Towards Improving Hurricane Intensity Forecasts

Florida State University, Tallahassee FL

Investigators

Abstract

This study is centered around a suite of mesoscale models that are to be used for research on multi-model superensemble for the improvement of hurricane intensity forecasts. This research complements prior work on multi-model superensemble for hurricane tracks and intensity where a suite of large-scale models was used. The prior research demonstrated that the superensemble-based intensity forecasts are somewhat (~20%) superior to all of the participating member models. Since the intensity of a hurricane is expected to have mesoscale signatures, the notion that a suite of mesoscale models may prove to be very useful in this regard is being explored in this research. The recent multi-aircraft reconnaissance flights and the development of newer generation mesoscale models provide a unique opportunity for carrying out such experimentation. The component areas of science in this endeavor include: i) data assimilation within global and regional models, ii) prediction experiments from a suite of mesoscale models, iii) definition of the training and forecast phase of a mesoscale multi-model superensemble and iv) execution of forecasts for these phases. The intent is to examine superensemble based forecast validations and interpretations of model biases towards hurricane intensity predictions. Broader Impacts: Being able to provide improved hurricane intensity forecasts has major societal impacts. Hurricane Katrina is an excellent (however unfortunate) example where the hurricane modeling community faced a major challenge. Although atmospheric scientists cannot prevent the loss of life and property, they can work towards improving advance warning capabilities from improved modeling. The Principal Investigator will continue his long-standing cooperation with forecasters and researchers at the National Hurricane Center.