Background
During the 2009 circulation model validation exercise it was noticed that there was a difference between the observed and modeled salinity. The most likely source of the error came from the freshwater input used by the circulation model. This led to an effort to improve the hydrological (freshwater) model.
Methods
Two new hydrological models were developed. One a coarse-resolution model that covered the Gulf of Alaska watershed. And a second high- resolution model that included additional transport processes. The models were validated through the collection of existing meteorological data and the collection of additional snowpack and stream flow measurements. Snowpack and glacial information were used to validate the snow and ice melt portions of the model. Stream flow data was collected on streams in glacial, snow and rain, and groundwater dominated systems and used to validate the model output.
Results
The high-resolution model was able to accurately predict the freshwater runoff in individual watersheds. We were able to identify the meteorological models that provided the inputs that best matched the observed ice and snow loss, as well as the measured streamflow. The model has also been integrated with the ocean circulation model used in PWS and there is better agreement between observed and modeled salinity levels. This is important because salinity helps drive the surface circulation that would transport an oil spill.