ABSTRACT
Evapotranspiration (ET) is an important process of the hydrologic cycle. Approximately 75% of the total rainfall on a continental basis is returned to the atmosphere by evaporation and transpiration [4]. The ET represents the quantity of water which must be supplied at all times to maintain a soil water balance. Knowledge of ET is necessary in planning and implementing soil water management systems. Numerous approaches can be used to estimate ET-, such as mass (water vapor) transfer, energy budget, water budget, soil moisture budget, groundwater fluctuations, empirical formulae, and combination (of energy budget and mass transfer) methods. These different techniques have been developed partly in response to the availability of different types of data for estimating ET. Each method has certain advantages and limitations (see Chapter 5). Availability of specific types of data is often a limiting factor in the choice of calculation technique for practical applications. The choice of calculation technique also depends on the intended use [6, 8] and on the time scale required by the problem. For example, irrigation management requires daily estimates of ET to allow producers to make rational decisions concerning the timing and amount of irrigation. Different types of vegetation and stages of growth must be considered because they have a considerable influence on the daily rate of ET. In contrast, basin-level planning may require monthly estimates of ET to project changes in water supplies and requirements during the year. The combination energy-balance mass-transport method [9] has displayed the best overall fit among the empirical methods for ET estimation [6, 7, 13]. However, the combination method [9, 10] requires a variety of climatological data such as net radiation, air temperature, relative humidity, and wind velocity. If sufficient climatological data are not available for using the Penman method, an alternative method should be chosen. Shih [14] suggested that in choosing an ideal ET equation, one should minimize input of climatological data without affecting the accuracy of estimation, so that not only the multi – collinearnity problem among the data can be eliminated but also the data availability can be improved considerably. For example, Shih et al. [13] used the modified Blaney-Criddle method [12] as follows to predict ET:
ET = k*f’
f’ = {[2.54*PR] * [1.8*Tavg + 32]} / [100] (1)
where, k = Coefficient for modified Blaney – Criddle method; Tavg = Mean monthly temperature in °C; PR = Percent of annual solar radiation during the month; ET = Evapotranspiration; and f’ = Monthly Et factor in mm.
