Wave Probabilities Consistent with Official Tropical Cyclone Forecasts

Abstract

Development of a 12-ft-seas significant wave height ensemble consistent with the official tropical cyclone intensity, track, and wind structure forecasts and their errors from the operational U.S. tropical cyclone forecast centers is described. To generate the significant wave height ensemble, a Monte Carlo wind speed probability algorithm that produces forecast ensemble members is used. These forecast ensemble members, each created from the official forecast and randomly sampled errors from historical official forecast errors, are then created immediately after the official forecast is completed. Of 1000 forecast ensemble members produced by the wind speed algorithm, 128 of them are selected and processed to produce wind input for an ocean surface wave model. The wave model is then run once per realization to produce 128 possible forecasts of significant wave height. Probabilities of significant wave height at critical thresholds can then be computed from the ocean surface wave model–generated significant wave heights. Evaluations of the ensemble are provided in terms of maximum significant wave height and radius of 12-ft significant wave height—two parameters of interest to both U.S. Navy meteorologists and U.S. Navy operators. Ensemble mean errors and biases of maximum significant wave height and radius of 12-ft significant wave height are found to be similar to those of a deterministic version of the same algorithm. Ensemble spreads capture most verifying maximum and radii of 12-ft significant wave heights.

Evaluation of Global Wave Probabilities Consistent with Official Forecasts

“…The WW3TCOFCL ensemble follows the algorithm published in Sampson et al (2016), except that the number of ensemble members has been reduced to 20 (the same number as in the FNMOC operational WW3 ensemble run using NAVGEM ensemble surface winds, hereafter referred to as the WW3NAVGEM ensemble) from 128.The WW3TCOFCL ensemble grid has also been expanded to a global 0.258 3 0.258 grid to match the operational WW3NAVGEM ensemble. These changes are made so that the algorithm adheres to computing and other resource constraints at FNMOC, and so that the algorithm could also be implemented within the current WW3NAVGEM ensemble job instead of as a completely separate algorithm.…”

“…To address this issue, the U.S. Navy's Fleet Numerical Meteorology and Oceanography Center (FNMOC) implemented a deterministic global wave model forecast that uses postprocessed winds from U.S. TC forecast centers as input to WAVEWATCH III (WW3; Tolman 1991;Tolman et al 2002;NCEP 2020). This algorithm is named for the WAVEWATCH III model (WW3) and its input TC winds from the U.S. TC forecast centers (OFCL), thus named WW3TCOFCL (Sampson et al 2013). Faced with deficiencies in both the forcing winds and resolution for forecasting TC generated waves in the Northwest Australian region, the Australian Bureau of Meteorology (Zieger et al 2018;Aijaz et al 2019) designed a postprocessing method that correct wind distribution biases associated with TCs in the NWP model ensembles used to force their high resolution (8 km) wave model.…”

“…Corresponding author: Charles R. Sampson, Buck.Sampson@ nrlmry.navy.mil ensemble and the waves. These consistency and resolution issues are important to operations, and as yet there is no operational global wave model ensemble consistent with U.S. TC forecast center forecasts, wind probabilities associated with TC forecasts (DeMaria et al 2013), and deterministic wave forecasts derived from U.S. forecast center forecasts (Sampson et al 2013).…”

“…This is an intentional design to be consistent with the current NAVGEM global wave model ensemble so that implementation is simplified, extra computational resources are minimal, and the wind postprocessing algorithm can be run independently of the NAVGEM global wave model ensemble. Sampson et al (2016) demonstrated that more ensemble members would be beneficial, but computational restrictions may not allow for expanding the ensemble. NRL has implemented the postprocessing algorithm with the WW3 ensemble, executed in real time for over a year, and gathered runs for this evaluation.…”

A Machine Learning Approach to Modeling Tropical Cyclone Wind Field Uncertainty

“…As a result, considerable resources have been invested in TC risk assessment and warning systems. Risk assessment (e.g., Vickery et al 2009) and real-time probabilistic forecasting (e.g., DeMaria et al 2009DeMaria et al , 2013Sampson et al 2016) typically employ large ensembles of TC simulations in an attempt to sample the full range of potential risks. Monte Carlo methods are often at the core of such systems as they allow for the accounting of uncertainties in the TC track, intensity, and surface wind field structure.…”