Abstract
<title>Abstract</title> <p> A major variable used in the irrigation scheduling process and in water resource management is reference evapotranspiration (ET₀). Most of the ET₀ estimates are based on the FAO-56 Penman Monteith method which is generally non-linear in meteorological variables, which may introduce systematic bias due to its non-linear response to meteorological inputs.. As more satellite-based meteorological data has been used, including NASA POWER, it has been necessary to know the implications of using hourly versus daily data. ET was estimated in this study based on hourly meteorological measurements that were summed up over 24 hours and mean meteorological measurements that had been calculated using NASA POWER data on a semi-arid agro-climatic area. The seasonal differences in hourly-aggregated and daily-computed ET were examined and the implication of this on the estimation of irrigation water requirement evaluated. Root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R <sup>2</sup> ) were used to assess the performance of the statistics. Daily-computed ET₀ systematically exceeded hourly-aggregated ET₀, with a mean bias of 1.2055 mm day⁻¹ (RMSE = 1.5497 mm day⁻¹). Correlation analysis revealed strongest association of aggregation bias with relative humidity (r = − 0.8337) and solar radiation (r = 0.7133), indicating energy-driven amplification under dry atmospheric conditions. The aggregation bias was worst in the high-energy periods, and the cumulative overestimation was high in the summer of 165.6 mm. The difference was greatly affected by the day-night radiation levels of the sun and wind speed. These results indicate that the Daily aggregation smooths peak hourly evaporative demand, but overall results indicate systematic overestimation compared to hourly-aggregated ET₀, which may influence irrigation decision-making. The paper emphasises the role of the time resolution in ET₀ determination to use meteorological data retrieved by satellites. </p>