Assessment Of The Main Methods For Estimating Potential Evapotranspiration At Samaru In Northern Guinea Savanna Zone Of Nigeria
Evapotranspiration is a major component of the water balance in any ecological system. Several methods have been proposed to estimate potential evapotranspiration in different parts of the world. However, the success of most of them has been limited because of uncertainties in the reliability of the methods under subhumid tropical environment. A study was therefore undertaken at the Institute for Agricultural Research Farm, A.B.U., Zaria (11° ll’N, 07° 38’E, 686m above sea level) to determine the most suitable method for estimating potential evapotranspiration in Samaru within the Northern Guinea Savanna Zone of Nigeria. To achieve this objective, predictive efficiency of five PET models (BMN, Hamon, Jensen Haise, Penman and Makkink) was examined against measured microlysimetric PET and open pan evaporation. The estimates of the five models were computed from the 30 years (1965-1994) climatic data and compared with pan evaporation. These five equations were also used to compute PET for two years (1996-1997) and compared with measured microlysimetric PET and pan evaporation during this period. It was observed that Hamon, method overestimated PET throughout the entire year. Jensen Haise and Penman methods over estimated slightly above the pan evaporation in some months. BMN method estimated closest to pan evaporation. Makkink method underestimated in some months and slightly below the pan evaporation. Only BMN method predicted magnitude of performance for the two measured PET was in the order: BMN > Penman > Makkink > Jensen Haise > Hamon. The methods developed by BMN, Penman and Makkink give comparable results which were good approximation of measured PET. The recmmended methods ( BMN, Penman and Makkink) are inspite of the observation that they do not exhibit the same degree of oscillation as measured microlysimetric PET and pan evaporation and thus the differences between the maxima and minima are relatively smaller than the other methods. To further improve on the efficiency of the three equations, adjustment factors were incorporated into them; so that they can estimate as precisely and accurately as the measured PET. Equations relating BMN, Penman and Makkink equations to measured PET were developed. The modified BMN, Penman and Makkink equations are given as follow: Y = 1.38+0.59 BMN R2 = 0.99 Y= 2.58 + 0.17 Penman R2 = 1.00 Y = 0.5 + 0.99 Makkink R2 = 1.00 Y =-0.08 + 0.99Pan Y = Microlysimetric PET For pan evaporation (x) x = 0.033 + 0.99BMN, R2 =1.00 x =-0.051 + 1.007Penman, R2 = 1.00 x = 3.79 + 0.23Makkink, R2 = 0.95 The BMN method is recommended for application as best satisfying the objective of the study. It is the simplest, most The BMN method is recommended for application as best satisfying the objective of the study. It is the simplest, most accurate and reliable of the three recommended methods. It was also observed that the PET is highest in March the warmest month of the year and lowest in August the wettest month in this zone. Path correlation analysis on the contribution of each of the six important climatic factors (relative humidity, air temperature, wind speed, soil temperature, sunshine hours and solar radiation) indicated that microlysimetric potential evapotranspiration was influenced in the following trends: sunshine hours (18.47%) > soil temperature (13.84%) > relatively humidity (11.49%) > wind speed (11.22%) > air temperature (8.29%) > solar radiation (4.41). Whereas pan evaporation was influenced as follows: soil temperature (39.19%) > sunshine hours (22.28%) > air temperature (7.56%) > wind speed (4.05%) > relative humidity and solar radiation (2.56%). A comparison of Penman and Monteith equations with measured soil evaporation beneath maize and sorghum crops showed that the Monteith equation is a better estimator of soil evaporation than Penman equation. However, adjustment factors were developed for both equations.