Modelling surface resistance from climatic variables?

For the Penman-Monteith equation to be used to predict crop evapotranspiration in a one-step approach, methodologies for determining surface resistance (r(s)) must be available. One usual approach to the modelling of r(s) is to compute it by inverting the Penman-Monteith equation and then relate it to the most important environmental variables (radiation, temperature, vapour pressure deficit) using the multiplicative model of Jarvis. In this paper, some results obtained for lettuce are presented to illustrate the pitfalls of this approach. It is shown that the same environmental variables and the same functional forms that are used in the Jarvis model are already considered when calculating r(s) as the residual term. One cannot thus expect to get a better insight on the behaviour of r(s) with the multiplicative model. Also, as r(s) includes information on the transport conditions inside the canopy and thus, is dependent on wind speed (or, indirectly, on the aerodynamic resistance), procedures that only contemplate stomatal functioning may be not adequate, The interactions between r(s) and latent heat flux are also discussed and indicate that future studies should be focused on the determinism and quantification of the energy partitioning. (C) 2000 Elsevier Science B.V. All rights reserved.