Dynamic response of land–atmosphere-coupling parameters to precipitation in the sparse-vegetated Asian summer monsoon transition zone

Land–atmosphere-coupling parameters usually relate to particular climatic conditions and seldom consider the influence of the inter-annual variability of precipitation. This results in large uncertainty in the estimation of land-surface physical variables. In the current study, observed data for five land–atmosphere-coupling parameters (surface albedo, soil thermal conductivity, aerodynamic roughness length, and the bulk transfer coefficients of momentum and sensible heat) were analysed and related to the inter-annual variability of precipitation on the Loess Plateau, China, from April 2006 to March 2013. This is an area with sparse vegetation. The results demonstrate that the land–atmosphere-coupling parameters are very sensitive to the inter-annual variability of precipitation. The surface albedo increased with increasing duration of snow cover. The other four parameters increased when annual effective precipitation (yearly sum of daily rainfall > 4 mm or daily snowfall > 0.1 mm) increased, and the sensitivity decreased when annual effective precipitation increased. Empirical relations between the five land–atmosphere-coupling parameters and the inter-annual variability of precipitation were derived via regression, with R2 values of 0.67, 0.85, 0.93, 0.99, and 0.95, respectively. The land-surface physical variables calculated with dynamic land–atmosphere-coupling parameters (considering the inter-annual variability of precipitation) were much closer to the observed data than those calculated with static land–atmosphere-coupling parameters. The relative error was reduced by > 80% in years with low precipitation. This indicates that the inter-annual variability of precipitation has a significant impact on the land-surface physical variables. The results demonstrate that land–atmosphere-coupling parameters, which take into account inter-annual variability of precipitation, provide a more realistic representation of the land surface.