declining. Vancouver’s three largest sewage treatment plants all discharge into critical chinook salmon habitat. The physical processes involved in this chinook-sewage-orca death spiral have become better understood in recent years thanks to research by Dr James Meador, an environmental toxicologist with the Northwest Fisheries Science Center in Seattle, and Dr Samuel Wasser, a research professor of conservation biology at the University of Washington. Since 2013, Meador and his team of researchers have published three studies that considered the impact of chemical contamination on juvenile chinook salmon during the period they reside in their natal estuary. Meador’s first study found that the survival rate of juvenile chinook that smolted in contaminated estuaries of rivers flowing into Puget Sound was cut in half compared with juveniles coming from a relatively uncontaminated natal estuary. Let me repeat that: Survival rate is cut in half. In his second study, Meador analyzed the discharge from secondary sewage treatment plants, located upstream from chinook estuaries, for the occurrence of 150 “chemicals of emerging concern,” or CECs. These are chemicals associated with pharmaceutical and personal care products, as well as industrial compounds. Many are known endocrine disruptors, which can affect hormonal balance and result in developmental and reproductive abnormalities. The researchers also analyzed the tissue of juvenile chinook and resident sculpin in the estuary for the presence of the selected CECs. That study became widely publicized in 2016 because cocaine and antidepressants—and many other chemicals—were found in both the treatment plants’ discharge and in fish tissue. Indeed, Meador’s team found unexpectedly high levels of certain CECs in the treated effluent. The study’s findings suggested that chinook juveniles have a significant vulnerability to bioaccumulation of CECs. Many contaminants that were found in juvenile chinook tissue were at concentrations below detection limits in the estuary waters. The scientists also observed higher levels of contaminants in juvenile chinook than in resident sculpin, even though the latter were permanent residents of the estuary. Meador’s team observed that the contaminants found in chinook tissue, although present in sub-lethal concentrations on a chemical-by-chemical basis, were, in some cases, present at levels that would be expected to cause detrimental physiological effects. The scientists noted the potential for a drug-cocktail effect: “The fact that we observed multiple pharmaceuticals capable of interacting with a variety of molecular targets in our two fish species, leads to the potential for mixture interactions on critical physiological processes. These interactions can be additive, synergistic, or inhibitory.” Meador noted that these effects could be responsible for the two-fold reduction in survival rate found in his earlier study. In a third study (click link below to download), released this past April, Meador’s team found that the contaminants were also causing metabolic dysfunction, which “may result in early mortality or an impaired ability to compete for limited resources.” Again, Meador noted that metabolic dysfunction induced by CEC contamination could contribute to the two-fold reduction in the survival rate of these juvenile chinook, compared with chinook migrating from the uncontaminated estuaries, that he had found in his first study. Adverse metabolic effects in fish exposed to contaminants of emerging concern in the field and laboratory.pdf The US EPA has listed Puget Sound chinook as a “threatened” species, and the decline of those runs has been even more profound than the Fraser decline. Historically, according to Jim Myers of the Northwest Fisheries Science Centre in Seattle, Puget Sound’s chinook runs were about 25 percent greater than the Fraser River’s. But by 2010, Puget Sound chinook returns had collapsed to only six percent of the size of the greatly-reduced Fraser River returns. Although the link between the abundance of chinook salmon in the Salish Sea and the physical health of the Southern Resident Killer Whale population has been known for some time, Wasser’s seven-year-long study, published in 2017, provided the first confirmation that low availability of chinook is suppressing the population’s birth rate and endangering the health of reproductive female orca. Wasser’s team collected orca poop and analyzed it for hormone measures of pregnancy occurrence and health. The scientists also looked for chemical indicators of nutritional and disturbance stress in the poop. By making the same measurements over time, they were able to distinguish between nutritional stress caused by low availability of chinook salmon, and disturbance stress caused by the presence of nearby boats. Wasser’s team correlated periods of nutritional stress with the timing and strength of the two main chinook runs that are keeping the southern orca alive: the Columbia River early spring run and the Fraser River summer and fall runs. They found that—depending on the timing of those runs, and how many fish were in them—the southern resident orca experienced more or less intense famines through the winter months and between the spring and summer runs. The scientists observed: “Low availability of chinook salmon appears to be an important stressor among these fish-eating whales as well as a significant cause of late pregnancy failure, including unobserved perinatal loss.” The scientists surmised that “release of lipophilic toxicants during fat metabolism in the nutritionally deprived animals may also provide a contributor to these cumulative effects.” Not only are the orca being periodically starved, but when a starved, pregnant orca begins burning off her fat reserves in response to the lack of food, toxins bioaccumulated in her fat reserves—such as PCBs and PBDEs—begin to have more of an impact on her health, such as a reduced ability to fight infections. This could contribute to the demise of the fetus and increase the risk to the mother’s life. As a consequence of these conditions, the study noted, “the 31 potentially reproductive females in the Southern Resident Killer Whale population should have had 48 births between 2008–2015. Yet, only 28 births were recorded during that period. The 7 adult females in K pod have not had a birth since 2011, and just two births since 2007. The 24 females in the remaining two pods (J and L) have averaged less than 1 birth per pod since 2011, with no births in 2013, but had 7 births in 2015. One of the two offspring born in 2014 died.” As of this writing, with the presumed death of “Crewser,” the population has dwindled to 75 whales. As recently as 1996 there were 98 orca in the 3 pods. Wasser noted, “Results of the Southern Resident Killer Whale study strongly suggest that recovering Fraser River and Columbia River chinook runs should be among the highest priorities for managers aiming to recover this endangered population of killer whales.” Let’s make the obvious connection between Meador’s and Wasser’s findings. Meador’s research strongly suggests that the chemical contamination in Puget Sound rivers that’s quickly bioaccumulating in juvenile chinook is coming from sewage treatment plants discharging into their natal estuary. Removing that contamination could double the number of chinook returning to those rivers as adults. Wasser’s study shows the Southern Resident Killer Whale population’s decline is strongly correlated with the availability of chinook and he recommends, for one thing, that managers of the Fraser River fishery make chinook recovery amongst their highest priorities. A rational conclusion, based on the two groups of scientists’ extensive research, would be that Fraser River fisheries managers should be determining whether the impacts Meador measured in Washington estuaries are at play in the Fraser estuary. But that’s not happening. DFO recently published “A science based review of recovery actions for three at-risk whale populations” that listed 98 specific actions. DFO acknowledges that only 2 of the 98 measures are “specifically directed toward recovery of chinook salmon stocks in Canada.” None of those 98 actions include examination, let alone reduction, of the impacts of chemical contaminants on chinook juveniles in the Fraser River estuary. DFO paper on SRKW recovery efforts.pdf DFO has been caught flat-footed on chemical contamination of the Fraser River estuary in the past. The Cohen Commission of Inquiry into the decline of sockeye salmon in the Fraser River produced a technical report in 2011 that stated: “There is a strong possibility that exposure to contaminants of concern, endocrine disrupting chemicals, and/or contaminants of emerging concern has contributed to the decline of sockeye salmon abundance in the Fraser River.” Despite that, the technical report noted, “Due to limitations on the availability of exposure data and/or toxicity thresholds” it could provide only a “qualitative evaluation.” Cohen Comm report on chemical contaminants re sockeye decline.pdf That was in 2011 and the information gap was related to sockeye. With chinook runs on the verge of collapse, you would think that Meador’s published research on chinook estuary contamination, only 200 kilometres away, would have prompted DFO to narrow the gap in their knowledge. We contacted DFO, but as of our press deadline a spokesperson had been unable to confirm whether or not any DFO-affiliated scientist was investigating the impact of the Fraser River estuary wastewater treatment plants, or other sources of chemical contaminants, on the survival rate of juvenile chinook. The presumption may be that because sewage effluent is being discharged into the Fraser River estuary through outfalls that achieve legally required dilution ratios, no further consideration is required. But the rivers Meador considered in Puget Sound are meeting similar if not higher requirements, and he found chinook survival rate is being cut in half. Meador has said it’s unlikely these contaminants can be effectively filtered out of the huge volume of wastewater that’s being flushed into Puget Sound. In the case of the Fraser River it seems possible that the three plants could be connected to a super outfall that diverts the discharge away from the estuary and into deeper marine waters. But without any examination of chemical loading of Fraser chinook juveniles being conducted by DFO, there will be no public pressure mounted for such a measure. Mr Floatie ought to find a new costume (Cocaine Man?) and relocate to Vancouver. Victoria’s deepwater marine outfalls, by the way, are located about 70 kilometres away from the nearest chinook estuary. While DFO wasn’t certain about what research is being done, it’s more certain about the magnitude of the chinook decline. In its 2018 outlook for the six different populations of chinook in the Fraser Basin, fisheries managers found that only one was at a level considered necessary to maintain a healthy population. David Broadland is the publisher of Focus. Related stories: The orca famine and Puget Sound's poisoned rivers Washington's phony sewage war with Victoria
  6. David Broadland

    The Wounded White Elephant

    We should call the new bridge what it is. VICTORIA'S NEW BRIDGE—variously dubbed “The Blew Budget Bridge,” “Fortin’s Folly,” and “The White Elephant”—opened on March 31, 2018. Notably absent from the group of officials presiding over the occasion were any of the former high-level City officials who lost their jobs as a consequence of the project’s long history of miscalculation and misrepresentation. The bridge’s original designer, Sebastien Ricard of Wilkinson Eyre Architects, who ended his connection with the project back in 2012, was nowhere to be seen. Nor was Joost Meyboom, the private engineer who first advised the City to fix the old bridge rather than build a new one, back in 2008, but then went on to become the City’s prime consultant on the new bridge project. Meyboom’s employer—formerly MMM Group, which changed its name to WSP—launched a legal suit against the City over the project earlier this year. Opening of Victoria's new bridge No wonder all these folks skipped Mayor Lisa Helps’ $42,000 opening-day celebration. Back in April 2009, when City council voted to replace the existing “historically-significant” double-bascule bridge, those officials assured Victorians a new bridge would cost $41 million and take 18 months to build. Nine years later, costs have almost tripled while major elements of the project remain unestimated, unfunded and unfinished. Millions in costs directly attributable to the project have been hidden. The project’s record of underestimation and miscalculation, though, may be just a prelude to greater embarrassment to come. Certain aspects of the new bridge’s design and construction are so at odds with engineers’ warnings that, although the new bridge can, at this moment, open for marine traffic just like the old bridge, it’s highly doubtful whether it will come anywhere near to matching the old bridge’s 95-year record of reliable operation and minimal ongoing expense. While “The Blew Budget Bridge” does capture the huge miscalculation in cost, and “Fortin’s Folly” makes it clear that good people made bad decisions, only “The White Elephant”—which signifies over-the-top original cost and unjustifiable ongoing expense—provides a fulsome enough characterization of the so-far nameless new bridge. But even that needs updating. Here’s why I am leaning toward “The Wounded White Elephant.” The new bridge 's opening span provides the same navigational channel width as the old bridge did, but at huge costs and with hundreds of "nonconformities" welded into its superstructure. The most eye-catching feature of the new bridge is its 50-foot-diameter steel rings. The counterweight lobes attached to the rings do happen to resemble an elephant’s ears, but that’s not the reason why this bridge should be named “The White Elephant.” You won’t find anything like those rings—or the gigantic machinery below them needed to support and rotate them—on any bridge anywhere else on Earth. Unfortunately, designing a bridge that required 1700 tonnes of structural steel in its moveable superstructure and hundreds of tonnes of machinery to support that—just to span a 41-metre-wide opening—is not the direction planet Earth needs to go. (Worse, largely unacknowledged by the project is the fact that the first attempt to fabricate the superstructure was scrapped and as much as 1700 tonnes of steelwork wasted; more on this later.) Good, common-sense arguments were made against building Ricard’s design by engineers, and it’s because all the decision makers over the years ignored those arguments that the bridge cost three times what it should have and took 108 months to complete instead of 18. Because of those officials’ dismal performance, the bridge promises to be an ongoing source of expense to City of Victoria ratepayers and hence deserving of “The White Elephant” moniker. Below, I’ll highlight just a fraction of what happened. To start with, though, let’s recall why Ricard put those rings into his design. This is key to understanding why the project flubbed. Just before the November 2010 referendum in which the City won elector approval to build a new bridge, Ricard explained those rings at a public presentation. He told a handful of people that the underlying design challenge, for him, was to create a moveable bridge that easily communicated to an observer how it worked. That’s it. That’s the entire argument for the rings. The rings were all about appearance and not about any substantive proven need particular to a crossing of Victoria’s harbour. Ricard wasn’t trying to reduce seismic vulnerability or to create a bridge less likely to be hit by a barge full of gravel and survive the blow if one did. Nor was he trying to reduce life-cycle costs or use scarce public resources more efficiently. No, it was enough for Ricard that a tourist from Iowa, standing on the Laurel Point walkway, would be able to gaze across the Inner Harbour and understand how the bridge lifted, at a glance. Perhaps there’s a similar, whimsical rationale at the foundation of every white elephant construction project, and maybe Victorians are no better or worse than any other community at sniffing out ideas that don’t have much merit. I don’t know. But what I have learned, after following Victoria’s project closely for 9 long years, is that there were real-life consequences that flowed from the project’s ill-considered underpinnings. Ricard’s whimsical central motivation radiated outward through the project with force, inflating engineering and construction costs, laying waste to political and public service careers, substantially increasing carbon emissions, straining the City’s coffers, and dividing the community. Ricard’s imagination even put a well-established Chinese company out of business. How did all this happen? Let me sketch in some details. Let’s start with a fundamental physical fact about the bridge that resulted directly from Ricard’s rings. Because of a quirk in Ricard’s design, every time the bridge opens, it has to lift and hold the span a full 2.5 storeys higher than it needed to. Indeed, people have observed that, when it is lifted, the new bridge is “so much higher” than the old bridge. Yet the new bridge provides the same navigational channel width as the old bridge. That extra height might be dramatic to observe—like a hopping car—but it’s otherwise pointless and that little moment of drama has come with a lot of negative long-term consequences. The mechanical design of the new bridge results in it lifting the weight of the superstructure 2.5 storeys higher than necessary, an engineering feat almost as pointless as a hopping car. For example, every part of the bridge that moves had to be stronger than it would have needed to be in a bridge that used a fixed central axle with conventional bearings located as close to the edge of the channel as possible. That extra strength was obtained by using far more steel for the bridge—in the rings and trusses—than would have been necessary in a more conventional approach. Extra steel in the trusses meant more lead and steel were needed in the counterweights to balance that extra weight. All that extra weight in the superstructure meant the machinery that supports and rotates it needed to be immense compared to the shaft, bearings and machinery needed to rotate a more conventional moveable bridge. The higher lift of the span also meant that it would experience greater pressure during strong winds, and so that force, too, had to be offset with more steel and heavier support equipment, all costing more than a conventional approach. All of these additional weights and costs affected the approach bridges, too. It meant that for a given budget, less money could be spent on the approach bridges. Originally, to satisfy high seismic performance requirements, they were going to be built of steel. Instead, because of the inflating cost of the lifting span, there was only enough budget to use less costly reinforced concrete. But in order to include the use of concrete approach bridges, City officials had to secretly agree, during the procurement process, to place a rider in the construction contract that specified much lower levels of seismic performance than had been recommended to the City. The rider clearly states that its stipulations of (lower) performance take precedence over the seismic performance requirements of any of North America’s highway bridge building codes. This loss of one of the fundamental objectives of the project—a legally enforceable contractual assurance of a high level of seismic performance by the bridge if Victoria is struck by a large earthquake, can be traced directly to Ricard’s choice of rings in the lifting mechanism and the extent to which they inflated the cost of the project. City officials, the ones who later lost their jobs, were well-warned by engineers about the risk of Ricard’s open-ring design inflating costs. For example, during bidding for the contract to build the bridge, participating companies were required to provide a critical review of the design MMM had developed with Ricard, and they were invited to “optimize” that design so that it could be built within the City’s $66-million “affordability ceiling.” The winning bid by PCL was the only proposal that utilized Ricard’s open-ring concept. The only other serious bid proposal received by the City, from Kiewit Infrastructure, rejected the axleless design and predicted what would happen if the City went ahead with Ricard’s design. Specifically, Kiewit told City managers it had contacted “a number of steel and machinery fabricators, who are experienced in movable bridge design and/or construction. All expressed the opinion that there were likely more cost effective mechanical concepts for a bascule bridge” than the open-ring design used by Ricard and MMM. Kiewit advised the City that “unknowns and/or unexpected costs” of Ricard’s “unconventional design” would “conflict with the City’s mandate to remain near or below the indicated Affordability Ceiling…Kiewit is of the view that the [design] may represent a fundamentally high risk and expensive design approach.” The company’s engineers noted that the counterweight in Ricard’s design was attached to the truss rings in a way that “would load the truss ring eccentrically, which could distort the ring—a highly undesirable condition.” The bridge proposed by PCL had the same eccentric loading of the rings that concerned Kiewit, but was going to have an added complication: In order for its bid to be within the City’s affordability ceiling, fabrication of the moveable part of the bridge would have to take place in China. In hindsight, it’s easy to see that the City listened to the wrong engineers, chose the wrong company to build a bridge, and built the wrong bridge. City officials were warned they were in danger of buying a White Elephant. Instead of heeding the warnings they insisted on having one as quickly as possible—and this meant hiding the critical reviews (which cost the City $150,000) from the public—and so Ricard’s whim rolled forward into the next phase. AS YOU MAY RECALL, the City of Victoria awarded a construction contract to PCL in late 2012 to build the bridge under a $63.2-million “fixed-price” contract. PCL made it clear it planned to have the moveable part of the bridge fabricated in China. This, apparently, raised no red flags at City Hall. At the time PCL won the construction contract, MMM Group were contracted to provide engineering, and it in turn subcontracted Hardesty & Hanover to provide engineering and design for the lifting span and the machinery used to raise that span. When PCL began construction in late 2013, the City of Victoria assured its ratepayers that the cost of the bridge could not rise since PCL had agreed to a “fixed-price” contract. But, by early 2014, PCL started to pepper the City with demands for more money. Those demands began soon after fabrication of the rings and trusses had started in China in March 2014. By September of that year, work in China had been halted. In January 2015, the City’s Project Director Jonathan Huggett reported that fabrication problems were so bad that “one of the rings is being replaced while the other is being repaired. The north truss steel will be replaced.” The first attempt to build Ricard's bridge at ZTSS's plant. Shown above are fabrication of the bridge's rings, trusses and deck components in July 2014. All of the steelwork done up to January 2015 was scrapped. Notably absent from Huggett’s reports from this era is any acknowledgment that the thing the Chinese welders were screwing up was actually very difficult to build. Neither did Huggett tell councillors that the City had been warned by Kiewit engineers that this was likely to happen if the City attempted to build Ricard’s design. Instead, Huggett persuaded the City that simply increasing quality control would produce rings and trusses with adequate strength and structural integrity. In spite of such hopes, fabrication problems in China continued to accumulate in the bridge components. Recently, Huggett admitted: “We rejected an entire bridge at one point.” The City’s project director seemed to see the scrapping of “an entire bridge” as a good thing, a sign that people were doing their jobs properly, that quality assurance procedures were working, and that Victorians could be confident that the project wouldn’t accept crap for a bridge. But think about that: An entire bridge wasted. If we take Huggett at his word, about 1700 metric tonnes of steel were scrapped. That’s the weight of structural steel for the superstructure specified in the City’s contract with PCL. (The City did not respond to repeated requests for confirmation of the amount of steel that was scrapped.) But we should add to that heavy burden all the human effort and other costs—including associated environmental damage—that went with throwing away the warm-up bridge. Who was going to pay for that waste? As it turned out, it wasn’t going to be PCL. The City had acknowledged PCL’s first demand for more money—$7.9 million in early 2014—but then demurred from providing information about subsequent demands. In early 2015, about the time “an entire bridge” was rejected, the City admitted it had entered a “legal mediation process” with the companies building the bridge. The second attempt to build Ricard's bridge, in March 2016 at ZTSS's plant near Shanghai. It wasn’t until April 2016, at the conclusion of the mediation, that the City acknowledged that PCL, MMM and H&H had demanded $27 million in additional costs. The details of that $27 million claim were never made public, but it is believed PCL’s share was about $25 million. After out-competing two other companies for the contract and assuring the City Ricard’s bridge could be built for $63 million, what circumstance could possibly have justified PCL’s demand for over 40 percent more money? The timing of the start of PCL’s demands, you may have noticed, coincided with the beginning of fabrication in China. As major components of the bridge were rejected, PCL’s claims against the City increased. The company may have realized that the lifting span being (badly) fabricated in China could carry a huge risk of future legal claims by the City. By demanding more money and halting work in China, PCL may have simply been creating the conditions for dumping all of that risk back on the City. And that’s exactly what happened. The City settled the $27 million in claims by agreeing to pay an additional $2.4 million and making changes to the terms of the contract. In a news report at the time, Mayor Helps claimed: “I think it’s better news than anyone could have hoped for.” But an FOI filed by johnsonstreetbridge.org revealed the City agreed to “release and forever discharge” PCL, MMM and H&H “from all debts, claims, demands, damages, expenses and costs (including without limitation, legal costs) of any nature or kind that are in any way related to the Project and either known or which ought to be known by the [City] as of [April 23, 2016].” This was hardly “better news than anyone could have hoped for.” Whatever problems have been built into the bridge by PCL, MMM and H&H are now City taxpayers’ problems. One of those problems was brought to the public’s attention in the last two editions of Focus (stories posted here, here, and here). And this is where the “wounded” part of “ Wounded White Elephant” comes into our story. MANY of the risks PCL adroitly shifted back onto the City arose directly from the bridge’s open-ring design. That such risks would have actual physical consequences became clear shortly after the rings were erected at the bridge site last December and Focus pointed out that the rings had already been repaired with metre-square bolted-on plates, apparently required because of a structural weakness in both rings. The bolted-on plates definitely eliminated any chance of the bridge winning any awards for excellence in engineering or construction. But much worse, they may signify a more pervasive problem with the lifting span. The City has refused to provide a full explanation for the plates, but we have since found a photograph taken during an open house at Point Hope Maritime’s shipyard last October that unintentionally captured details of the repair. The repair was made in Victoria after the rings had been shipped from China. The photograph (see the close-up below), taken before the bolted-on plates were added, reveals not only the make-shift nature of the repair but also at least two holes cut into the “fracture critical” steel with a cutting torch. That damage may have created the need for the plates, at least in part. This photograph shows the repair that was made to the north ring at Point Hope Maritime’s shipyard in October. The lines of small holes were drilled in China and would later allow the bolted-on plates to be attached. The trapezoidal-shaped opening cut into the ring was made at Point Hope. The holes circled with yellow are believed to be “rat holes” cut into the rings by an unknown welder in China. These rat holes may be part of the reason why bolted-on plates were added to both rings. The City’s Project Director Jonathan Huggett has acknowledged that the bridge has hundreds of such “non-conformances.” I emailed the photograph to Martin Bache, a 40-year veteran of the heavy steel fabrication industry in Canada, most recently with Canron as a project supervisor. About the burned-in holes that seem to have created the need for the bolted-on plates, Bache commented, “The cuts are similar to what are termed ‘rat holes’ in steel fabrication. These allow continuous welding of two members to take place through the member with the hole. But, I have never seen two rat holes coming together in two planes as these appear to be. A welder in China may have just taken a torch and cut out two large rat holes to make life easier for himself, but damaged the structural integrity in the process. But that would not seem to require such large bolted-on plates to correct, so I really don’t know what the real story is.” Around the time the photograph was taken at Point Hope Shipyard, someone had removed a trapezoidal-shaped section from the ring and had added some light steel supports for two edges of the bolted-on plates. Of the repair that was done in Victoria, Bache observed, “Not only this bizarre rat hole but also the other pieces of steel in the photo appear to be butchered to an astounding extent. No competent steel fabricator works this way. So, what the hell is going on here?” The steel members of the bridge that were cut into by both the Chinese welder and the workers at Point Hope are considered “fracture critical.” That designation, according to the US Federal Highway Administration, applies to “any steel member in tension, or with a tension element, whose failure would probably cause a portion of or the entire bridge to collapse.” Given the apparent low quality of the repair evident in the photograph, Bache is concerned the repair has not been executed properly. “Any modifications or repairs done to fracture critical bridge components must be performed to detailed procedures approved by the Engineer of Record (EOR) and must be inspected by the EOR or his agent to confirm 100 percent compliance with the procedures. It seems inconceivable that Hardesty & Hanover are accepting all of this butchery,” Bache wrote. Butchery. Wounded. Get it? Bache added, “With all due respect to shipyards, they are not generally expected to work to the same standards of quality and accuracy as bridge fabricators. I would have needed a lot of evidence to persuade me that a shipyard could handle modifications to a fracture critical bridge. Which party approved Point Hope as capable of doing this?” Bache had difficulty understanding who was/is looking out for the City’s interests: “Regarding third-party inspectors, they range from highly competent individuals with substantial levels of practical experience on fabrication shop floors, all the way down to people with absolutely no knowledge of steel and no ability to read drawings but are tasked only with receiving paper reports such as steel mill certificates and weld test reports prepared by others. In 40 years of fabrication I never heard of Atema, so I googled them. They appear to sell inspection equipment and offer to train others in how to run quality control programmes. They make no mention of having vast hands-on, shop-floor experience which would be necessary for confirming that complex fabrications are being made exactly to approved drawings and specifications. So, I don’t know how good a job Atema did in China but I have reason to be very suspicious. I know PCL very well and its hard to believe they would not have hired top level practical inspectors to go to China, but who knows? I wonder at what stage MMM ceased to be of real practical help to the City, including fabrication monitoring. After that its doubtful that [the City’s] interests were being handled by anybody.” Unfortunately, Focus can’t provide the answers to any of Bache’s concerns. The City has dismissed any such concerns about this repair, explaining only that it was the result of a “fabrication challenge.” This is just one of over 150 similar “non-conformities” recorded by the project, according to Huggett. The City’s idea of providing the public with information about the issue has been, in effect: “Why worry us about that one problem? The bridge contains hundreds of them.” The City continues to refuse to release records related to this one repair that were requested by Focus back in mid-December through BC’s access to information legislation. Until the City provides the basic communications about the issue between the City and the engineers who were responsible for resolving the issue, we will keep insisting on seeing those records. Martin Bache’s final comment was this: “What an absolute disaster that this bridge was not made in BC.” This raises an interesting point. PCL based its 2012 bid on a quote from a Chinese fabricator and that allowed it to sneak under the City’s affordability ceiling. That miracle required everyone involved to pretend that a 4 percent contingency would cover any errors in cost estimation and that Chinese labour really was “lower-cost.” If a few people had been smarter, Ricard’s rings would never have been built. Instead, Victoria got a disaster. Ironically, ZTSS did even worse. That company suffered significant financial losses during the time it was building Victoria’s bridge. (It was a publicly-traded company, so its financial performance is a matter of public record.) The cost of having to build the bridge twice, along with bad international publicity about “cracked welds,” no doubt harmed ZTSS’s ability to get new work. By November 2016, trading of the company’s shares had been halted. In August 2017, as Ricard’s wounded rings were finally arriving in Victoria and being readied for repairs at Point Hope, ZTSS announced it planned to sell the operation that had fabricated Victoria’s bridge. By January 2018 the company had undergone a corporate name change and was transformed into Beijing-Kaiwen Education Technology Co., Ltd. With a such a history, it’s unlikely that anyone would want their name on Victoria’s new bridge. What it deserves is a nickname that truthfully reflects its troubled 9-year-long birth. I respectfully propose “The Wounded White Elephant.” David Broadland is the publisher of Focus.
