The connections between climate and hidden hydrology is a continuing theme, and inevitably will result in more examples that can be documented making the connections between present and future impacts and historical ecological systems. Each of these has a specific context, which influences the extent of impacts (urban/rural) and the hydrological dynamics (lakes, rivers, shorelines). The recent flooding in November of this year in the town of Abbotsford, which lies near the border of Washington State and British Columbia is one recent example. The connection was made via Twitter by @tornadodc (Lisa Genialle), who posted the following image of flooded highway overpass with the quote “Anyone who knows their local Fraser valley history, this used to be a lake back in the day. Before it was drained to make way for settler farming.”

I had little reference for this particular region or the Sumas Lake specifically, however was supplied with more context via the subsequent thread of responses, many of which yielded Chad Reimer’s 2018 book “Before We Lost the Lake” which explores the ‘biography’ of the lake and its ‘natural and human history’. This provided some much-needed context for what existed and how the flooding takes the shape of this lost lake.

Reimer outlines a similar story found in many locations, a waterway that had been a vital and ecological benefit to the surrounding environment and a resource to milennial of indigenous people, and was summarily ‘reclaimed’ for farmland through drainage, canalization and pumping in 1924. From the review in the Chilliwack Progress: “For thousands of years, Sumas Lake sat near the centre of local culture and life. Then, nearly 100 years ago, engineers drained Sumas Lake, built a canal to harness the Vedder River, and, in the process, radically transformed the area.” A big missing piece of the narrative also covered by Reimer’s book is the displacement of the indigenous people, who have sought compensation for the lost cultural resource, both as a place of hunting, fishing, and gathering that was eliminated when the lake was filled.

Beyond the Sumas First Nation people, the lake is forgotten by most, or at best a historical footnote. As mentioned on the publisher’s site: “Today, few people are aware that Sumas Lake ever existed. The only reminder is a plaque erected on the old lakeshore, at a rest-stop along the Trans-Canada Highway just east of Whatcom Road, on the historic trail blazed to BC’s gold fields. Yet for millenniums, Sumas Lake was a dynamic, integral part of the region’s natural and human landscape.”

The narrative that hidden hydrology is gone but not forgotten is both a cultural phenomenon (in histories and place names for instance), and ecological one (the loss of ecosystem services by removal of function and resilience), and a hydrological one (systems impacted yet still wanting to function and flow as they once did). The extreme storms exacerbated by climate change take shape in these old patterns. For Sumas Lake, a series of maps show the devolution and the impact of the most recent flooding. The first from Wikipedia is an animation of the lakes removal over the years between 1827, the filling in 1924 and the development through the 1940s. It includes the extent of devastating 1894 flood, which was the impetus for the eventual drainage.

From the Fraser Valley News, an undated map (pre-1924) of the lake shows the original extent of the historic shoreline (referenced here as Lake Sumass) which covered between 9,000 and 11,000 acres inside the valley.

Map of Lake – image from Chad Reimer – via Fraser Valley News

The area had been developed over time, starting with transformation into agricultural fields and while mostly rural productive lands, have also since been developed with some other residential and commericial uses. The map below from an extensive analysis of the flooding by the Tyee as part of an article “Mapping the Flood in Abbotsford“, show the original lake, and also showing the dashed line outlining the extent of the 1894 flood within the confines of the valley as well.

The extent of flooding in 2021 came from the Nooksack River and breached the Sumas River dike. The amount of water overwhelmed the series of pumps, which are able to move 250,000 gallons per minute of water, which became unable to keep up with the sheer amount of flooding and the area was quickly inundated, cutting off roadways and stranding people and livestock. The shape of this water corresponds to the former lake bed, with significant depths estimated in a 2020 study to map worst-case flood scenarios, proved to be a hint at what impacts were to come in an extreme event.

The actual flooding is not as bad as the map scenarios above, but did expand into many areas outlined in the former lake bed, the deepest sections potentially inundating more than 3 meters. The post from the Fraser Valley Current shows many of the scenes of flooding near Abbotsford (seen below) which cut off some routes of evacuation, with some people rescued via helicopter. Sparsely populated, the impacts were severe but could have been much worse in a zone of higher population density.

