Search

Daily time series from the Climate Prediction Center (CPC) Africa Rainfall Climatology version 2.0 (ARC2), Climate Hazards Group InfraRed Precipitation with Stations (CHIRPS) and Tropical Applications of Meteorology using SATellite (TAMSAT) African Rainfall Climatology And Time series version 2 (TARCAT) high-resolution long-term satellite rainfall products are exploited to study the spatial and temporal variability of East Africa (EA, 5S–20N, 28–52E) rainfall between 1983 and 2015. Time series of selected rainfall indices from the joint CCl/CLIVAR/JCOMM Expert Team on Climate Change Detection and Indices are computed at yearly and seasonal scales. Rainfall climatology and spatial patterns of variability are extracted via the analysis of the total rainfall amount (PRCPTOT), the simple daily intensity (SDII), the number of precipitating days (R1), the number of consecutive dry and wet days (CDD and CWD), and the number of very heavy precipitating days (R20). Our results show that the spatial patterns of such trends depend on the selected rainfall product, as much as on the geographic areas characterized by statistically significant trends for a specific rainfall index. Nevertheless, indications of rainfall trends were extracted especially at the seasonal scale. Increasing trends were identified for the October–November–December PRCPTOT, R1, and SDII indices over eastern EA, with the exception of Kenya. In March–April–May, rainfall is decreasing over a large part of EA, as demonstrated by negative trends of PRCPTOT, R1, CWD, and R20, even if a complete convergence of all satellite products is not achieved. This study was supported by the European Union’s Seventh Programme for research, technological development, and demonstration under Grant Agreement 603608 (eartH2Observe).

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.