A new methodology is presented that allows the upscaling of land surface parameters of a Soil-Vegetation-Atmosphere-Transfer (SVAT) Model. Focus is set on the proper representation of latent and sensible heat fluxes on grid scale at underlying subgrid-scale heterogeneity. The objective is to derive effective land surface parameters in the sense that they are able to yield the same heat fluxes on the grid scale as the averaged heat fluxes on the subgrid-scale. A combination of inverse modelling and Second-Order-First-Moment (SOFM) propagation is applied for the derivation of effective parameters. The derived upscaling laws relate mean and variance (first and second moment) of subgrid-scale heterogeneity to a corresponding effective parameter at grid-scale. Explicit upscaling relations are exemplary derived for a) roughness length, b) wilting point soil moisture, and c) minimal stomata resistance. It is demonstrated that the SOFM-Method yields congruent results to corresponding Monte Carlo simulations. Effective parameters were found to be independent of driving meteorology and initial conditions.