European Soil Water Content at Field Capacity and Saturation using 3D-CMCC-FEM soil hydraulic formulations and SoilGrids mapping dataset

3D-CMCC-FEM – Soil water balance

The 3D-CMCC-FEM simulates soil water dynamics using a single-layer bucket representation, with the water balance updated at a daily time step. Water is added to the soil water pool through precipitation reaching the soil surface after canopy interception, snowmelt, and irrigation when prescribed. Water is removed through evapotranspiration, including canopy transpiration, canopy evaporation, and soil evaporation. The available soil water (ASW, mm) is therefore updated as:

ASWₜ = ASWₜ₋₁ + Pₜ + Iₜ − ETₜ

where (Pt) represents daily precipitation or snowmelt reaching the soil, (It) is irrigation, and (ETt) is total evapotranspiration. Snow is represented separately from the soil water pool while present and can accumulate under freezing conditions, subsequently contributing to soil water through melt. Snow sublimation can also occur under sub-freezing conditions depending on incoming solar radiation.

When the amount of incoming water exceeds the soil water storage capacity, the excess is removed from the soil water pool and is assumed to contribute to surface or subsurface runoff and/or drainage to deeper soil layers. These processes are not explicitly represented as separate hydrological compartments in the model, and the excess water is therefore no longer available for plant uptake.

Soil water availability is expressed as volumetric water content (VWC), calculated from ASW and the prescribed soil depth (SD, cm):

VWC = ASW / [(SD / 100) × 1000]

The soil hydraulic properties are determined from soil texture. In the map, the percentages of sand, silt, and clay used by the soil-water module were derived from the corresponding SoilGrids layers. SoilGrids is a global digital soil mapping system providing 250 m resolution maps of key soil properties, including sand, silt, and clay content, across six standard soil depth intervals. It combines soil profile observations with environmental covariates using machine-learning methods (SoilGrids).

Following the soil-water relationships implemented in 3D-CMCC-FEM and based on Clapp and Hornberger (1978) and Cosby et al. (1984), soil water matric potential (ψ, MPa) is calculated as:

ψ = ψₛₐₜ × (VWC / VWCₛₐₜ)ᵇ

where (VWCsat) is the volumetric water content at saturation, (ψₛₐₜ) is the soil water matric potential at saturation, and (b) is a texture-dependent parameter. The matric potential at saturation is calculated from soil texture as:

ψₛₐₜ = −[exp((1.54 − 0.0095 × %sand + 0.0063 × %silt) × ln(10)) × 9.81 × 10⁻⁵]

The exponent (b) is calculated as:

b = −(3.10