Modelling and Mapping
Once soil data is collected, analyzed, archived, and organized, the soil data can be used for modelling and mapping in the next step of the development of a SIS. Soil databases typically contain observations and measurements taken at sampling (point) locations. Many users though, require information in the form of soil maps for their applications. Therefore, developing soil maps from soil (point) data stored in a database is a logical next step in the soil information workflow. For soil mapping, two mapping approaches are often used:
- conventional soil mapping (also known as landscape-based soil class mapping)
- digital soil mapping.
In conventional soil mapping soil mapping, soil surveyors take observations in the landscape and then, often supported by aerial photography, delineate soil bodies that are homogeneous in terms of morphology and composition. These soil bodies are subsequently classified based on a soil classification system. The result is a traditional soil class map that is often accompanied by a survey report that provides descriptive information and quantitative data on the soil bodies.
Since the beginning of the 21st century, more maps are often developed using statistical methods for modelling and mapping soil spatial variation. These methods are referred to as ‘digital soil mapping’ (DSM). Nowadays, DSM is an accepted practice for soil mapping and more often used than conventional soil mapping. The widely used Soil Survey Manual of the U.S. Department of Agriculture now contains a chapter dedicated to DSM. With the advent of DSM, not only the way soil maps were produced changed, but also the type of maps. DSM typically produces gridded maps of quantitative soil properties at a specific spatial resolution (grid cell size), while conventional survey typically produces polygon maps of soil types and associated soil properties at a certain (cartographic) scale level.
This workflow step presents an overview of methods, tools, and standards for developing soil maps from soil observational data. Other type of models, such as dynamic or process models, that use soil data as input to predict functional properties of the soil (that often cannot be directly measured such as carbon sequestration potential) or other type of agricultural or environmental variables (such as yield potential based on soil nutrient status and soil depth) are considered in step on Applying soil information.
This tutorial shows how to process and prepare input data for digital soil mapping (DSM) using the statistical software R. The processing steps are illustrated with a sample dataset from North Macedonia. The tutorial consists of three parts: …
This tutorial shows how to map a soil property with geostatistical (kriging) approaches using the statistical software R. A sample dataset from the Geul river valley in the south of the Netherlands is used for this purpose. The tutorial consists of …
This is a video recording of the lecture on ‘Geostatistics for soil mapping’ that was given by Prof. Dr. Gerard Heuvelink during the ISRIC spring school 2015. The lecture is of introductory nature. No knowledge of geostatistics is …
This tutorial shows how to fit a random forest model for soil data and use this model to spatially predict a soil property of interest across a mapping area using the statistical software R. The processing steps are illustrated with a sample dataset …
The Sourcebook is designed to provide a conceptual foundation for soil organic carbon measurement and monitoring in croplands and grazing lands or rangelands. It provides methods and simple step-by-step guidance to produce reliable soil carbon …
SoilGrids ™ (hereafter SoilGrids) is a system for global digital soil mapping that makes use of global soil profile information and covariate data to model the spatial distribution of soil properties across the globe. SoilGrids is a collections of …
The latest SoilGrids release can be accessed through the following services: WMS: access for visualisation and data overview. Instructions for using WMS with commonly used GIS software can be found here WCS: best way to obtain a subset of a map and …