TC Satellite Research
Satellite Applications Development: GOES imagery and derived products are being used to improve tropical cyclone (TC) forecasts under the satellite applications developer project supported by the GOES-R program. The current emphasis of this project is to improve statistical tropical cyclone intensity forecast models. Two sub projects are summarized below.
Investigation of high resolution GOES imagery: The National Hurricane Center’s operational Statistical Hurricane Intensity Prediction Scheme (SHIPS) and Logistic Growth Equation Model (LGEM) both use predictors from GOES-16/17/18 IR window channel imagery around TCs. However, the current predictors use very little information about convective asymmetries and only use input every 6 hr. Methods are being developed to use additional information from the IR window channel, including higher temporal resolution and tendencies as well as asymmetric structure, to improve SHIPS and LGEM.
Outflow layer diagnostics from GOES Derived Motion Winds (DMWs): Most of the predictors used in SHIPS and LGEM that measure the tropical cyclone environment are derived from global model fields, such as the Global Forecast System (GFS). To help provide model-independent input, GOES DMWs are being used to develop new diagnostics of the TC outflow layer. The DMWs over 3 h intervals are being composited in a storm-relative coordinate system in the 150-300 hPa layer. The DMW coverage in that layer is usually very good due to the presence of cirrus generated by the deep convection near the TC center and in rainbands, so that a gridded two-dimensional horizontal wind field can be obtained using an objective analysis method.
The figure below shows an example of the DMW coverage for a case from Hurricane Ida (2021) as it approached the Gulf coast, and the corresponding gridded wind field. Several diagnostics such as area-averaged symmetric and asymmetric components of the divergent wind are being calculated. Results show that these new predictors have the potential to improve the statistical intensity forecasts, and an experimental intensity prediction model is being developed to demonstrate the utility of the DMW predictors.