HU-SoilHydroGrids provides nationwide information on the most commonly used soil hydrological properties (saturated water content, water content at field capacity and wilting point, saturated hydraulic conductivity and Mualem‐van Genuchten parameters for the description of the moisture retention curve) at a spatial resolution of 100 metres, to a soil depth of 2 meters for six standard layers (0-5 cm, 5-15 cm, 15-30 cm, 30-60 cm, 60- 100 cm, 100-200 cm).

Quantitative data on the hydrophysical properties of the soil, which are spatially explicit and also describe depth conditions, are an extremely important input for hydrological modelling. During the development of the Institute’s soil spatial data, these requirements were constantly taken into account and attempts were made to satisfy them to the maximum extent possible, depending on the available technology and data background. The Institute’s involvement in the National Laboratory for Water Science and Water Security consortium gave a significant boost to our developments. The continental scale EU-SoilHydroGrids 3D Soil Hydraulic Database of Europe, developed earlier with the significant contribution of the Institute’s staff, has proven its usefulness at European level over the past decade. Building on this precedent, we have developed a national 3D Soil Hydraulic Database for Hungary based on similar principles, but with several improvements and larger spatial resolution. HU-SoilHydroGrids is the Hungarian equivalent of the continental-scale EU-SoilHydroGrids, based on domestic data with greater reliability, as

• pedotransfer functions developed exclusively on Hungarian soil observation data,
• primary soil property maps provided by the renewed national soil spatial data infrastructure (https://dosoremi.hu/) and
• a large number of additional environmental predictor variables, as well as
• combining the results of several machine learning methods (ensemble modelling).

We also performed spatial uncertainty analysis for each soil hydrological thematic layer of the HU-SoilHydrogrids 3D database. The estimation uncertainty, which was based on the weighted average of statistically significant machine learning methods, i.e. the “cubist” and “ranger” ensemble models, was modelled using quantile regression. We characterised the estimation uncertainty of the HU-SoilHydroGrids maps with 95% and 5% quantiles, as recommended in the international literature, and prepared uncertainty maps.

HU-SoilHydroGrids provides unique support for research aimed at analysing national-scale environmental problems related to the hydrological cycle (such as drought, inland flooding, water retention, etc.).

The most important publications related:

Szabó B, Mészáros J, Laborczi A, Takács K, Szatmári G, Bakacsi Z, Makó A, Pásztor L: From EU-SoilHydroGrids to HU-SoilHydroGrids: A leap forward in soil hydraulic mapping, Science of the Total Environment, 2024, 921:171258, https://doi.org/10.1016/j.scitotenv.2024.171258

Szabó B, Kolcsár R, Mészáros J, Laborczi A, Takács K, Szatmári G, Makó A, Rajkai K, Benyhe B, Barta K, Pásztor L, Bakacsi Z: National soil hydrologic groups map for environmental applications using data-driven and expertbased methods, Scientific Data, 2025, 12(1):1590-1601, https://doi.org/10.1038/s41597-025-05853-5