Ethiopia lies within the tropics, about 30N to 150N and 330E to 48E. The climate of Ethiopia is very much the reflection of its diversified topography and its location in the tropics. Climate seasons of the country are mainly controlled by the annual migration of the ITCZ and associated atmospheric circulation, which are modulated by the complex topography of the region. Based on existing studies and methodologies about five climate classifications have been developed for Ethiopia, namely: traditional, Koppen’s, Thronthwaite’s, rainfall regimes, and agro climatic zones. These studies have clearly identified the reach and diversified climatic resource of the country. The basis for any kind of climate related study is observational data obtained from meteorological stations network in the country, which began to take some shape after 1943. However, the distribution of the meteorological stations’ network in Ethiopia has never been homogeneous in space and time. This has resulted in a non-homogeneous meteorological data set for the country. In this study attempt has been made to address the in homogeneity in the rainfall dataset using the robust statistical technique, called kriging, Climatological Normal’s for the period 1961-90 determined from the kriging analysis for Rainfall Regimes A, which covers Northeast, Central, southern and eastern highlands of Ethiopia, are plotted in a 10x10 grid lattice for all months of the year. The results indicate that estimates from kriging analysis agree well with climatology for a substantial portion of regime A for wet seasons (spring and summer). However, substantial errors in estimation are evident in north east corner of the region during all months, and also for the whole regime A during the dry months by when relative speaking there is a little or no rainfall. Comparison has also been made with point rain gauge data for Addis Ababa Observatory (9.02N, 38.46E, altitude 2304 meters above sea level) and Gewane (40.63E, 10.15N, altitude 605 meters above sea level) with corresponding grid box results, the kriging outputs have captured well the inter annual and inter seasonal patterns over those stations. Attempt also been done to verify the results of the kriging analysis by comparing it with the ERA-40 reanalysis dataset. In general ERA-40 has captured the main future of the inter annual variability in most months in of regime A. However, in terms of rainfall amounts ERA-40 underestimates during the dry months (October to January) and over estimates during the wet months (March to September). Still ERA-40 has a better representation of the rainfall amount estimates after 1980 as compared to the period 1961-1979. With the availability of rain gauge data from more number of stations, comparison of the results of the block kriging with ERA-40 reanalysis datasets indicates that there is a good potential to verify the accuracy of the ERA-40 output and other related global rainfall datasets. The results of this analysis are promising to continue work for remaining parts of Ethiopia.
Key words: Ethiopia, rainfall, rainfall regimes, climate classification, kriging.