Journal of Environment and Earth Science www.iiste.org ISSN 2224-3216 (Paper) ISSN 2225-0948 (Online) Vol. 3, No.9, 2013 84 Rainfall Occurrences, Precipitation Deficit and Surplus Analysis for Rainfall Water Harvesting and Management in the Central highlands of Ethiopia, the Case of Bishoftu District, Oromia Region 1 Fitsume Yemenu (Corresponding author) Ethiopian Institute of Agricultural Research , Debre Zeit Center PO box 32, Debre Zeit, Ethiopia E-mail: fitsum_yimenu@yahoo.com Desalegn Chemeda 2 School of Civil Engineering and Built Environment; Central University of Technology, Free State (CUT); 20 Pres Brand Street, Private Bag X20539; Bloemfontein 9300 This research is finance by the Ethiopian Institute of Agricultural Research Abstract Rainfall remains the crucial component of the weather elements for improving agricultural yield in Ethiopia. Rainfall occurrence analysis is extremely helpful in planning of water resources and agricultural development. A study was conducted to asses the potential of sufficient rainfall occurrences and precipitation surplus and deficits in the central highland of Ethiopia for a selected district based on thirty three years of weather record data. The FAO(1978) and Reddy (1990) models were employed to set the threshold limits and the Weilbul frequency formula was used to calculate the probability of occurrences during the two growing seasons, belg(shorter) and kiremt (main). The results showed that the probability of occurrences of the sufficient amount of rainfall during the decades of main rainy season is promisingly stable while belg is observed to suffer from fewer occurrences of the sufficient amount even at the lower probability levels (25% probability of occurrences). Thus rainfall water harvesting during the main rainy season (Kiremt) is promising either for double cropping practices or other domestic uses. Key words: Frequency, Weibull, belg, kiremt 1. Introduction Rainfall is the dominant source of water for agriculture through out Ethiopia in general and the central highlands of Ethiopian in particular. Recent studies reported that the rianfed agriculture is still contributing more than more than 50% to GDP and about 60% to foreign exchange earnings and provides livelihood to more than 85% of the population (Goodswill etal., 2007). But the rainfall variability, unreliable occurrences in sufficient amount and delay in onset dates are reducing the crop yields with reasonable amount almost in all parts of the country. Contrary to those facts, large amount of increase in crop yields is expected to come from the rainfed agriculture to fill the food demands of the world as whole and the country in particular. This is because there are limited land resources to bring in to medium and large scale irrigation projects and also the investment required for the same is too big (WDR, 2007. Thus improving the rainfall water management for sustainable crop production remains the question to address and the priority to be taken to least maintain the optimum level of crop yields. Some studies conducted in sub-Saharan Africa indicated that there is a potential opportunity for increasing crop yields in the region. An assessment made by Baron (2004) showed that the cereal crop yields could reach as high as 3.5 t ha -1 against the existing yield 1 t ha -1 yield estimated by Rockstrom (2003). This wide gap suggests that there is an enormous opportunity to raise crop yield from rainfed agriculture. This is entirely linked to focusing attention on maximizing yield per unit of water. But this requires of adapting to the prevailing climate variability and rainfall occurrences. Appropriate quantification methods of the historic climatic data have been very helpful for planning in crop production and thereby improving the desired achievements, increasing the yield per unit water (Mersha, 2005). Analysis of Rainfall probability occurrences is becoming an important work especially in water resources sector and risk associated rainfed agriculture (Sayang et.al., 2008). Integration of the rainfall occurrences with the crop water requirement is observed to be more practical and sound for many of crop rainfall water management activities. Different researchers and organizations defined some threshold values of rainfall or water amount for the crop to grow and perform well in a growing season. Accordingly FAO (1978, 1998) developed simple soil water balance model that relates the minimum threshold value of rainfall with potentail evapotranspiration. In this model, the rainfall amount in a certain growing season should at least be equal half of the potential evapotranspirtation of the place the crop is to grow. This is from the plant physiological point of view in that the