Isotope effects in the evaporation of water: a status report of the Craig-Gordon model.

The Craig-Gordon model (C-G model) [H. Craig, L.I. Gordon. Deuterium and oxygen 18 variations in the ocean and the marine atmosphere. In Stable Isotopes in Oceanographic Studies and Paleotemperatures, E. Tongiorgi (Ed.), pp. 9-130, Laboratorio di Geologia Nucleare, Pisa (1965).] has been synonymous with the isotope effects associated with the evaporation of water from surface waters, soils, and vegetations, which in turn constitutes a critical component of the global water cycle. On the occasion of the four decades of its successful applications to isotope geochemistry and hydrology, an attempt is made to: (a) examine its physical background within the framework of modern evaporation models, (b) evaluate our current knowledge of the environmental parameters of the C-G model, and (c) comment on a general strategy for the use of these parameters in field applications. Despite its simplistic representation of evaporation processes at the water-air interface, the C-G model appears to be adequate to provide the isotopic composition of the evaporation flux. This is largely due to its nature for representing isotopic compositions (a ratio of two fluxes of different isotopic water molecules) under the same environmental conditions. Among many environmental parameters that are included in the C-G model, accurate description and calculations are still problematic of the kinetic isotope effects that occur in a diffusion-dominated thin layer of air next to the water-air interface. In field applications, it is of importance to accurately evaluate several environmental parameters, particularly the relative humidity and isotopic compositions of the 'free-atmosphere', for a system under investigation over a given time-scale of interest (e.g., hourly to daily to seasonally). With a growing interest in the studies of water cycles of different spatial and temporal scales, including paleoclimate and water resource studies, the importance and utility of the C-G model is also likely to grow in the future.