thumbnail	notes
	That's how the system works with the big Bearspaw South Feeder running. This presentation is how hard we tried to avoid feedermain failures. We gave so many presentations on that program to other cities that were following along in our wake. I was able to do this whole section is just slides taken from presentations between 2007 to 2014. This is the title slide from 2014.
	We always started with this slide, of the whole intersection of McKnight Boulevard, and 36th Street, turned into a lake. On January 28, 2004. That little dot at the upper right...
	...is this guy's car. The fire department had to pull him off the roof in his business suit. It was -39C, on a Sunday. If it had been Monday, a whole lot of cars might have piled into water that stopped them cold, pardon the expression, and we'd have probably lost people. It was the very first time a large main had ever failed in Calgary; we'd started to regard them as simply bullet-proof. A colleague asked me in 2003 why we were inspecting small mains, but not large ones. I told him none had ever broken, so there was no budget. But that a single break would immediately hand us a million a year. This guy's shivering got us a nearly blank cheque.
	What the hell had happened to it? Concrete doesn't rust; it's very passivating to any corrosion that the soil tries to inflict on metal even touching the very alkaline concrete.
	We hired an engineering professor from the University of Calgary, who patiently explained to us that even "sulfate resistant" concrete only resists up to a few thousand parts per million - and this pipe had been sitting in soil with three times that much sulfates. The sulfates turn the concrete into a different chemical called "Thaumasite", which has all the strength of chalk, and is porous, letting the sulfates through to the steel wires and pipe barrel beneath. The pipe was only 20 years old.
	This graphic is from the very good explainer piece that Lily Dupuis of CBC did with the drawing contributed by City of Calgary. Lily interviewed myself and others, got her facts straight, I recommend it. The upshot is that everything depends on the thin steel can made strong by a wire wrap. The cement mortar just holds the wires tight and protects them.
	This 2009 slide from a presentation to a Saskatchewan conference shows us frantically doing digs up and down the main from the break, checking the concrete, doing soil tests everywhere. The problem was actually fairly isolated. Do note that "historical slough/pond" brown polygon, and the orange 100m buffer zone the GIS drew around it. The bad main, only where high-sulfate soils were, ran from one of those soil-map polygons to another, and stopped.
	There's a few more of these in the presentation slide from 2007. We were using a soil map from Alberta Research Council, which recently converted from paper to a file we could load into our GIS. The map polygons were estimated from thousands of samples taken decades back by the ARC. Not accurate within 100m or so, it did predict where we found the high-sulfate soils, so we used it to look further afield.
	Soon, our maps of the feedermains - all the bad-soil polygons overlaid on our big mains, with 100m buffers around them in orange. Any PCCP main - on the east, basically all of the feedermains - in brown or orange was high-risk, and an inspection candidate.
	How much pipe were we suddenly having to worry about? Well, if you zoom out from the break site to the whole Northeast, there's a lot of kilometres of big pipe in that part of Calgary, all PCCP-type pipe.
	And what does that Alberta Research Council map look like, around there? I've changed the colour scheme from the brown we used back then to my favourite highlighting colour, red. The term for that soil type is "pond sediments", that is, the black, organic, sulfur-rich soil that you love to see in your garden, rich with nutrient ions for your plants - and deadly to metal it touches. Or even concrete, in this case. The whole northeast of Calgary, this big flat plain far from our hills, was lousy with swamps and ponds and life..and sulfates. Note the deep blue colour for the soil polygons that are "Fluvial Gravel", that is, river gravels way up on a hill, because Calgary was put through multiple mix-master passes by glaciers that put every soil everywhere. I coloured the GIS soil layer so that the larger the particle size, from clay to silt to sand to gravel, the cooler a colour. Very general rule, the larger the particles, the less corrosive. And extra-corrosive soils, the pond sediments, I made red hot. The general green all around the gravel and pond sediment blobs is called "Crossfield Till", about equal parts of sand and silt and gravel, and not very corrosive; we didn't have to inspect everywhere - not immediately.
	With that colour scheme in mind, zoom out to the whole of Calgary. All those large-particle sands and gravels in blue are common, up the hills, and bluest of all in the river valleys, were a flood every hundred years assures most soils are washed and gravelly. The orangey-yellows are silts and some clays that can be fairly corrosive to my little metal mains, but big concrete pipes, not so bad as the hot-soils of pond bottoms. I'll come back to this map, and the spot I do wish we'd looked at harder at. But the main message from this is obvious: only an idiot would have spent millions per year, to inspect just ten or twelve kilometres of main per winter, anywhere but in the red freckles of doom all speckled over the east side of the city. So that's where we started.
