Skip to main content
ARS Home » Plains Area » Temple, Texas » Grassland Soil and Water Research Laboratory » Research » Publications at this Location » Publication #376870

Research Project: Contributions of Climate, Soils, Species Diversity, and Management to Sustainable Crop, Grassland, and Livestock Production Systems

Location: Grassland Soil and Water Research Laboratory

Title: Updating the national soil map of Nepal through digital soil mapping

item LAMICHHANE, SUSHIL - University Of New England
item KUMAR, LALIT - University Of New England
item Adhikari, Kabindra

Submitted to: Geoderma
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 2/18/2021
Publication Date: 3/11/2021
Citation: Lamichhane, S., Kumar, L., Adhikari, K. 2021. Updating the national soil map of Nepal through digital soil mapping. Geoderma. 394. Article 115041.

Interpretive Summary: Soil maps are compiled through rigorous process of soil survey which demands a significant amount of resources and time. Maps thus produced are the first-hand legacy vector maps, generally at course spatial scale, which are becoming less desirable to solve environmental problems in the modern digital age. Therefore, to increase the reliability and applicability of such maps, digital soil mapping techniques can be applied, and maps can be updated. This study applied Disaggregation and Harmonization of Soil Map units Through Resampled Classification Trees (DSMART) algorithm and updated the national soil map of Nepal (1:1,000,000 scale) using a wide range environmental variable including climate, topography, vegetation, and geology as predictors. The updated raster map showed more detailed information on the distribution of soil types in comparison to the original vector map. We believe that the new national soil map of Nepal can be a valuable resource, and is more reliable to use in land resources planning and management decisions in Nepal.

Technical Abstract: While most legacy soil maps are available at coarse spatial details with composite mapping units, high resolution and detailed soil maps are desired for various land resource applications. With the advancement in remote sensing and computing power, it has now become possible to use advanced algorithms and a multitude of covariates to segregate individual soil types from such composite units. This type of work is more important, especially in a country such as Nepal, where rugged terrain and harsh weather conditions make fine resolution soil sampling prohibitive. In such time and resource-constrained circumstances, the application of disaggregation methods and modelling approaches that capitalize on existing less detailed soil maps is important for a more rapid generation of soil maps at finer resolutions. A legacy soil map of 1:1,000,000 scale for Nepal was therefore disaggregated using a modern digital soil mapping approach to reveal the most probable locations of individual soil classes from the composite soil map units held in the legacy soil map. The soil map disaggregation was carried out using “Disaggregation and Harmonization of Soil Map units Through Resampled Classification Trees” (DSMART) algorithm with the C5.0 classification tree algorithm and an area proportional virtual sampling technique. Environmental covariates sourced from remote sensing, digital elevation model, climatic databases, and national databases were used for predictive mapping of soils. The predicted most probable maps were found to show more detailed soil maps in comparison to the original soil map. The areas under predicted soil classes were quantified for different physiographic regions and land covers. Accuracy assessment of the 1st most probable soil map with independent datasets showed that the overall accuracy of prediction was 51.2% while considering the level of Reference Soil Groups only and 32.6% for the soil groups with 1st principal qualifier. Confusion and Shannon Indices were calculated to show the uncertainty of prediction and the diversity of probable soil classes. Geology was the most frequently used covariate, followed by the minimum temperature of the coldest month, elevation, valley depth and land cover. DSMART was found to be an effective technique for spatial disaggregation of the legacy soil map of Nepal. The predicted most probable maps can be used as a more detailed version of the legacy soil map of Nepal. Knowledge of soil types is essential to plan and manage agricultural and environmental activities. Amidst the scarcity of spatially explicit detailed soil information, this disaggregated soil map can be a useful resource for individuals concerned with planning and managing land resources at the scales ranging from local administrative units to regional and national ones. Environmental covariates found to rank among the top ones in this study could be suggested for the improved accuracy of the disaggregation of soil maps in similar landscape scenarios.