Assessing the sensitivity of standardized precipitation evapotranspiration index to three potential evapotranspiration models in Nigeria
Akinwale T. Ogunrinde, David A. Olasehinde, Olotu Yahaya
Abstract
Drought indices are very essential in monitoring and planning for drought occurrences. Though many indices have been established over the years, it is expected that proper investigation should be done using these indices to know the ones that can better characterize the drought condition of a region or country. The current study evaluated the performance of three potential evapotranspiration models (Hargreaves, Penman-Monteith and Thornthwaite) as inputs into Standardized Precipitation Evapotranspiration Index (SPEI) in regards to identified historic drought events for five agroecological zones in Nigeria. The computation of the indices was done using either one or more of the monthly time series of rainfall, minimum, maximum and mean temperature, sunshine hours and wind speed from 1981 – 2015 at five timescales (1, 3, 6, 12 and 24 months). Although all the models correlated significantly with one another, SPEI-H (Standardized Precipitation Evapotranspiration Hargreaves) vs SPEI-P (Standardized Precipitation Evapotranspiration Penman-Monteith) show higher correlation for all timesteps than SPEI-H vs SPEI-T (Standardized Precipitation Evapotranspiration Thornthwaite) and SPEI-P vs SPEI-T. The efficiency of the indices in identifying historic drought episodes: 1983 – 1985, 2007 and 2011 was examined using information obtained from Emergency Events Database (EM-DAT) and previous literature on drought studies in Nigeria. Drought onsets, severity and duration determined by the indices were compared with that of the EM-DAT. All the three indices show earlier onset for both 3- and 12- month timesteps than that of EM-DAT, but indicated a significant difference in severity and duration during the 1983 – 1985 drought episode for 12-month timestep. Though SPEI-P has a higher comparative predictive ability as indicated by EM-DAT, it was discovered that SPEI-H is a good alternative for regions where there is a paucity of data.