The Impact of Weather Model Resolution on the Simulated Hydrodynamics of Lake George, NY, in June 2017

The Jefferson Project at Lake George, NY is a collaborative project between IBM Research, Rensselaer Polytechnic Institute and The FUND for Lake George. A core component of the project is to develop a one-way coupled atmosphere-lake circulation modeling system capable of robust, very high resolution operational forecasts. An important challenge is determining what value is provided by very high resolution weather forecasts driving a hydrodynamic simulation. Towards this end, we present an analysis of a 1-month case study on the impact of the horizontal spatial resolution (dx) of daily weather forecasts on the representation of the hydrodynamics in Lake George, NY. It is hypothesized that for Lake George, which is shallow (< 20 m average depth) and surrounded by complex topography, it is critical to drive the hydrodynamic model with realistic and highly-localized wind fields.

The Weather Research and Forecasting (WRF) model was used to generate daily operational weather forecasts at four horizontal resolutions of dx = 9 km, 3 km, 1 km and 333 m. WRF was one-way coupled with the ocean model, SUNTANS, which has an unstructured grid of dx = 15-40 m, to produce four 1-month simulations using the different resolution weather forcing. To assess the confidence in model outputs, we used observations from three onshore and two offshore weather stations, and from two vertical profilers measuring the lake temperature and chemistry. Discrepancies in atmosphere and lake dynamics in the four simulations have been quantified using a principal component analysis to identify the major modes of the surface wind field over the lake (viz. wind stress) and the water velocity field at several layers. We have also characterized the variation of baroclinic energy flux in the lake, along with other statistical measures of variance in each experiment.