Trends, sensitivity and estimation of daily reference evapotranspiration ET0 using limited climate data: regional focus on Burkina Faso in the West African Sahel

• Published in: Theoretical and applied climatology Vol. 153; no. 1-2; pp. 947 - 974
• Main Authors: Yonaba, Roland, Tazen, Fowé, Cissé, Mansourou, Mounirou, Lawani Adjadi, Belemtougri, Axel, Ouedraogo, Vincent Alligouamé, Koïta, Mahamadou, Niang, Dial, Karambiri, Harouna, Yacouba, Hamma
• Format: Journal Article
• Language: English
• Published: Vienna Springer Vienna 01-07-2023; Springer Nature B.V
• Subjects: Agriculture; Aquatic Pollution; Atmospheric Protection/Air Quality Control/Air Pollution; Atmospheric Sciences; Climate; Climate change; Climate science; Climatic data; Climatology; Developing countries; Earth and Environmental Science; Earth Sciences; Evapotranspiration; Evapotranspiration estimates; LDCs; Mass transfer; Maximum temperatures; Methods; Minimum temperatures; Negative feedback; Performance indices; Radiation; Radiation data; Relative humidity; Solar radiation; Solar radiation data; Temperature; Temperature effects; Trends; Waste Water Technology; Water balance; Water Management; Water Pollution Control; Wind speed; Burkina Faso
• ISSN: 0177-798X; 1434-4483
• DOI: 10.1007/s00704-023-04507-z

The precise estimation of reference evapotranspiration (ETo) is critical for water planning in agriculture. However, with the scarcity of data in developing countries, the use of the reference Penman–Monteith FAO-56 (FAO56-PM) equation to estimate ETo is difficult. Besides the quantification of this water balance component, its trends and sensitivities to climate variables are poorly assessed in such countries. In this study, an examination of ETo is carried out over the period 1988–2017 in Burkina Faso, a landlocked country in the West African Sahel. A gap-filling procedure is first used to estimate missing solar radiation data, in combination with bias-corrected climate data from MERRA-2 reanalysis. The gap-free dataset is therefore used to assess annual, seasonal trends in ETo and its dependant variables, as well as the sensitivities of ETo to its dependant variables. The results show a significant decrease (at Dori, northern region) and increase (Gaoua, southern region) in ETo. Also, solar radiation (rs) and maximum temperature (tx) have the highest effect on ETo, followed by relative humidity (rh) (with negative feedback), wind speed (ws) and minimum temperature (tn). Finally, 49 alternative methods of ETo estimation are calibrated (over 1988–2007), validated (over 2008–2017) using FAO56-PM as a reference and ranked out by performance index (PI). In general, combinatory methods (such as Kimberly–Penman equation) are the most accurate (PI = 0.38–0.61), followed by radiation methods (such as Trajkovic–Stojnic equation) which are satisfactory (PI = 0.21–0.76). Temperature methods (TMP) (such as Penman–Monteith temperature equation) and mass transfer methods (MTF) (such as Rohwer equation) perform worse in comparison (TMP: PI = 0.14–0.69/MTF: PI =  − 0.49–0.40), with large underestimation by MTF methods (PBIAS =  − 15.2–0.0%). Calibrated alternative equations such as Trajkovic–Stojnic, Penman–Monteith temperature and Kimberly–Penman are therefore recommended after FAO-56 PM for daily ETo estimation in Burkina Faso.