Hydrometeorology of the Amazon in ERA-40

The hydrometeorology of the Amazon basin in the ERA-40 reanalysis for 1958–2001 is compared with observations of precipitation, temperature, and streamflow. After 1979, the reanalysis over the Amazon has a small cool bias of the order of −0.35 K, and a small low bias of precipitation of the order of −0.3 mm day−1. In the early years (1958–72), there is a large upward drift in reanalysis precipitation and runoff associated with an upward drift in the atmospheric water vapor in the analysis, and a somewhat smaller downward drift of temperature as precipitation increases. In the presatellite data, there are inhomogeneities in the radiosonde and surface synoptic data, and there were problems with the variational analysis of humidity once satellite radiances were introduced. Approximate bias corrections can be made for precipitation and runoff on an annual basis, but this also removes some of the interannual variability. The reanalysis runoff–precipitation relationship is similar to the observed streamflow–precipitation relation, on an annual water-year basis. Compared to observations, ERA-40 precipitation for the Amazon is low by about 1.3 mm day−1 in the rainy season, and high by a smaller amount in the dry season. The precipitation bias produces a temperature bias in ERA-40 of the opposite sign on the annual time scale. The reanalysis has a small cold temperature bias after 1967, but on an annual time scale it reproduces the interannual variability of the observations. Although the biases in temperature and precipitation in recent decades are small, the difficulties with the analysis of atmospheric water vapor lead to large uncertainty in long-term trends of the water cycle.