	What did large-pipe inspection mean, back in 2005, when we took our keen new budget and threw it at pipeline inspection companies trying to get a new market, new technology started? Pipeline inspection for oil and gas was a few decades old by then, much easier to budget for when the product is a hundred dollars a barrel, whereas water is just pennies per barrel - and doesn't poison nearly all life it touches. These folks market in water was basically limited, at the time, to cities that had just recently had a major disaster, like the one in Annapolis, Maryland in 2004 that had cars getting washed down the street. Even they were impressed at a new customer that wanted, not a frantic inspection of one pipe, but a whole new program to eventually tour the city. We were thinking of that from the start, got a lot of praise for it.
	That's why I could only laugh, to not cry, when voices rose in 2024, beating us up for not doing "more frequent inspections", which is like beating up James Cameron for not doing enough CGI movies. We were early adopters, new-technology funders. This 2009 slide shows the state of the art, for commercial, reliable work, at the time. The inspections not only required the main to be de-watered, meaning thousands of tonnes of water to be dechlorinated and wasted... they required four guys pushing it along, with eight bike wheels cradling it.
	Accompanied, obviously, by a laptop, so they could see if it was getting good signal. Solving these problems, and automating their solution down to something the device did automatically and independently, was, for the inspection industry, like NASA having to develop astronautics mission by mission. With we early-adopters paying a lot of R&D costs. It was patriotic, for Albertans; the two main companies in the world for pipline inspection were PICA of Edmonton, and Pure Technologies, of Calgary. This 2009 presentation is our first to mention the eventual fully-submerged submarine inspection tool, the DipeDiver, then starting development.
	We BSc engineers were a little out of our depth on the science, as you can imagine when we got presentations on X-ray diffraction, X-ray emission spectroscopy, and nuclear magnetic resonance technologies. We basically got the idea of remote-field eddy current, at least; and acoustic monitoring - that an optical fibre would also be a microphone kilometres long, tiny vibrations in the water changing the signal down the fibre. It was all very cool.
	This slide is from a 2007 presentation, showing the immediate area, the part all covered with more red freckles. The pipe in the areas is shown mostly inspected, and part of it replaced. It was $1.2M to replace that first 340 meters, but back then, just inspecting 1.7 kilometres more had cost over $3M. We were one of several cities with real yearly programs ongoing, that were hauling those prices down by funding the development of better and better tools.
	We also used some cheap, quick-and-dirty inspections, like the SmartBall from Pure Technologies. It just rolled freely along with the flow in the main, no cable, just pings to let you find it again at the other end. All you got was an SD chip with hours of sound recordings on it, and you'd listen for the hiss of a leak - have, from the time, the approximate location. But we roughly "inspected" over 20 kilometres of main that went clear to Cochrane in a day. Steel mains are entirely different in failure: just a leak at first, smaller than a dime, and audible. Parenthetically, I predict a bright future for steel feedermain sales in Calgary in future. Maybe for generations.
	What we needed, for a quick inspection of the whole system, was costs more like these: a tenth of the cost to inspect, as it would be to actually replace. Much more than that, and they start talking about doing replacements by guesswork, to save money for the replacement itself.
	This is around the time where I was moved to other job areas, and two younger engineers took over, whom I'd had the privilege of training, the previous five years. They managed a comprehensive program that was looking all across Calgary for other trouble spots. This 2009 slide shows all those "Pond Sediment" polygons, removed from the rest of the soil map, where they were within a few hundred metres of any large main, at left. This informed the soil testing program, that's the cheapest thing we could do across hundreds of kilometres of mains. You augur down a few metres and grab a bag of soil, send it to a lab. Event that, I'm afraid, costs a thousand bucks a sample.
	There's another hundred dots on this soil map from a year or so later, most of them thankfully green, for "reassuringly low sulphates". It was only another hundred samples a year, but that's fifty kilometres of checking up on soils we assumed to be good, from the map, and then another hundred grand on just 5 kilometres of soils the map said were more worrisome.
	The soil sampling drove the vastly-more-expensive and inconvenient inspections program, so that if you look at where the soil sample locations are on this map and go over to the inspection program history map by 2014:
	...you find that the inspected locations follow the soil tests, wherever the tests confirmed the provincial soils map.
