Talk to us

Send us a message

Replies usually arrive within 1–3 working days.

Find us

  • Office
    National Soil Science Research Center (NSSRC) Nepal Agricultural Research Council (NARC)
  • Address
    Khumaltar, Lalitpur, Nepal Visit during working hours: Sun–Fri, 10:00–17:00
  • Email
    info@narc.gov.np General enquiries
  • Website
    narc.gov.np Parent organisation

Frequently asked questions

Digital Soil Map (DSM) is a computer-assisted production of digital maps of soil properties. This is developed by using mathematical and statistical models that combine soil information from laboratory analysis with environmental variables (soil-forming factors).

The maps were prepared using soil information from 23,273 soil samples, collected from 56 districts covering seven provinces. These soil properties are combined with a stack of 168 remote sensing-based soil covariates (SRTM DEM derivatives, climatic images, vegetation index, etc.). Spatial predictions were generated using a machine learning method (random forest). Reference: Hengl T et al. (2017). SoilGrids250m. PLoS ONE 12(2).

Soil texture (sand, silt and clay), soil pH, organic matter, total nitrogen, available phosphorus, available potassium, zinc and boron.

Environmental factors that affect soil properties. Covariate layers are generated from satellite imagery (raster data) and include topographic data, vegetation, precipitation, temperature, soil parent material, land-cover type and landform classes.

Spatial resolution of all DSM rasters is 100 m.

Soil profile data were collected from various government projects including the National Land Use Project, the Irrigation and Water Resources Management Project, the Central Agriculture Laboratory (previously the Soil Management Directorate), and the Nepal Agricultural Research Council (NARC).

DSM uses advanced computational algorithms that combine both soil-sample data and environmental variables. It doesn't only use spatial autocorrelation to interpolate, but also considers soil-forming factors. The process can be automated, so newer versions of the map can be developed faster once new samples are collected. Combining samples with environmental variables lets DSM work with fewer samples than a conventional map.

Published studies show random-forest DSM consistently out-performs linear regression, with RMSE reductions of 15–75% across soil properties and depths. Other studies show DSM is more accurate than the deterministic pedological approach.

Use the point or polygon tool on the soil map. Click on your farm or draw a polygon over it; the output (soil properties) appears in a downloadable table.

Caution: values are model predictions. Consult experts for interpretation, or run detailed soil analysis before making operational decisions.

Yes. Open Maps → Soil map, then use the drawing tools in the toolbar on the right-hand edge of the map to draw a box or polygon around your area of interest. The result covers the whole area you drew, and can be downloaded as CSV.

Yes:
• APIs: three public REST endpoints, with a free API key. The soil endpoint can also return XML via ?mode=xml; the crop endpoints are JSON only.
• Direct download: every dataset page in the catalog links a zipped GeoTIFF you can download directly; no account needed.
• Interactive tools: draw a point or polygon over the web map and export the result as CSV.

• Identifies domains with similar soil properties and the fertility status (deficiency / sufficiency of plant nutrients).
• Used by farmers, policy makers, extension workers, agri-input retailers and resource managers as a decision-support tool, gives location-specific access to soil properties and estimates of fertilizer needs.
• Feeds models that extrapolate domain-specific fertilizer recommendations and estimate total fertilizer required for a given crop / season.
• Locates areas with deficient nutrients so that site-specific fertilizer formulations (blended nutrient grades) can be designed.
• Helps design programmes on soil fertility, organic-matter management, micronutrient deficiency correction and acidity control.

Not to try it. The soil, BMP and pest & disease endpoints answer without a credential, at a limited number of requests per hour per IP address. A free API key raises that limit substantially and is the right choice for anything running unattended. Create an account, then generate a key in your dashboard.

Rice, wheat and maize. The soil endpoint returns doses for rice (OPV and hybrid), wheat (OPV) and maize (OPV and hybrid), based on the district and physiographic region of the point you queried. Best-management practice guides cover a wider set: rice, wheat, maize, cauliflower, tomato and onion.

The maps cover Nepal, roughly 80.0–88.2°E and 26.3–30.5°N, in EPSG:4326 (WGS 84). A request for a point outside that area returns a 404 rather than an estimate.

We ask that you cite any data taken from this portal in your outputs, websites and publications. The terms are:

• Share: copy and redistribute the material in any medium or format.
• Attribution: credit NARC, link back to the licence, and indicate any changes you made.
• Non-commercial: you may not use the material commercially.
• Non-derivative: modified material may not be redistributed.

Full terms and the citation format: /license

Yes. Soil properties change over time; we recommend updating the map every five years.

Please write to us using the contact form above.

Contribute soil data. Organizations that have been involved in data collection, let us know and we will integrate your dataset and acknowledge you as a contributing organization. Use the contact form to start the conversation.