Climate change and variability is posing adverse impacts on the economic, social and political livelihoods of mankind. The main objective of this study is to examine the impacts of climate change and variability on agriculture following an agro ecological based approach in the Tan abeles Sub basins (TBSB) of the Upper Blue Nile basin, Ethiopia. specifically this dissertation is designed to characterize the observed climate trends and variability of TBSB during 1979-2015; to classify the sub basins in to agro climatic zones with uniform weather and climate characteristics; to examine the soils and their characteristics in the TBSB; to define the farming systems in the TBSB and to analyze the agro ecosystems in the TBSB and assess the suitability of the zones for agriculture. To achieve these objectives a range of datasets namely observed time series as well as Gridded climate (Rainfall and Temperature) data, soil type and soil properties, terrain parameters (Altitude, slope and aspect), crops data and other agro ecological parameters are collected from global and local servers. The observed rainfall and temperature data is collected from the National Meteorology Agency of Ethiopia (NMA) and the gridded data was obtained from the WorldClim2 data base. The soil data was obtained from the Harmonized World Soil Database (HWSD1). Whereas, the rest of data were collected from regional agricultural and hydrological offices and through interviews and sample surveys .The study used a number of analysis tools such as Arc Gis, Arc Envi, Climpact2, Matlab, R and SPSS; and techniques such as extreme event analysis, homogeneity test, trend test, variability analysis, Agro climate zoning, soil classification, farming classifications and agro ecosystem analysis, to analyze the data. The climate trend and variability analysis for the study period of 1979-2015 reveals rainfall in the TBSB shows no clear trend however it shows high variability. The amount of variability in rainfall was 44%, 44%, 10% and 10% during Bega, Belg, Kiremt and the annual times respectively. The extreme rainfall indices also show trends like the mean changes; when averaged over the TBSB among the selected indices only the maximum 1-day precipitation (Rx1day), show a negative trend which is not significant. Whereas, the number of very heavy rain days (R20mm), Simple Daily Intensity Index (SDII), very wet days exceeding the 95th percentile (R95p) and annual total wet days precipitation (RCPTOT); shows a significant vi increasing trend. The amount of significant increase in PRCPTOT is about 4.15 mm/yr. Similarly the mean annual rainfall shows a significant increase of 5.01 mm/yr. On the other hand the mean temperature among its three variables (Tmax, Tmin and Tavg) shows a significant increasing trend except the Tmin which was significant for only Kiremt. The amount of increase in Tavg is 0.02, 0.04, 0.03 and 0.03oC/yr for Bega, Belg, Kiremt and annual period's respectively. The extreme temperature evnts analysis also reveals negative trends of cold temperature events such as; cool days (TX10p), cool nights (TN10p), and Cold Spell Duration Indicator (CSDI) at many parts of the TBSB. Conversely, marked warm temperature events such as; Hotest day (Txx), warm days (TX90p), warm nights (TN90p) and Warm Spell Duration Indicator (WSDI), show a positive trend during the study period. The frequency of cold days/nights had reduced by 0.325/0.381 days/yr; whereas the frequency of warm days/nights has increased by 0.911/0.830 days/yr. We also delineated the TBSB in to nine agro climate zones with homogenous climate units appropriate for agricultural practices. Generally the Moist Weina Dega, Wet kola and Wet Weina Dega are found to be the dominant zones covering 33.85%, 30.90% and 17.31% of the TBSB total area while the Wet alpine Wurch (0.03%) is the smallest zone. moreover the results reveal that a range of crops such as maize, sorghum, teff, wheat and barley can adapt in the dominant agro climate zones where as the highlands above 3700m are no known for a specific crop, rather the use as protected forest areas or grazing lands for animals. In addition to these based on the layers of Agro climate zones, soils types, terrain, farming systems in the area we classified the study area in to seven agro ecosystems Moist Lowlands (AES1), Moist Midlands (AES2), Moist Highlands (AES3), Sub Humid Lowlands (AES4), Sub Humid Midlands (AES5), Sub Moist Mid Lands (AES 6) and Sub Moist Highlands (AES7); AES2 is the dominant agro ecosystem covering an about 50% area of the TBSB, while AES7 covers only 0.2 % of the total area of the TBSB. The study has also classified the suitability of each agro ecosystem for agriculture; hence, the Moist Midlands (AES2) is highly suitable, the Sub Humid Lowlands (AES4) and Sub Humid Midlands (AES5) are moderately suitable, the Moist Lowlands (AES1), Moist Highlands (AES3) and Sub Humid Midlands (AES6) are marginally suitable and the Sub Moist Highlands (AES7) is unsuitable. Generally The climate analysis in the study area shows an increasing trend of temperature where as rainfall shows high variability, hence, it is recommended to intervene the actual farming system by cropping short term and drought resistant crops (to tolerate the temperature stress). Key Words: Trend test; climate extremes; climate variability; Agro climate zonation; Soil properties; Farming systems; Agro ecosystem anal