The influence of horizontal resolution and boundary forcing in simulating hurricanes over the South Atlantic Ocean using WRF

A hurricane is a threat to socio-economic activities in coastal communities bordering the South Atlantic Ocean (SAO). Hurricanes rarely form over this region and as such these communities are not prepared for them. Previous studies have suggested that anthropogenic warming may lead to more frequent hurricanes over the region and have demonstrated the capability of the Weather Research and Forecasting model (WRF) in capturing the impacts of the warming on hurricanes. However, none of the studies have investigated how the model's horizontal resolution and boundary forcing could alter the characteristics of simulated hurricanes. The present study used WRF to perform a series of experiments to simulate two hurricanes (Hurricane Catarina and Hurricane Anita) over the SAO at three horizontal resolutions (3.3 km, 10 km, and 30 km), using two reanalysis datasets (ERA-Interim (hereafter ERAINT) and NCEP CFSR (hereafter CFSR)) as the boundary forcing data. The performances of the reanalysis and WRF are compared with observational data from the International Best Track and Archive for Climate Stewardship. The results show that both reanalyses datasets give a good representation of the two hurricanes, but they grossly underestimate the intensity thereof. CFSR gives a better representation than that of ERAINT. However, both reanalyses also suggest that the South Atlantic Convergence Zone may be the moisture belt for hurricane formation over the SAO. WRF gives a credible simulation of the hurricanes. In simulating Hurricane Catarina, WRF performs best at a 10-km resolution; but in reproducing Hurricane Anita, the model performs best at a 3.3 km resolution. For both cases, the model performs better when forced with ERAINT than with CFSR. Hence, the study shows that increasing the resolution of the model may not necessarily improve the simulated hurricane over the SAO, and that, the quality of the simulated hurricane depends on the dataset that provides the boundary forcing. The results of the study have improved the understanding of hurricane characteristics in the SAO, and have shown the potentials of WRF to forecast and project future events as well as for downscaling the potential impacts of future climate change on hurricanes over the SAO.

Grantor dc:publisher.institution

Department of Environmental and Geographical Science

Year dc:date.issued

2017

Author dc:creator

• Bluff, Gemma Kendall Pelton

Advisor dc:contributor.advisor

• Abiodun, Babatunde Joseph

Language dc:language.iso

eng

Handle dc:identifier.uri

http://hdl.handle.net/11427/25188

OAI identifier oai:identifier

oai:open.uct.ac.za:11427/25188