  7. The City is refusing to provide records that would show who knew what, and when they knew it. FOLLOWING OUR STORY LAST EDITION about the surprise appearance of bolt-on plates on the new bridge that Victorians had been promised would be “world-class” and “iconic,” the reaction from ordinary people who don’t receive a regular cheque from the City of Victoria was consistently forthright. An English bridge designer, who has written about such projects all over the world for the past 10 years noted: “The latest reports from Focus cover issues with the bridge’s steel fabrication. They highlight the discovery of a problem with the steelwork, which appears to have been covered over with a truly awful-looking bolted plate, a real bodge if ever you see one…Focus is quite right to criticize the detail. It’s clear from the photographs that nothing this awful should be considered acceptable as part of the finished structure.” Controversial bolt-on plates on Victoria's brand-new $115-million bridge. The larger photo above shows the plate on the south-side ring. On Vibrant Victoria, a local online discussion forum, “jonny” noted: “I am absolutely gobsmacked that our NINE FIGURE shiny new bridge has two, seemingly haphazard and last minute, bolted-on steel plates that look like they were envisaged and put together by a 9th grade metalworking student.” “G-Man” responded, “Couldn’t agree more. It makes me want to puke. I could not care less whether or not an engineer says it’s okay. The brand new bridge should not have this. It is unbelievable. I am embarrassed as a Victorian.” Several days later, G-Man posted a photograph of the bolt-on plate on the north ring. Somehow a bolt had worked its way free from somewhere inside the ring and was trying to escape through a large gap between the ring and the bolt-on plate. A rusty bolt caught in the opening between the bolt-on plate and the defective north-side ring. Photo by G-Man. Martin Bache, a 40-year veteran of Canada’s structural steel fabrication industry, and a project supervisor with Canron in Vancouver before retiring to Victoria, wrote to Focus and commented: “I have never seen such an appalling patch.” Bache agreed that the plates would “promote corrosion” in the structure. He had contacted EGBC, BC’s association of professional engineers, which confirmed that the association’s bylaws require a third party independent review of the patches since they are on fracture-critical steel. No such review has been brought forward, or even mentioned, by either the City or the bridge’s American designers, Hardesty & Hanover. I covered the initial response from City Hall in a second story posted at focusonvictoria.ca. To put that response as succinctly as possible, the City claimed our story contained “serious factual errors and inaccuracies,” but was unwilling—or unable, to say what those errors and inaccuracies were. On January 25, Project Director Jonathan Huggett gave council his quarterly update on the troubled project. Huggett commented on the bolt-on plates: “There has been this inference by some that somebody found a piece of scrap steel, slapped it on as an afterthought, and put a few bolts in place. Whoever makes those statements clearly has no experience in engineering. As engineers we take great pride in our work. Nothing happens quickly or suddenly, and without due process and proper sign-off.” Huggett also told the CBC our story was "an attempt to scare people unnecessarily." Presumably Huggett meant that there was an implication in our story that the plates were a public safety issue. We didn't, in fact, say or imply any such thing. The issue we raised is whether or not the plates represent a significant decline in value to taxpayers. Will the plates promote corrosion and therefore increase maintenance costs? Will they reduce the useful life of the bridge and thereby increase lifecycle cost? Do the plates not make a sham of the City's claim to a "world-class" or "iconic" bridge and raise questions about the huge amount of money wasted in pursuit of that futile endeavour? The bridge engineers themselves may have metal fatigue concerns—that's why they added the plates—but Focus raised no red flags on that point other than to mention the project's own concern about fatigue. Huggett's claim of "an attempt to scare people unnecessarily" is simply deflecting attention away from the real issues. The “pride” Huggett claims has gone into this project is hard to see when you examine closely the two patches on the new bridge. And, if they are any indication of the pride with which the rest of the bridge has been built, Victorians could be in for more embarrassment. But it’s Huggett’s claim that “nothing happens…without due process” that is the focus of my attention this time. What has become evident is that Huggett may not have informed anyone at City Hall about the problem that led to the bolt-on plates, thus making it impossible to consider options that would have prevented the delivery of a defective bridge. With Mayor Helps and Huggett refusing to respond to our questions, Focus requested relevant records under access to information law. So far, Huggett and the City have been uncooperative and Focus has filed a complaint with the Office of the Information and Privacy Commissioner. In my initial story I raised these questions: Were City councillors informed about the situation that led to the bolt-on plates? If so, were they given any options? It’s vital to have answers to these questions. The plates reflect an unacceptable diminishment of the expected quality of the bridge. According to Huggett, people are comparing the bridge to scrap metal. The plates also reflect a lower-cost solution to the problem they were intended to address than a proper refabrication, but the parties building the bridge would have been responsible for any additional cost. Someone’s choice to overlook the public interest and accept a defective bridge has saved the companies building the bridge a lot of money, perhaps millions of dollars. Are councillors responsible for this bad decision? Or were they kept in the dark by Project Director Huggett? Thus far, the only indication of what happened that led to the bolt-on plates has been the minimal response from Huggett that I reported in my first story, and a letter from Hardesty & Hanover’s Keith Griesing sent to the City on January 8, shortly after our story was published. After reading our story, Griesing “felt it would be helpful if I gave you a brief summary of why those plates are there and how their use came to be.” Griesing is the project’s engineer of record. Griesing disagreed with our characterization of the circumstance that led to the bolt-on plates as a “design flaw.” His letter stated: “There was no ‘design flaw’ by Hardesty & Hanover nor any other of the City consultants involved; it was assembly by the fabricator that did not conform to the design plan requirements nor to the applicable detailing and fabrication standards required in the specifications” that led to the need for the bolt-on plates. Griesing’s need to make a distinction between a “design flaw” and “assembly by the fabricator” is understandably important to Hardesty & Hanover. If the bolt-on plates resulted from some error made by Hardesty & Hanover, they could become defendants in a legal suit if City councillors realize a world-class bodge has been foisted on City taxpayers. If the cause was solely attributable to an error made by the fabricator, then the company the City contracted to build the bridge—PCL—would be the defendant. But Hardesty & Hanover’s concern is not equal to the public interest. Just because Griesing claims Hardesty & Hanover aren’t responsible for the weakness in the rings that required the plates doesn’t mean the City hasn’t received a defective bridge. City councillors ought to be focussed on which companies the City should consider suing, rather than resorting to talking points designed to relieve them of any responsibility for their failure to protect the public interest. And just because Huggett tells City councillors there was “no design error” doesn’t mean his apparent concealment of the issue isn’t an issue. Councillors need to examine carefully the role Huggett played in the delivery of a world-class bodge. An examination of what information has been provided by the project shows none of the questions about who did what—and when and why they did it—have been answered. The bridge builders seem to have the support of Victoria City council in avoiding any financial or professional accountability for providing a defective bridge. Why? In his letter to councillors, Griesing attributes the need for the bolt-on plates to errors made by the Chinese company ZTSS, hired by PCL to fabricate the moveable part of the bridge. Griesing states: “In the course of our routine quality inspections in the steel fabrication plant in China, [PCL’s] quality control team [Atema] discovered a violation of fabrication and welding standards in the particular area in question. This determination was confirmed by the City’s Quality Oversight consultant.” According to Huggett, this discovery was made on December 9, 2016. What was found? Huggett provided Focus with a single sentence from Atema’s report. It stated: “Weld access holes in MW1 and MW3 to MF1 and TF1 at MW2 were unnecessary, not clearly detailed and may not have been evaluated to proper fatigue design category, and not fabricated to code requirements”. That’s largely incomprehensible to most of us, but here’s the essential part: Atema found “unnecessary” weld access holes in steel parts close to where the bolt-on plates were eventually added. Weld access holes are openings into otherwise closed chambers inside the rings that allow welders to complete welds within those closed chambers. Why would ZTSS cut “unnecessary” holes if it didn’t need them? With Huggett refusing to provide any information, I sought insight from the aforementioned Martin Bache, who has 40 years of experience in heavy steel fabrication. Bache described the process that would have been used for determining where such holes are needed: “Weld access holes in fracture-critical members must be designed by the Engineer of Record [Griesing]. Competent detail draftspersons would be expected, during preparation of the shop drawings, to identify closed chambers where the EOR may have forgotten to show on his plans weld access holes without which the required welding cannot be performed. They would then issue an RFI [request for information] pointing this out, and asking the EOR how they should proceed.” According to Bache, then, Griesing would ultimately be responsible for the design of every weld access hole that was required, since every steel member in the rings was designated “fracture-critical.” Griesing has told the City that the fabricator was responsible for the weld access hole violations. In that case, the bad holes wouldn’t have appeared on the shop drawings Griesing was required to approve. That means they should have been discovered quickly by any robust quality assurance (QA) program. If found quickly, those holes would still be accessible and could be fixed immediately. Bache noted: “Under what we must imagine would be rigorous QA on this second attempt to fabricate a bridge, we would expect an error to be spotted very soon after the occurrence.” But according to Griesing, “Because of its location in a critical area of the structure, this non-conformance was particularly difficult to correct.” Why, exactly? If the QA teams were as diligent as Huggett claimed in his quarterly reports, why would “unnecessary” access holes just cut by fabricators end up being “particularly difficult to correct.” From what Huggett has told Focus, we know that Hardesty & Hanover’s decision on how to address these unnecessary holes was delayed for six to seven months. During that time, fabrication of the bridge continued. It appears that Hardesty & Hanover dithered on fixing the unnecessary weld access holes, which were made inaccessible by subsequent work and couldn’t be fixed. Did Griesing forget to tell someone to do something? Bache wrote: “What amazes me is the tremendous time gap between the Atema non-compliance report and the attempted fixes. It sounds as though no one at Hardesty & Hanover could decide what to do, but the work continued and the bridge was shipped anyway to try to keep to a schedule.” Griesing’s explanation to the City noted: “The design team and fabrication team designed and reviewed numerous mitigation options. We even consulted two internationally known experts in fabrication and welding for their input. After reviewing all options, the project team unanimously agreed that the bolted plates were the best option, all factors considered.” Griesing, obviously, did not factor in jonny or G-Man. Maybe he should have. G-Man and jonny seem to represent the values and priorities of ordinary Victorians better than either Huggett or Helps. It wasn’t until after the rings had arrived in Victoria, late last summer, that large holes were chopped in the rings and plates bolted over the holes. That work was done at Point Hope Shipyard in Victoria. The need for these large holes is unclear. Were they needed to allow someone to get inside the rings so bolts could be inserted from the inside and tightened? If so, what happened to the tightener? Hey, we just want to know. Griesing’s letter provided no explanation for why a fix wasn’t made immediately in December 2016 when the unnecessary access holes would still have been accessible. So while Huggett and Griesing have successfully focussed the City on shooting the messenger, more important questions that need to be answered are being ignored. Let me, just for the sake of thoroughness, offer an alternative story to that being told by Huggett, Griesing and Helps. Let’s start with Atema’s report. Although we’ve been provided with only one sentence from that report, let’s presume that sentence is the whole report and that Atema did find weld access holes that were unnecessary and that those unnecessary holes are the entire reason bolt-on plates were required. All of those assumptions are leaps of faith, but let’s jump. In that case, PCL would have been responsible for the cost of any refabrication necessary to meet the City’s agreed-upon specifications defined in the contract. If the City had been given all the facts about this when it happened, the City would surely have insisted on refabrication rather than accepting a bridge that would forever wear “truly awful-looking” bolt-on plates. But wait. According to Griesing, the City did know about the issue. In his letter, Griesing wrote, “City Staff was fully involved in arriving at the best solution, particularly with respect to public safety, cost and schedule impacts.” The “was” in that sentence suggests a single person from the City was involved, but we don’t know for sure. Who did Griesing mean by “City Staff”? Did he mean just Huggett? Or did he mean Huggett and other people at City Hall? Again, we don’t know the answer to this yet, because Huggett has refused to respond to a legal request for his records on the issue, and Helps won’t respond to questions. But this is vital to understand because if Huggett didn’t inform anyone else at the City of Victoria, we would have to ask why he kept that information from his client. Until we see Huggett’s record of communication on the plates, no judgement can be made as to his conduct. But at this point, with Huggett appearing to have not properly informed his client, the City may need to seek advice about the implications of the plates from someone not involved in the project. Griesing’s claim that it wasn’t a “design flaw” that led to the bolt-on plates is an open question until detailed information about what Atema found, and why it took six or seven months for Griesing to act, is released. But there is a broader issue that deserves comment. In one sense there is no question that the bolt-on plates are the direct consequence of a design flaw. The design flaw was the open rings themselves. The choice of that particular design approach to creating a movable bridge made the structure unnecessarily complex, difficult to build and overly expensive. Of the three companies originally bidding for the project, two rejected the open-ring design and based their bids on designs that had proven track records. Kiewit’s engineers had concluded that the open rings posed “a fundamentally high risk and expensive design approach.” Bizarrely, the City’s scoring of the bid proposals actually penalized Kiewit and Walsh for not using the risky design. As part of PCL’s bid, Hardesty & Hanover embraced this risky design. Victoria taxpayers have been paying the costs ever since. For example: two additional years of construction are attributable to difficulty in fabricating the open rings and fitting them to the trusses. Those extra two years of construction have made people in Victoria frustrated. That sense of frustration, especially in an election year, is not something politicians like Helps and her councillors want to aggravate with further delays. Their public promise to deliver the bridge by such-and-such a date meant that if any problem arose that would cause further delay, councillors were going to favour whatever solution was quickest. They telegraphed that to Huggett and Griesing. So that’s what councillors got, but in spades. So when Griesing tells councillors that the bolt-on plates are not the result of a design flaw, he’s overlooking his company’s responsibility for promoting a design that other engineers warned the City not to build. Hardesty & Hanover’s risky and hard-to-build design created a whole chain of connected events that led inevitably to the bolt-on plates. Don’t take my word for it. Huggett has already confirmed that the City's hired technical advisors have given it bad advice on the project. Last summer, in a rare moment of self-reflection in which councillors had an opportunity to openly consider why the project had encountered such difficulties, Councillor Pam Madoff offered the following: “I remember very specifically having this conversation [with the bridge’s designers and engineers] about the mechanics, you know, the—in simplistic terms—the cogs, the wheels, how it was going to lift. I remember at the time saying, ‘Is this basically just a larger version of the Meccano sets that we played with as kids, in terms of its actual mechanical operation?’ And, again, that was the assurance. To me it comes down to: how far does one have to go? We felt like we asked the right questions at the time. It turns out they may not have been the right answers.” In response, Huggett told councillors: “You were not given good advice.” A question councillors might ask themselves right now: Why are we still accepting bad advice? Perhaps a sloppily-executed sign with those words on it could be hung from each of the bolt-on plates. With or without such signs, though, each time the bridge lifts and the bolt-on plates descend to the level of pedestrians waiting for the bridge to reopen, those present will be reminded of the bridge’s dubious origins. David Broadland is the publisher of Focus. Hardesty & Hanover's letter to the City of Victoria with its explanation of the bolt-on plates:
  8. Did Mayor Helps conceal a serious bridge design flaw from other councillors and the public at a critical moment? Only the expeditious public release of pertinent records will show what happened. TWO BOLT-ON PLATES DEFACING THE FRACTURE-CRITICAL RINGS of the new Johnson Street Bridge aren’t a problem, according to Victoria Mayor Lisa Helps. The real problem, Helps stated in a Facebook post, were “a number of serious factual errors and inaccuracies” made by me in my story about the issue in the January/February edition of Focus. Below her Facebook statement Helps endorsed comments posted anonymously on the social media site reddit. On reddit, anyone can call themselves an “engineer” by providing an email address to a computerized registration system. Helps’ and her Facebook fans were moved by the assurances of reddit “engineers” HollywoodTK and thisguy86 that there was nothing unusual about a new $115-million bridge sporting repair patches before it even opened. My own article on the issue, on the other hand, even though it is likely subject to the careful scrutiny of libel lawyers working for the companies and professionals named in the article, is, according to Helps, untrustworthy. I will come back to Mayor Helps’ role in the City’s response to the issue, but first let me report on information that came in after publication of my original story. Firstly, City of Victoria Councillor Jeremy Loveday confirmed that he had not been informed about the issue before he read our story. Loveday’s statement seems to suggest that Project Director Jonathan Huggett, a professional engineer, did not properly inform his client—the City of Victoria—about a significant structural issue that had arisen during construction of the rings in China. However, it’s also possible that Loveday is the only person at City Hall that wasn’t told. Secondly, photos taken in Victoria show the work of cutting steel out of the rings and then adding the bolt-on plates took place at Point Hope Shipyard in Victoria in October. Thirdly, engineers and experts in steel fabrication have expressed concern that the bolt-on steel plates will likely create a corrosion problem that could increase maintenance costs and shorten the useful life of the bridge. While social media comments have focussed on the way in which the steel plates diminish the structure’s aesthetic value, the plates may end up costing City of Victoria taxpayers tens of millions of dollars as a result of premature loss of use. Professional engineers and steel fabrication experts that have contacted Focus have confirmed that the concerns we identified in our story are reasonable. Even with only one sentence of the Atema report that first identified a weakness in the rings during construction in China, engineers confirmed that at least partial responsibility for the issue likely lies with the rings’ designers, Hardesty & Hanover. Until the full Atema report is released, the full extent of Hardesty & Hanover’s responsibility for the weakness in the rings is unknown. If the City had insisted on rings that did not have bolt-on plates, whatever additional costs were incurred would have been borne by the various parties to the extent they were responsible for the weakness in the rings. The extent of blame assigned to each of the parties involved is unknown. What we do know is that Hardesty & Hanover’s Engineer of Record for the project was able to sign off on a cheap, bolted-on plate solution even though he was the Engineer of Record at least partly responsible for the structural weakness that needed to be addressed. The record of how all this played out needs to be made public since there seems to be an inherent conflict of interest at work in what occurred, with City of Victoria taxpayers coming out on the losing end. Following publication of our story, a concerned steel fabrication expert asked Engineers and Geoscientists of BC (EGBC) to confirm that the addition of bolt-on plates to the fracture-critical rings needed to be approved by an engineer other than the Engineer of Record. The EGBC confirmed that such an approval would have been required and directed the expert to Hardesty & Hanover’s Keith Griesing, the Engineer of Record, for Griesing’s confirmation that such a review took place. In response to advice from one professional engineer, we checked EGBC’s online membership directory to confirm that Griesing is a registered professional engineer in BC. The EGBC did not confirm his membership. Griesing has not yet responded to a request for information from Focus. The expert in steel fabrication told Focus, “I believe it is not necessary for the Engineer of Record to be registered as a member of EGBC provided that he is registered as an Engineer in a jurisdiction acceptable to EGBC.” Lastly, we have learned that the public statements issued separately by Helps and Loveday—the same statements, word-for-word—were provided to them by City Manager Jocelyn Jenkins. Since Jenkins is not an engineer, the claim Loveday and Helps made that what we reported in our story as a “design flaw” should have been called a “fabrication challenge” had to come from Huggett. (Loveday has since apologized for not making it clear that his statement was copied from a briefing note. Mayor Helps’ has made no such clarification.) The entire attempt to build architect Sebastien Ricard’s unproven design has definitely been a “fabrication challenge,” but the specific way in which a structural weakness had been engineered into the rings remains a design flaw until further, more complete information proves otherwise. Aside from the important issues of safety, lowered life expectancy and diminished aesthetic value, there are other questions involving professional and political conduct that need to be examined. If it isn’t clear to you already, let me outline why the City’s characterization of our story as “a number of serious factual errors and inaccuracies” ought to be seen as obfuscation—a non-denial denial, as I predicted in my initial story. The weakness in the rings was first identified on December 9, 2016 in China. At the time, the rings were still being fabricated. Reinforcing the problematic section of the rings in a way that would not create long-term corrosion problems or diminish the aesthetic value of the bridge was still possible. Since the cost of that refabrication would have been the responsibility of those companies whose work had contributed to the structural weakness in the rings, the best interests of the City of Victoria would have been served by refabrication. But that didn’t happen. Why not? On the surface, it appears that no one in Victoria was told, so there was no opportunity for the City to consider its options. If the City had been told, and it had insisted on refabrication—and why wouldn’t it?