Factors influencing the skill of synthesized satellite wind products in the tropical Pacific

Abstract

Given the importance of tropical Pacific winds to global climate, it is interesting to examine differences in the mean and trend among various wind products, and their implications for ocean circulation. Past analysis has revealed that despite the assimilation of observational data, there remain large differences among reanalysis products. Thus, here we examine if satellite‐based synthesis products may provide more consistent estimate than reanalysis. Reanalysis product winds are, however, typically used as a background constraint in constructing the synthesis products to fill spatiotemporal gaps and to deal with satellite wind direction ambiguity. Our study identified two important factors that influence both the mean and trends from synthesized wind products. First, the choice of background wind product in synthesized satellite wind products affects the mean and long‐term trends, which has implications for simulations of ocean circulation, sea level, and presumably SST. Second, we identify a clear need for developing a better understanding of, and correcting differences between in situ observations of absolute winds with the satellite‐derived relative winds prior to synthesizing. This correction requires careful analysis of satellite surface winds with existing colocated in situ measurements of surface winds and currents, and will benefit from near‐surface current observations of the proposed Tropical Pacific Observing System. These results also illustrate the difficulty in independently evaluating the synthesis wind products because the in situ data have been utilized at numerous steps during their development. Addressing these identified issues effectively, will require enhanced collaborations among the wind observation (both satellite and in situ), reanalysis, and synthesis communities.