Salmon & Sediment Science: How Floods Affect River Sediment and Pink Salmon Spawning Success

Hi. My name is Henry Hocking and I currently live in Deep River, Ontario. My project is about alluvial river systems and how floods affect pink salmon spawning success. I was inspired to do this project because my grandparents live on the east coast of Vancouver Island right next to the Englishman River system where pink salmon spawn annually. My dad and my grandfather both fly fish they observe that in some years there are lots of fish and other years very few. I wondered how the life cycle of pink salmon are affected by floods in alluvial river systems. To answer my question I built this replica river in my basement so I could test different water flow rates, then I would see how different discharge rates would affect three types of sediment and replica roe in a simulated salmon spawning bed. Then I compared these results to real world data of river discharge rates and salmon return numbers.

My grandparents live near the Englishman River on the east coast of Vancouver Island (Figure 1, [10]). The river has very low water discharge rates (volume of water flowing per second) in the summer (Figure 3) but it has extreme flooding events in the fall and winter (Figure 2; [5], Video 1). There is also a side channel that was built to help all types of salmon spawn in more controlled water flow environments (Video 2, 3). My dad and grandfather fly fish for pink salmon; in some years there are a lot of fish, but in other years far fewer. I wondered how the river flooding might affect the spawning process.

Many people and animals rely on a healthy river and salmon fishery, including, commercial and sport fishermen, predators and the salmon themselves. The results of this project can help conservationists design spawning channels and protect salmon spawning beds (redds).

For the experimental part of the project I predicted that:

Higher discharge rates would cause more sediment disruption and smaller sediment (sand) would move further than coarser sediment (pebbles) at the same rate.

More of the simulated roe will get washed away from the pebbles with progressively higher discharge rates. And when sand is added, higher discharge rates will cause more sand to smother the roe.

As a result of my experiments I expected that:

High water discharge rates (and sediment transport) would negatively impact spawning beds and salmon return numbers in the Englishmen River.

I built an apparatus to simulate a river. The experiment has three parts:

Determining the water discharge rates with different starting water volumes;

Investigating how increasing discharge rates affect sediment transport; and

Understanding how increased water discharge disturbs spawning beds.

I then compared my findings to pink salmon spawning data and river flow conditions from the Englishman River in British Columbia.

The artificial river was built with a wooden frame and tower. An angled tray on the frame acted as the river bed and four water bottles were mounted on the tower. Plumbing parts were used to control the flow of each bottle independently, including hose spigots and inline ball valves connected with vinyl hose, silicone and hose clamps. The hose fed a funnel that flowed onto a piece of curved plastic and onto the ramp. Clay was used to fill gaps and smooth the transition.

Water discharge data was collected by recording the time and measuring the water volume before and after discharge (fiver water volumes, three trials each). Three different types of sediment were weighed before and after each timed run of the 5 water volume trials. Hydrated chia seeds were used to replicate salmon roe, as they are approximately the same size and have a sticky surface similar to salmon roe. The number of chia seeds in a tablespoon was estimated and mixed with pebbles and placed on the tray to simulate a spawning bed. After water was released, the number of chia seeds that washed through the pebbles was recorded.

The height of the water tower and slope of the tray were constant. Water volumes were measured before and after each test, sediment was weighed and chia seeds were counted. Measurements were recorded in a spreadsheet.

I developed a working model of a river to simulate different water discharge conditions.

This allowed me to examine a broad range of conditions and the correlation between discharge rate and sediment transport and spawning bed disruption. The water discharge rate on the Englishman River changes by up to 30 times between summer and peak winter floods and 3-5 times (m3/sec) between average winter discharge and peak floods. The model allowed me to test a range of rates (ml/sec) from 1.3 to 6.2 times the lowest level. My findings showed that higher discharge rates cause more sediment disruption, and that finer sediment is disrupted and transported further, than coarser sediment, except clay, which is so fine that it sticks together and doesn't move. I also observed that under high flow conditions sand could smother spawning beds and reduce salmon survival rates [11].

I created X-Y scatter plots to examine relationships between variables in the model: water volume and discharge rate (Figure 1), discharge rate and sediment disruption (Figure 2), and discharge rate and the number of disrupted (simulated) roe (Figure 3). The relationships were generally linear.

Then, I analysed and plotted data for discharge rates and salmon spawner return numbers on the Englishman River (Figure 4 and 5). The Englishman River has flood events in the fall and winter. I looked at the spawner returns and flood events when those fish would have been in the egg or at early life stages. I selected the peak flood event for either fall or winter and compared that value to the returning spawning numbers. For example, for fish returning in 2011 I looked at flood events in the fall of 2009 and the spring of 2010 when those fish would have been eggs or fry [5]. For the seven years (odd years from 2011 to 2023) the spawner return numbers ranged from 942 to 20,364 fish [10]. When comparing the return numbers to the relevant flood periods I observed three clusters:

Low flow (122 m3/sec) associated with the largest return (20,364)

Moderate flow rates (229 to 257 m3/sec) with variable return numbers (4,114, 18,889, and 19,692)

Very high flow (>289 to 303 m3/sec) associated with a very low return of fish (942, 1,060 and 2,279)

Under low or moderate flow conditions the data show that spawner return numbers were variable but relatively high, but when the discharge was very high (> 289m3/sec) the spawner return numbers were very low. This may be because the C.W Young side channel is protected until very large discharge events flood it; researchers have documented flooding in this channel [4].