The impacts to property and life are not to be dismissed and my goal here was not to extensively cover the events at Sumas Lake. The bigger picture was that we do need to make the connection between areas where the historical draining and filling waterbodies for progress and development (and often, ironically for flood protection) does have the potential to give hints at the future impacts of extreme events such as the flooding at Sumas Lake, beyond the loss of habitat and other ecosystem services and the lost resource for indigenous people. The photo below, circa 1900 shows the original lake from Daily Hive (which also is a great overview of the lake’s history), which looks remarkably similar to the photo above of flooding here in 2021 (with another major flood in 1990).

The case in point, that water will inevitably ‘find the level’ and create massive impacts, the memory recall I used to outline this post, and this has implications for life and property with real economic and social impacts. As noted in the Daily Hive story: “A 2016 study by the Fraser Basin Council on the Lower Mainland’s flood management strategy estimated that a repeat of the severity of the 1894 flood would currently cost about $23 billion in damages, including $9 billion in losses to residential, commercial, public, and institutional buildings; $7.7 billion in interrupted cargo shipments; $4.6 billion in infrastructure losses; and $1.6 billion in agricultural losses.”

It’s probably too early to tell how large of an impacts was absorbed by the events fall, but history can instruct us in perhaps knowing the location of impacts and using this as a guide for prevention. This is a more rural example, and I’m working on similar urban examples, which, as climate change continues to impact rainfall and other weather events, will be more and more useful in helping us understand, and plan, for what’s next.


Header Image: Sumas Lake circa 1920 prior to drainage – Agassiz-Harrison Observer

A quick one to show a map that captures the essence of the story of Seattle – this 1996 “Map of landfills, regrades, and cuts” outlines significant changes to the landscape of the city over the formative decades from 1896 to 1930. The source is the Seattle Municipal Archives, with a note of a source  Seattle Engineering Department Digest, Special Commemorative Edition. A few interesting ones that aren’t labelled, such as the northernmost ‘Regrade’ which aligns in the vicinity of Northgate Mall and the wetlands around North Seattle College in the Thornton Creek drainage, and what I assume is also filling of the wetlands in the middle of Magnolia (just below the letter H).  Also, while I get that there was some manipulation of Green Lake, I don’t think it’d be technically correct to refer to that as a fill, at least in the traditional sense of ‘making land’.  It’s interesting to see these all captured in one figure. The impacts, of course, are well known.

 


HEADER: Denny Regrade – circa 1910, via Wikipedia entry “Regrading in Seattle”
\\ h/t to folks on the Reddit Seattle Group that linked to the map originally.

GREEN LAKE

I’ve been wanting to write about Seattle’s Green Lake, which is an addition to round out the post these small Seattle lake stories, and supplement the coverage of the larger lakes Union and Washington.  Green Lake has a special place for me, having lived close to it our entire time in Seattle, it’s been a place for fun, recreation, and even protest.  This really cool cross-time image from Then & Again shows the juxtaposition of the current with the old, and Green Lake spanning this , here with “…the majestic USS Macon gliding above Seattle’s Green Lake on August 22, 1934. The airship was traveling to its new station near San Francisco but took a leisurely route with time for a number of photo ops along the way.”

The history of the lake goes back to similar era to the smaller Bitter and Haller Lakes and the larger Lake Union and Washington, as mentioned on the Seattle Parks website: “Geologists say the Vashon Glacial Ice Sheet, which also formed Puget Sound and other area lakes, formed Green Lake 50,000 years ago. Dredgings of Green Lake have produced volcanic ash from an eruption of Glacier Peak that occurred about 6,700 years ago.”  The original lake was lowered 7-8 feet as park of early 20th Century Park improvements, and this 2014 article from Seattle Greenlaker  ‘Olmsted and the Origin of Green Lake Park‘ offers a good introduction to the modern incarnation of the park and this process. In that post, it links to this great map from 1907 from the National Association of Olmsted Parks, which shows the development of Green Lake Boulevard and the areas near the lake as part of this process, and the first evolution of Green Lake as part of the overall Park System.

Via Seattle Greenlaker – Caption: Courtesy Frederick Law Olmsted National Historic Site, National Park Service, 02714-21. Used with permission.

