Assessing surface heat fluxes in two new generation atmospheric reanalyses with a decade of buoy measurements at the Kuroshio Extension Observatory (KEO)

Dongxiao Zhang, JISAO/University of Washington and NOAA/PMEL, Seattle, WA, United States, Meghan F Cronin, NOAA Pacific Marine Environmental Laboratory, Seattle, WA, United States, Caihong Wen, NOAA/CPC college park, College Park, MD, United States, Yan Xue, Climate Prediction Center College Park, College Park, MD, United States, Arun Kumar, NOAA/NCEP, Climate Prediction Center, College Park, MD, United States and Dai C McClurg, University of Washington/PMEL NOAA, Seattle, WA, United States
Abstract:
Surface air-sea fluxes are the most direct ocean climate indicators of how the ocean influences climate and weather, and their extremes, and how the atmosphere forces ocean variability. Previous studies have found large biases and uncertainties in the air-sea fluxes from Numerical Weather Prediction (NWP) models. These errors must be identified and reduced in order to make progress with improved weather forecasts and climate projection. NOAA Ocean Climate Station (OCS) buoys measure upper ocean properties and state variables from which air-sea fluxes can be computed. Fluxes from the Kuroshio Extension Observatory (KEO) at 32.3°N, 144.6°E provide a very challenging test for NWP due to the large range of meteorological and oceanic conditions experienced there. In this analysis, air-sea flux assessment at KEO is performed on two new NWP reanalyses, the NCEP’s Climate Forecast System Reanalysis (CFSR) and ECMWF Reanalysis-Interim (ERA-I). Heat fluxes provided by the reanalyses are compared to those computed from OCS data to determine mean bias, RMS errors and correlations. State variables are also compared in a similar way to determine their contribution to the errors. Our assessments at KEO suggest that the two new generation reanalyses significantly improved the representation of surface fluxes when compared to NCEP Reanalysis 1 (NRA1) and 2 (NRA2). Although the annual mean biases of total surface heat flux have significantly reduced, the RMS error of total surface heat flux in CFSR and ERA-I remain large. The main cause of the biases in total heat flux are due to the latent heat flux, while RMS errors are mainly due to latent heat flux and short wave radiation errors in the reanalyses. Both reanalyses overestimate the winds associated with winter storms and underestimate specific humidity in summer. It is however the bulk algorithm that is largely responsible for overestimates of winter heat release from the ocean resulting in a larger bias of annual mean in CFSR.
Back to: Air-Sea Exchange Processes in Western Boundary Current Systems and Marginal Seas: Their Local and Remote Climatic Implications I Posters