	The other bit of self-appointed expertise, was the notion that the Bearspaw South Feeder was this well-known problem type of pipe; you can find the paper on the Internet that notes a cluster of failures that happened in the US northeast. The material that turned up a lot was middle pipe-type in the group, C-301-E. We have just 11 kilometres - the Bearspaw South Feeder was our only purchase of it. Mostly,we had the type that failed on McKnight, C-301-L, which is very similar, but has slightly thicker steel wires, under less stress. The problem with beating us up over that paper, is that I can find you a paper like that, for nearly every pipe-type ever sold. They all did badly, sometime, somewhere. One paper, I wrote, about ductile iron, trashing its performance in favour of PVC plastic. So I was surprised to come to Vancouver 15 years later,and find ductile iron still in use. Calgary, Winnipeg, other prairie cities with black soils now hate ductile iron, but rocky, non-corrosive Vancouver never saw the problem. The pipe that works so well in Calgary, is PVC plastic - but you can find papers from Australia about how brittle PVC was, how badly it worked in their swelling clay soils. Turned out that Australian factories were using a more-brittle formulation, and putting it in the worst soils for that problem.
	The other problem with crapping on us for not being alert to the special problems of C-301E was the evidence we had that went the other way. We had cut across a private lot, decades before, with a main that served the Rundle community. Whether the proximity to a concrete foundation, or pipes, or wires nearby, was the cause, the pipe picked that one spot on private property to really explode - and tear apart his garage. The Calgary Sun had a great day, we did not.
	The pipe that had broken so spectacularly was the opposite of C-301-E. C-303 is actually called "bar-wrapped" because the wires, at no tension, are much thicker than the newer tensioned products. It was the least susceptible to corrosion at all, but if the soil is bad enough, that just doesn't matter.
	So soil, ever more so, continued to be the driver of the program. C-301-L and C-303, the bar-wrapped pipe, were equally on the agenda, but there were other issues with inspecting the one 11km piece of C-301-E.
	All those complaints about difficulty of inspection went quadruple for the Bearspaw South Feeder. Really hard to get along without, even back then, and really hard to inspect. We were reassured that the PipeDiver was coming, we were helping to test it, and we figured the Bearspaw South Feeder could wait for that new miracle technology to deal with it. This slide is from that 2013 presentation that was the last I was involved in, while the feedermains-inspection engineer concluded his portion with the triumphant first trial of the PipeDiver, with its airlock insertion mechanism that cost something like a million to install. That was a short run under controlled conditions, in a pipe that had been inspected before, with known technology, so we could compare the PipeDiver's accuracy to established baseline. Over the next few years, before I left, they progressed to about to start the first PipeDiver commercial inspection, I think in 2016.
	The actual state of the art in 2013, itself, is Charles' earlier slide: they still had to dewater the main to inspect it dry, but they were down to just one guy walking it along. Any hope of doing something to a 2-metre diameter main, 11 kilometres long, was years in the future - like SpaceX doesn't plan a Mars mission with their first rocket.
	Almost done, since I have no data about Waterworks to share with you from after 2016; but I have to come back to one thing I just spotted when I did that simplified cold-to-hot soil map for this presentation. Looking again at the full city map, you see mostly cold soils on the west side of town. But, simplified down to silt and clay soils being warmer colours, there's one chunk of it - right on the Bearspaw South Feeder.
	Zooming in on it, that soil type gets a purple colour, because "Fluvial Fan Sediments" might start with "Fluvial" for "River" but it includes not just gravel, but "Silt, Sand, and minor Clay"..and the blob of it is right where a lot of the trouble happened. Oh. Em. Gee. Did I miss something? But if I did, how did everybody else miss it for years after I left?
	No, they didn't miss it. Eventually, the soil testing program got around to that even that lower-priority area. By the time I left, in fact. The were tests run in 2014, and they weren't alarming, within range of expectations anywhere in Calgary that wasn't pond sediments. Look around the rest of the map, and you see sulphate numbers up around 1000 and 2000, over in the freckled east, and even one 1493, just a half-kilometre north on Crowchild. But the sulphates on the south feeder itself are all non-scary, down in the low hundreds.
