Comparison of techniques for estimating annual lake evaporation using climatological data

Abstract

Andersen and dobson [1982] have estimated mean annual evaporation for 30 lakes by using a published version [Morton, 1979] of what are now referred to as the complementary relationship lake evaporation, or CRLE, models and by using an evaporation map prepared by the U.S. Weather Service [Kohler et al., 1959]. The two estimates for each lake were compared with evaporation values that have been reported in the literature over the past 31 years. Andersen and Jobson concluded that the evaporation map provides annual evaporation estimates which are more consistent with observations than those derived from the CRLE model. However, there are two reasons why neither the comparisons nor the conclusion have any value. The first reason is that the CRLE model estimates computed by the authors are obviously in error. The second is that the basis of comparison is inadequate in that the majority of the values reported in the literature are alternative estimates and not observations. These inadequacies are discussed in more detail hereinafter. Table 2 [Andersen and dobson, 1982, p. 633] are obviously too high. This is because they used the potential evaporation outputs of the CRLE model rather than the lake evaporation outputs. When advised of this they prepared a revised version of Table 2, which is presented in the response to this discussion. Table 1 provides an evaluation of their "revised" results for 13 of the lakes. The "comparable" values are also CRLE model estimates, but the inputs are the monthly mean values of maximum and minimum temperatures, relative humidities at 6-hour intervals, and ratios of observed to maximum possible sunshine duration that were published in the two volumes of Climates of the States [NOAA, 1974]. The vapor pressure was estimated to be the average of the saturation vapor pressure at the maximum temperature multiplied by the minimum of the relative humidities and the saturation vapor pressure at the minimum temperature multiplied by the maximum of the relative humidities. Table 1 shows that the difference between the revised and comparable estimates is less than 25 mm/yr for seven of the lakes and less than 70 mm/year for nine of the lakes. Such differences can be attributed to differences in input data from one set of years to another and thus indicate that the authors have used the CRLE model correctly. However the differences for the other four lakes are too large to be ignored. The difference of the 192 mm/yr between the revised and comparable estimates for Lake Okeechobee can be attributed to the use of data at different locations. Thus the revised value is based on sunshine duration records at Key West, which is approxi- The CRLE [Morton, 1979] model estimates presented by Andersen and Jobson for the last 14 lakes (symbols I to Z) in

Comparison of artificial neural networks and empirical equations to estimate daily pan evaporation

“…Indirect methods include those that use meteorological data to estimate evaporation from other meteorological variables through empirically developed methodologies or statistical and stochastic approaches, in addition to mass-balance-based formulations. Both direct and indirect methods have been used for evaporation estimation studies by many researchers (Stewart and Rouse, 1976;Warnaka and Pochop, 1988;de Bruin, 1978;Andersen and Jobson, 1982;Morton, 1979;Choudhury, 1999;McKenzie and Craig, 2001;Vallet-Coulomb et al, 2001;Abtew, 2001;Rotstayn et al, 2006). Singh and Xu (1997a) evaluated and compared 13 evaporation equations that belonged to the category of mass-transfer methods and a generalized model form for that category was developed.…”

Application of support vector machine and relevance vector machine to determine evaporative losses in reservoirs

“…Therefore, the determination of Evaporation Losses ( E ) in reservoir is an imperative task in hydrology. Researches use different methods for prediction of E in reservoirs (Stewart and Rouse, 1976; Bruin, 1978; Anderson and Jobson, 1982; Abtew, 2001; Murthy and Gawande, 2006). Recently, Deswal and Pal (2008) have successfully employed Artificial Neural Network (ANN) for determination of E in reservoirs.…”

Analyses of equations for free water evaporation estimates

“…Of the criteria considered above, a relatively small amount of information exists on the ability of various equations to define the variation of evaporative losses. Most comparisons of methods have considered average annual evaporative losses [e.g., Andersen and Jobson, 1982] rather than the statistics describing the variation of these losses. Allen and Wright [1983] have compared estimates of evapotranspiration using a modified combination equation and the Blaney Criddle formula.…”