E-Weather™ Service Modeling Information

SkyBit uses a dynamical-statistical approach to generate the E-Weather™ products. SkyBit's numerical modeling system first takes advantage of atmospheric output from global and regional models located in government centers, such as the U.S. National Weather Service's National Center for Environmental Prediction (NCEP). The global models have an effective grid spacing of 80 km or larger, while regional models utilize a smaller grid cell size (typically about 30 km). Even the finest-scale government model does not have a small enough grid resolution for site-specific forecasting.

Grid-point output from government models serves as the "first guess" input for SkyBit's "mesoscale" models (Figure 1). The grid resolution for the mesoscale model is between 1 km and 10 km. As part of the model's data preparation, the grid-point output values are combined with surface data from in-house archives (i.e. terrain, elevations, and land use) and real-time atmospheric observations (surface and upper air observations). The resultant output from the data preparation module serve as the necessary input data sets, which specify values for the initial and boundary conditions of the forecast domain (target area).

Figure 1. SkyBit Modeling Diagram.

After receiving this input, the physics-based mesoscale model is executed and its output is a prediction of the future state of the atmosphere. These predictions are still too crude for site-specific products due to the spatial resolution of the input data, inadequacies of the model physics, and computational limitations. Therefore, it is necessary to employ additional models to optimize the quality of the site-specific forecast required for specific applications, such as plant disease forecasting.

These models can take the form of physics-based "canopy" models that account for the physical details of a vegetated surface, or "statistical" models, which empirically predict local surface conditions (Figure 1). The statistical models, unlike the physics-based models, require onsite data collection in order to develop the statistical relationships for a specific location. The output from either the canopy or statistical model is then used to drive plant disease models, which in turn, provide the forecasts for the SkyBit's E-Weather™ products.