Satellite Estimates and Shipboard Observations of Downward Radiative Fluxes at the Ocean Surface

Abstract

Abstract Clouds and the Earth’s Radiant Energy System (CERES) uses a suite of instruments on the Terra and Aqua satellites combined with analyzed weather data and information on surface conditions to estimate surface radiative fluxes. CERES estimates for the Terra satellite were compared with measurements of the surface radiative fluxes collected with the research vessels (RVs) Wecoma and Thomas G. Thompson radiometers for cruises off the Oregon coast undertaken during 2000–03. To assess the shipboard measurements, the radiometer observations were analyzed to identify cloud-free conditions characterized by ∼1–2 h of relatively stable radiative fluxes. Fluxes for the cloud-free conditions were compared with those calculated using profiles of temperature and humidity from analyzed meteorological fields for the times and locations of the measurements and broadband radiative transfer models. For summertime conditions along the Oregon coast, and assuming a marine aerosol having 0.55-μm optical depth of 0.05, modeled and observed values of the shortwave flux agreed to within 1%–2%. Similar comparisons for the downward cloud-free longwave flux were within 1%–3%. This agreement also held for the CERES surface radiative flux estimates with CERES cloud-free fields of view for ocean scenes within 50 km of the ship being compared with 30-min averages of the shipboard measurements centered on the times of the Terra overpass. Using the CERES observations to identify cloud-free conditions for the Wecoma revealed that in some cases the shipboard measurements of the shortwave flux varied erratically. Criteria were adopted to avoid such behavior, yielding periods in which the surface radiative fluxes were reasonably stable for a range of cloud-free and cloudy conditions. With the criteria applied, the absolute magnitude of the mean differences between the shipboard measurements and the CERES estimates for the downward shortwave flux were within 2%, with RMS differences less than 6% within each month of CERES–shipboard matchups. The absolute magnitude of the mean differences for the downward longwave flux was less than 2%, with RMS differences less than 5%.