|M.Sc Student||Nussboim Shulamit|
|Subject||Using the Kalman Filter for heat storage and evaporation|
estimation in Lake Kinneret
|Department||Department of Civil and Environmental Engineering||Supervisors||ASSOCIATE PROF. David Broday|
|DR. Alon Rimer|
Information available from short-time measurements of a lake's water temperature profile reveals large fluctuations. Hence, such measurements are difficult to use when calculating surface fluxes and heat balance at the lake surface. The advantages of using such measurements as additional information to model predictions of surface fluxes can be revealed however by using the Kalman Filter (KF) algorithm. In order to estimate the current value of variables of interest, e.g. evaporation rates and heat storage change, this algorithm uses the measured time series of heat according to lake temperature, statistics of system’s noise at measurement and the uncertainty in the dynamics of the heat balance model. The Kalman Filter procedure is applied here for studying the energy balance at the Lake Kinneret surface at high temporal resolution (time steps of 10 min). We demonstrate the KF operation using different algorithms for the surface fluxes, and examine its performance in light of the seasonal variations of disturbances associated with meteorological and lake temperature conditions. It was found that for the spring and summer, when the lake is more dynamic, the weighting factor of the measurements correction, the Kalman filter gain, K, becames low, which indicates high noise level of the heat storage data. On the other hand, during calm lake conditions in the autumn the gain became high. The proposed method is especially suitable for cases where information on high temporally resolved evaporation fluxes is required online.