Errors in measuring evaporation

Surface evaporation is notoriously difficult to measure accurately, as are precipitation and areal precipitation estimation. Root-mean-square differences between 30-minute averages of measurements with pairs of adjacent instruments at six sites in Oklahoma were around 0.5 mm/day (1). One sort of equipment involved measuring

1. the covariance of upwards wind and atmospheric humidity above the surface (i.e. the 'eddy correlation' method) and
2. the net radiation to the surface.

There were seven similar systems for comparison, and three others measuring the Bowen Ratio. Such experimental errors resemble those in estimating lake evaporation rates, depending on the period of averaging (2,3). Clearly, it is not possible to determine the accuracy of estimates with more precision than that of the measurements against which they are compared.

1. Twine, T.E. 1998. Underestimation of eddy covariance fluxes over a grassland. M.S. Thesis, Dept. of Atmospheric and Oceanic Science, University of Wisconsin, Madison, Wis. (J.M. Norman - Advisor.)

2. Linacre, E.T. 1993. Data-sparse estimation of lake evaporation using a simplified Penman equation. Agric. Forest Meteor., 64, 237-56.

3. Linacre, E.T. 1994. Estimating U.S. Class A pan evaporation from few climate data. Water International, 19, 5-14.