—who would have had to pay? Hardesty & Hanover and/or PCL. Somehow, Victoria got a defective bridge and PCL and Hardesty & Hanover got a free pass. What happened? Huggett should have been informed about the Atema report’s findings shortly after December 9, 2016. If he was, it’s not clear whether he even notified the City. The evidence that he didn’t tell his client, so far, is the absence of any mention of the issue in his public reports, and Councillor Loveday’s public statement that our story was the first he had heard of the issue. So let’s pursue—cautiously—the hypothetical case in which Huggett told no one at City Hall. What would be the implications of that? Keep in mind that Huggett is paid approximately $300,000 each year by taxpayers to watch over the City’s interests on the project. If Huggett had told no one, the main beneficiary of such a concealment would have been Hardesty & Hanover and/or PCL. But Huggett’s client is the City of Victoria. If this was how things happened—Huggett telling no one—how would we expect a sensible mayor to act when the existence of the design flaw was publicized by Focus? A sensible mayor would see that if Huggett had kept the City in the dark, that would have allowed Hardesty & Hanover and/or PCL to avoid the higher cost of refabrication as compared with bolt-on plates. A sensible, cautious mayor would, on first hearing of this issue, understand that Huggett’s apparent failure to inform her would require the immediate production of all the records that could show exactly what took place during the nearly eight months between the Atema report and shipment of the rings to Victoria. Otherwise, public trust in civic government would plummet. A sensible mayor would demand: “Release the records.” But that didn’t happen. Rather than acting swiftly to push for release of those records, Helps parrotted Huggett’s statement, assuring the public that the real problems plaguing the bridge project were serious factual errors and inaccuracies in the observations of the guy who first noticed the bolt-on plates. So, given that Helps is a reasonably sensible mayor who is perfectly capable of sniffing out corruption, we can likely reject the hypothesis that Huggett didn’t tell anyone at City Hall. That leads us, inevitably, to the only other reasonable hypothetical possibility—that Huggett informed one or more officials at City Hall, and that between them they decided that the best course of action was to keep the issue concealed from Loveday (and probably other councillors) and settle for a quick, cheap fix that kept the bridge on schedule for completion well before next November’s civic election, bolt-on plates and all. Let’s cautiously explore this possibility. As a reporter, I’ve found that when public officials won’t answer direct questions, they are usually trying to avoid public embarrassment. It’s awful to be publicly embarrassed, but public embarrassment is a powerful and legitimate tool that has been traditionally used to hold people accountable for their actions when they screw up some decision they had to make. In preparation for my initial story, after Huggett declined to say whether he had informed the City, I emailed questions to Mayor Helps, including whether she had been filled in by Huggett on the issue. The questions were simple and could have been answered with a “Yes” or a “No.” I also asked her for important dates when things might have happened. The mayor did not respond to any of five emails sent over a one-week period. Then, following Helps’ release of the Huggett-Jenkins statement on her Facebook page and her implicit endorsement of the anonymous reddit engineers, I emailed her a request to itemize the “serious factual errors and inaccuracies” she had referenced in her statement. Normally, a public official that makes such a claim would have proactively provided that information without being asked. That’s the process: We make a mistake, the official tells us about the mistake we made, and if they are correct we acknowledge our error. So I asked the mayor to make those mistakes clear. Then something peculiar happened. Mayor Helps’ inadvertently copied me on a “proposed response” to my questions that she had meant to send only to Jenkins and Huggett and one other City staffer. “Do you see any downfalls in this approach?” the mayor asked Huggett and Jenkins. Later, realizing what she had done, Helps emailed me: “David there you have my response. Sent before my morning meditation and copied to you inadvertently. But truth may walk through the world unarmed. So please feel free to use what I have said.” She had written: “I trust all of the reporters at the Times Colonist. I trust all of the reporters at Vic News. I trust all of the reporters at CBC and CFAX. I trust all of the reporters at CTV, CHEK, and GLOBAL. This trust has come through hard conversations, good reporting and relationship building. I do not trust you. As such I feel that however I answer your questions you will use the answers to suit your own needs, not to serve the public good.” Mayor Helps made no attempt to point out even a single error or inaccuracy. The mayor’s insistence that Focus needs to negotiate stories with her before she will provide factual information is an interesting issue all by itself, but it’s not the issue at hand so let’s not be diverted by it. Why wouldn’t the mayor respond in a straightforward manner and provide the “serious factual errors and inaccuracies”? Added to her failure to answer questions for the first story, my reporter’s nose tells me Mayor Helps is hiding something. Here’s what now appears to me to be the most likely chain of events: Atema issued its report in December 2016. Huggett informed then-City Manager Jason Johnson. Johnson informed Helps and perhaps City engineering staff. Between them they decided to accept the quickest fix to the weak-rings problem and to conceal the issue from the other councillors and the public, perhaps thinking that no one would notice the bolt-on plates. Now the City is busily trying to hide their miscalculations and errors in judgement to avoid embarrassment. If I’m wrong, and neither Helps nor Huggett have anything to hide, all they need to do to prove that is to release the full Atema report, the record of all Huggett’s communications about that report and the bolt-on plates, and the required independent third-party review of the proposed fix, if that was done. Then all local media can share that information with the public, which will then be better able to gauge whether the public interest—or a corporate, political or personal interest—was served by the actions of whoever was involved. Sunshine is the best disinfectant. David Broadland is the publisher of Focus Magazine. He has been, reluctantly, following the bridge issue for about nine years.
  9. The latest cover-up on the $115-million project raises the question: What needs to change at Victoria City Hall? LIKE MANY VICTORIANS, I visited the Johnson Street Bridge construction site in early December to check out the newly-erected rings. My attention was immediately drawn to two large, heavily-bolted plates attached to the underside of each of the rings at the 12-o’clock position. Uh-oh. As you may know, I have been watching this project closely, for nine years. No such plates had ever appeared in any of the detailed construction drawings or project photographs that I had seen over the past five years of construction. I snapped a few photographs. At home, blown up, the photos showed that the welded steel rings—which took three years to fabricate in China—had recently been cut open. Steel plates, angle steel and hundreds of bolts had then been placed over the openings. This assemblage had a “quick-and-dirty” appearance, the kind of short-term repair you might expect to see on a bridge deemed to be near the end of its useful life—not at the start. As a result of a flaw in its structural design, the signature feature of the new bridge—the rings—required the addition of external bolt-on plates (inset). I sent my photos to Project Director Jonathan Huggett and asked him for an explanation. Over Huggett’s nearly four years on the job, I’ve sent him questions several times. Before this, he hadn’t answered a single question. In his last non-response, he had explained, “I am very busy trying to deal with a multitude of issues right now.” I didn’t expect to hear from him this time, either, but he surprised me. In an email, Huggett revealed that Atema—the quality-control company hired by the City of Victoria to monitor fabrication in China of the large steel parts of the bridge—had issued a “non-compliance report” (NCR) on December 9, 2016 after an inspection of the rings. Atema’s report indicated the structure contained a design flaw that could leave the rings vulnerable to metal fatigue. In response to discovery of the design flaw, Huggett says, “Lengthy discussions occurred in China and North America during the first half of 2017 and a number of different options to remedy the comments in the NCR were presented and reviewed. After discussions involving many experts in steel fabrication, the Engineer of Record agreed to design a bolt-on steel plate to ensure that the rings had not only the required strength, but also met the fatigue design requirements for the opening and closing of the bridge. This amended design was carried out and signed off by the Engineer of Record.” Wow. That’s a dramatically understated admission that the project had gone dangerously off the rails. After three years of fabrication, the rings had to be hacked into with cutting torches and hastily repaired. Yet not one of Huggett’s public reports to City councillors even hinted at such a problem. Huggett apparently had no intention of publicly acknowledging the design flaw, or the repair, unless someone else brought it up. Were those his instructions from the City? One question that immediately occurred to me: Is this the structure’s only design flaw? Huggett, a private engineering consultant, was appointed project director in 2014 by the City of Victoria after a report he authored condemned the project for its lack of leadership. He billed the City about $300,000 for his services, including expenses, in 2017. When pressed for more information, including the date he had informed City of Victoria officials about the design flaw, Huggett simply responded: “We have no additional information to provide.” If Huggett had informed anyone at City Hall about the design flaw, it most likely would have been City Manager Jason Johnson, who hired Huggett in 2014. But Johnson was fired by City council shortly after the rings arrived in Victoria, so I was unable to confirm whether Huggett told Johnson about the design flaw and repair. Five emails to Victoria Mayor Lisa Helps asking her to confirm whether or not Huggett had informed City councillors all went unanswered. I’ll come back to the question of why City Hall is reluctant to acknowledge what has happened, but first let me describe more exactly what was done to the rings after the design flaw was discovered. (If some readers have a hard time wading through this account, my apologies. I am hoping that an engineer with bridge design and/or bridge construction experience will come forward to comment on the repair that has been done to the new bridge.) I provided Huggett with a written description of what appears to be a hastily-executed repair that has been made to both rings and asked him to correct any part of my description so that it would accurately reflect the “amended design” for the public record. Huggett provided no correction. Sometime after the trial fit-up of all the major parts of the bridge in China in March 2017, significant, identical alterations were made to each ring. This included cutting out a section of steel plate from the inside flange of each ring. Steel appears to have been removed from the centre of each ring right out to their outer edge. This removal included about one metre of steel along the edge of the rings, including the weld. This project photo of the south ring in March, 2017 shows the intended design. Despite having known about the design flaw for over three months, the project then proceeded in such a way as to make it impossible to back-track and properly address the issue. Later, a large section of steel (in the area indicated by the yellow circle) was cut away from both rings and then covered over with bolt-on plates. These cutouts in the rings would have allowed access to the interior of the ring. Work may have been done inside the rings to address the fatigue issue identified by Atema. A photograph of the rings taken during their fabrication (see below) shows an abrupt narrowing of the structural steel in the same area where, later, the bolt-on plates were installed. This abrupt transition in the structure, along with an internal access port, may have prompted Atema’s report. The south ring during fabrication in China in July 2016. The yellow circle indicates the area of the ring later red-lined by Atema’s non-compliance report. Whether or not any steel was then added to the internal structure of the rings is unknown. If not, the next step would have involved attaching the external plates, which are about two metres by two metres in size. That required drilling 180 approximately three-quarter-inch-diameter holes into each ring, with matching holes in the plates. The plates are bolted along the edge of each ring to a steel angle that protrudes from the gap cut in the rings. The angles are bolted to the inside of the rings’ side webs. Filler pieces roughly support the plates at their forward edges where the plates span an uneven surface. The plates appear to be deformed (bent) across this uneven surface. One question that arises: Wouldn’t drilling a large number of holes, in a small area that had already been identified as having a weakness, further weaken the rings? The rings are considered “fracture-critical,” which implies their failure could lead to collapse of the bridge. As well, gaps and joints between the rings’ original steel and the bolted-on steel parts, evident in photographs, seem to make it possible for moisture to get between the steel surfaces and from there into the bolt holes. If that happens, corrosion would occur. The plates, angle steel and bolts introduce the need for careful, ongoing inspection, additional maintenance and future repair that would not have been required if a properly fabricated structure had been delivered. While many questions require answers from the City, what is known seems straightforward and damning: The design flaw was pointed out by a company whose actual job was to certify the grade of steel being used, monitor the quality of the welding, and ensure fabrication proceeded according to drawings that originated with Hardesty & Hanover, the company that engineered the steel lifting section of the bridge. Its drawings were supposed to be checked and approved by the City’s project manager, MMM Group, which has billed the City for close to $20 million for its services on the project since 2009. Fabrication of the lifting section began in China in early 2014. So it took nearly three years before anyone noticed this flaw in Hardesty & Hanover’s design, and then it was discovered by someone not responsible for the engineering of the structure. The structural integrity of this part of the bridge was judged to be so far below standard that an extraordinary intervention was required. It then took, according to Huggett, another six or seven months before a decision was made about how to address the flaw. Part of that decision included choosing to conceal the problem from the public. Another part of the decision was to do a quick-and-dirty repair. Is that because the rings had already been shipped to Victoria, precluding a proper repair at the steel fabrication plant in China? So many questions with no answers. While the Engineer of Record may have “signed off” on the bolt-on plates, the Engineer of Record works for the same company—Hardesty & Hanover—that engineered the structural flaw into the design in the first place. As engineers, their work is now suspect and their stamp of approval on their solution to a problem they created seems fraught with potential for conflict of interest. Wouldn’t City of Victoria councillors have wanted to obtain an independent, disinterested assessment of the proposed fix? Did they? If councillors had been made aware of this flaw and its proposed remedy, and agreed to accept a substandard bridge anyway, they have a lot to answer for—public oversight of the project appears to have failed. Until the City of Victoria makes it clear whether or not Huggett informed City officials of the circumstances related to the design flaw, it ought to be assumed that he did. If that’s the case, City councillors will need to explain the basis for their decision to accept a bridge that needed to be repaired. At the very least, they ought to provide public answers to the following: 1. When were City councillors informed about the design flaw? 2. When were they informed about the proposed fix? 3. Did the City of Victoria obtain an opinion from an independent professional engineer—one with no previous involvement with any of the parties undertaking the project—as to whether the City should agree to the proposed fix? 4. In return for accepting a substandard bridge, has the City of Victoria obtained a long-term guarantee from the builder (PCL), beyond the limited two-year warranty previously agreed to, that the damaged rings will be replaced by the builder if the repair shows any sign of deterioration or failure over the expected life of the bridge? 5. Were councillors planning on informing the public of the design flaw and repair before the coming civic election? It has taken 9 years and, if we’re honest, about $115 million to build a 156-metre-long bridge that needed to be repaired before it could be opened.Why has this happened to our city? Long before this particular design flaw emerged and its cover-up commenced, the project had repeatedly reduced the value of the bridge being built, each time concealing that fact from the public. Focus has documented this sad history, right from the project’s origins in 2008. This seems an appropriate moment to recount why this troubled project has turned out the way it has. THE LONG RECORD OF CONCEALMENT OF PROBLEMS with the bridge’s design and construction seems to be a natural consequence of the project’s controversial origins, and the haste with which a conceptual design was chosen. The project was born at the height of the world financial crisis in late 2008 and early 2009, when governments around the planet rushed forward with gigantic plans for infrastructure spending to stimulate the global economy. In Victoria, the possibility of a big federal-provincial grant appeared just after the City had received an engineering assessment of the condition of the 86-year-old Johnson Street Bridge. This unfortunate coincidence determined the fundamental nature of the project that followed: It was hurried, and therefore ill-conceived. To justify going after a big grant, whose application deadline was only weeks away, City officials had to quickly manufacture a plausible rationale for replacing the Johnson Street Bridge. They did that by abruptly announcing that the Johnson Street Bridge had a serious seismic vulnerability. On top of that, the City claimed that repairing the bridge would require lengthy closure—at great economic cost to Downtown businesses. Since that repair would be only marginally less expensive than building a new bridge at $40 million, the City argued, building a new bridge was the best choice. But before the City took that position, it had been advised, unequivocally, by two professional engineers on two separate occasions, to repair the double bridge rather than replace it. The first engineer to provide that direction, Joost Meyboom, told the City in 2008 that an adequate repair, including seismic upgrading, would cost $8.6 million. The second engineer, Mark Mulvihill, gave the same advice in 2009. Mulvihill based his recommendation on the structure’s “high and significant” heritage values. But Meyboom’s and Mulvihill’s professional recommendations were concealed by the City, and were only revealed through FOIs filed well after City council had committed to a new bridge. That’s how the project started. Founded on a fundamentally deceptive approach to providing information about the project, City managers went on to repeat—for the next nine years—that same pattern of misrepresentation and concealment in response to every major challenge that came along. Instead of following Meyboom’s and Mulvihill’s recommendations, the City placed its bet on a back-of-the-envelope concept created by Sebastien Ricard at the British architectural firm Wilkinson Eyre. Inexplicably, Ricard’s design depended on a novel open-ring (no axle) lifting mechanism that had previously been used for only two small bridges in the Canary Wharf development in London. Just a few years old, the bridges had almost no record of performance or durability. Nor was there any proof that the open-ring design could be successfully scaled up to the size proposed for Victoria. By July 2009 the City was estimating the project would cost $63 million. When it tried to proceed without electors’ consent, a counter-petition—mounted in the middle of a cold winter by indignant Victoria citizens—successfully forced the City to put its plan to a referendum. The City’s response to that setback, in preparation for a vote, was to spend heavily on creating the perception that building a new bridge would be less expensive than repairing the existing structure, and that Ricard’s design would allow a number of highly desirable features: dedicated bicycle lanes, rail, a high level of seismic protection, a wider navigational channel and a “signature” structure with high-level architectural qualities that would make the bridge “world class” and “iconic.” Sebastien Ricard’s glamorous, but hastily-conceived, 2010 design was approved by voters in a borrowing referendum. Critics of the project, like Ross Crockford, a director of the watchdog organization that had forced the City to hold a referendum, pointed to the unproven, experimental nature of the design. To Crockford— who, unlike the City, had sought out the advice of bridge engineers not involved in the project—the design presented an unnecessary financial risk to City taxpayers. The design flaw discovered by Atema is exactly the kind of risk critics like Crockford warned the City about, before and after the referendum. The City ignored those warnings, and so did the majority of City voters. In the November 2010 referendum, electors approved the City’s now-$77-million-plan. Soon after the referendum had been won, project engineers and City staff quietly began stripping away most of the promised elements of the project’s scope, even as the project’s cost continued to climb. The first things to go were rail and a wider navigational channel. Ricard’s renderings of the bridge from 2010 all show a bridge wide enough to accommodate rail and long enough to allow a navigational channel 47 metres wide. But records obtained by Focus showed that project engineers suspected Ricard’s open-ring design couldn’t actually accommodate either. By early 2011, MMM engineers were gathering evidence to help convince City managers, behind closed doors, that the City should build a much smaller bridge. By mid-2011 the City had signed a design contract with MMM that, contrary to promises made before the referendum, eliminated rail, reduced the opening span from 47 to 41 metres, and reduced the required life expectancy of the approach bridges from 100 years to 75 years. There was no proactive disclosure of these latter two reductions in quality and scope. They only became known to councillors and the public later, through FOIs filed by Focus. The shortage of truthfulness wasn’t confined to the engineers. Just before the civic election in 2011, City Manager Gail Stephens announced that the project “continues to be within the budget of $77 million and the March 2016 timeline.” But, as we learned much later, she was hiding the truth from both councillors and the public. An FOI filed in 2012 showed Stephens had been warned months before by City staff that the project was definitely over budget. Those staff advised her that councillors should be informed. Stephens failed to do so. As for her claim of being on schedule for completion by March 2016, the truth of that is now evident. In mid-2012, while the City was working with three companies short-listed to bid on the bridge’s construction contract, two significant changes were made to the project’s scope. Each of these changes was made to lower the cost of the project after the three companies bidding on the contract made it clear the City’s recently-expanded $93-million budget would not cover the cost of even the shrunken bridge it wanted them to build. The first of these changes was a decision to leave the support piers of the existing bridge in place. That would eliminate the cost of removing and disposing of the piers, but this also resulted in losing one of the primary objectives of the project: a wider navigational channel