Estimating low-frequency variability and trends in atmospheric temperature using ERA-Interim (Revised 2014)

Estimating low-frequency variability and trends in atmospheric temperature using ERA-Interim (Revised 2014)
Date Published
ERA Report Series
Document Number
P. Poli
D.P. Dee
P. Berrisford
Hans Hersbach
S. Kobayashi
C. Peubey
Event Series/Collection
ERA Report
Abstract Low-frequency variability and trends in temperature from 1979 to 2012 are examined. Observational improvements are noted and near-surface behaviour of the ERA-Interim reanalysis is reviewed. Attention is then focussed on how closely ERA-Interim fits the upper-air data it assimilates, the bias adjustments it infers for satellite data, and its agreement with the ERA-40 and MERRA reanalyses and with longer-term model simulations. Global-mean fits to independently homogenised radiosonde temperatures and variationally adjusted satellite brightness temperatures vary mainly within about 0.1K over the troposphere, with some degradation over time from assimilating varying amounts of aircraft and rain-affected microwave-radiance data, and from a change in source of sea-surface-temperature analysis. Lower-tropospheric warming appears to be slightly underestimated. Temperature variations in the tropical upper troposphere match those at the surface, but have more than twice the amplitude, in agreement with modelling. Specific humidity varies in concert; relative humidity is largely uniform, but dips during El Niño events. Agreement among the reanalyses and with model simulations is particularly close over the lower stratosphere, where radiance data and the background model constrain analyses to cool slightly more slowly than indicated by the homogenised radiosonde data. Global means are perturbed by at most 0.2K due to underestimation of warming following the El Chichón and Pinatubo volcanic eruptions and by assimilating recent GPS radio occultation data. Middle-stratospheric variations are more uncertain, but recent cooling appears to have been underestimated by assimilating increasing amounts of unadjusted radiosonde data. Strong upper-stratospheric cooling in agreement with model simulations is analysed despite occasional jumps due to unadjusted bias changes in high-sounding satellite data.