NPZ Model Validation in Sound Predictions
NPZ Model Validation in Sound PredictionsRob Campbell, Prince William Sound Science Center
Scope of Work:
In July/August of 2009, a model validation exercise will take place in Prince William Sound (PWS), to validate a large scale physical oceanographic model (Regional Ocean Modeling System: ROMS) with observational data. There has also been a Nutrient-Phytoplankton-Zooplankton (NPZ) model embedded within the ROMS model, but no provision has been made to evaluate that model. Most of the parameters in the NPZ model are not specific to the PWS area, and it uncertain how well the model reproduces observations. The objective of this project is to collect a limited amount of biological and biogeochemical data that may be used first as a reality check for the NPZ model, and which may be used in later refinements of the model.
This project addresses the OSRI Understand goal, by providing basic information that may be used to test a model that is being developed to better understand the biological processes in PWS. It also addresses the Respond goal, because one of the reasons that the physical models have been developed is to provide predictions for spill response. A biological model could also be used to develop predictions in the event of a spill.
The evaluation of the NPZ model will be done by comparing model predictions to in situ observations, using profiled instruments and a small number of samples. Profiling instruments (added to a larger instrument package) that will measure oxygen concentrations, nitrate concentrations and in situ chlorophyll fluorescence; water samples will be collected for nutrient analysis (nitrate, phosphate and silicate), extracted chlorophyll, and CHN analysis; net samples will be taken to measure mesozooplankton concentrations.
Ocean models have exploded in complexity and diversity in recent years, and testing of their adherence to reality is important and worthwhile. This work will help in the refining of current models, and inform developments in modeling in the future.