Posted November 7, 2012
In 1985 the Province did a study of water levels in Shawnigan Lake and Shawnigan Creek. That study evaluated the capacity of the creek to carry water out of the lake. Since that time much has changed around the lake and in the creek.
Three residents - Brent Beach, Grant Price, and Graham Ross-Smith - have been discussing the factors affecting the lake level during the rainy season. We do not always agree. We thought that we might agree if we could get some more information about how the creek behaves during high water flows.
Grant and Graham, working through the Shawnigan Residents Association, were strong supporters of the Creek Cleanout initiative. Brent felt the specified target of the initiative was wrong and wanted the cost to be borne by all residents of Area B (as did Graham).
That initiative passed (was not opposed by enough people to fail) and this fall the CVRD excavated parts of the creek bed near William Rivers Park.
Grant and Graham thought the work should have been done nearer the lake, Brent felt work closer to the weir was more important. We have agreed to work together to collect the evidence to determine where any further work should be done.
Note: Click on any image for a larger version.
What has Changed
Both the forest in the watershed and the creek itself have changed since the 1985 study.
Here are two images of our area, the first from 1999 the second from 2012.
The first was taken from an airplane, the second was captured from Google Earth images.
The airplane image appears to be a compilation of a number of photos. If you look closely, you can see changes in colour. For example, just above the bottom of the lake the picture shades a little darker as you go up.
While involving a number of pictures, all the pictures were taken at roughly the same time, in 1999 (date in 1999 unknown).
This composite image then can be taken to be an accurate representation of the forest cover in 1999.
I have outlined the watershed in blue. This is an approximate line. The watershed extends to the west, out of the image, to include the headwaters of McGee Creek.
In 1999, most of the watershed was forested. The exception is an area south and west of the lake. This area became the Cougar Ridge development.
The long pale line along the west side (left) of the watershed is the west side power line. This power line runs almost along the ridge that separates the Shawnigan Lake watershed from the Sooke Lake watershed.
The next image shows exactly the same area to the same scale.
The google earth image is also a combination of satellite images that were taken at different times and stitched together. Unlike the previous photo, the various bits of a google satellite image can be taken at very different times. If you look closely at the image you can see a slight darkening near the bottom that goes from Sooke Lake on the left almost to Finlayson Arm. Except for the small area below that, all of this image was from a single satellite scan done on Sep 3, 2012.
The part of the watershed below that darker area (and Devereaux Lake is just below that) is from an older scan. The state of the forest today may be different from what is shown in this part of the picture.
In 2012, most of the watershed has little or no forest cover. South of the lake there are only two patches. One corresponds to the Cougar Ridge development - so the trees there, while appearing green, are only 13 years old. The second corresponds to the Elkington property, right at the south end of the watershed.
The bad news - almost the entire watershed has been logged. What little is left is crown land. Formerly considered safe, Crown land may now be used in treaty negotiations and be subject to logging in the near future. An area on the east side of the lake has already been slated for logging as part of a First Nations land claim settlement.
Here are the two pictures side-by-side, for easier comparison.
As we lose forest within the watershed, rain runs down the hills to the lake much more quickly. Snow melts more quickly and enters the lake sooner. The result - lake levels rise more quickly, increasing the potential for quick floods. On the other hand, the hills dry out sooner resulting in lower lake levels in summer.
We have seen, in the last weeks, that we are in a period of more intense weather. Whatever the cause, more intense storms seem to be the norm. Unfortunately for us, a watershed without a forest is less able to handle extreme weather events.
The Creek Bed
Not only have the hillsides around the lake changed, the creek bed has changed as well.
This picture was taken under the railway bridge, about 150 metres from the lake. Reflections on the surface make this a little hard to see clearly.
What you are seeing is the bottom of the creek. All those branches you see pointing to the right are embedded in the sediment of the creek bed. They were live branches on shrubs around the lake and in the creek. They died, floated along the creek, became waterlogged, and sank to the bottom Sediment built up on top of them. Over time the creek becomes shallower, more restrictive to the flow of water.
These branches are exposed because this is the upper limit of the creek bottom excavation done this fall. The creek bed on the left is the unexcavated creek bed. It is about 70cm (24 inches) deep. The creek bed on the right is the excavated section. It is about 150cm (60 inches) deep.
The excavation removed more than waterlogged branches and other recent sediment. The large round rocks on the bottom at the right would not get there through through a sedimentation process. They could be remnants of the last ice age.
Over time this silting process happens all along the creek above the weir, because the creek bed is flat and has very low flow during the dry season.
This combination of hillsides without trees and a creek bed that is slowly silting up together increases the danger of a flooding event in the lake.
Goals of Study
Our goal is that future work on Shawnigan Creek be based on facts. We can collect the necessary facts by collecting accurate water level information along Shawnigan Creek. This water level information will let us pinpoint the parts of the creek which are bottlenecks - where water is held back. Remediation work in future can be done in these bottlenecks.
Work done on the creek this year was done without this factual basis. The CVRD admits it dredged the creek beside Williams River Park because it owned the park and could work there without getting permission from any land owner.
