Advances in data assimilation to improve the quality of weather and environmental forecasts

点击翻译按钮获取中文摘要

Data assimilation is concerned with using information from all available observations obtained from a variety of instruments to estimate the current state of the atmosphere to improve the quality of weather forecasts. Based on statistical estimation principles, an a priori estimate of the atmosphere is provided by a short term forecast obtained from a numerical weather prediction (NWP) model. Its accuracy is represented by the background-error covariances, which depend on the meteorological situation. In cases associated with the onset of atmospheric instability, the initial conditions need to embed the dynamical structure that can trigger the development of a weather system. This proposal is concerned in making advances in data assimilation by introducing flow-dependent structures within the assimilation process that are in a position to improve the quality of forecasts of high impact weather events in the short, medium and long range. This will also be applied to adaptive data selection to improve the use of the vast amount of satellite data received daily at operational weather centres.