under the bridge. The width of the channel was limited by the existing piers which were 39 metres apart. Leaving them in place meant the navigational clearance would be virtually the same, with no reduction in the risk of marine traffic hitting the bridge. Project managers hid this change, too, from councillors, who were left to learn about it from the pages of Focus. At the same time, in mid-2012, City managers secretly accepted a lower standard of seismic performance for the bridge. While no engineer can, with great certainty, guarantee that a bridge will be accessible to traffic after a large earthquake, MMM engineer Joost Meyboom had convinced the City that it should buy the highest level of seismic protection possible. Meyboom put the cost of that protection at $10 million and, during the 2010 referendum, electors were told the bridge would include that high level of protection. However, after it had been established (in secret) by the three companies bidding for the construction contract that MMM’s estimate of cost was too low, MMM introduced a document into the procurement process that accepted a much lower level of seismic performance than Meyboom had previously advised the City to accept. This document’s reduced seismic design criteria allowed for the replacement of the planned all-steel approach bridges with more economical—but more seismically vulnerable— concrete structures. Again, councillors were left in the dark. I’ll come back to the lowered seismic design criteria in a moment, because the way this issue was manipulated by the City when it was made public in these pages is a good indicator of how the City will respond publicly to the design flaw issue. But first, let me refresh your memory about the warnings about the design that were provided by the companies in their bids for the construction contract. Two of the bid proposals rejected Ricard’s open-ring design outright as too risky in terms of cost, reliability, and repairability. The third bid, from PCL, rejected a part of Ricard’s design and altered what remained in a way that allowed PCL to meet the City’s price ceiling. But that alteration also resulted in material changes that PCL expected would reduce the life of the bridge before major repairs would be needed. PCL admitted its proposal would result in a bridge in which parts that were “subject to wear” would last only 30 years. Senior City managers kept all these warnings out of sight of elected officials. Records obtained by Focus show that at a critical in camera meeting soon after the bids were received—a moment in which councillors could have been fully apprised of the companies’ warnings before committing to Ricard’s design—City staff didn’t even mention them. In light of the design flaw discovered by Atema and its warning of the risk of metal fatigue, it now seems possible that one of those parts “subject to wear” is the entire section of the bridge built in China. LET'S GO BACK AND PICK UP THE THREAD about the project’s reduced seismic design criteria. The document mentioned above later became part of the City’s contract with PCL. Its presence in the contract protects PCL from any future legal claim from the City of Victoria in the case that the bridge suffers unrepairable damage—or is unusable by emergency vehicles—following a much smaller seismic event than that for which Meyboom had recommended the City prepare. Keep in mind that Meyboom had put the value of that additional protection at $10 million, and the City had agreed to pay for this extra protection in exchange for an implied guarantee that the bridge would stand up well in a large earthquake. That $10 million had been included in the “$77 million” estimate in 2010. That extra $10 million was meant for such features as all-steel approach bridges, which have much better seismic performance than concrete. Recall that questions about the seismic vulnerability of the existing double bridge had been the primary rationale for replacing it. Ironically, all four of its approach bridges were steel. But inclusion of the Seismic Design Criteria document in PCL’s contract meant the City had, in effect, agreed to a lower level of seismic performance, so concrete approach bridges could now be used in the new bridge. None of this was divulged to councillors when they were asked to approve a contract with PCL. When the issue was brought to light by Focus in 2015, Huggett, by then project director, provided an extensive non-denial denial that carefully avoided even acknowledging the existence of the contract document that contains the lowered seismic design criteria. For readers unfamiliar with the expression “non-denial denial”: This is a term coined by journalists to describe a response from a subject that sounds like a refutation of facts, but, on careful examination, doesn’t actually refute anything specific in the reporting and doesn’t provide any evidence that disproves the report, yet isn’t, itself, untruthful. EACH OF THE ABOVE DECEPTIONS was first divulged to the public in the pages of Focus. The City has never presented any evidence that what we have reported was inaccurate or untrue. Yet, in almost every case, some City official—often the mayor of the day—has appeared at other Victoria media outlets with vigorous non-denial denials of our reports. The City hasn’t limited its defensive tactics to traditional obfuscation, though. They’ve been ground-breakers on keeping the record opaque. When Focus filed an FOI that sought evidence that Stephens had been advised the project was already over-budget in 2011, the City employed a legal maneuver—used against a media outlet only once before in BC’s history—that allowed it to delay responding to our FOI. On the very day the City was required to provide evidence to the Office of the Information Commissioner to support its tactic, the City withdrew its claim. Such self-inflicted wounds to the City’s credibility have not been without cost. One cost of the serial deceptions has been a continuous loss of top-level City managers closely associated with the project: City Manager Gail Stephens, Director of Operations Peter Sparanese and Director of Engineering Dwayne Kalynchuk all “resigned” suddenly—or were fired. Others, too, have disappeared. As well as that huge loss of senior personnel, the serial deceptions have had a corrosive effect on the community’s trust in its civic government. Why didn’t City councillors put a halt to the repeated cycle of beating down the value of the project and concealment of their actions? The majority on council went along with the original rushed decision in 2009, and concealment of the project’s problems provided those seeking re-election in 2011 and 2014 with cover for their original error in judgment. To be fair, in many of the cases in which City staff reduced the scope of the project in significant ways just to keep Ricard’s open-ring design alive, councillors were simply not informed. In some cases, once those issues were made public, senior staff soon resigned or were fired. But getting rid of project managers didn’t have any effect on the basic underlying problem: The initial decision to proceed had been rushed, and in that rush a difficult-to-build and under-priced design had been chosen. That brings us back to the current issue of the design flaw discovered by Atema and concealed by…well we don’t know who yet, but when we do, we’ll let you know. What we will likely hear from the City now, if past behaviour is any predictor, is an adamant non-denial denial. Regardless, Victoria is now stuck with a badly degraded version of Ricard’s problematic design, and the only recourse for electors seeking accountability is to get out and vote in November’s election. UPDATE: A follow-up story has been posted here: Victoria City Hall continues cover-up of bridge design flaw (This story was edited in June 2018 to reflect information about the physical size of the plates obtained by FOI. The plates are each about two metres by two metres in size, not one metre by one metre as we originally reported.) David Broadland is the publisher of Focus.
  10. Recent scientific studies show how resident orca populations are affected by diminishing chinook runs and—critically—why the chinook are disappearing. RIVERS RUNNING INTO PUGET SOUND have perennially low returns of chinook salmon—currently estimated at just 10 percent of their historic levels—even though many of them are enhanced with hatcheries. Last year, scientific research connected this decline to secondary sewage treatment plants discharging partially-treated effluent into Puget Sound. Last June, a group of Washington scientists published a study showing the extent to which the decline in the birth rate of the Southern Resident Killer Whale population, listed as “endangered” by both the Canadian and US federal governments, is linked to the precarious state of the Salish Sea’s chinook salmon. Puget Sound chinook, which were given “threatened” status under the US Endangered Species Act in 1999, have become a cross-border issue. Recovery of both Puget Sound chinook and the Southern Resident Killer Whale population would require investment of many billions of dollars by Washington State in new sewage treatment infrastructure. While taking action to protect both the orca and chinook is required by US federal law, Washington State currently has no plans to make that investment. Is our southern neighbour ignoring its responsibility to be a good environmental steward? Killer Whales can be long-lived (“Granny,” above, lived past 100), but their birth rate is dependent on chinook salmon, a threatened species in Puget Sound. (Photo: markmallesonphotography.com) LAST JUNE, A BRILLIANT SEVEN-YEAR-LONG STUDY that correlated the declining birth rate of the Southern Resident Killer Whale population with falling chinook salmon numbers, mercilessly compared what’s happening to the remaining orcas to the mass starvation of the Dutch population at the hands of German Nazis during World War II. The authors stated: “The Nazis closed off the borders of Holland between October 1944 and May 1945, causing massive starvation over a 5–8 month period, with good food conditions before and after. There was a one-third decline in the expected number of births among confirmed pregnant woman during the under-nutrition period. Conceptions during the hunger period were very low. However, women who conceived during the hunger period had higher rates of abortion, premature and stillbirths, neonatal mortality and malformation. Nutrition had its greatest impact on birth weight and length for mothers experiencing hunger during their second half of gestation, when the fetus is growing most rapidly.” The inclusion of the word “Nazis” in a peer-reviewed scientific study on the reproductive dynamics of an endangered whale population may strike some as odd, but the Dutch Famine, as the above events are known, was highly unusual: it took place in a well-developed, literate population that kept excellent health records and the vast majority of those affected survived. Thus it was one of the first events in human history for which scientists had accurate, reliable records to help them understand what health impacts occur when a population of mammals is starved. The orca scientists found that a similar dynamic between food availability and birth rate has been impacting the Southern Resident Killer Whale (SRKW) population, but with one big difference: For the orca, this is not a one-time event. For them, a months-long famine now occurs almost every year. Dr Samuel Wasser, the study’s lead author, is a research professor of conservation biology at the University of Washington. Wasser’s team gathered evidence from 2008 to 2014. They found that 69 percent of detectable pregnancies in the SRKW population failed during that period. Of those failed pregnancies, the scientists found, “33 percent failed relatively late in gestation or immediately post-partum, when the cost is especially high.” That high cost included an increased risk of mortality for the would-be mother. The scientists observed: “Low availability of chinook salmon appears to be an important stressor among these fish-eating whales as well as a significant cause of late pregnancy failure, including unobserved perinatal loss.” They added: “However, release of lipophilic toxicants during fat metabolism in the nutritionally deprived animals may also provide a contributor to these cumulative effects.” In other words, not only are the orca being starved, but when a starved, pregnant orca begins burning off her fat reserves in response to the scarce supply of food, toxins bioaccumulated in her fat reserves—such as PCBs and PBDEs—begin to have more of an impact on her health, such as a reduced ability to fight infections. This could contribute to the demise of the fetus and increase the risk to the mother’s life. As a consequence of these conditions, the study found “the 31 potentially reproductive females in the SRKW population should have had 48 births between 2008–2015. Yet, only 28 births were recorded during that period. The 7 adult females in K pod have not had a birth since 2011, and just two births since 2007. The 24 females in the remaining two pods (J and L) have averaged less than 1 birth per pod since 2011, with no births in 2013, but had 7 births in 2015. One of the two offspring born in 2014 died.” As of this writing, the population has dwindled to 76 whales. As recently as 1996 there were 98 orca in the 3 pods. How did the scientists determine that 69 percent of all pregnancies failed? After all, many of the pregnancies terminated early on, and there would have been no visible signs that the females had been pregnant. How does one detect whale pregnancies? Detection dogs. Tucker, one of Wasser’s orca poop detection dogs (Photo: University of Washington) Over the seven years of the study, the scientists intermittently followed J, K and L pods through the Salish Sea and used specially-trained dogs stationed at the bow of the research vessel to sniff for orca poop, and then point out its location to the scientists. The poop was collected and later genotyped (associated with a known individual whale) and analyzed for hormone measures of pregnancy occurrence and health. The scientists also looked for chemical indicators of nutritional and disturbance stress in the poop. By making the same measurements over time, they were able to distinguish between nutritional stress caused by low availability of chinook salmon, and disturbance stress caused by the presence of nearby boats. Fisheries scientists had previously estimated that 70 to 80 percent of the SRKW population’s year-long diet consists of chinook salmon. The whales are thought to prefer chinook over other species of salmon partly because they use echolocation to find their prey. Since adult chinook are physically larger (they can weigh as much as 55 kilograms) than adults of other salmon species, chinook might be easier for orca to find. As well, there are runs of chinook returning to spawn in different river systems in the spring, summer and fall (sockeye, coho and chum return only in the fall). In the past that meant a reliable, almost year-round supply of chinook. And chinook may be preferred by the orca simply because of its higher fat content compared to other salmon. Canada’s Department of Fisheries and Oceans (DFO) estimates that reliance on chinook rises to 90 percent during July and August as the resident orca target returns to the Fraser River and rivers flowing into Puget Sound. Although the link between the abundance of chinook salmon in the Salish Sea and the physical health of the southern resident population was known, Wasser’s research provides the first confirmation that low availability of chinook is suppressing the population’s birth rate and endangering the health of reproductive females. Wasser included comparison over the seven years of the study of the two main chinook runs that are keeping the southern orcas alive: the Columbia River early spring run and the Fraser River summer and fall runs. Depending on the timing of those runs, and how many fish were in them, the southern resident orca experienced more or less intense famines through the winter months and between the spring and summer runs. Estimating how many more chinook would need to be in the Salish Sea to make up for the southern orcas’ nutritional deficit wasn’t part of Wasser’s research. But in 2010, DFO estimated the nutritional requirement of the southern resident orca population, which then numbered 87, at about “1200 to 1400” chinook per day. Over the five-month period the orca occupy their critical habitat in the Salish Sea each year, that would amount to 180,000 to 210,000 chinook. Wasser’s research shows the whales weren’t catching enough chinook in 2010 and the deficit is threatening the population. Yet in the Salish Sea in 2010, the total number of chinook caught by commercial and sport fisheries, plus the number of chinook that escaped to spawn, was about 500,000 fish. (These numbers are from the US EPA and the Pacific Salmon Commission.) Of those, 320,000 returned to their natal rivers to spawn. The 180,000 fish taken by commercial and sports fishers were split roughly in half between Canada and the US, even though 94 percent of the spawning fish were headed for the Fraser River in Canada. Only 6 percent were headed for rivers in Puget Sound. Note that the total catch taken by humans is roughly equivalent to the catch required by orca. The quickest way to end the orca famine would be to end the commercial and sports fisheries for chinook in the Salish Sea, and Canadian scientist David Suzuki called for that action following the release of Wasser’s study. To recover chinook populations, however, will require a deeper understanding of why they are declining. A comparison of the Southern Resident Killer Whale population with their northern cousins helps in that understanding. Wasser noted the “significantly lower” fecundity rate of SRKW compared to the Northern Resident Killer Whale (NRKW) population. From a 2011 study by Ellis, Tower and Ford, we know that in 1974 there were 120 whales in the NRKW population; by 2011 that had risen to 262. According to Canada’s Species at Risk Registry, the population grew to 290 by 2014. DFO used this number in its 2017 reports. Above: Both NRKW and SRKW populations feed primarily on chinook, but one population of whales is growing while the other has stagnated since 1974. Data from DFO and The Center for Whale Research. Over that same period, though, the SRKW population went from 70 to a high of 98 in 1996 and then dropped to the current 76. Although both resident groups experienced a decline in population after 1996-1997 following significant declines in chinook runs, the northern population then recovered and grew steadily while the southern population has languished. As mentioned above, scientists have determined that both orca populations prey heavily on chinook as they return to spawn. It’s also known that, while their territories overlap, the northern orca rely on chinook returning to spawn in rivers north of the Salish Sea. The relative strength of the northern population compared to the southern, then, suggests the low availability of chinook that’s limiting growth of the southern orca population is a result of something that’s happening to the southern chinook that’s not happening to the northern chinook. What could that be? The most dangerous period in a chinook salmon’s life, according to fisheries scientists, is its first year. Research scientist Dr James Meador, an environmental toxicologist with the US National Oceanic and Atmospheric Administration (Fisheries) in Seattle, estimates the current first-year survival rate of Pacific Northwest ocean-type juvenile chinook salmon at 0.4 percent. That’s four-tenths of one percent. Another way of stating that is that 99.6 percent of ocean-type chinook salmon die in their first year. That year is spent in their natal river, their natal estuary and marine waters not too far from that estuary. Since this is where almost all of the mortality occurs, it follows that any substantial recovery of chinook numbers would require conditions in these areas to improve. A doubling of the current rate of survival in that first year—so that only 99.2 percent of them die—could double the number of fish that return to spawn. We’ll come back to Meador later. Wasser and his University of Washington team concluded their paper with this noteworthy comment: “Results of the SRKW study strongly suggest that recovering Fraser River and Columbia River chinook runs should be among the highest priorities for managers aiming to recover this endangered population of killer whales.” What about Puget Sound, where chinook runs are listed as “threatened”? Historically, according to Jim Myers of the Northwest Fisheries Science Centre in Seattle, the Puget Sound chinook runs were about 25 percent greater than the Fraser River’s. But in 2010, according to the US EPA and Pacific Salmon Commission, Puget Sound returns were only 6 percent of Fraser River returns. The much bigger hole in chinook numbers is in Puget Sound. Shouldn’t international attention be focussed there? Instead of accepting responsibility for the role it has played in the orca famine, Washington State has shifted public attention away from its lack of action, thereby reducing the chances of the Southern Resident Killer Whales’ survival. Now the situation is getting critical. The EPA recently downgraded the endangered whales’ survival status from “neutral” to “declining.” Time is running out. Wasser, on sabbatical, was unavailable to explain why the recovery of Puget Sound chinook stocks shouldn’t be a priority in the effort to recover the southern population of killer whales. However, an examination of two scientific studies published by Meador shed light on why Wasser and other fisheries researchers might not regard recovery of the Puget Sound runs as a likely prospect to save the orca. The decline of the Southern Resident Killer Whales may be linked to the low survival rate of juvenile Chinook salmon in contaminated Puget Sound estuaries. (Photo by Roger Tabor, US Fish and Wildlife Service) IN 2013, DR JAMES MEADOR published the study “Do chemically contaminated river estuaries in Puget Sound affect the survival rate of hatchery-reared chinook salmon?” Meador was with the Ecotoxicology and Fish Health Program at the Northwest Fisheries Science Center in Seattle. NFSC is a division of NOAA. In that study, Meador observed: “Ocean-type chinook salmon that rear naturally or are released from a hatchery migrate in the spring and summer to the estuary as subyearlings (age 0+) and reside there for several weeks as they adjust physiologically to seawater and increase in size and lipid content before moving offshore to marine waters… Conversely, juvenile coho salmon spend their first year in freshwater and migrate to the estuary in the spring or summer as yearlings (age 1+), generally spending only a few days in the local estuary before proceeding to more open waters. This major difference in life history can have a large effect on the degree of toxicant exposure in contaminated estuaries, which can affect fish in several ways, including impaired growth, altered behavior, higher rates of pathogenic infections, and changes to physiological homeostasis, all of which can lead to increased rates of mortality.” The physiological process of a juvenile salmon acclimatizing to saltwater is known as “smolting.” The juveniles become “smolts.” Meador examined the records from hatcheries on major rivers flowing into Puget Sound over the 36 years between 1972 and 2008. Some of the rivers had contaminated estuaries while others were considered uncontaminated. He determined the difference in the chinook smolt-to-adult return rate between rivers with contaminated estuaries and those with uncontaminated estuaries. Meador noted that the smolt-to-adult return rate is the “primary metric to assess life-cycle success.” He did the same analysis for hatchery coho in these rivers. Coho pass quickly through their natal estuaries and so would be far less impacted by contaminants in that estuary. The coho data, Meador clarified, “was used as another line of evidence to test the hypothesis that contaminated estuaries are one of the main factors determining the rate of survival for chinook.” And that’s what he found: Coho survival was not substantially impacted by contamination in their natal estuary. Meador noted that “Salmonid survival is dependent on a large number of factors, many that co-occur. The analysis presented here is simplistic, but highlights an important relationship between hatchery chinook survival and contaminated estuaries. Because this analysis examined the smolt-to-adult survival rate in fish from a large number of hatcheries and estuaries over several years in one geographical location, many of these factors were likely accounted for and therefore had less of an effect on the overall results.” As mentioned earlier, mortality in the first year of an ocean-type chinook is high. Meador described this as follows: “Survival for first-year ocean-type chinook in the Pacific Northwest has been estimated at 0.4 percent. Rates of survival over successive years are considerably higher for 2-, 3-, 4-, and 5-year-old fish at 60 percent, 70 percent, 80 percent, and 90 percent, respectively. Clearly, first-year survival is important for chinook, and most of the mortality for first-year ocean-type chinook is attributed to predation, poor growth, pathogens, starvation, and toxicants.” Meador determined whether or not a particular estuary was “contaminated” or “clean” based on existing records of contaminants found in juvenile chinook tissue in that estuary, records of sediment contamination, and whether or not the estuary had been listed as a contaminated site. He noted that most of the data on contaminants he was able to access had focussed on polychlorinated biphenyls (PCBs) and polycyclic aromatic hydrocarbons (PAHs). The scientists did not make their own measurements of contaminants in the estuaries, nor did they speculate on the possible sources of such contamination. They simply compared the statistical differences in survival rates for chinook smolts between apparently contaminated estuaries and apparently uncontaminated estuaries. Meador concluded that “when all data were considered…the mean survival for juvenile chinook released from hatcheries into contaminated estuaries was 45 percent lower than for fish outmigrating through uncontaminated estuaries.” In other words, a contaminated natal estuary causes a nearly two-fold reduction in survival compared with uncontaminated estuaries. Wow. That was quite a discovery: A single factor that doubled the mortality of a threatened species of fish that was known to be the cornerstone of the diet of an endangered species of whale. Meador’s data was confined to juveniles that came from hatcheries. Does his conclusion apply to river-reared chinook? Meador’s study reported that, except for the Skagit River, Puget Sound rivers are all dominated by hatchery-bred chinook. But, for juveniles whose parents spawned in rivers, the effect of contaminants may be even greater than for hatchery-bred fish. Meador noted that “wild juvenile chinook spend approximately twice as long in the estuary as do hatchery fish, which would likely increase their exposure to harmful chemicals.” If the incidence of a contaminated natal estuary was limited to one or two of Puget Sound’s smaller rivers, this effect might not be of too great consequence. But that’s not the case. Some of the Sound’s largest river systems have contaminated estuaries. For example, the Snohomish and Puyallup rivers have the second and third largest drainage areas in the Puget Sound Basin, an indication of their potential for rearing chinook. Two forks of the Snohomish—the Skykomish and the Snoqualmie—have, according to Washington fisheries scientists, the potential for up to 84,000 spawners. But over the last four decades these rivers have been averaging only 4,500, a mere 5 percent of this river system’s potential. Meador’s research suggests this and other rivers’ collective capacity to provide nourishment for a healthy Southern Resident orca population is being cut in half, year after year, by the contamination in their estuaries. But what contamination? The Puyallup River—which once hosted one of the largest chinook salmon runs in Puget Sound—now hosts the Tacoma Central Wastewater Treatment Plant, which is permitted to discharge up to 10,000 kilograms of suspended solids per day into the river’s estuary, habitat critical to juvenile chinook. IN 2016, MEADOR PUBLISHED “Contaminants of emerging concern in a large temperate estuary” in the scientific journal Environmental Pollution. The CECs targeted in the study included a long list of pharmaceutical and personal care products, hormones, and a number of industrial compounds. Many of these substances, the authors observed, “are potent human and animal medicines.” They considered their targets to be just a “representative subset” of CECs in the environment, not a comprehensive list of what’s actually there. The scientists estimated there are over 4000 CECs leaking out into the ecosphere. Meador referenced his earlier study, noting that “juvenile chinook salmon migrating through contaminated estuaries in Puget Sound exhibited a two-fold reduction in survival compared to those migrating through uncontaminated estuaries.” His choice of targets suggests that he suspected secondary sewage treatment plants might be the source of the contamination that is causing that two-fold reduction in juvenile chinook survival. He noted that “some CECs are poorly removed by wastewater treatment plant processing or are discharged to surface waters, including streams, estuaries, or open marine waters due to secondary bypass or combined sewer overflows.” Having found no other research by other scientists along this line of investigation, Meador noted that “bioaccumulation and comparative toxicity to aquatic species constitutes the largest data gap in assessing ecological risk” posed by CECs. Meador’s team targeted 150 contaminants. They focussed on three estuaries, two considered to be contaminated and one uncontaminated. The two contaminated estuaries (Puyallup River and Sinclair Inlet) each had one or more secondary sewage treatment plants discharging treated effluent into the rivers on which they were located. The third, the Nisqually River estuary, which doesn’t have a sewage treatment plant above it, was intended as a reference—an uncontaminated estuary to establish the extent to which the other two were contaminated. The researchers took water samples from the estuaries and effluent from the treatment plants and analyzed each for the 150 target contaminants. As well, they netted juvenile chinook and Staghorn sculpin from the estuaries and whole-body tissue analyses for contaminants were performed. Eighty-one of the CEC’s were found in effluent being discharged from the treatment plants; 25 were detected in the estuaries. To the surprise of the researchers, nine (9) of the CECs were detected in the water column of the Nisqually estuary, which they had supposed was uncontaminated. Their data indicated an even more disturbing situation: “Collectively, we detected 42 compounds in whole-body fish. CECs in juvenile chinook salmon were detected at greater frequency and higher concentrations compared to Staghorn sculpin.” Finding more CECs in fish tissue than estuary water meant juvenile chinook were quickly bioaccumulating the CECs. Moreover, the chinook were absorbing a higher dose of toxins in just a few weeks than were the Staghorn sculpin, which spent their entire life in the estuary. Of the targeted contaminants, 37 were found in chinook. This included, from A to Z: Amitriptyline, Amlodipine, Amphetamine, Azithromycin, Benztropine, Bisphenol A, Caffeine, DEET, Diazepam, Diltiazem, Diltiazem desmethyl… well, you get the picture. How might multi-contaminant doses lower the survival rate of juvenile chinook? The scientists found “several compounds in water and tissue that have the potential to affect fish growth, behavior, reproduction, immune function, and antibiotic resistance,” all of which could lead to early mortality. But they also noted that even if individual contaminants weren’t at a lethal level in tissue or organs, the cumulative effect of so many different contaminants in the juvenile chinook at the same time could very well be lethal—the drug-cocktail effect that so many humans experience, sometimes with fatal results. The scientists put this finding in the context of Puget Sound as a whole: “The greater Puget Sound area contains 106 publicly-owned wastewater treatment plants that discharge at an average total flow about 1347 million litres per day (Washington Department of Ecology (2010)). Our study examined two of these with a combined total of 71 million litres per day. The output for these two wastewater treatment plants alone was on the order of kilogram quantities of detected CECs per day into estuarine waters of Puget Sound. Considering the low percentage of commercially available pharmaceutical and personal care products analyzed in this study and the amount of effluent discharged to Puget Sound waters, it is possible that a substantial load of potentially harmful chemicals are introduced into streams and nearshore marine waters daily. If the concentrations from the two studied effluents are representative of that from other wastewater treatment plants in Puget Sound, then it is reasonable to assume that inputs to streams and nearshore waters are substantial and likely on the order of 121 kilograms per day (approximately 44,000 kilograms annually) and even higher if secondary treatment bypass, permitted flows, maximum outputs, unmeasured compounds, septic system contributions, and transboundary contributions are considered.” Some of Puget Sound’s largest secondary sewage treatment plants. There are 106 publicly-owned sewage treatment plants in the Puget Sound Basin. Many are located on or near to the natal estuaries of threatened chinook salmon runs. All of Puget Sound is considered to be an estuarine ecosystem. The data the scientists collected contained another ominous finding. The concentrations of the targeted contaminants found in the effluent from the treatment plants were unexpectedly high, by American standards. Meador found that “a large percentage of the chemicals detected in Puget Sound effluents are among the highest concentrations reported in the US, which may be a function of per capita usage of these compounds or the treatment processes used at these wastewater treatment plants.” One final, noteworthy point: In the estuary that was thought to be uncontaminated—the mouth of the Nisqually—the researchers found 9 of the targeted contaminants in estuary water and 13 in chinook. Meador observed, “Based on our water and fish data, the Nisqually estuary was more contaminated than expected, which highlights the difficulties of establishing suitable non-polluted reference sites for these ubiquitously distributed CECs.” This observation has an interesting implication with respect to Meador’s earlier study, mentioned above, in which he was comparing the survival rates of juvenile chinook between contaminated estuaries and those considered uncontaminated. The Nisqually estuary was on the “uncontaminated” side of the ledger in that study, but on investigation it was, in reality, merely less contaminated. Would Meador’s finding of double the rate of mortality have risen if he actually had a number of pristine estuaries to compare with those that are contaminated? IN AN EARLIER STORY (“Washington’s phony sewage war with Victoria,” Focus, May 2016) we reported on the 32.4 million kilograms of suspended solids permitted to be discharged by 77 of Puget Sound’s largest wastewater treatment plants each year. Attached to those solids are many contaminants, including PCBs and PBDEs, not targeted by Meador’s study, but known to have a negative impact on the health of fish and their sources of food. The additional impact on chinook smolts, after they leave their natal estuaries and migrate through this near-shore chemical soup—dubbed “Poisoned Waters” by the 2005 PBS film of that name—is hinted at by the Puget Sound Basin’s 10-fold decline in chinook returns from historic numbers. As the urbanization of Puget Sound’s shores has spread, and the daily recontamination of marine and estuarine waters has grown, the chinook and the Southern Resident Killer Whales have been pushed closer and closer toward extinction. This intense urbanization—right beside the critical habitat of both whales and their prey—is not occurring for the Northern Resident Killer Whale population, and that difference may be the deciding factor in the different birth rates of the two populations. Given the seriousness of the situation and the headlines in the media about drugged fish in Puget Sound, one might have reasonably expected that Washington State’s political leaders would respond to Meador’s findings. After all, what Everett-Seattle-Tacoma residents were flushing down their toilets into Puget Sound by way of sewage treatment plants was doubling the rate of mortality of a fish already listed as threatened under the Endangered Species Act. They did respond, but apparently only to deflect attention away from Puget Sound’s contamination from sewage plants. To do that they pointed at…Victoria. Just two days after an embarrassing drugged-chinook story appeared in the Seattle Times, Washington State Representative Jeff Morris boldly announced a proposal to ban Washington State employees from claiming travel expenses for trips made to Victoria until Victoria built a sewage treatment plant just like the ones around Puget Sound. A week later, Morris sent a letter to Victoria Mayor Lisa Helps claiming that “chemical loading” from Victoria’s marine-based sewage treatment system poses a “long-term risk” to “our shared waters.” Morris’ letter was signed by 36 other Washington legislators whose districts border on Puget Sound. The legislators’ letter informed Helps: “We recognize the shared risk in short-term loss of tourism activity on both sides of the border from publicity surrounding [Victoria’s lack of secondary sewage treatment]. However, we believe the long-term damage to marine mammals, in particular, but all marine wildlife, does more long-term damage to ecotourism.” Washington State Representative Jeff Morris Morris’ idea that extinctions should be prevented because they’re bad for tourism highlights the gap between a politician’s level of understanding of this critical issue and the depth of knowledge that has been created by scientists like Wasser and Meador. If State legislators were drawing up an action plan for the recovery of Puget Sound, they could do worse than to put on their list: “Read some science about contamination.” The Washington legislators’ proposal to discourage State employees from travelling to Victoria—a move they didn’t follow through on—wasn’t the only action precipitated by Meador’s science. There was a bureaucratic response as well. The Puget Sound Partnership (PSP), which describes itself as “the State agency leading the region’s collective effort to restore and protect Puget Sound,” undertook two related “actions” after Meador’s study had been published. One of those was “Action 0156,” which directed the University of Washington to conduct an “analysis of impacts…from Victoria, BC sewage.” Nowhere to be found on PSP’s long list of actions was any analysis of the impacts from the 106 publicly-owned sewage treatment plants around the Sound that are permitted to discharge over 32.4 million kilograms of suspended solids each year. The PSP also committed to “Action 0048,” which was “Identifying sources of contaminants harmful to juvenile salmon.” PSP reports that after the expenditure of $273,000, the project is “off-schedule.” Contacted by Focus, the Washington State Department of Ecology—the agency responsible for undertaking the analysis—clarified that the study “was not actually funded.” It appears that little else on the “Action” list for the Sound’s recovery is funded, either. PSP estimated its list of “Actions” for 2016 would cost $130 million, but acknowledged that only $17 million of that had been found. Washington’s Department of Ecology confirmed that, as of 2016, the State had no plans to upgrade or relocate any of the existing large sewage treatment plants on Puget Sound. Washington State says it’s commited to the recovery of Puget Sound. That would require the State to act on its scientists’ findings about the ecological impacts of ongoing contamination from its sewage treatment facilities. Unfortunately, the State’s current course doesn’t appear likely to produce anything that the Southern Resident Killer Whales will be able to chew on. David Broadland is the publisher of Focus Magazine.
  11. A shadowy group has launched complaints with the RCMP and several other public agencies. IN EARLY SEPTEMBER, Metchosin Mayor and CRD Director John Ranns, in the midst of an on-air conversation with CFAX’s Adam Stirling about the CRD’s sewage treatment project, dropped a bombshell: “There’s a group of engineers and lawyers that have put together the most remarkable chronology of events through this whole process and have filed complaints with the police. I have been interviewed by the police. They filed complaints with the ombudsman, with the society of engineers, I don’t know where that complaint has gone. Nobody wants to touch this.” Ranns later introduced me by email to the person acting as the voice of the mysterious “group of engineers and lawyers.” In exchange for a promise not to reveal Voice’s name, I was forwarded all the documents that had been sent by the group to the Victoria division of the RCMP’s Federal Serious and Organized Crime office, the Association of Professional Engineers and Geoscientists of BC, the CRD Board Chairperson Barb Desjardins, the BC Attorney General’s office, and Ombudsperson Jay Chalke. Voice also provided me with an email exchange between RCMP Constable Erin Bajic and Voice. The documents confirmed who was behind the complaints. While Ranns had obviously been impressed by the information presented, the RCMP was blunt in its assessment. Staff Sergeant Steve Wetter of “E” Division Federal Serious and Organized Crime Section informed Voice by letter, “In thoroughly reviewing the correspondence you provided, and in speaking with witnesses you have named, our review has found no evidence of a fraud, breach of trust by a public officer, or other offence in this matter.” Read the RCMP's response to the complaint Why did Wetter come to that finding? He told Voice, “You have indicated that it is your belief, based on your research and received sewage treatment plan proposals from Pivotal, a competing company to Stantec, that the sewage treatment plan proposed by Stantec is not financially prudent to the taxpayer.” Wetter seems to be suggesting that the complaint of criminal and professional wrongdoing had actually been motivated by commercial competition between Stantec and Pivotal. The “Pivotal” Wetter referred to is Pivotal IRM, a Victoria-based company that has represented both a gasifier manufacturer and a Norwegian company that builds wastewater treatment plants. Pivotal isn’t, strictly speaking, a “competitor” to Stantec in the sense that Wetter seemed to mean. Stantec was appointed as the CRD’s treatment program management consultant. Its role was to guide the CRD toward decisions on what technology to use and the most suitable capacity of the treatment plant, sludge processing facility and conveyance systems. Stantec created the “indicative design” for a secondary treatment plant at McLoughlin Point and advised the CRD to use anaerobic digestion to reduce the sewage sludge that plant would produce. Stantec’s consultant role excluded them from being involved in any consortium of companies that would eventually bid to build the system. Pivotal, on the other hand, represented one company that wanted to play a major role in building the treatment facility, and another that wanted to supply gasifiers to burn off the sewage sludge created by the treatment process. From mid-2013 onward, Pivotal’s Chief Executive Officer Chris Corps criticized the program that Stantec had developed, both in public and in private, and promoted a distributed, multi-plant system equipped with gasifiers to burn off sewage sludge. By mid-2014, even though the contract to build a plant at McLoughlin Point in Esquimalt had already been awarded, the CRD board was forced to consider other options after BC Minister of Environment Mary Polak declined to require Esquimalt to allow the project to proceed. Esquimalt had voiced several objections to the project, and one of Esquimalt’s appeals to Polak was that there appeared to be cheaper alternatives than the one the CRD had chosen. A grassroots organization had emerged—The Rite Plan—that promoted Corps’ vision of many small distributed plants and gasifiers. Led by Richard Atwell, The Rite Plan claimed Corps’ approach would be significantly less costly than Stantec’s McLoughlin-Hartland plan. Corps appeared to be closely involved with that group and Atwell. That could reasonably be seen as an attempt by Pivotal to use a grassroots organization to apply pressure on CRD board members to shift the project away from Stantec’s recommendations so that the technologies that Pivotal represented might be considered for the project. So while Staff Sergeant Wetter’s assessment that Pivotal was in competition with Stantec was presented in blunt terms, it wasn’t wrong. I recently asked Corps if he had instigated the complaints. In an email Corps wrote: “It’s not coming from me.” Here’s the “group of engineers and lawyers” own summary of their complaint to the RCMP, verbatim: • CRD staff have known, and been repeatedly advised, that the sewage project is over-scoped, which now calculates to at least 75 percent. Staffs are documented agreeing in writing that the project is over-scoped. Challenges by CRD Directors on the scope were overridden, even while multiple independent engineers documented the excess; • Staff failed to act on evidence from potential providers that the project could be done more cheaply. Published evidence by CRD’s own advisors illustrates the excess scope is at least $280M ($192M conveyance avoidance plus $88M savings through gasification rather than digestion); • Staff stated they did so in order to keep federal and provincial funding levels even though they knew the scope was falsely high, yet there are provisions at CRD and provincial level that require bureaucrats be fiscally prudent and minimize costs for the taxpayer - unless there is a reasonable justification to do so; • Recently, the CRD’s Project Board was appointed by the province to verify the project. Not only did they ignore what the engineers have said, but they also based their decision on the original and flawed CRD mandate to arrive at their conclusion; • The Project Board then forced acceptance on CRD, they direct-awarded and sole-sourced the project (via the original group) ignoring all standard procurement transparency and fairness practices and refusing to go out to tender, which prevented the issue from being exposed. • Because the staff and Directors were aware of the flawed calculations and potential savings, this is a Breach of the Public Trust. Now each of these claims can be refuted with evidence, but it’s not necessary. There’s an easier way to show that the overriding complaint—taxpayers are paying too much as a result of CRD staff’s wrongful actions—is at odds with reality. Consider the following: At the same time as the CRD was responding to a federal deadline for creating a land-based secondary sewage treatment system for Victoria, Metro Vancouver Regional District was doing exactly the same thing. It had been ordered by the federal government to build secondary sewage treatment for the municipalities of North Vancouver and West Vancouver by 2020. Both the project on the North Shore and the one here are now at the same stage—construction contracts have been awarded and some work has begun. Because of similarities in the timing, size and technologies used in the two projects, it’s easy to compare their relative overall cost. If, as the “group of engineers and lawyers” claim, Victoria’s project is greatly “over-scoped” and more costly than it need be, then that would be obvious in any comparison with the Lions Gate project in Vancouver. The Lions Gate project will serve an area with a residential population of about 158,000 (2016). The estimated project cost is $700 million. On a per capita basis, that’s $4430 per resident. The McLoughlin-Hartland treatment project will serve areas with a residential population of about 298,000 (2016). The estimated project cost is $765 million. On a per capita basis, that’s $2570 per resident. If Victoria’s project has been “over-scoped” by “at least 75 percent,” then Vancouver’s has been “over-scoped” by twice that percentage. Yet no one is claiming that project is too large or that there has been fraud or breach of trust in Vancouver. The contracts for both projects were won—in separate, open, competitive procurement processes—by the same consortium of companies, Harbour Resource Partners. In Victoria, the initial cost estimate and indicative design was produced by Stantec. Stantec was not involved in the Lions Gate project. Specific differences between the two projects suggest the Victoria system should, in fact, be considerably more expensive, per capita, than the Lions Gate project. For example, the Lions Gate anaerobic digester will be located right beside the wastewater treatment plant, whereas McLoughlin’s digester will be located 19 kilometres away at the end of a pipeline that is estimated will cost $90 million. As well, the Victoria project involves a costly under-the-harbour force-main tunnel. The Lions Gate project has no such tunnel or cost. If anything, the CRD has likely underestimated the required capacity of the system and underestimated construction costs. We shall see. Any responsible authority considering the “remarkable chronology of events” outlined in the complaint, in the context of the wider world of wastewater treatment construction projects, would come to the same conclusion that Staff Sergeant Wetter did: There’s no evidence of a fraud, breach of trust by a public officer, or other offence in this matter—unless it’s in Vancouver. CRD staff and most elected directors did, however, commit an act of utter foolishness. They ignored what local marine scientists and public health officials told them about the existing, already-paid-for, tidal-powered, source-controlled marine-based treatment system and instead followed orders given by senior government bureaucracies who had little knowledge of the actual environmental conditions here. Those scientists and health officials made it clear that no evidence had been provided that the approximately $800 million project would produce an environmental or public health benefit. The CRD proceeded anyway. Now that should be a crime. David Broadland is the publisher of Focus.