If you were, like me, temporarily disoriented for a bit, it helps to rotate this drawing so north is up, and the more familiar shapes and lobes of the lake become clearer. I’ve also highlighted the old shoreline (dashed line) and the new shoreline (solid line w/ blue fill), so it’s clearer where the drawdown left space for the trails and more usable open spaces, with amenities like Boat Houses and beaches.  The map also has information on the inlets and outlets, which I’ve highlighted below in a bit more detail for reference.

If you remember back a bit to my post of the exploration of Licton Springs, the sketch above is instructive, as this inlet #1 shows an inlet with a specific reference: the “Brook inlet box culvert 2’x6′ from Licton Min. Sprg.”, showing the subsurface connection to the spring that had been filled in previously but was still flowing from the north.  A Bathhouse is shown, which does not currently exist, but there is a swimming beach access which is still in place today, along with the splash play and some open spaces.

There is another pair of inlets #2 a bit further west, near Corliss Avenue, with a label “Inlet box 1 1/2′ x 2′” and near there is one with a note “Old Inlet 18″x18” wood box” which has a note “very little flow now – some sewerage seems to enter it now.”

The final inlet #3, is to the west, showing a connection to an “Inlet-brook, 6′ wide, 3″ deep” that comes from the northwest, running under the streetcar track in a culvert before coming into the lake.

On the opposite side of the Lake, heading southeast, there is the Outlet, which is marked #4, near 4th Avenue (a few blocks from the proposed boulevard which would become current day Ravenna Boulevard), which was the natural drainage of Green Lake into Ravenna Creek.  The reconfiguration of this zone and the shift of the shoreline created a larger area that now has sports fields, as well as a boat house and what is a popular beach access spot today.  This is also adjacent to the larger commercial zone which is the hub of activity adjacent to the lake.

 

These flows in and out correlates somewhat with the 1850s maps, but does closely align with the the USGS Topo Map from 1894 (see below) of Green Lake that identify three inlets from the north and the Ravenna outlet heading southeast – which does line up with the hydrology shown on these 1907 drawings.

Historical Topographic Map Collection

The formal plan as presented to the Seattle Parks Commission in 1910 mirrors much of the modern day condition, with the lowered water levels providing for perimeter trails, new plantings, a new island, some amenities such as beaches and boathouses, and the boulevard that rings the park (the western half of which was transformed with the routing of Aurora Boulevard (Highway 99).

 

And another version, this one from 1925 showing a more colorful version of this, “Proposed Plan for the Development of Green Lake” via the Seattle Municipal Archives Digital Collections.  This map dashes in the existing and proposed shorelines

I posted previously about the fun bathymetry maps, which included Green Lake. Another map I like is this one ‘Showing Depth Contours of Green Lake’ via the Seattle Municipal Archives Flickr page, which was done in 1938 as part of the Sanitary Survey by the WPA and featured in the “Report on Green Lake Algae Control”, which highlights perennial water quality issue . It reinforced that the lake is relatively shallow, with maximum depths no greater than 25′ feet on the western edge.  It also identifies some of the hydrology, including overflows, intake from the City resevoir, and to the west, a “permanent inlet from deep springs” which is a fascinating addition both due to it’s mystery and also it’s location, which is not shown on later maps but does appear in the 1950s map.

The history of the lake beyond that Olmsted plan has many facets and this Chronology is helpful to see the evolution.  There many tales (and History Link is a great resource here) that connect with Seattle history, such as in 1869, when David Denny “…kills what is likely the last elk in Seattle, near Green Lake. The elk weighs 630 pounds.”, or 1893 when a cold spell froze the lake completely over. including  of Hydroplane boat races in the 1930s, as well as a cleanup and redesign in 1936.  Lots of history and evolution I won’t get into here, as it’d take days, but my favorite lost part of the Green Lake history, which I only discovered by accident after visiting the park many times, is the Aqua Theater, built in 1950 on the south edge of the lake as a 5,500 seat performance venue, built in a little more than two months coinciding with the first of what is now an annual Sea Fair.

