Evaluation and intercomparison of clouds, precipitation, and radiation budgets in recent reanalyses using satellite-surface observations

摘要: Atmospheric reanalysis datasets offer a resource for investigating climate processes and extreme events; however, their uncertainties must first be addressed. In this study, we evaluate the five reanalyzed (20CR, CFSR, Era-Interim, JRA-25, MERRA) cloud fraction (CF), precipitation rates (PR), top-of-atmosphere (TOA) surface radiation budgets using satellite observations during period 03/2000–02/2012. Compared to annual averaged CF of 56.7 % from CERES MODIS (CM) four reanalyses underpredict CFs by 1.7–4.6 %, while 20CR overpredicts result 7.4 %. PR Tropical Rainfall Measurement Mission (TRMM) is 3.0 mm/day PRs agree with TRMM within 0.1–0.6 mm/day. The shortwave (SW) longwave (LW) TOA radiative effects (CREtoa) calculated EBAF (CE) are −48.1 27.3 W/m2, respectively, indicating net cooling effect −20.8 W/m2. Of available results, CFSR MERRA CREtoa values CE 1 JRA-25 ~10 W/m2 more negative than result, predominantly due underpredicted magnitude LW warming in reanalysis. A regime metric developed vertical motion field at 500 hPa over oceans. Aptly named “ascent” “descent” regimes, these areas distinguishable characteristic synoptic patterns predominant cloud-types; convective-type clouds marine boundary layer (MBL) stratocumulus clouds. general, overpredicted (underpredicted) ascent (descent) biases often larger descent regime. both regimes; however observed an order those regime, different types exist two regimes. Based upon Radiation Program ground-based CM observations, as well 15 higher Azores site Nauru (70.2 vs. 55.2 %), less SW (~20 %) transmitted surface, (~60 W/m2) emitted back surface. Also, seasonal variations fluxes much smaller site. dichotomy between atmospheric regimes good measure determining which parameterization scheme requires improvement (convective MBL clouds) reanalyses.