Search

Estimating space-time variability of precipitation is an important task in East Africa, considering the observed increased frequency of extreme events, drought episodes in particular. These events deeply affect the population with implications on agriculture and consequently food security. Daily accumulated precipitation time series from satellite retrieval algorithms, ARC, CHIRPS, TAMSAT, TMPA-3B42, and CMORPH are exploited to study the spatial and temporal variability of East Africa (EA - 5°S-20°N, 28°E-52°E) precipitation during last decades (1983-2013 for ARC, CHIRPS, and TAMSAT; 1998-2014 for TMPA-3B42 and CMORPH). The analysis is carried out by directly investigating the precipitation time series and computing the time series of the joint CCI/CLIVAR/JCOMM Expert Team on Climate Change Detection and Indices (ETCCDI, http://etccdi.pacificclimate.org/index.shtml), i.e. RX1day, RX5day, CDD, CWD, SDII, PRCPTOT, R10, and R20, at the yearly and seasonal scales. The purpose is to identify the occurrence of extreme events (droughts and floods), and extract precipitation spatial patterns of variation by trend analysis (Mann-Kendall technique). Prior to the analysis satellite time series are checked for the possible presence of inhomogeneities due to variations in rain gauge density and/or in the satellite retrieval algorithms. Preliminary results relative to the trend analysis of the yearly precipitation indices reveal a consensus of the different satellite products on a positive trend of the annual total precipitation (PRCPTOT) over Somalia and East Ethiopia. More problematic is the situation over West Ethiopia, Sudan, and Uganda, where is not possible to converge to a common type of trend (e.g., negative trend for CMORPH, ARC, and TMPA-3B42 and positive trend for TAMSAT and CHIRPS over west Ethiopia). Similar behaviors are found for other indices (e.g., SDII - average precipitation on wet days over a year, and R1 - annual number of precipitating days). As for CDD (maximum number of consecutive dry days) ARC, CMORPH, and TAMSAT agree on a negative trend over Somalia, East Ethiopia, and most of Kenya. CHIRPS exhibits a predominance of weak positive trends almost over all EA, whereas the contrary is true for TAMSAT. For this parameter only very limited areas are characterized by high significance levels of the trends.