Grant and Graham believe that the bottleneck occurs at Halhed Bridge - the highway bridge near Mason's Beach. If this is indeed correct, then dredging the creek by William Rivers Park was of little value.
Water Level Measurements
Shawnigan Creek, the only outflow from the lake, exits the lake in the Northeast corner, flowing almost directly north under Halhed Bridge near the intersection of Renfrew Road and Shawnigan Lake Road.
It travels along a relatively flat creek bed for about 400 metres, crosses the weir, then starts to drop. This first 400 metres is the part of the creek that restricts the flow of water out of the lake.
The amount of water that flows under Halhed Bridge is exactly the same as the amount that flows over the weir. This simple fact is key to understanding how we can isolate restrictive sections of the creek using water level alone.
Flow along any section of a creek depends on how restrictive that section of the creek is and on the difference in water level between the up-stream and down-stream ends of the section. The more restrictive the section, the greater the difference in water level required to produce a specific flow.
Increasing the capacity of a section that is not a bottleneck won't change the flow down the creek. The flow between the lake and the weir cannot exceed the flow through the most restrictive section of the creek. Unless the section of creek beside William Rivers Park was the most restrictive section, dredging it will not affect the flow between the lake and the weir.
We intend to find out how restrictive each section is by measuring the water level at several points along the first 400 metres of the creek.
We have divided this 400 metres of the creek into 6 sections. To do this we added 5 new level gauges to the two existing gauges (which show the level on the lake and at the weir). After this high water season - October to March - we will know how restrictive each of these 6 sections is. If necessary we can install additional gauges for the next rainy season to further isolate the most restrictive sections.
This is one of the new gauges. Grant Price, on the left, is bolting it to a piece of rebar. Graham Ross-Smith, on the right, is clearing the dead wood from the area so the gauge can be photographed to record water levels.
This is the gauge on the weir. It is currently reading about 15.81. I will discuss how we interpret readings from these scales later on.
Since the weir was holding back the creek when we installed these gauges, the level at the weir and the level at the lake were very nearly the same. If lots of water were flowing over the weir, the difference between the water level at the lake and the water level at the weir would be larger. We will see those differences when we get our gauge measurements during the high water wet season. When we installed the gauges, the flow was quite low. There was a bit of water flowing through the fish ladder and a bit seeping through the weir gate.
Here I have inserted a scale into the picture beside the gauge to help get a slightly finer reading than possible with the gauge alone.
The exact water level is a little hard to determine, but a reasonable estimate of the reading using the scale is 15.81.
What does that mean? Well, you have to add the hundreds digit to get 115.81. That in turn means that the water level here is 115.81 metres above (mean average) sea level.
The bottom of the creek at the weir is 115.10 metres above sea level. The water level could be 0.71 metres lower here if the weir gate was allowed to fall all the way down and drain the lake. (To give readers of my generation a feel for what this is - 0.71 metres is just under 28 inches.)
Now compare the current water level on one of our gauges with the official gauge above. This gauge reads between 15.84 and 15.85, indicating a level of between 115.84 and 115.85. The level at the weir and in the lake is 115.81. Our placement of the gauge was slightly in error.
We knew when we were installing these gauges that we could not get them to exactly match the official gauge. What we also knew was we could take a picture of each gauge and of the official gauge, at the same water level. We could then apply an appropriate correction to each gauge, based on a comparison of the pictures.
Adding the scale to the picture and doing a little work in a graphics program, the scale reads 15.846, which corresponds to 115.846 metres above sea level.
Comparison with the known water level, 115.810, tells us that we got the placement of this gauge wrong by .036 metres (about 1.4 inches). We don't need to correct the position of the gauge. We just have to apply this correction to each reading we take on THIS gauge.
Our last gauge is a special case because it is below the weir. The water there is not at the same level as the lake. It is below the lake level by the amount the weir holds back water.
You can see the weir in the background of this picture. If you look closely, you can almost see that the water below the weir is lower than the water above the weir.
We measured the difference when we installed the gauges and found that it was 0.468 metres (18.4 inches).
We now know that with almost no flow in the creek, the water level below the weir is 0.468 metres below the level of the lake, or 115.334 metres above sea level.
This is the tricky bit! This is the gauge below the weir, with the inserted scale to allow more precise readings.
The current reading is 115.80.
At first it would seem that we are pretty close. In fact, we are almost 18 inches off. The actual water level here is 115.334 so the error for this gauge is 0.466 metres.
Each time we read this gauge, we have to subtract 0.466 metres to get the correct height above sea level.
Data Collection over the Winter
We have correction numbers for each gauge.
We intend to read our gauges regularly over the wet season, particularly after heavy rain. We want to know how much the lake level rises after each rain. We also want to know how those higher lake levels affect the water level at each gauge.
With this information, we will be able to isolate the sections of the creek that slow the water down and the sections of the creek that allow the water to flow more freely.
A web page explaining how water level and water flow within the creek are related is in preparation and will be available shortly.