  12. To create a realistic pathway to a low-carbon regional transportation system, science—not activism—needs to lead the way forward. IT HAD LONG BEEN MY UNDERSTANDING that cycling—all on its own—would become a significant part of the solution for reducing local transportation emissions. However, when I used the Capital Regional District’s most recent comprehensive travel survey to estimate the relative amount of work done by each form of transportation at the regional level, I was flabbergasted to find that cycling accounts for such a tiny share: 1.5 percent in 2011. The amount of work done by each transportation mode can only be compared when you consider the total distance travelled each day by CRD residents using each type of transport. Replacing the work done by fossil-fuelled automobiles is essential if we’re going to reduce emissions. But how much of that work can be replaced by humans exerting themselves by cycling or walking instead of driving? More than is currently the case in our region, no doubt. But when we consider how to shift enough of the work done by automobiles to more energy efficient modes of transportation, like walking, cycling, and transit, the magnitude of the challenge facing us becomes clear. There has to be a huge shift in how people move around, quickly. Why time is such a critical part of the equation should be obvious, and the Trudeau government’s announcement late last year of a mid-century emissions goal establishes the rate of descent for making reductions. The perplexing question is: What do we shift to? Cycling and walking are part of the solution, but there needs to be a massive shift of the work done by cars to public transit. If other places that have already made this change are any indication of what Victoria will choose to do, the role of cycling and walking will largely be for making the first short leg of a trip made by public transit. While we’re seeing local governments create isolated pockets of inordinately expensive improvements for cycling, there’s little evidence that the region is on the verge of making sensible (let alone massive) investments in public transit. I pointed this out in the last edition in “Mayor Helps’ 1.5 percent solution,” which was subtitled, “Local government’s response to reducing transportation emissions may be wishful thinking. Or foolish.” New two-way protected cycling corridor in Downtown Victoria Responding at a local level to the existential threat posed by climate change, rising sea level and ocean acidification—all caused by carbon emissions—will be a transformative, Herculean task that requires constant, difficult conversation about the path we should be on. If we Earthlings don’t do this work—including the conversations—we’re cooked. What is the task facing us? According to the CRD, 55 percent of emissions generated in the region come from fossil-fuelled vehicles. Unless there is a significant and quick decline in their use, the planet will be at increasing risk of runaway warming. We simply can’t take a long-term approach to this shift. How rapidly do we need to act? The Trudeau government’s overall emissions goal is to lower them by 80 percent (compared with 2005 levels) by 2050. As yet, no targets have been set for individual economic sectors, but it’s reasonable to assume that the transportation sector’s contribution would have to be on the order of 80 percent, give or take a few percentage points. To be on the most gradual descent that would get Canada to that goal, transportation emissions, and those from other sectors, would need to be reduced by about 34 percent over the next 12 years. Canada’s mid-century emissions target, announced in late 2016, means an overall emissions reduction of 34 percent by 2030—12 years from now. To put that time frame into perspective, consider that the City of Victoria started the process to replace the Johnson Street Bridge in 2008. It will, hopefully, open for traffic in 2018, ten years later. The amount of time left before 2030 is only a little longer than the City of Victoria needed to build a 156-metre-long bridge. What would this rapid transformation mean for drivers of fossil-fuelled cars in Victoria? Collectively, over the next 12 years, we will have to either drive 34 percent less distance each day, get new vehicles that use, on average, 34 percent less fuel, shift 34 percent of our travel to non-fossil-fuel modes of transportation, or employ a strategy that combines some or all of these. What is the CRD’s plan for responding to the goals announced by the Trudeau government in late 2016? In its already-outdated 2014 Regional Transportation Plan (RTP), the CRD noted: “Long-term transportation planning efforts and investments are therefore needed to help reduce GHG emissions and adapt to a changing climate—both requirements are fundamental principles to all of the themes elaborated in this RTP. This means focusing on integrating land use and transportation planning to support sustainable transportation choices and reduce trip distances.” The CRD’s short-term plan is to double ridership on public transit by 2030 and build more cycling and pedestrian infrastructure. Will this suffice to meet our national emission reduction target? The short answer is a definite “No.” I’ll show you the arithmetic for that conclusion later on. In “Mayor Helps’ 1.5 percent solution” I used the CRD’s most recent and most comprehensive survey of the region’s transportation system, done in 2011. It showed that autos accounted for 88 percent of the distance travelled in the CRD each day. By comparison, public transit accounted for 7.1 percent, walking 1.7 percent, and bicycles 1.5 percent. I questioned whether the CRD’s plan would be able to significantly shift the share of the work being done by the various modes of transportation enough to significantly reduce emissions. These numbers baffled cycling advocates, who were more familiar with “mode share” to describe cycling’s contribution to our transportation needs. Mode share is a way of comparing the number of individual trips made by each form of transportation in a day. Using mode share, both a 3-kilometre trip on a bicycle and a 10-kilometre drive in a car are given equal weight. Although the CRD’s 2011 information shows bicycling had a mode share of 2.8 percent in the region, in certain places and for certain trip purposes, such as commuting to work in the City of Victoria, cycling’s mode share can be considerably higher. The Victoria area isn’t much different from Vancouver, where cycling accounts for about 1 percent of total distance travelled. Notably, Metro Vancouver’s equivalent of the travel study done by the CRD includes such information, whereas the CRD does not. Share of total distance travelled by each mode of travel (Source: 2011 Metro Vancouver Regional Trip Diary Survey Analysis Report) Presenting basic information about the work done by components of transportation systems in this way might be discouraging to cyclists. However, when the primary consideration is reduction of emissions, “mode share” provides no useful information. As laid out in the CRD’s emissions reduction plan, the task will be to shift some fossil-fuelled auto use to a combination of transit, cycling and walking. Only by including the distance travelled, which reflects all the current realities about where people live, study, work and play and how far they have to travel each day to accomplish what they need to do, can we gauge how much energy needs to be shifted from autos to other modes. To put it as plainly as possible, a 34 percent reduction in emissions would require, after factoring in small increases in fuel efficiency and a small shift to electric vehicles, a shift of about 25 percent of the distance travelled in fossil-fuelled autos to non-fossil-fuelled modes over the next 12 years. I’ll elaborate on this later. AS MENTIONED ABOVE, my use of “total distance travelled” to compare the current energy contribution of different modes baffled cycling advocates. Former City of Victoria councillor John Luton, who has played a lead role in promoting cycling infrastructure projects in the region, wrote on Facebook, “Stories emerging from unreliable sources claim that CRD numbers show that only 1.5 percent of trips in the region are bicycle trips.” Luton went on to state, “Promoters of this theory are dishonest or unable to understand statistical information…The premise used to sell this fairy tale is that total mileage equals number of trips. That is false.…lying about the numbers is not a useful contribution to these discussions.” Edward Pullman, president of the board of directors of the Greater Victoria Cycling Coalition, responded to Luton: “Spot on John. By focussing exclusively on total distance travelled, folks that commute long distances become more important than those that live closer to their destinations. It’s a bizarre perversion of commuter choices.” Contacted by email, neither Luton nor Pullman could explain what their comments had meant. The story did not propose that “total mileage equals number of trips,” as Luton claimed. Former MLA and cycling advocate David Cubberley asserted: “There are no useful analytics involved in focussing on total distance travelled.” In a letter to Focus, Paul Rasmussen wrote, “Using the percentage of total miles travelled by mode… seems designed to minimize the positive impact of cycling.” The idea that our story was intentionally “designed to minimize the positive impact of cycling” occurred to other readers, as well. Transportation planning consultant Todd Litman wrote a lengthy response to our story in an online blog in which he claimed I had written that bicycle lanes were “wasteful” and “unfair to motorists.” On the basis of those claims—neither of which were made in our article, or intended—Litman continued on to assert what possessing such beliefs must indicate about the writer, including this zinger: “Critics like Broadland imply that cycling facilities only benefit a small number of serious cyclists—those who ride expensive racing bikes wearing lycra.” Nothing like that, though, was either stated or intended in our story. Luton, Pullman, Cubberley, Rasmussen and Litman are all in a position to influence the CRD’s plan for reducing emissions and the expenditure of many millions of dollars in public resources, yet none of them seemed able to understand what the CRD’s own numbers say about the magnitude of the energy shift that will be required to meet the federal target. Instead, they mounted a defense of cycling on the basis of other details we reported—or didn’t report—about the new Pandora Avenue protected bike lanes. Litman complained: “By extrapolating the Pandora bike lane cost to other Downtown arterials, Broadland estimates that Victoria’s cycling program will cost $16 million, which is almost certainly an exaggeration since the first project is always more costly than those that follow.” But the City’s record of underestimating and hiding project costs is a matter of public record. For example, when City councillors voted to replace the Johnson Street Bridge in 2009 they understood the project would cost $40 million. It’s now close to triple that. A more prudent reporter would have pushed the City’s bike lane estimate much higher. I simply extended the City’s actual cost per kilometre for the Pandora lanes—which was higher than the City’s budget estimate—to the full length of the protected corridor it plans to build. Merely reporting the likely cost of the planned Downtown protected network was, it seemed, enough to set the cycling advocates’ sense of fairness on fire. Rasmussen wrote, “Broadland criticizes the cost of the project—which he claims will be $16 million—over twice as much as the City says it will cost. In the eight years I’ve lived in Victoria, this is the first time that any entity has spent any significant amount of money on bike infrastructure. Meanwhile, just off the top of my head, I can count three significant projects for automobile traffic within the CRD in the last few years—the McTavish Interchange at $24 million, the Johnson Street Bridge project at $100 million and counting, and the McKenzie Interchange project at least $85 million. So that’s at least $210 million for car infrastructure just in major projects. Maybe even $16 million for something that promotes a clear social good isn’t so much?” Rasmussen could have included the $30-million Leigh Road Interchange (aka The Bridge to Nowhere) in Langford on that list, but let’s examine his claim a little more closely. The cost of the new McKenzie interchange, for example, includes the cost of space for cyclists, pedestrians and public transit. The new Johnson Street Bridge also includes space for those three non-car modes. In fact, 53.5 percent of the bridge’s available deck space is dedicated to pedestrians and cyclists. If the final cost of the bridge is $115 million—which it will be once hidden and as-yet undetermined costs for landscaping and additional protective fendering are included—should 53.5 percent of that cost be assigned to cycling and walking? That would be $62 million. Moreover, the public record of how this project unfolded shows that cycling advocates greatly overstated the extent to which the old bridges were being used by cyclists and their exaggerations helped to inflate the project into the public works nightmare it has become (See “Juking the stats,” Focus November 2011). Comparison of the space for autos (red) and cyclists and pedestrians (green) on the new Johnson Street Bridge (Source: PCL drawing) In Litman’s response to our story he wrote, “Cyclists just want a fair share of public resources (transportation funding and road space). What would be fair? You could argue that it should be about equal to cycling’s mode share: if 5 percent of trips are by cycling then it would be fair to invest 5 percent of public resources in cycling facilities. But this is backward looking since it reflects the travel patterns that occur under current conditions, ignoring ‘latent demand,’ the additional cycling trips that some travellers want to make but cannot due to inadequate facilities. To respond to these demands it would be fair to invest the portion of money and road space that reflects the mode share after those programs are completed; if comprehensive planning is likely to result in 10 percent cycling mode share, it would be fair to invest 10 percent of transportation funds and road space in cycling facilities.” Litman’s point isn’t particularly relevant to a discussion focussing on whether proposed bicycle and LRT infrastructure will effectively address emissions reduction, but it’s worth exploring. The record at the City of Victoria shows that transportation infrastructure decisions have been wonky, but not in the direction Litman claims. Again, consider the new Johnson Street Bridge. In the only reliable survey comparing trips across the bridge—published in a 2010 economic assessment used by the City to promote a new bridge—cycling and walking accounted for about 6 percent of mode share during periods of the year when those modes are at their peak. In the winter that share drops. Yet the new bridge will provide them with over 53 percent of the available deck space. So far there is no evidence to suggest mode share for cycling and walking will ever reach 53 percent, but they got it anyway. The City of Victoria Engineering Department's traffic counts on the Johnson Street Bridge used in a 2010 economic impact analysis to support a new bridge: Autos on left, buses centre, bicycles on right. Reading the various responses to our story, I got the strong impression that cyclists were not willing to consider the story’s core idea: Transportation infrastructure decisions need to more strongly reflect the urgent need to reduce transportation emissions, and we need better, more timely information on vehicle use in the CRD in order to gauge the effectiveness of the strategies that are being employed to reduce emissions. By “better” I mean more trustworthy information, the gathering of which is insulated from the influence of special interest groups like the Greater Victoria Cycling Coalition, engineering and project management corporations, or current and former politicians. In email exchanges with Litman and others, it emerged that, in their minds, Focus had written the wrong story. The cycling advocates were furious that our article focussed so narrowly on the issue of emissions reduction rather than fully explaining all the other benefits that more cycling infrastructure would bring, such as cleaner air, greater personal safety for cyclists and a reduction in vehicle congestion. Litman wrote, “Public investments should be evaluated based on total benefits and costs. My report, ‘Evaluating Active Transportation Benefits and Costs’ (vtpi.org/nmt-tdm.pdf ) provides a framework for doing just that: it identifies about a dozen categories of impacts (benefits and costs) that should be considered when evaluating walking and cycling policies and programs, including direct impacts on users, and indirect impacts on society. Your column only considered two benefits: increased user safety and climate change emission reductions. That is grossly incomplete and undervalues cycling improvements.” Our story, in fact, made no attempt to examine “increased user safety” beyond presenting Mayor Helps’ publicly stated position. Nor was it our purpose to present any of cycling’s other benefits. Our focus was on emissions reduction and getting better information. Litman encourages us to evaluate cycling infrastructure on the basis of total benefits and cost, but this would be an exceedingly speculative endeavour. Consider cost. The 2011 CRD Pedestrian and Cycling Master Plan—the only plan for building cycling infrastructure in the member municipalities of the CRD—estimated the cost of a region-wide bicycle network at $275 million. But that plan didn’t include any cycling improvements on Pandora Street. Yet it’s still the “Master Plan.” Indeed, the plan estimated costs of $3.3 million for 22.7 kilometres of “priority” bike lanes in the City of Victoria. But that’s a lower cost than the actual cost incurred for only 1.4 kilometres of protected bike lanes on Pandora (which wasn’t in the plan). And, optimistically, the plan estimates the cost of “all projects” (54.7 kilometres) in the City of Victoria at $12.4 million. Yet that won’t even cover the four legs of the 5.3-kilometre-long protected network in the Downtown core. The plan’s estimates for other municipalities seem even wilder, if that’s possible. For example, it put the cost of 26.5 kilometres of bikeway in View Royal at $36 million. Why would $36 million be spent way out in View Royal and only $12.4 million in Victoria? By the way, the consultant who wrote the CRD’s Master Plan lived in Oregon. Even if we did have a good grasp of the benefits an advanced cycling network might provide, the cost estimating that has been done so far is deeply flawed. So how can a useful cost-benefit analysis be conducted? Again, the CRD needs more trustworthy information gathered by a process that’s insulated from special interest groups. In any case, cyclist-centric claims about mode share, costs and fairness—and the backlash from other parts of the community those claims generate—are diversions for which we no longer have time. Shouldn’t the choice about how to transform our transportation system be simpler than that? Shouldn’t it be: Are we going to make a serious attempt to meet the federal emissions target or not? If we are, what do we need to do to accomplish that? Personally, I’m not interested in writing about all the benefits of a “sustainable” transportation system if that system won’t come anywhere close to meeting our 2030 emissions reduction target. So here’s the crux of the problem: The emissions reduction potential of an improved cycling network, if that’s all that’s executed, is limited. A paper published by Litman quoted results from “a detailed study of five US communities with active transport improvements” which found the improvements resulted in a reduction of “one to four percent of total automobile travel.” A “one to four percent” reduction would be the equivalent of rearranging the deck chairs as the ship is sinking. We need a 34 percent reduction in 12 years. Let’s shift back to what our regional transportation system would need to look like by 2030 so that we could meet that target. To get a clearer picture, let’s start in the Netherlands. The Netherlands has invested billions of dollars in public transit and infrastructure for bicycles and pedestrians. Is this a solution for Victoria? STATISTICS NETHERLANDS REPORTS that, in 2015, with 1.1 bicycle for each of its nearly 17 million inhabitants, that country had “the highest bicycle density in the world.” Featured prominently in its depiction of that country’s transportation system is a chart showing the percentage that each different mode contributed to transportation of people on land—bicycles, cars, buses, trains, walking, etcetera. Percentage of what? The percentage of the total distance travelled: Domestic distance travelled by transport mode in the Netherlands (Source: Statistics Netherlands) According to Statistics Netherlands, cars accounted for 73 percent of the total distance people travelled within their country. Public transit provides 12 percent, bicycles 7 percent and walking 3 percent. The City of Amsterdam, considered to have the greatest regional participation in cycling of any large European city, also publishes comparisons of the extent to which each transportation mode is used within that city, both by mode share and total distance travelled: Mode share (left) and share of total distance travelled (right) in the City of Amsterdam (Source: City of Amsterdam) The combined mode share for cycling and walking amounts to 54 percent (30 + 24). Yet when the total-distance-travelled lens is applied, together they account for 14 percent (12 + 2). The Dutch, rightfully proud of their extensive use of bicycles for transportation, have no problem being transparent about how much of the work of transporting people is done by each mode. Cars, at 54 percent, still account for the majority of the work done. (According to TomTom, an Amsterdam-based company that measures vehicle congestion all over the globe, Amsterdam’s traffic congestion is increasing; it’s already at a level higher than many American cities.) In the CRD, 88 percent of that work is being done by cars. The 34 percentage points of difference between Victoria’s and Amsterdam’s reliance on fossil-fuelled cars to transport people is, completely coincidentally, equal to the shift Victoria would need to make by 2030 to be on a path that would meet the federal mid-century goal. In other words, Victoria would need to become Little Amsterdam (Amsterdam has a metropolitan population of 1.6 million, Victoria’s is 368,000) within 12 years—the equivalent of a moonshot. Amsterdam’s achievements, it should be noted, include extensive bus, tram, metro and railway networks which provide the means to extend the length of a trip that a person starts and ends as a pedestrian or a cyclist. This achievement has taken many decades and many billions of dollars. For example, the city’s 73 kilometres of underground metro lines have a current value of $30-40 billion. Amsterdam’s highly developed public tramway, metro and railway system. Bus routes aren’t shown. Estimated cost? Unknown, but the 9.5-kilometre North-South Line (shown by the blue line), a new metro line currently under construction, will cost the equivalent of $4.6 billion CAD. What would Victoria need to do to knock 34 percent off its emissions tally? Let me take you through that exercise, but keep in mind that this is an arithmetical exercise performed only to provide you with a sense of the magnitude of the challenge we face. To do it we need to start with some basic assumptions. First, let’s assume 4 percent of fossil-fuelled auto travel in the CRD shifts to electric cars over the next 12 years (it’s currently less than 1 percent). That would take care of 4 percent of transportation emissions and our reduction requirement would fall to about 30 percent. If there’s a quick breakthrough in super-capacitor technology, which could replace the lithium ion batteries currently used in electric vehicles, this shift could eventually be much higher. But even such an unexpected breakthrough wouldn’t have a big impact over the next 12 years. Secondly, let’s assume there will be only minor emission reductions as a result of people using cars with higher fuel efficiency. In the USA earlier this year, Trump ordered a review of Obama’s regulations requiring much greater fuel efficiency by 2025. There’s broad expectation in the US that those standards will be rolled back, partly because car manufacturers have made the case that Obama’s regulations can’t be met without making cars unaffordable. Canada harmonizes with the US on such matters, so higher fuel efficiency seems like a long shot. Still, let’s include a conservative five percent reduction in car emissions due to fuel efficiency gains by 2030. Now we’re down to the need for a 25 percent reduction from taking other actions. Most people are aware of the need to reduce emissions and believe they already limit their travel to only what’s essential. That leaves government only one option: somehow persuading drivers to replace 25 percent of their current auto travel with a combination of public transit, bicycling or walking. How will we be persuaded? There would be no need for a carbon tax if people would voluntarily limit their auto use to the level governments told them was necessary. But we’re not like that, so implementation of a much higher carbon tax to start pushing the most cost-sensitive drivers out of their cars would have to occur soon. The Province’s account of BC’s emissions shows the current level of the carbon tax doesn’t appear to be having much bite, especially with gas prices as low as they are. So our last assumption is that much more serious fuel-cost persuasion will begin soon. With current total travel by autos in the CRD running at approximately five million kilometres each day, 25 percent of that—or 1.25 million kilometres per day—would need to be shifted from cars to buses, walking and cycling. However, in reducing the distance driven by autos by 25 percent, we would also likely displace 25 percent of the 1 million kilometres travelled in autos by passengers each day. So the shift to public transit, walking or bicycles would need to amount to about 1.5 million kilometres per day. Doubling the mode share of buses by 2030—the CRD’s stated goal—would cover about 500,000 kilometres of the required shift. The remaining 1 million kilometres of the shift would fall to walking and cycling. When added to their current levels, that would mean that cycling and walking would account for about 1.2 million kilometres each day, or about 18 percent of the total distance travelled—in just 12 years time. Now compare that with Amsterdam. Its combined total for bicycles and walking is 14 percent of the total distance travelled—a level that has taken several decades and billions of dollars invested in infrastructure for walking, bicycles, buses, subways, trams and commuter rail. Moreover, Amsterdam has packed 1.6 million people into an area about the same size as Victoria’s metropolitan area. That high population density, over four times Victoria’s, is essential for the financial viability of Amsterdam’s expansive, complex and costly public transit system. For the CRD’s vaguely-outlined plan to work, the distance travelled by cycling and walking would have to increase by about 600 percent (over levels in 2011) within 12 years. For a City with a steadily aging population and a so-so transit system, is this realistic? Has the CRD come up with the moonshot plan that will reduce the region’s transportation emissions by 34 percent within 12 years? So far, only minimal information has emerged into public view about how the region’s public transit system will evolve so its mode share doubles by 2030. What seems evident is that the rationale stated in the CRD’s Regional Transportation Plan for very expensive rapid transit is much more of a response to brief periods of traffic congestion—along the Trans Canada Highway out to Langford, and the Pat Bay Highway out to Sidney, during peak commuting periods—than it is a response to the need to cost-effectively reduce emissions throughout the day. The assumption that such congestion will continue on the Trans Canada, even after the new McKenzie Road interchange is complete, is founded on the debunked theory that most future growth in the region will occur in Langford. The 2016 census data shows that over the past 15 years—Langford’s glory years—the Core’s share of the metropolitan population has hardly changed, dropping from 68 percent to 65 percent. That strongly suggests the best place to focus future investment in public transit is where most of the people already live—in Victoria and Saanich. Instead, the CRD could be the first government in history to plan for an LRT to Nowhere. After the next 12 years, of course, the same rate of shift from autos to public transit, cycling and walking would have to continue—right through to 2050. Keep in mind, too, that transportation emissions in Canada amount to about 24 percent of total emissions, so to be on the most gradually descending path to 2050, all the other sectors would need to be reducing their emissions as well. That will impact all of our lives in ways that, at this point, we haven’t yet imagined. But unless we do it—according to the world’s best scientific minds—we’re cooked. Is Victoria’s political culture up to the task of getting us through this daunting challenge? The short answer may lie in the record of the attempt to build a new Johnson Street Bridge. An even more chilling possibility is hinted at by the misplaced effort to convert Victoria’s safe, source-controlled, low-cost, tidal-powered marine-based sewage treatment system to a land-based system that will cost Victorians billions of dollars over the life of the infrastructure that’s being built. According to DFO scientists, land-based sewage treatment will have negligible effect on environmental conditions in the Strait of Juan de Fuca. The existing marine-based system was endorsed by an overwhelming number of Victoria’s marine scientists and current and former public health officials. One of the DFO scientists I spoke with during those deliberations was Sophie Johannessen, the lead author of the peer-reviewed study that found land-based treatment would have a negligible environmental effect on environmental conditions in the Strait. I asked Johannessen if there was anything the community could do that would have a more positive effect on marine ecosystems than moving Victoria’s marine-based sewage treatment system onto land. “I think so, yes,” Johannessen said. “We could reduce our greenhouse gas emissions, enact source control for persistent contaminants, and reduce other local pressures on the marine biota.” The local political culture didn’t listen to the scientists. Instead it followed Mr Floatie to Seattle and started the never-ending process of flushing billions of dollars down our toilets. On atmospheric emissions, the scientists have spoken loudly and clearly: there’s a pressing need to act. In response, will our politicians be led by special interest groups? Or will their decisions be based on science and evidence? David Broadland is the publisher of Focus Magazine.
