Comparing different methods for estimating the soil moisture supply capacity of a soil series subjected to different types of management

Abstract

The capacity of soils to supply water to growing plants was expressed in a static and dynamic manner by calculating ‘available water’ (AW) and the ‘soil moisture supply capacity’ (MSC), respectively. Four methods of increasing complexity were compared for the AW and two for the MSC, using measured moisture retention and hydraulic conductivity curves and data derived from class- and continuous pedotransferfunctions. Calculations were made for a prime agricultural soil in the Netherlands, a fine, mixed, mesic Typic Fluvaquent (genoform). Three phenoforms, defined by long-duration management, were distinguished: BIO (biodynamic); CONV (conventional, high-tech) and PERM (permanent grassland). Values for AW for a given treatment were, in most cases, significantly different when using measured data or pedotransferfunctions. AW can be used to rank different soils but does not reflect the amount of water the crop can take up. Simulation models are needed to estimate MSC which cannot directly be measured. Simple empirical models, using AW for the rootzone in a ‘tipping-bucket’ approach, did not produce realistic values because upward flow from the relatively shallow water table could not be distinguished. The more complex mechanistic WAVE model, also including hydraulic conductivity data, but still operating under the implicit and incorrect assumption of soil homogeneity and isotropy, produced soil water contents that were too high. Realistic values were only obtained when considering bypass flow, internal catchment and accessibility of peds for rooting, using a modified WAVE model incorporating hydraulic data derived from continuous pedotransferfunctions. Values for AW and MSC were, in most cases, significantly different for the three phenoforms, illustrating the need to distinguish phenoforms, rather than only genoforms, when reporting basic physical data for soil series.