Numerical weather forecast models run on a grid, or regularly spaced group of points that cover the particular area of interest. These models combine weather observations with equations that explain the motions in the atmosphere and ocean. Two types of models are used to make the most accurate forecasts possible. Global models analyze and forecast the weather over the entire planet, but because of the large area, the points on the grid are relatively far apart, making local forecasts difficult. Regional models produce forecasts over a relatively small area of the globe, and because of the limited area, the points on the grid can be close together. Since the weather in the regional model depends in part on what is happening outside the region, the regional models rely on information from the global model. In other words, the current weather in the regional model is reset because the global model is believed to be more accurate. This practice makes it difficult to discover problems that may be present in an individual model, so the models may misuse observations or the equations may contain errors that degrade forecasts without anyone knowing! This study proposes a road map for designing models that avoid this problem. As a proof-of-concept, the authors modified NOAA’s operational regional hurricane model, the Hurricane Weather and Research Forecast (HWRF) model, to work with minimal resetting from the global model. The authors show how careful design of forecast models can allow scientists to assess any deficiencies in models themselves, in the observations, and in forecasting how accurate the forecast is likely to be. This study is timely, as NOAA upgrades older models and implements new ones.
For more information, contact email@example.com. The full text can be found at https://journals.ametsoc.org/view/journals/wefo/aop/WAF-D-20-0204.1/WAF-D-20-0204.1.xml.
All news and articles are copyrighted to the respective authors and/or News Broadcasters. eWeatherNews is an independent Online News Aggregator
Read more from original source here…