Confidence and sensitivity study of the OAFlux multisensor synthesis of the global ocean surface vector wind from 1987 onward

Abstract

This study presented an uncertainty assessment of the high-resolution global analysis of dailymean ocean-surface vector winds (1987 onward) by the Objectively Analyzed air-sea Fluxes (OAFlux) project. The time series was synthesized from multiple satellite sensors using a variational approach to find a best fit to input data in a weighted least-squares cost function. The variational framework requires the a priori specification of the weights, or equivalently, the error covariances of input data, which are seldom known. Two key issues were investigated. The first issue examined the specification of the weights for the OAFlux synthesis. This was achieved by designing a set of weight-varying experiments and applying the criteria requiring that the chosen weights should make the best-fit of the cost function be optimal with regard to both input satellite observations and the independent wind time series measurements at 126 buoy locations. The weights thus determined represent an approximation to the error covariances, which inevitably contain a degree of uncertainty. Hence, the second issue addressed the sensitivity of the OAFlux synthesis to the uncertainty in the weight assignments. Weight perturbation experiments were conducted and ensemble statistics were used to estimate the sensitivity. The study showed that the leading sources of uncertainty for the weight selection are high winds (>15 ms 21 ) and heavy rain, which are the conditions that cause divergence in wind retrievals from different sensors. Future technical advancement made in wind retrieval algorithms would be key to further improvement of the multisensory synthesis in events of severe storms.

Characterization of Sea Surface Temperature and Air‐Sea Heat Flux Anomalies Associated With Mesoscale Eddies in the South China Sea

“…The data sets of LHF and SHF and related surface meteorological variables are from the newly developed high‐resolution satellite‐derived analysis by the OAFlux project (Yu, 2019; Yu & Jin, 2014, 2018). Compared to the 1° gridded OAFlux analysis (Yu & Weller, 2007), the 0.25° gridded OAFlux‐HR analysis has made two major improvements.…”

“…It is different from the 1° gridded OAFlux that is obtained from objective synthesis of the output from atmospheric reanalyses. The second improvement is that wind speed and direction are derived simultaneously from 16 satellite sensors including scatterometers and microwave passive radiometers (Yu & Jin, 2014). By comparison, the OAFlux‐1° analysis has wind speed but no wind direction.…”

“…The present study aims to use the newly developed 0.25° gridded satellite‐derived LHF and SHF products (Yu, 2019; Yu & Jin, 2014, 2018) to quantify and characterize the atmospheric thermal response to oceanic eddies via LHF and SHF. The study focuses on developing a statistical characterization of two main mesoscale features: the pattern and magnitude of LHF and SHF anomalies associated with the eddy‐induced SSTA in the SCS.…”

Evaluating and Extending the Ocean Wind Climate Data Record

“…Typically, the solution method to find the minimizing analysis is based on the conjugate gradient approach. Using these techniques, several institutions have produced gridded wind products, which include the cross-calibrated multiplatform (CCMP, [66], the ERA* (a specialized version of ECMWF reanalysis, [15]), and the OAFlux product [69], [70].…”

Evaluation of Satellite Surface Winds in Relation to Weather Regimes Over the Indian Ocean Using DYNAMO Observations

“…In situ surface wind measurements are needed to validate and calibrate the satellite wind retrievals. Moored buoy observations have been extensively used to characterize the overall global errors and biases associated with satellite surface wind estimates and derived surface wind products over the ocean (Bentamy et al, ; Yu & Jin, , ). However, the global and/or long‐term mean statistics tend to de‐emphasize the importance of impactful weather systems.…”