Daily evaporation from Lake Ontario during the International Field Year for the Great Lakes (April 1972 to March 1973) was computed by a modified mass transfer technique. Turbulent energy flux was calculated by using upwind !and station data and the surface water temperature at 88 grid points on the lake. Wind speed and humidity at each grid point were determined through multiregression equations that take into account stability, fetch, and water temperature. Daily, monthly, and annual evaporation amounts obtained in this study are compared to amounts obtained by conventional mass transfer, energy balance, and terrestrial water budget methods. Discrepancies in the results of the various approaches are examined and discussed.
INTRODUCTION
One of the objectives of the International FieldYear for the Great Lakes (IFYGL) was to improve current formulations used to estimate evaporation from large, temperate lakes. Several projects were designed to arrive at independent estimates of time-averaged lake evaporation for periods ranging from 1 day to 1 month. Evaporation from Lake Ontario was compared (1) as a residual in the water budget, (2) as a residual in the energy budget, (3) from an atmospheric water budget for the autumn season of the year, (4) from aerodynamic equations applied to meteorological buoy data, and (5) from mass transfer formulae. This paper describes details of a modified mass transfer technique in which regression equations requiring fetch and surface water temperature over that fetch, as well as the thermodynamic variables, are used to estimate the flux at a number of grid points over the lake. The spatial distribution and temporal variation of the turbulent energy flux, i.e., sensible and latent, from large lakes are important information useful for forecasting mesoscale phenomena such as lake effect snowstorms and lake-land breeze circulations and for predicting the time and extent of ice formation and dissipation for specific locations. In addition, a more accurate estimate of water loss would improve predictions of lake level fluctuations. To this end, comparisons are made with the results of a number of approaches to determine evaporation from Lake Ontario during IFYGL.
M^ss TR^SSFER: L^iCE H•FS•R FORMULAOne indirect method of computing evaporation consists of a modified application of Dalton's law, where evaporation is considered to be a function of the wind speed and the difference between the vapor pressure of saturated air at the water surface and the vapor pressure of the overriding air. Many variants of the Dalton formulation have been developed to compute evaporation, and these are frequently referred to as mass transfer equations. The terms 'bulk aerodynamic' or 'bulk transfer' have also been used.