Format: Practical workshop (multi-day)
Host: Iranian Soil & Water Research Institute
Instructor/materials: Ruhollah Taghizadeh (R scripts + exercises + slides)
GitHub materials: https://github.com/RuhollahTaghizadeh/Workshop-DigitalSoilMappingwithR

Overview

This workshop is a structured, hands-on introduction to Digital Soil Mapping (DSM) using R. It guides participants through the complete DSM pipeline: understanding DSM concepts, preparing soil and environmental datasets, generating covariates from remote sensing and DEMs, fitting predictive models (from simple baselines to advanced machine learning), incorporating geostatistics, and producing final soil property maps.

The workshop is organized into daily modules with reproducible code and practice datasets, designed for students and professionals who want to build practical DSM skills and a solid understanding of methodological choices and common pitfalls.

What participants learn

By the end of the workshop, participants are able to:

• understand the DSM concept and how soil–environment relationships support spatial prediction
• work confidently with spatial data in R (vector and raster)
• preprocess geodatabases and prepare modelling-ready datasets
• derive and manage covariates from:
  • remote sensing imagery
  • digital elevation models (DEMs) and terrain attributes
• fit and evaluate spatial prediction models for soil properties using:
  • simple statistical models (baseline approaches)
  • advanced machine learning models
• apply geostatistical concepts for mapping and handling spatial structure
• produce final prediction maps and document a reproducible workflow

Workshop structure (modules)

The repository is organized as a step-by-step training sequence:

1. Introduction to Digital Soil Mapping (DSM)
2. R basics (foundation for reproducible analysis)
3. Preprocessing geodatabases, remote sensing, and DEMs
4. Extracting spatial information & prediction with simple models
5. Advanced machine learning for DSM (Part 1)
6. Advanced machine learning for DSM (Part 2)
7. Geostatistics for mapping soil properties
8. Advanced topics + “DSM from 0 to 100” overview

Tools & topics

• R programming for reproducible workflows
• spatial data handling (vector/raster), covariate engineering
• model training, validation, and performance assessment
• machine learning for environmental prediction
• geostatistics (intro-to-applied) for spatial mapping
• map production and reporting

All workshop materials (code, module structure, and learning content) are openly available on GitHub at the link above.