Quantifying variance due to temporal and spatial difference between ship and satellite winds

Abstract

Ocean vector winds measured by SeaWinds can be validated with comparison in situ data that are within a certain time and space range to the satellite overpass. The total amount of random observational error is composed of two primary components, which are quantified in this study: the uncertainty associated with the data sets and the uncertainty associated with the temporal and/or spatial difference between two observations. The variance associated with a temporal difference, which can be translated into a spatial difference using Taylor's hypothesis, between two observations is initially examined in an idealized case that includes only Shipboard Automated Meteorological and Oceanographic System (SAMOS) 1 min data. The results show that the amount of variance in wind speed and direction increases as the time difference increases, while the amount of variance in wind speed increases and direction decreases with larger wind speeds. Collocated SeaWinds and SAMOS observations are used to determine the total amount of variance associated with a temporal (equivalent) difference from 0 to 60 min. For combined differences less than 25 min (equivalent) and the selected wind speed bins, the variance associated with the temporal and spatial difference is dominated by small changes in the wind speed distribution, and the sum of the observational errors is approximately 1.0 m2 s-2 (12 deg2) and 1.5 m2 s -2 (10 deg2) for wind speeds between 4 and 7 m s -1 and 7-12 m s-1. For larger combined differences, the observational error variance is no longer the dominant term; therefore, the total variance is seen to gradually increase with increasing time differences. Copyright 2011 by the American Geophysical Union.