First Advisor

Scott Wells

Date of Publication


Document Type


Degree Name

Master of Science in Civil Engineering (MSCE)


Civil Engineering


Limnology -- Washington (State) -- Coldwater Lake -- Computer simulation, Water quality -- Washington (State) -- Coldwater Lake -- Computer simulation



Physical Description

1 online resource (3, xi, 107 p.)


Coldwater Lake is a new lake formed when a massive mudflow down the Toutle River Valley, caused by the eruption of Mount St. Helens on May 18, 1980, blocked the natural outlet of Coldwater Creek. This research utilizes physical, biological and chemical data collected at Coldwater Lake during the sun1mers of 1989 and 1990 to calibrate and verify the one-dimensional computer models CE-THERMRl and CE-QUAL-Rl for Coldwater Lake. CE-THERM-R1 was used to simulate thermal characteristics in Coldwater Lake during the summer stratification periods of 1989 and 1990. The model was calibrated to 1989 data and was verified with 1990 data. The model performed well with respect to typical stratification features such as depth and temperature of the epilimnion, gradient of the thermocline and temperature of the hypolimnion. The 1990 verification simulation indicated a lack of heat in the epilimnion and metalimnion towards the end of the summer. This is thought to be a product of inaccurate cloud cover data. Model simulations predicted vertical eddy diffusion coefficients (Ez) throughout the water column. These were compared to Ez values in the hypolimnion calculated from temperature data collected by Kelly (1991). Model simulated Ez values in the hypolimnion were near molecular diffusion while field calculated values were one to two orders of magnitude greater than molecular diffusion. The model simulation assumed no lake inflow or outflow so the hypolimnion was more stable than the natural system. The amount of photosynthetically available radiation (PAR) to Coldwater Lake was determined from output derived from the model simulation. This will be useful in determining primary productivity within the lake. CE-QUAL-R1 was used to simulate water quality in Coldwater Lake. The model was calibrated using dissolved oxygen (D.O.) data collected in 1989. The model adequately predicted the D.O. profile in the hypolimnion but tended to over predict D.O. concentration in the epilimnion by 1.0- 3.0 mg/1. This may be caused by an under estimation of the vertical diffusion coefficient in the model simulation. Mean phytoplankton concentrations were similar to field data in the surface layer assuming a 1 mg/1 phytoplankton to 10 ug/1 chlorophyll a ratio. However, concentrations at 10 m and 20 m were under predicted. The phytoplankton - chlorophyll a comparison may not be valid for these lower regions because a significant portion of algal cells within this region are non-viable and are found as particulate detritus in various stages of decomposition. Model simulated nutrient concentrations were in good agreement with the field data. N03 concentration in the hypolimnion increased slightly throughout the model simulation due to decay of the assumed initial condition of 2 mg/1 refractory dissolved organic matter which was contributed to the lake during the eruption of Mt. Saint Helens on May 18, 1980. The model simulation predicted that the phytoplankton were nitrogen limited. An analysis of potential changes in water clarity and water temperature due to the addition of fish in the summer of 1989 was made. The 1990 model simulation predicted that the addition of 200 kg/hectare of Rainbow Trout had little change on water temperature and water clarity. Collection of further site specific data such as; cloud cover estimates, primary productivity rates and quantity of dissolved organic matter contributed by the May 18, 1980 eruption of Mt. Saint Helens would be useful in building a reliable water quality model at Coldwater Lake.


If you are the rightful copyright holder of this dissertation or thesis and wish to have it removed from the Open Access Collection, please submit a request to and include clear identification of the work, preferably with URL

Persistent Identifier