Comparing ERA5 reanalysis and in situ wind databases to characterize windstorms causing Tehuelche scallop beach strandings in San José gulf, Patagonia.

Abstract

Windstorm patterns associated with beach strandings of the commercially important Tehuelche scallop, Aequipecten tehuelchus, in San José Gulf, Patagonia, were analyzed to understand the windstorms recurrence given their potential impact on the dynamics of the resource. Although the phenomenon of scallop strandings has been recognized for a long time, the lack of environmental records has made it difficult to thoroughly study the meteorological conditions that trigger these events. The availability of reanalysis data has the potential to address this data gap; however, its feasibility must first be validated. To address this, the first step was to characterize and compare hourly wind intensity (m/s) and direction (degrees) data from the fifth-generation global climate reanalysis (ERA5), provided by the European Centre for Medium-Range Weather Forecasts, with in situ wind data for the 2013-2020 period. The study focused on wind conditions during reported Tehuelche scallop strandings along the northern coast of the gulf. Subsequently, a historical search for southerly windstorms with characteristics similar to those responsible for past strandings was conducted using the ERA5 database, covering the period from 1950 to 2023. Overall, ERA5 accurately captured the dates, durations, intensities, and directions of southerly windstorms at high temporal resolution (hourly), though with a slight tendency to underestimate wind intensities of higher winds. The estimated mean annual frequency of ERA5-detected southerly windstorms with intensities higher than 11 m/s was 5.4 storms per year, with durations ranging from 12 to 74 h. These windstorms can occur throughout the year, with an increased frequency and duration during winter. ERA5 proved to be a valuable tool for characterizing windstorms, complementing in situ data and providing insight into past stranding events where no observational data were available.