  13. Local government’s response to reducing transportation emissions may be wishful thinking. Or foolish. IS THE CITY OF VICTORIA’S STRATEGY to create protected bike lanes in the Downtown core a well-thought-out strategy to make bicycling safer, relieve vehicle congestion and move Victoria in the direction of a low-carbon future? Or is it another case—like the Johnson Street Bridge Replacement Project—of the City unintentionally displaying its proven tendency toward decision-based evidence-making? The first component of the strategy—a $3.5-million, 1.2-kilometre-long corridor on Pandora between Cook and Store—became operational in May. By mid-June the City’s PR team announced “the number of cyclists using the new bike lanes is very encouraging” with “nearly 40,000 bicycle trips” made along the corridor in its first month of operation. That timeframe, and the numbers, included Victoria’s Bike to Work Week, an annual outpouring of temporary enthusiasm. A second protected bicycle corridor—1.2 kilometres of Fort from Cook to Wharf—was approved by City of Victoria councillors on June 8. Construction is scheduled to begin in September. The City plans to expand these corridors to Wharf, Humboldt and Cook. At the cost per kilometre of the Pandora corridor, the 5.3-kilometre-long Phase 1 would cost about $16 million—and depends almost entirely on the availability of grants from the Gas Tax Fund. The rationale behind the protected lanes—as opposed to cyclists sharing the existing infrastructure with automobiles—is to increase the safety of cyclists. But creating protected lanes has resulted in removal of auto parking space, already in short supply in the Downtown core much of each day. The Pandora corridor removed 43 auto parking stalls; another 30 will be removed on Fort Street. At that rate of parking space removal, Phase 1 would see about 175 spaces disappear. Before construction of the protected corridors began, the City had less than 2000 on-street parking spaces Downtown. So Phase 1, originally planned to be complete by the end of 2018, will see the loss of nearly 10 percent of on-street parking in the Downtown core. The City’s aim appears to be to quickly replace a significant fraction of motorized individual transport with unmotorized individual transport. For people who drive a car, truck or van Downtown and don’t see themselves as likely to ever switch to a bicycle, the new situation feels like an attempt to force them to make a change they can’t or don’t want to make, and carries a whiff of social engineering. Some Downtown businesses have expressed concern that making vehicle parking Downtown less available will discourage potential clients and impact their businesses. But Victoria Mayor Lisa Helps has argued that protected bicycle corridors will make auto parking more available, not less. Her theory is that by making biking around Downtown safer, people who in the past would only travel there by auto will now be encouraged to come by bicycle. Victoria’s 40-ish mayor is an avid cycle commuter and she now has a protected corridor that runs most of the 1.4 kilometres from her home in Fernwood to her place of work at City Hall. Implicit in the City’s decision to proceed along this course is the belief that the number of cyclists, especially those commuting to work, needs to be encouraged and allowed to grow far beyond current levels. Why are they doing that? Here’s the City’s official position on “why”: “Encouraging cycling, along with walking and transit use, is an important strategy to manage expected population growth and support community health, affordability, economic development, air quality and climate action objectives. As the City grows in population, we will need to shift some of our trips to transit, cycling and walking because these are much more efficient modes of transportation than single occupancy vehicles.” The City supports its position with data that it attributes to the 2011 Census that indicated 10.6 percent of people living within the City of Victoria cycle to work. That’s Canada’s highest per capita incidence of commuting by bicycle. It’s hard to argue with federal census data that counts (almost) every single person in the country and has a margin of error close to zero. The City is hoping to build on that encouraging number and calls its plan “Biketoria.” While the City’s vision sounds progressive and smart, the best available data about transportation in Victoria calls into question the City’s emphasis on cycling and walking—and perhaps transit, too. Let’s start with Victoria’s claim to fame, that 10.6 percent of Victorians who cycle to work. It turns out that number wasn’t obtained directly from the 2011 Census. Instead, the “10.6 percent” figure comes from the 2011 National Household Survey, which was voluntary and produced data with a margin of error much higher than zero. Since good transportation planning requires good transportation data, it’s important to understand why one of the fundamental numbers supporting the City’s Biketoria initiative is probably flawed. The National Household Survey asked participants only one question about transportation: How did the person filling out the survey “usually get to work.” There were 11 modes of transportation listed (auto driver, auto passenger, transit, walking, bicycle, etc...) and the respondent could choose only one. How did multi-modal commuters decide how to respond? We don’t know, but it’s well-known that commuter cycling ebbs in the darker, colder, wetter half of the year, so it’s reasonable to assume that some cyclists must be using other forms of transportation to get to work at least part of the year: walking, transit, some might even drive an auto. But the National Household Survey didn’t allow for such complexity. Nor did it attempt to gauge the distance people had to travel to work. As a guide for transportation planners, then, the National Housing Survey doesn’t really qualify as a reliable tool for making multi-million-dollar transportation decisions. Yet it is attributed as one of the primary sources upon which the City based its case for building protected bicycle corridors. The other source the City cites is the 2011 CRD Origin-Destination Household Travel Survey. But a careful read of the data in that survey, especially when compared with the data the survey produced in 2006, raises questions about the City’s direction. According to the CRD’s 2011 survey, only 3.8 percent of trips within, into and out of the City of Victoria over a 24-hour period were made by bicycle. When the average distance of trips made by different modes of transportation are factored in, bicycles accounted for less than 2.5 percent of the total distance travelled using all modes. Moreover, the Origin-Destination survey didn’t capture trips that were made to move goods or to provide services—like taxi drivers, social workers, delivery services, healthcare providers, transit drivers—it’s a long list and almost none of it is done by bicycle. If bicycles currently account for only a tiny fraction of the total distance travelled each day in the City of Victoria, how realistic is it that large numbers of Victorians will soon become cyclists? While Copenhagen’s large contingent of cyclists is held up as a model for Victoria to aspire to, the average age of a person living in Copenhagen is 35.9 years and has been falling for many years. In the City of Victoria, the average is 44.5 and is projected to rise for many years. As people get older, they generally spend a lot less time on bicycles, especially in hilly places like Victoria. Perhaps that’s one reason why the Origin-Destination surveys for 2006 and 2011 show that, for the whole CRD, the daily mode share for bicycles dropped slightly over those five years, from 3.2 percent to 2.8 percent. Yet the official goal in the CRD is to raise that to a regional level of 15 percent by 2038. Is this realistic? The answer to that becomes clearer when we consider the cumulative distance travelled each day by residents of the Capital Regional District (see table below). According to data in the 2011 Origin-Destination survey, about 6.6 million kilometres are travelled within the CRD each and every weekday (note that’s each day, not week). Of that travel, 72.7 percent was as the driver of an auto and 15.6 percent as a passenger in an auto. That means that about 88 percent of all travel within the CRD relies on autos. Share, by transportation mode, of total distance travelled within the CRD on a weekday Source: 2011 CRD Origin-Destination Household Travel Survey, conducted by Malatest and Associates Ltd. The survey did not capture commercial traffic or traffic originating outside of the CRD. The 2011 survey is the most recent data available. Only 1.5 percent of the distance travelled is by bicycle. As noted above, the survey does not capture commercial trips made to move goods or to provide services. If commercial traffic was included, bicycles would likely drop to little more than one percent. So bicycles currently account for a tiny fraction of the actual distance people cover in getting from point A to point B in the CRD. Again, is it realistic to think that bicycles—currently providing about 1 percent of the work being done by our regional transportation system—will supply 15 to 20 times as much work in 20 years? Without improving the cycling network, City and CRD transportation planners won't know whether their long-term goal is achievable. Unfortunately, though, as they experiment, the protected lanes may unintentionally increase emissions by delaying vehicles making right-hand turns off Pandora, resulting in hours of additional engine idling each day (see the short video below). Unless use of the corridor increases dramatically, it could be argued it's doing more harm than good most of the day. The Pandora Avenue protected bicycle corridor includes new traffic signals that delay right-hand turns off Pandora by 25 seconds at each of six intersections. As is shown in this video, this will increase emissions from autos even though there are few cyclists using the lane. You might be wondering why I am quoting a study done in 2011. The Origin-Destination surveys are done every five years, but the 2016 survey has been delayed. John Hicks, senior transportation planner at the CRD, told Focus the 2016 version, which would normally have been released about now, was pushed back a year so that 2016 federal census data could be used more directly in determining required sample sizes. A call for credentials was issued by the CRD in April and the survey will be conducted during the same months as the 2011 survey. It should be released in March 2018. So, for now, we are dependent on the 2011 data, and that shows bicycles only provide a tiny fraction of the travel needs of people throughout the CRD. Regardless of whether the loss of nearly 10 percent of the Downtown core’s on-street parking is or isn’t a reasonable trade-off for a greater level of safety for bicyclists, the claim that these corridors will play a significant role in reducing carbon emissions seems like a refusal to accept the obvious: Most people prefer to use four-wheeled motorized personal transport. So at least some of the CRD’s and the municipalities’ efforts in transportation planning ought to include how that strong preference can be incorporated in a transportation system that evolves toward a low-carbon future. For example, why not incorporate charging stations for electric cars into the protected bicycle corridor infrastructure? Providing electricity for free would create an incentive for electric vehicles Downtown. Even so, such ideas would be little more than civic acknowledgment of the need to reduce emissions since the vast majority of motorized vehicles depend on fossil fuels and likely will for many years to come, according to auto industry experts. To produce a significant reduction in CRD transportation emissions, a more sophisticated approach than painting bicycle lanes on roads will be needed. That will, at least to begin with, require helping auto drivers and auto passengers reduce their use of fossil-fueled vehicles, while accepting their choice for how to get around. How can that use be downsized? The data the CRD has collected, if it’s accurate, contains some interesting possibilities. Comparing the 2006 and 2011 surveys, it appears two shifts in the use of autos were underway. One was good news, the other bad. First the bad news. According to the CRD’s Origin-Destination surveys, between 2006 and 2011, about 78,000 fewer trips were taken as auto passengers each day. Where did the passengers go? It appears that many of them might have become drivers. In 2006, drivers accounted for 59 percent of non-commercial trips. But by 2011 that had climbed to 64 percent. If nothing else had changed, this would have meant more vehicles travelling each day—and higher emissions. But—and this is the good news—regional transportation emissions per person may have declined in spite of the trend of passengers becoming drivers. That’s because the average number of daily trips per person in the CRD decreased after 2006 by 4.8 percent. That translates to CRD residents driving about 43,000 fewer kilometres each weekday than in 2006. These two factors—the incidence of single-occupancy vehicles and the average number of daily trips taken by CRD auto users—suggest possibilities for emissions reduction that don’t involve converting car drivers to cyclists. (Again, this is only true to the extent that the data collected for the Origin-Destination surveys for 2006 and 2011 is accurate.) The CRD needs to gather more data that provides decision-makers, elected officials and media answers to basic questions, such as: Why did CRD residents reduce the number of their trips between 2006 and 2011? Is there some way to incentivize that shift in behaviour? If the City of Victoria can spend $16 million on safer bicycling for a few thousand bicyclists—using money collected from auto drivers through the Gas Tax Fund—why can’t the CRD refund a few million a year back to auto drivers who can prove a significant reduction in the miles they travel each year or switched to an EV? And what were the factors that turned auto passengers into auto drivers? What would it take to reverse that trend? Can local government, especially the CRD, play a role in facilitating that reversal? With the huge growth in the use of cell phones, ipads and apps, why does the CRD not have its own high-profile regional rideshare system in place that can connect car drivers who are about to make a similar trip? The absence of timely, deep, reliable data on transportation in the CRD will make it difficult for the community to make sensible decisions based on evidence. One example of how badly politicians can steer the public interest—when they make a decision and then look for evidence that supports it—is the CRD’s controversial LRT initiative. That began in 2009 as the provincial NDP’s response to the then-Campbell government’s transit initiatives in Vancouver. Local NDP MLA Maureen Karaganis stated back then: “The Campbell government’s transit plan focuses almost entirely on projects in the lower mainland while the rest of BC, including Victoria, has been ignored. The Capital Region seeks to avoid sprawl by building an innovative, high quality public transit system with LRT between Downtown and the western communities.” By 2012, that we-want-one-too logic had ballooned into a live, billion-dollar proposal to build an LRT between “Downtown and the western communities.” Note that the western terminus of such a system wouldn’t have been the “western communities,” but rather Langford. When politicians start pounding the drum for some large infrastructure project, which they hope will distinguish themselves from their political competitors, the only thing that might prevent them from making a big, expensive mistake is credible, accurate, up-to-date information. With the LRT proposal, if a billion was going to be spent anywhere, should it really be used to connect Downtown with Langford? The 2011 Origin-Destination survey included a graphic of the “Desire Lines” in the CRD (see graphic below). These represent the most heavily-travelled routes people take in moving around the CRD each day. In the illustration, you can see that, by far, the strongest flows are between Downtown and south Victoria, from Downtown to Uptown, and from all three of those areas out to the University of Victoria. Notice the feeble desire line out to Langford. Desire Lines in the CRD, from the 2011 Origin Destination survey. The illustration excludes trips that originate outside of the CRD. For example, trips from anywhere north of Langford, which contribute much of the traffic on the at-times congested Trans Canada Highway. The LRT the NDP was proposing would likely not be used by such travellers. The most prominent desire lines show where an LRT should be located if the goal was to reduce emissions and create a more compact community. A 19-kilometre-long loop that connected Downtown, Oak Bay, the Shelbourne Valley, UVic and Uptown would follow arterial roads that already pass within a kilometre or so of tens of thousands of existing homes. Over time, the presence of a transit line would encourage even greater population density in those already-developed areas. According to the desire lines, a 15-kilometre (one way) route from Downtown to Langford wouldn’t make sense. Yet the Regional Transportation Plan’s rationale for LRT sees that route “as a possible means to significantly curb pressure on auto infrastructure in high growth areas.” By “high growth area” the CRD means Langford, which has the highest relative population growth rate in the CRD. But in terms of absolute growth—which includes growth in population, commercial and institutional development and employment—the area of Victoria and Saanich already heavily criss-crossed by desire lines is experiencing more than twice the growth of Langford. With the NDP about to become government, will an LRT to Langford be resurrected? Quick! Using Google Earth, someone needs to count all the dwellings and places of employment within walking distance of the above two routes. Why isn’t that information already available? Choosing the wrong first route for LRT would have a devastating effect on the long-term prospects of reducing emissions in the CRD. What the CRD really needs, before spending countless millions on pet transportation projects that address a tiny fraction of the CRD’s emissions problem, is credible and comprehensive information about what it would take to get people who live here (as opposed to Danish twenty-year-olds) to change their travel behaviour. Obtaining that information would cost money, of course. One way to fund such data gathering would be to use the Gas Tax Fund. Unfortunately, that huge chunk of cash—which is taken from drivers of vehicles that run on fossil fuels—is used almost exclusively for cycling infrastructure or non-transportation-related projects in the CRD: an agricultural strategy here, a tennis court there, water system upgrades all over the place, even a fire hydrant in Shirley. The tax isn’t being used, however, for any initiative that might one day seriously lower carbon emissions. Perhaps that’s because actually reducing the use of fossil-fuelled vehicles would diminish the flow of money to the Gas Tax Fund, and that, in turn, would start to dry up funding for local politicians’ pet projects. It’s an interesting dynamic, one addiction feeding another. How do we get free of it? Please let me know what you would do. David Broadland is the publisher of Focus Magazine. An earlier version of this story implied that the CRD's 2038 goal for cycling's regional mode share was 25 percent. Its actual goal is 15 percent. The goal is for 25 percent in "urban areas." 2012 CRD bike counts.pdf
  14. To not be misled by experts into making bad decisions, elected officials need to ask hard questions. Voters need to elect prosecutors, not patsies. IN MID-JUNE, City of Victoria councillors were given another update on the Johnson Street Bridge Replacement Project by Project Director Jonathan Huggett. They learned the project was experiencing new technical problems and were given yet another date at which completion might occur. Yes, that sounds just like the last update, the one before that and the one before that. But the real purpose of Huggett’s report was to show the City was including “lessons learned” from the bridge project into its next foray into fiasco avoidance. He delivered to councillors a report on fumbles made during that project which, hopefully, could be avoided if council’s decision to replace the Crystal Pool at a cost of $70 million goes forward. I’m going to focus on a single lesson learned that Huggett unintentionally highlighted: the tendency of some current councillors, in their meetings with costly experts like Huggett, to ask questions or make observations that have little or no meaning or value, thus inviting meaningless advice from those experts—like Huggett. Why do Victoria councillors ask so many dumb questions of experts? Who really knows? But the impact of not being up to sniffing out inaccuracy, underestimation, and overselling inevitably leads to millions of extra dollars of cost—and some hidden liabilities that one day may thrust themselves into the faces of local taxpayers. Aftermath of the 1995 Kobe earthquake. At a critical moment in this project, did elected officials ask dumb questions instead of grilling engineers? Following Huggett’s report, in which he laid the responsibility for everything that had gone wrong with the project on unnamed persons who were no longer sitting in the room, councillors spoke. Councillor Pam Madoff, who has been sitting in the room from the very beginning, expressed her view that she had done her due diligence on the project and, repeatedly, asked “how far” councillors had to go in their public oversight role in such projects. Here’s an excerpt of what she said: “I remember very specifically having this conversation [with the bridge’s designers and engineers] about the mechanics, you know, the—in simplistic terms—the cogs, the wheels, how it was going to lift. I remember at the time saying, ‘Is this basically just a larger version of the Meccano sets that we played with as kids, in terms of its actual mechanical operation?’ And, again, that was the assurance. To me it comes down to: how far does one have to go? We felt like we asked the right questions at the time. It turns out they may not have been the right answers.” If an elected official asks a professional engineer if a bridge is like a Meccano set, what should happen? Does the engineer say something along the lines of: “Well, you know, Councillor Madoff, a bridge is nothing like the Meccano sets you played with as a child. I’m worried about your question because it seems to indicate you have little understanding about what we’re considering doing here. I think we ought to postpone the rest of this meeting while you—and perhaps some of the other councillors—go back to your rooms and think harder about what your responsibilities are in this situation, and what questions you should be asking. People’s lives, scant public resources, the City’s reputation—all these could be put in serious jeopardy if you fail to provide adequate oversight of the public interest on this project. Is a bridge like a Meccano set? Well that’s the dumbest question I’ve ever been asked.” Huggett, of course, didn’t respond so impolitely. Instead, he said: “There is no question that you were not given good advice.” In much the same manner that previous experts had politely given City councillors bad advice, Huggett had just done the same thing. When you are collecting $20,000 a month for writing report after report about why the new bridge is still stuck in China—as Huggett has been doing since 2015—you obviously don’t bite the hands of those who are nervously signing your cheques. But Madoff’s unanswerable question about whether this bridge was like a Meccano set allowed Huggett to pile bad advice on top of older bad advice and yet appear to be sympathetic and sage. Later, Councillor Ben Isitt—who I hasten to point out does not ask dumb questions—provided the meeting with his own list of lessons learned. One of those was about the “lack of public oversight by council.” He recalled an attempt by the City “to quash investigative journalism into this project.” Isitt reminded his fellow councillors that “in the autumn of 2012, just as we got reports that there were three bidders, only one of which was within the City’s affordability ceiling, the City attempted to quash attempts by Focus Magazine to file FOIs on this project.” (The City was supported in that effort by current councillors Madoff, Charlayne Thornton-Joe, Chris Coleman and Marianne Alto.) Isitt told the meeting the incident showed “the need for public oversight of council and the City administration.” Councillor Jeremy Lovejoy picked up on Isitt’s point and asked Huggett, “Do you see lessons learned in terms of public communications on the project?” Perhaps feeling uncomfortable by Isitt’s promotion of the virtues of Focus’ reporting on the project (Huggett has refused to answer any questions from this magazine over a 3-year period), Huggett responded to Loveday’s vague question by referring to “a long, semi-public debate about seismic design. It becomes extremely difficult to have a meaningful debate in the public arena about something as complex as the seismic design of a bascule bridge. I understand that it is legitimate…freedom of the press to raise these issues. But it’s a very difficult issue to debate. It’s highly specialized and, you know, I don’t think you would appreciate it if I told you I’d spent last week debating with a journalist about seismic design.” Huggett didn’t actually answer Loveday’s question. The question’s vagueness allowed Huggett to provide, instead, excuses for why he never provided City council and the public with an explanation of why there is a rider in the bridge’s construction contract that will limit the liability of those involved in the project if the bridge fails to perform to expectations. To explain that rider would have been “extremely difficult,” “complex,” and would have taken a week of “debating with a journalist,” which, at the rate he is billing City taxpayers, would have meant a cost of $5000. This could have been Loveday’s moment to shine a light on a critical issue of public safety and the City’s legal liability. Focus first raised questions about the contract rider back in March 2015 after obtaining the document—Johnson Street Bridge Seismic Design Criteria—through an FOI. The document was secretly introduced into the project’s procurement proceedings in August 2012. The document stipulated much lower levels of expected seismic performance than had been recommended to the City by engineers. The City had earlier agreed to pay an extra $10 million for the recommended level of protection. Attached to the construction contract, the document would provide proof that the City had accepted this low level of seismic performance if the situation ever arose that the bridge was badly damaged, or people were killed on it, as the result of a large earthquake. The rider put all the risk back on the City’s plate. In case there’s anything too “complex” about this issue for the reader, let me make a comparison between seismic performance and paint performance. Suppose the City had been advised to paint the new bridge with a type of paint that would last for twenty years, and suppose the City had agreed to pay the extra cost for paint that would last 20 years instead of, say, five years. Now suppose that after that agreement had been signed, Focus found out that there was a secret rider in the City’s construction contract that stated that the City accepted that the paint might only last five years, even though it was still paying for paint that would last 20 years. Is that issue too “complex” for you to understand? Of course, poorly performing paint wouldn’t jeopardize public safety, and a new paint job is a relatively straight forward procedure. With seismic performance, though, it’s baked into the bridge and can’t be changed without very expensive material changes to the structure. With seismic performance, people’s lives and huge financial liabilities to City ratepayers are at stake. In the last two years, geoscientists have confirmed that two active faults, capable of producing an earthquake more energetic than that which toppled the elevated highway in Kobe, Japan in 1995, lie beneath Victoria. The Devil’s Mountain Fault, which surfaces a few kilometres south of Downtown, actually dips below the City itself. Victoria has the highest seismic hazard of any Canadian city. This is a real issue that, if Victoria’s elected officials ignore or ask dumb questions about, will have a profound impact on whether people live or die when a big earthquake here finally happens. So, before I get back to Councillor Loveday’s response to Huggett’s answer, let me pose a question to you, the reader. Would it be your expectation that City councillors should ask for an explanation of why that rider was inserted into the construction contract? If your answer is “Yes,” you’re thinking like Councillor Isitt. When Focus raised this issue back in 2015, Isitt point-blank asked Huggett to come back to council with an explanation of the document. Over the course of the following two months, Huggett did everything but what Isitt had requested: Provide an explanation of what the document was. In multiple reports to City of Victoria council and a media briefing, he provided an elaborate non-denial denial. Except for one brief reference to the document as a “memo,” his reports didn’t acknowledge its existence, let alone provide any explanation for how it might impact future liability for the City. After it became clear that Huggett was scrupulously avoiding providing councillors with an explanation of the contract rider, Focus filed an FOI for his communications with the other parties involved in designing and building the bridge. The documents released to us by the City showed two things: First, that Huggett never asked the creator of the rider, MMM Group, for an explanation of why the document had been created. At least, not in writing. Secondly, we found that none of the communications between the engineers denied that any of the provisions of the rider were, in fact, in play. Indeed, the former City employee who signed off on the rider document, Dwayne Kalynchuk, confirmed in writing to Focus that the provisions in the rider had been incorporated into the bridge’s design and construction. It should have been obvious to all of the councillors that Huggett had never provided them with an explanation of what the rider was. So when Huggett went back to the issue in his response to Loveday and told him the issue was too complex, would require debating in public with a journalist, and would use up a week of his expensive still-delayed-in-China-reports time, Loveday had his big moment to act in the public interest. To perform that service, he would have had to respond to Huggett something like this: “Hey, wait a minute, you never provided us with an explanation of what that contract rider is. Why was it created and how does it impact the City’s legal position in the case of a catastrophic event in which people are killed on the bridge or it can’t be used by emergency vehicles?” Instead, all Loveday could muster was, “I agree that’s not how we’d like you to spend your time.” David Broadland is the publisher of Focus Magazine. He’s been filing regular reports on the bridge project for eight years.
  15. The project seems to be a complete fiasco. But is that just a perception created by something in the air? IN A REPORT HE DELIVERED to Victoria City council in late March, Johnson Street Bridge Project Director Jonathan Huggett did a 180-degree flip-flop on one of the project’s costly screw-ups. Before I tell you about that, though, I have to provide the reader with a caveat-emptor kind of warning about my story. The fact is, I may be suffering from a mind-altering overdose of carbon dioxide. I don’t think I’m making this up, but I might be. I came to realize this was a real possibility after coming across a 2015 study by research scientists at Harvard, State University of New York, and Syracuse University. I was earnestly googling away for what might be in the air that could possibly explain the widespread mental confusion we’re seeing south of the border these days. Is it something in the water? No, it’s in the air. These scientists reported that human cognitive abilities are significantly and adversely affected by the concentration of carbon dioxide that we are now regularly exposed to inside many buildings. Their work confirmed two previous but smaller studies that had come to much the same conclusion. The cognitive functions most severely impacted, the research found, were the ability to use information and the ability to strategize. So I need to warn you: I wrote this story while sitting inside a building. Moreover, my subject—Mr Huggett’s flip-flopping report—was presumably also written while the author was inside a building. Even worse, because of the likelihood of elevated levels of carbon dioxide wherever you are right now, your ability to process my potentially confused reporting of a potentially confused report could be compromised. By the end of this story, you may be completely dazed and confused. Before venturing into that minefield, consider this: The only real solution to adverse levels of indoor carbon dioxide is thorough ventilation with fresh, outdoor air. But, as the level of carbon dioxide outdoors continues to increase as a result of carbon emissions from human activity, ventilation will increasingly fail to make any difference. How bad could this get? The worst-case scenario is that global concentration of atmospheric carbon dioxide will one day reach the levels that significantly affect human cognition. Confusion begins around 800 to 900 parts per million. Currently, outdoor levels are about halfway there and rising. Donald Trump seems eager to get all of us fully there, but being even halfway seems to allow for craziness enough. So reader beware, and let’s look—for the billionth time—at the Johnson Street Bridge project, whose nine-year history so far provides plenty of circumstantial evidence that carbon dioxide levels during City council meetings in Victoria need to be carefully investigated. JONATHAN HUGGETT, it turns out, is the most highly-paid official currently working for the City of Victoria. At $20,000 per month, he’s making more than even City Manager Jason Johnson, who hired him. Including expenses and taxes, Huggett is billing Victoria taxpayers approximately $295,000 per year. Not bad for a guy who lives in Surrey, only needs to report to City council four times a year, and isn’t required to answer questions from reporters. Since he’s so highly paid—by taxpayers—and since some of his claims about the project have seemed to be at odds with the public interest, Huggett’s reports beg for a detailed examination by local media. He has told Focus he’s too busy to answer our emailed questions, although he has made frequent appearances on local talk-radio programs. As Huggett’s open-ended contract with the City notes, the City also has a highly-paid “designer and project manager,” MMM Group. Since 2009 the City has paid MMM about $16 million for its services. As the “owner’s representative,” MMM, supposedly, would insure the City’s interests were given top priority by the company building the bridge, PCL Constructors Westcoast Inc. So why does the City need Huggett? Can’t MMM be trusted to do its job? According to his contract, Huggett was brought in by City Manager Johnson in April 2014, to “undertake an independent review of the Project, including assessment of the relationship between the City, MMM and PCL, to evaluate the current status of the project and potential risks to its successful completion.” But after undertaking that review and providing a report in July 2014, Huggett was appointed “Project Director.” He has spent the time since then providing quarterly reports formerly written by City employees in collaboration with MMM. The breakdown in trust between the City and its project manager became public in 2014 when both PCL and MMM began to present the City with claims for additional costs even though the City had been assured that project costs had been capped by a “fixed price” contract. Huggett’s first report to City councillor’s assured them that the City didn’t have a fixed-price contract. For some reason, councillors liked what they heard and Huggett’s monthly cost then escalated. With an extended period of legal battles likely to follow physical completion of the bridge, Huggett can expect to receive a monthly cheque from the City at least through 2018. If that’s the case, his own work on the project will add roughly $1.3 million to the cost of the new bridge. It’s unclear whether that amount has been fully included in any of the quarterly updates Huggett has delivered to City councillors. It should also be noted that Huggett does not track the bridge’s costs. That’s done by the City’s finance department. As well, the City is represented by an outside law firm—as well as its own highly-paid legal staff—on legal issues related to the project. Huggett has stated publicly a number of times that his job is to make sure the project gets completed. But Huggett has sometimes presented opinions to City council and the public that haven’t been based on facts. His use of—let’s call them alternative facts—have had the effect of protecting the reputations of professional engineers who have screwed up on this project rather than protecting the public interest. One good example of Huggett’s use of alternative facts was his response to a story Focus published about how the level of seismic protection stipulated for the bridge—the seismic design criteria—was secretly downgraded from the level that MMM had recommended. The essential facts of that story are these: MMM recommended to the City in 2010 that the new bridge be able to withstand a magnitude 8.5 earthquake and the City agreed to pay an additional $10 million for that recommended higher level of protection. However, after initial estimates from the construction companies bidding to build the bridge were received in 2012, project engineers realized that the bridge would cost much more than they had hoped. At least one of the companies also expressed concerns about the unusual design’s inherent seismic risk. For whatever reason—whether it was to reduce costs in an attempt to save a failing project or because the engineers realized the peculiar design could not withstand a magnitude 8.5 earthquake without irreparable damage—the project’s target seismic protection level was lowered. The decision to build the bridge to a lower seismic standard was made in secret—that is, without City council’s knowledge—and that broke the agreement City managers had made to seek elected officials’ consent to change the project’s scope. More importantly, the downgrading of the seismic design criteria meant the bridge could be more easily damaged by an earthquake. It also made it more likely the bridge would be unrepairable following a smaller earthquake. When Focus published a story pointing this out, Huggett’s response was to obscure what had occurred. His explanations never acknowledged the existence of the Johnson Street Seismic Design Criteria document which proved the change had been made. This document was an integral part of the construction contract the City signed with PCL. Instead, Huggett provided City councillors with a report in which a critical paragraph of the building code governing construction of bridges had been altered so that it appeared that the lower standard to which the bridge had been built was in accord with the requirements of the (altered) code. This was a truly remarkable sleight of hand, and I have wondered whether carbon dioxide might have been involved. What else could explain Victoria City council’s utter lack of ambition to look more closely at the issue? The City was in a position to demand that MMM return $10 million of its $16 million payment for its failure to provide a bridge with the level of seismic performance it had recommended. And what explains Huggett’s course of action? Instead of pursuing MMM, he misquoted the bridge code. A partially-redacted email (it was obtained by FOI) from an MMM employee to Huggett following the creative rewrite of the seismic code, expressed MMM’s relief “since the seismic issues appear to be contained for the time being.” Huggett never publicly admitted that such “issues” even existed, but it’s apparent that MMM expected the issue might resurface. So now we come to Huggett’s 180-degree flip-flop. (Also see the slideshow: Seismic rip-off on the Johnson Street Bridge) I RECENTLY REPORTED WHAT HUGGETT has said about the issue of fendering on the north side of the bridge. Fendering is the protective barrier placed around the support piers of a bridge to minimize the damage that could be done if a ship or barge accidentally hit the piers. Huggett told councillors in July 2015 that more extensive fendering was needed on the north side of the bridge than had initially been planned because, as it turned out, “The new bridge is somewhat less robust than the existing structure.” In explaining why this would add significantly more cost to the project than had been stipulated in the so-called “fixed-price” contract, Huggett told councillors that the north-side fendering had been “clouded-out” in a contract drawing. That indicated, he said, “It is not in the original contract.” But a review of the “fixed-price” contract by Focus strongly suggested that the cost of the fendering had been included, even if the final design of the north side fendering had not been fully worked out. In response to an FOI request from Focus, the City said it could not find the “clouded-out” contract drawing that Huggett had referred to, further eroding the credibility of his claim that the contract did not include the north side fendering. In spite of these facts, Huggett continued to maintain that the additional cost of the north side fendering could be substantial and would have to be borne by City taxpayers. The cost has been rumoured to be as high as $10 million. A rendering of fendering on the north side of the new Johnson Street Bridge from Jonathan Huggett’s March 2017 quarterly report to Victoria City Council, in which it was described as “one option.” One Victoria engineer estimated the installation could add $10 million to the cost of the project. Who was Huggett representing by taking this position? He is being paid $20,000 each month by Victoria taxpayers. Shouldn’t his positions reflect that? Let me boil this down to two points. First, why would Victoria be getting a bridge that was “less robust” than the existing bridge? Questions raised about the ability of the existing bridge to withstand the forces exerted on it by even a minor earthquake was the very rationale used for building a new bridge. Yet, according to Huggett, the new bridge would be less robust than the old bridge. Rather than openly accepting this apparent project failure, shouldn’t Huggett have been advocating for a better outcome? Secondly, why didn’t Huggett take the position that the cost of all fendering was in the PCL contract? In his report to City council in March, Huggett reversed his position and admitted that PCL’s fixed-price contract was “supposed to cover all fendering costs.” Huggett also provided details about the issue that have been kept secret for two years. Huggett revealed two errors that were made. One was made before the construction contract was negotiated with PCL and one afterward. Both subsequently “impacted” the design of the fendering, and hence its cost, Huggett reported. The first error was the relocation in early 2012 of an underwater duct bank containing numerous telecommunications cables, including fibre optic cables connecting CFB Esquimalt to the world. That $1.6 million project was engineered and overseen by MMM. According to Huggett, though, the duct bank “was not moved sufficiently far enough to allow for easy construction of fendering systems. Without additional protection measures, piles cannot be driven close to the duct bank as in the event of a ship collision the piles might move and damage the duct bank.” Unbelievable but—according to Huggett—true. By the way, the duct bank was relocated even before the City had a final bridge design, let alone a signed construction contract. At the time, City managers insisted such work needed to proceed in order for the project to meet its March 2016 completion deadline so that federal funding would not be lost. (Arbitrary deadlines and high levels of carbon dioxide are a truly awesome combination of conditions under which City councillors are asked to make important decisions, don’t ’ya think?) The second error identified by Huggett involved the City’s property at 203 Harbour Road. According to Huggett, “The City sold 203 Harbour Road to Ralmax as it was assumed the land was not needed for the construction of the bridge. This impacts an economical design since access to the water side frontage of 203 Harbour Road must be preserved.” That’s not quite true, though. The City actually transferred 203 Harbour Road and other adjacent properties to the Province in 2014 in exchange for the Crystal Garden property on Douglas. The Province then sold the Harbour Road properties to Ralmax. Regardless, Huggett is implying that whoever negotiated the transfer of 203 Harbour Road to Ralmax apparently neglected to obtain an agreement that would have allowed a minor intrusion on its riparian access to 203 Harbour Road to allow economical fendering for the bridge project. Wow. I bet the negotiating room had poor ventilation. Following delivery of Huggett’s March report to councillors, he appeared on CFAX. Among other things, Huggett told listeners the City hoped to recover, through legal action, the additional cost of fendering from the bridge’s “designer.” In Huggett’s contract with the City, the bridge’s “designer” is identified as MMM. A review of what MMM committed to in writing on the design and cost of fendering suggests that the City will have little chance of recovering that cost from MMM. But still, this is a complete flip-flop from Huggett’s previous position that the cost of north-side fendering was explicitly excluded from the original contract—and so the City would have to suck it up. Could he also flip-flop on the seismic issue and assist the City in getting MMM to return $10 million for that fiasco? Not likely. To flip-flop on the seismic issue would require that Huggett explain why he rewrote the bridge seismic code for a council report. That would be awkward for him to explain. Perhaps he could invoke a carbon-dioxide defence. SPEAKING OF CARBON DIOXIDE, one of the original premises used to justify building a new bridge in 2009 was that the existing double-bascule bridge presented a daily discouragement to thousands of would-be cyclists who, promoters claimed, were just waiting for a new bridge so they could abandon their daily commute by car. That would reduce carbon emissions, they said. Bicycle access across the railway bridge was eliminated in April 2011. If the bridge was a choke point before then, it has been even worse in the six years since. The prolonged disruption of vehicle traffic—with long waits on both sides of the bridge only adding to overall vehicle emissions—was never part of the bridge promoters’ calculations. The longer the bottleneck lasts, the more ridiculous the claim of reducing carbon emissions becomes. When will it end? The project has been on hold for months, waiting for completion in China of the lifting part of the bridge, which will span the remaining 41-metre gap. So far, fabrication of that one section of the bridge has taken over three years. How is that going? Explaining the project’s schedule—and why the bridge won’t be finished anytime soon, has been a major part of Huggett’s $20,000 per month assignment. In his September 2016 report to the City, Huggett said that Chinese fabricators had been working at fitting the rings to the trusses in preparation for a “trial fit-up.” “Painting of the structure will commence shortly,” Huggett reported. Three months later Huggett’s report noted that Chinese fabricators experienced difficulty fitting the first ring to the first truss, but Huggett expressed optimism that what the fabricators had learned would speed up fitting the other ring and truss together. It didn’t. Almost four months later, Huggett presented photographs that showed most of the major components had been fitted together, although there was no photographic proof that the north-side truss and ring had been matched. Photos published by the City showed Chinese workers apparently ready to lift the north-side truss into place on March 16. The photographs suggest painting of the bridge parts might be weeks—if not months—away. Yet Huggett had reported six months earlier that painting would “commence shortly.” So when is Victoria getting its new bridge? According to PCL’s original construction schedule, it would take slightly more than six months between the date the steel components were delivered to Victoria and the date the bridge could be opened for traffic. It would take another three months after that before the Blue Bridge could be removed and the project completed. So far, PCL hasn’t completed any of the tasks on its original schedule in less time than predicted. So, with the final shipment of steel components not expected to get to Victoria until September—according to Huggett—six months after that would put the bridge opening for traffic in February 2018, and project completion in early May 2018. One has to wonder: If those Harvard scientists are right about carbon dioxide affecting human cognitive function, did Shanghai’s notoriously dirty air play a role in the Chinese fabricators’ stumbling performance on Victoria’s new bridge? That seems possible. And there’s plenty of evidence of mental confusion at play on this project right here in Victoria, too. If there’s something in the air that’s making it more difficult for people to make good decisions, it’s a global phenomenon. Which means, of course, I, too, could be dazed and confused on the Johnson Street Bridge. How about you? David Broadland is the publisher of Focus.
×