Nation-wide open soil property maps of up to 100 cm depth for Sri Lanka

Published: 12-01-2021| Version 2 | DOI: 10.17632/5sc7njfcyn.2
Eranga M. Wimalasiri,
Ebrahim Jahanshiri,
T.A.S.T.M. Suhairi,
Ranjith B. Mapa,
Asha Karunaratne,
Lal Vidhanarachchi,
Hasika Udayangani,
Sayed Azam-Ali


The first ever national level digital soil database of Sri Lanka. This dataset contains interpolated maps for 9 soil properties; pH, bulk density (gcm-3), cation exchange capacity (cmol+ kg–1), organic carbon content (%), volumetric moisture content (VMC) at 0.33 bars level (drainage upper limit – mm mm-1), VMC at 15 bars level (wilting point – mm mm-1), clay (%), sand (%) and silt content (%). The data are available at 5 different depths as 0-5 cm, 5-15 cm, 15-30 cm, 30-60 cm and 60-100 cm. Observed data from 122 locations that were distributed throughout the country were used for interpolation mapping. The choice of interpolation method can generate different results, therefore several interpolation methods were used in the interpolation process. The least Root Mean Square Error (RMSE), through the process of cross-validation or leaving-one-out of determined the final method used for mapping each property. For sand, silt and clay, the Radial Basis Function interpolation was used while all other properties were interpolated using the Empirical Bayesian Kriging method. The resolution of the interpolation maps is approximately 1 km. Other than the soil properties, standard error maps of respective properties were also made available. It can be used to explain the variation of the final results of simulation models. The dataset is mainly designed for process-based crop models, but it can be used for any soil-related studies in Sri Lanka.