For Africa's developing countries the agricultural system is among the most vulnerable due to extensive use of rainfed crop production, presence of droughts and floods that affect crops as well as initial poverty of population that limits the capacity to adapt. In this study were realized the analysis of long-term rainfall data and its impact on main crop products in Rwanda. Some rainfall data was infilled for the period of 1926-2013. It was done using the monitoring data of a neighbor weather station with relatively the same elevation above sea level and with a monitoring record of no less than 40 years. The neighboring station with the best correlation was selected for the infilling. The missing rainfall data was infilled for all the stations with resulting regression coefficients ranging from 0.55 to 0.80. This indicates the acceptability of the performed regression. Also were constructed different-cumulative curves of rainfall and sort out cycles of decline and increment of rainfall. Similar different-cumulative curves were constructed for main crops in Rwanda. Correlation and regression analysis were used to determine the relationship between rainfall, arable land expansion, fertilizer use and crop yield. Particularly for Rwandan conditions, the rainfall variations are determinant for the crop yield increment. The intensification of extreme flood's and, as rule, flooding of agricultural lands in connection with rainfall augmentation was also allocated.

Analysis of the climate parameters over Tanzania were done for period 57-61 years. Rainfall and wind speed had greater variability than air temperatures. Positive significant trend of dry air temperature, maximal and minimal air temperatures were detected almost for all weather stations. As well positive trends of wind speed were detected just for stations Dodoma, Arusha, Mtwara and Mwanza. While rainfall trend analysis just for station Mtwara showed negative significant trend. In absolute values wind speed increased by 0.7-1.1 m/s, rainfall decreased by 89.3-136.1 mm, dry air temperature rise up 0.5-1.0 °C, maximal and minimal air temperatures values have grown by 2.0-3.6 °C.

The flow rate is one of the most important elements of the river’s water regime. Establishing patterns in river flow rates’ changes within a year and by season is of great importance for the hydrological forecasts for spring floods, rain floods and hydrological calculations. By now, there are few publications on this area, both in Russia and abroad. The initial data were divided into 5 hydrological seasons – winter and summer low water, rise and fall of spring floods and autumn rain floods. For each season, an analysis of river flow rate over a period (1936-2013) was done. The average flow rate in rivers varied greatly depending on the water regime’ phase and the type of terrain in which the river flows. For all seasons, the amplitude of fluctuations in the flow rate on the plain rivers is less than on the mountain ones. So, in mountain rivers, flow rates vary from 0.04 to 2.42 m/s, and on flat rivers from 0.02 to 1.43 m/s. As well the pairwise and multiple correlation coefficients of flow rates for each morphometric and catchment characteristics were calculated. The highest correlation coefficients were obtained for the dependency between the average flow rate and the watershed elevation (0.30-0.75) and the watershed slope (0.42-0.76). The multiple correlation coefficients between the average flow rate and all catchment’s morphometric characteristics varied from 0.97 to 0.99. Results verification on an independent data demonstrated that the relative error was on average 2%.

Every year flooding of Songwe River has brought severe problems in the lower part of its catchment located in Kyela District. Floods lead to loss of people’s lives, destructions of infrastructures and properties. This study made floods analysis and short-term prediction of water stages. The study used rainfall data of Tenende weather station and water levels data at Kasumulu gauge station for identification of the flood patterns. Floods were grouped into three categories: floods with single peak, floods with several peaks and sprawled floods. The study recommends for the introduction and implementation of flood prediction practices, flood policy, and flood fight education, as well as continuous training of local residents on the best suitable traditional and modern ways of flood management practices and how to mitigate the floods effect.