Аннотация:At the present stage of soil hydrology development, the water retention curve, i.e., the dependence between the capillary-adsorption water pressure and the volume soil water, appears to be one of the most important hydrological parameters of soils. However, there is still no commonly accepted standardized experimental methods of soil sampling. However, it is unknown how significant is the sampling factor for the subsequent analyses of soil hydrological conditions including for use the prediction models.
The water retention curve (WRC) of medium-loamy gray forest soil horizons was studied in cylindrical-shaped samples of disturbed and undisturbed structure and different size. The soil monoliths were sampled in three directions: vertically, along the slope, and across the slope of intrasoil paleorelief formed by the cone-shaped second humus horizon. The experimental curves were approximated with the van Genuchten equation. The statistical analysis of WRC approximation parameters proved them to differ significantly in loose soil and monoliths, and for the different sampling directions, most often, for sampling across the paleorelief slope. It is shown that for the soils with marked structure and texturally differentiated profile, the sampling procedure should be obligatorily taken into account upon determining WRC and deriving pedotransfer functions. The noted variation in WRC approximation parameters may influence substantially the predictive estimation of soil water regimes. This fact suggests the necessity to indicate strictly the sampling procedure, in particular, in respect of the soil profile structure, asymmetry in soil properties, and sample sizes used for analyzing the hydrological properties of structured soils.