| description abstract | AbstractConvection-permitting numerical weather prediction models are a key tool for forecasting tropical cyclone (TC) intensities, intensity changes, and precipitation. The Met Office has been routinely running a regional (4.4-km grid spacing), explicit convection version of its Unified Model (UM) over the Philippines since August 2014, driven by its operational global model. The principal aim of this study is to assess the performance of this model relative to the driving global model. By evaluating over a year?s worth of operational TC forecasts, it is shown that the Philippines regional model offers clear benefits for TC forecasting compared with the Met Office global model. In particular, it provides much improved predictions for the intensities of strong storms (category 3 and above) and can successfully capture some rapid intensification (RI) events, whereas the global model cannot predict RI at all. The spatial location of rainfall within intense TCs is also more skillfully predicted by the regional model, and the statistical distribution of rain rates is closer to that observed. Although the regional model adds value, notable biases are also identified, highlighting areas for future work to develop and improve the model. | |