	I have heard one thing about this recent break: the soil didn't have high sulfates, but, for the first time, it was chlorides. Scary is about the same number for any kind of corrosive ion: as you get past a thousand parts per million, get worried. There are some high numbers around town, here and there: 2800 north of Nose Hill park, 1200 and 1020 around 52nd St east...but nothing over 500 on those dozen tests done down by the river. I just can't find fault with holding the priority on it down low..
	My last graphic, from 2015, finally has the BearsPaw South Feeder on the priority list, not because of any soil tests, just because it was already so critical... ...and, of course, because the first PipeDiver full test in that very year made the idea even thinkable. As you may have heard, the Mars Landing of Pipe Diver inspections, the Big Kahuna at last, the inspection of the Bearspaw South Feeder, was planned for October, 2024 - this week, in ironic fact. I should mention something that Hugh gigged me on - I told Don Braid of The Herald that the pipe was shut down in 2007, when the whole Bearspaw plant was being renovated in that winter. Braid apparently repeated my misinformation on National TV when Hugh was watching, and Hugh blanched.
	The main wasn't shut down in 2007. The main was running backwards, not just as far as Shaganappi, like in 2024, but running backards all the way to the Bearspaw plant. Glenmore water, pumped across half the city, was dumped into the Bearspaw clearwell. That's a plant's final tank, where the big pumps could take it uphill to those North and Northwest feeders. That's how the northwest was provisioned, with the entire Bearspaw plant shut down. As Hugh commented, we not only didn't inspect that main in 2007, we couldn't have; indeed, we had to trust it to run backwards. So, summing up, I think that City staff are going to do all right in the upcoming inquiry. We were doing more than most cities were, or even could do, to avoid this. And they nearly made it. The fault that can be found, comes in with how critical and irreplaceable that main was allowed to become, when there had been earlier plans to reinforce that part of the system with more resilience.
	This is the project that didn't happen. The 2016 plans for the "North Calgary Water Servicing Project" would have run a new main from the North Feeder, one of the two big ones going north from Bearspaw, right across Nose Hill Park, through some city streets and parks, and reached the huge mains at Beddington Reservoir, which then has large-main connections to the rest of the North and North-East of the whole city. With that main in the system, we could have shut down the Bearspaw South Feeder without water restrictions almost anywhere. It was cancelled in 2021, when it should have been complete by the end of 2022 at latest.
	Hugh put hours of work into this one timeline that lets everything overlap. How did an already-delayed, badly-needed project, just ... not happen? Let's start with that catastrophic fall in the price of oil in 2014, when Mohammed bin Salman of Saudi Arabia, ably assisted by the rest of OPEC, and by Vladimir Putin, not wanting to lose revenue, also opening taps, when bin Salman tried to drive the American frackers broke. Herald columnists later seemed to blame the layoffs and cancelled projects in the oil industry on either Trudeau, or Notley, or both, though both were not elected for over a year after the price fell, and the cuts started. It wasn't just industry - Hugh notes that $660M in public projects were cancelled over the next five years, even though the price of oil began to recover after less than 18 months. It was a bit of a jobless recovery, you see, and they were driven by population growth, which they continued to imagine would crater, even as Hugh's blue line shows Calgary growth steadily proceeding apace, without oil to drive it. The growth-predictors at the very top of the city, the people estimating needs for water, roads, buildings, everything, still believed that 2014 had been the end of the world, somehow. But even 2020 didn't slow it down; in fact, growth took off in 2022. But how could they know? Hugh discovered that in 2019, they quit using the census data. 2020 didn't even have a census, but they didn't start using it again in 2021 or 2022. We heard a lot of criticism - actually nothing but - for the huge re-organization started by CEO David Duckworth. As with most of the re-organizations for the last few decades, it seemed to take away people who knew a subject area, from their subject area, and moved in more of the "professional manager class" who mainly know how to manage people - and budgets. There was a new head of all "Infrastructure Delivery", Michael Thompson, who led all infrastructure projects from any department, and we suspect it was pressure from that high up that made Water cancel the North Calgary Water Servicing Project, after 3 years of work that included some $5M in engineering design and planning. Water projects managers were ready to pull the trigger and break ground, when the plug was pulled instead. Meanwhile, it wasn't just Calgary that jumps into accelerated growth at the end of Hugh's timeline: so had Airdrie, which we service from that Beddington reservoir, and Chestermere, served by a main running out east from the Glenmore zone. They planned as if for little or no growth, continued with that plan as growth obviously continued, right through 2018 and 2019. And then there was a growth spurt. All that growth made the Bearspaw South Feeder particularly devastating.