The venue hosted a range of events included the annual Aqua Follies, which included ” Water ballets, diving exhibitions and clown acts took place in the pool and on the stage behind. Many of the Aqua Follies mermaids were recruited in Minneapolis before June 1, and began practicing before Seattle area college students finished their school term.”  There were some notable music shows including 1969, which featured Led Zepplin and The Grateful Dead, which was one of the final shows at the venue before it was shut down.

The lake as a locus for recreation has stayed consistent over the years, with lots of walkers and joggers circling the 3 mile loop, along with water access via boat rentals, rowing, and use of adjacent open spaces sports fields, and even a Par 3 golf course.  Water quality issues are a perpeutual issue, but it doesn’t stop it from being the busiest park in the state of Washington, with over a million visitors a year.

Postcard circa 1950s – via Seattle Greenlaker – https://www.seattlegreenlaker.com/2017/06/green-lake-seattle/


HEADER: 1987 Aerial view of Green Lake – via Seattle Public Archives

Some news on the project front, which partially explains the slow output on this end lately in terms of hidden hydrology updates:  I’m moving from Seattle back to Portland.  As regular readers know, the project origins are firmly rooted in Portland, including plenty of documentation and expansion of ideas around Tanner Springs Creek (seen below), and maybe I will finally track down one of those elusive ‘I Kayaked Tanner Creek‘ t-shirts of legend.  Anyway, happy to announce this news, and Portland folks, let me know if you’re interested in some exploring in coming months.

There’s also a plethora of other areas to explore, and also to compare and contrast the unique dichotomy of Portland as a river city and Seattle as more of a ocean & lake city, and what that means/meant for development.  On that note, one item I’ve not announced is some of the work figuring out the best format for a Hidden Hydrology Atlas that will span both Seattle and Portland – so stay tuned for more of this as technology and funding aligns.  For now you can see the early version of the online example of interactive maps I’m testing out using a combination of Mapbox and my GIS database of information.  Early days, but the potential is there, and it will expand into something more comprehensive and multi-media.

While I did get to explore a number of Seattle hidden streams, there’s so much more to do and lots to document for Ravenna, Yesler, and Green Lake, and hopefully coming back up to do more investigations.  In the interim, one of my explorations I documented here in Seattle from last summer, Licton Springs, was the departure point for an essay I wrote recently for The Nature of Cities that was just published this week. Read ‘Map and Explore: Hidden Hydrology’ for some thoughts on exploring our places and connecting with our culture, geography and ecology.

So, stay tuned as projects, posts, and explorations will all pick up over the summer months.  And as always, thanks for reading.  See you all in Portland soon.

-Jason

I’ve been inspired by the work many others have done to capture the qualities of coverage of waterways at national scale both in the US and the UK, and beyond the mapping, appreciate their investigations into the unique distribution of place names, or toponyms.  The language of the waterways informs more local hidden hydrology endeavors, and understanding regional vernacular variations provides a snapshot into our varied relationships with water.  While a glance at the Pacific Northwest via these other maps shows that the predominant name for waterways is probably going to be either creek or river, I wanted to dive a bit deeper to see what other names are used to denote waterways.  To accomplish this, I spent some quality time with the USGS National Hydrography Dataset (NHD) to unlock a bit of the secrets of regional variations.

For starters, the NHD is an amazing resource of information, pulling together a comprehensive collection of data on flowlines, watershed basins, and more and the ability to get data from a variety of formats for small to large basins and states.  From their site, the purpose of the data is to: “define the spatial locations of surface waters. The NHD contains features such as lakes, ponds, streams, rivers, canals, dams, and stream gages, in a relational database model system (RDBMS). These data are designed to be used in general mapping and in the analysis of surface water systems.”  The first steps are a bit daunting, as the State of Washington included data with over 1.3 million flowlines, seen below in aggregate. The flowlines aren’t any one single waterway, but are the individual segments that make up each creek.

While the data preserves local basins shapes by sprawling outside state lines, I wanted to make this unique to Washington, so needed to clip it to the state boundary.  This ended up being a bit of a task for my rather slow computer to crank out the clipping, so I had to think of some alternatives to simplify the dataset.  Interestingly enough, over 80 percent of the flowlines (around 1.1 million of them) are unnamed, and while I’m sure are perfectly lovely bits of creek and river, they don’t help in our purpose in terms of deriving place names.  Eliminating them also serves the dual benefit of reducing the size of our working dataset quite a bit.  After trimming to the state boundaries, we ended up with a nice workable set of around 170,000 flowlines that have names, seen below.