I plotted the Pacific Decadal Oscillation (PDO) for the 18 months when the spawning fish were in the ocean (Figure 5, [9]). It's a measure of ocean conditions but I did not observe clear trends in the data suggesting that flooding (high discharge rates) may have the largest effect on spawning returns.

The Englishman River is an alluvial river system (Figure 1, [10]). The bed and banks of an alluvial river are comprised of sediment that can be shaped and moved by the flow of the water. This means that its channels are shaped by the magnitude and frequency of its floods [1,2].

In recent years 90% of the surrounding rivers' watersheds have  been logged out; this causes problems for the river and its fish because it promotes winter flooding and reduced summer water flow rates [4]. A side channel called the C.W Young side channel which was constructed in 1992 to increase salmon spawning by providing a controlled inflow to protect fish from flooding (Figure 2 [6] - inset map on Figure 1). However, it has been observed that during the highest water level the side channel is also flooded; local scientists have observed the impacts of flooding in this channel [4].

The results of this study show a compelling correlation between flooding periods and low salmon returns in subsequent salmon-life-cycle years. I also investigated the PDO values, which are an index that can reflect ocean temperature in the Pacific Northwest, which may indicate favourable or unfavourable ocean conditions for salmon, but I did not observe a correlation with spawner returns [9]. If further research were to verify these observations and rule out other variables, like predatory activity or overfishing, then it could help change the design of man-made spawning channels or encourage more responsible logging in the watershed.

If I were to continue this project, I would try to source commercial fishing data around the mouth of the Englishman River. I would also try to get data on predator populations, like sea lions. Understanding the return data for other species of Pacific salmon could be useful but the other species have variable return cycles, which would make the analysis more difficult. I would also consider making a proposal to the regional park and Department of Fisheries and Oceans about installing a webcam to monitor the side-channel and starting a longer-term project to further study pink salmon spawning.

I would like to thank my father for his help on this project. He supervised the building of my experimental river, he also helped me with turning valves, measuring water, weighing sediment, counting roe, and helping me with editing. He also took me to the hardware store to buy materials, and paid for them.

Books:

[1] Davidson, J.P., Reed, W.E., Davis, P.M. (1997) Exploring Earth, An Introduction to Physical Geology. Prentice Hall

[2] Press, F., Siever, S. (1998) Understanding Earth. W.H. Freeman & Company.

[3] Sen, G. (2001) Earth's Materials: Minerals and Rocks. Prentice Hall.

Articles:

[4] Ohlman, Z., Waite, T., & Black., E. (2023). Water Quality and Stream Invertebrate Assessment of the C.W. Young Channel, Englishman River, British Columbia. For Dr. Philip Morrison, Vancouver Island University. Retrieved from: https://wordpress.viu.ca/rmot306/files/2024/09/C.W.Young-Side-Channel-Report-2023.pdf

Webpages:

[5] Daily Streamflow for Englishman River near Parksville retrieved from:

https://bcrfc.env.gov.bc.ca/Real-time_Data/Interactive_Q_process/InteractivePlots/dailyQ_interactive/08HB002_dailyQ.html

[6] Englishman River Regional Park Map retrieved from:

https://rdn.bc.ca/sites/default/files/2025-06/RDNGeoPDF_EnglishmanRiverRP_May2025.pdf

[7] National Ocean and Atmospheric Association (NOAA) Fisheries: Salmon Life Cycle and Seasonal Fishery Planning retrieved from: https://www.fisheries.noaa.gov/west-coast/sustainable-fisheries/salmon-life-cycle-and-seasonal-fishery-planning

[8] National Ocean and Atmospheric Association (NOAA) Fisheries Pink Salmon retrieved from:

https://www.fisheries.noaa.gov/species/pink-salmon

[9] National Centers for Environmental Information National Oceanic and Atmospheric Administration Pacific Decadal Oscillation (PDO) retrieved from:

https://www.ncei.noaa.gov/access/monitoring/pdo/

[10] Salmon Explorer, including maps and data for spawning salmon river returns, clear-cut locations and hatchery location retrieved from:

https://salmonexplorer.ca/explore/data/spawner-surveys/e-vancouver-island-mainland-inlets/all-species/all-cus/englishman-river-1106/

[11] Sedimentation of Salmon Spawning Habitat - University of British Columbia blog on Management of Forests for Salmon Survival retrieved from: https://blogs.ubc.ca/mlws/sedimentation-of-salmon-spawning-habitat-2/