Long term (1901‒2016) temperature based potential evapotranspiration and aridity index analysis for lower eastern region of Kenya

Climate change has caused unreliable precipitation (P) and temperature (T) changes influencing global, regional and local water availability, agricultural production hence food insecurity. Precipitation and temperature are key climatic parameters among others in estimating potential evapotranspiration (PET). The main goal of this paper was to estimate long term (1901‒2016) potential evapotranspiration time series using seven temperature- and two temperature and terrestrial radiation-based methods and compare the values estimates within four counties of Kenya. Data was sought from the World Bank database constructed based on CRU TS (Climatic Research Unit Timeseries). The estimates were computed and Thornthwaite method was used as a reference in comparison with the other methods. The study established that Hamon III and Thornthwaite equations gave lower annual PET estimates than the other methods. For instance, in Makueni county, the two methods gave potential evapotranspiration estimates ranging from 899±49.6 mm/year to 1131±49.6 mm/year and 918±76.3 mm/year to 1274±76.3 mm/year respectively for the whole time period. Jensen-Haise and McGuinness-Borden PET estimates were the highest ranging from 2782±119.9 mm/year to 3343±120.0 mm/year and 2040±70.7 mm/year to 2371±70.7 mm/year in Makueni, respectively. Thornthwaite method depicted an increasing linear trend in all stations. Further, the Aridity Index (in our case P/PET based on the Thornthwaite method) varied from one climatic period (30 years) to the other with a large range of 0.3 to 1.2 in Makueni county. The study also established that Kitui county had the highest estimate values of PET from all methods used. The study concludes that different temperature- and temperature and terrestrial radiation-based methods perform differently and give different PET estimation based on the parameter specific requirement. The study is suitable in selection and comparisons of best temperature-based, temperature and terrestrial radiation methods in PET estimation in the tropical, data scarce regions in Kenya.