Per the NHD FAQ page, “Many features also are labeled with the geographic name of the feature, such as the Ohio River. The feature names must be approved by the Board of Geographic Names (BGN) in order to qualify for inclusion in the NHD.”  More on the BGN and the wonderful assortment of place names that exist in these lists beyond their descriptor (which is perhaps the fuller idea of toponyms), in this case we break down the list and see what comes to the top.   Not surprising, but the use of the terms Creek and River dominate the landscape of Washington, accounting for 98% of all named flowlines.

Of the totals, creeks truly dominate, with around a 75% chance that a trickle of water in the state will be referred to as a creek.  The larger, less numerous rivers make up 23% of all flowlines, and the map above paints a wonderful portrait of the density of waters.  Separated out by type, you see the branched structures of trunk and stem that pumps water through most of the mountainous west side of the state, with the larger, drier plains to the east more open.  All total the combined length of these equals over 30,000 linear miles.

1. CREEKS

2. RIVERS

So we live in a creek and river area of the world.  Amidst these dominating toponyms are a distributed layer of types of flowlines that make up the remainder of the story of Washington, that final 2 percent, emphasized in a darker blue below.

The secondary naming of these includes the most common, isolated and color coded, with a legend denoting the eight most common alternative flowline names.

The relative percentage as a portion of that slim 2% of state flowlines, include:

  1. Slough (30%)
  2. Fork (16%)
  3. Canal (16%)
  4. Ditch (9%)
  5. Wasteway (4%)
  6. Branch (4%)
  7. Run (4%)
  8. Stream (3%)

The remaining 14% are composed of small portions that include Lateral, Brook, Drain, Slu (a variation of Slough), Gulch, Channel, Siphon and it’s alternative spelling Syphon, Washout, Waterway, Swale, Glade, Pass, Gate, and Range.  Many of these as we see, are geographically located towards the center of the state where agricultural landscape has created larger modifications and creation of waterways (described in the NHD data under the names like Artificial Path, Canal Ditch, and Connector).  There’s a split between more traditional waterway name variations (i.e. Slough, Fork, Branch, Run, Stream) and those that mostly utilitarian, capturing the poetry of industrialization (i.e. Wasteway, Ditch, Canal, Siphon, Lateral).  Removing the background landform, you see the composite of the different stream types as a whole, with creek/river in blue and the remainder by color.

For a more local view, the NHD data is a bit less sparse, not capturing the same amount of complexity is smaller urban waterways, plus without the other water bodies like lakes the geography seems somewhat off.  The purple to the west in the Olympic Pennisula shows a density of flowlines referred to as streams, and the darker red denotes a number of local sloughs that exist in local river systems.  It’s harder to see, but you can catch the Ship Canal in this group, and the slightly lighter red fork in the center is the infamous Duwamish Waterway, the lower stretch that runs through Seattle and ‘lost’ its designation as a river – interestingly enough it’s the only flowline in the state with that moniker.

I was expecting the dominance of creeks and rivers in the nomenclature, but was also really surprised that these combined to make up so many of the collective flowlines. Perhaps early settlers and place-namers lacking a bit of creativity.  It was also a good surprise to find a wealth of other place names in Washington, albeit many used to describe man-made features, including the most poetic name of wasteway, but enough fun to find an occasional branch, fork, brook, and run, which are more common elsewhere in the United States, per the other US maps.

These are pretty basic graphics exported from GIS just to give a feel for the data, so I’d like to play around more with representation, perhaps some sort of heatmap.  Also I’m eyeing Oregon for a comparison, and maybe wanting to dive into the waterbodies as well beyond linear flowlines, so more fun to come.  Who knows, an atlas of the whole country with a top ten of their most common names of each state.  Or maybe not…


HEADER:  Excerpt of River and Stream Composite Map – data from ESRI, NOAA, USGS – Mapping by Jason King – (all maps in post same attribution, © Jason King, Hidden Hydrology, 2018)