Background and Methodology

Background

Land degradation refers to the decline or loss of the land’s biological or economic productivity¹. It is largely driven by human activities, often intensified by natural processes, and closely linked to climate change and biodiversity loss². Food systems have been at the center of this challenge. While they are a primary cause of land degradation, their own capacity to produce healthy and nutritious food is also undermined by this deterioration³. Within this context, pastureland are fundamental to food security as they provide the forage that sustains livestock. At the system level, well-managed grassland buffer drought and climate variability, maintain soil carbon and water regulation that steady fodder availability. Conversely, widespread pasture degradation erodes productivity and resilience⁴. However, by implementing a comprehensive set of solutions across food systems, we can shift them from being a major driver of land degradation to becoming a key catalyst for land restoration and recovery. This requires a fundamental shift in thinking, one that goes beyond conventional approaches. It involves recognizing the potential of restoration investments and strategically identifying areas where interventions can deliver both economic benefits and solutions for food security while reversing land degradation.

The RAIZ Mapping Tool, developed by the G20 Global Land Initiative Coordination Office of the UN Convention to Combat Desertification (UNCCD) with technical support from the Food and Agriculture Organization (FAO), the Food and Land Use Coalition (FOLU), and the Alliance of Bioversity International and CIAT (part of CGIAR), is designed to enable this transformative shift. The tool aims at providing essential data to a wide range of stakeholders including investors, policymakers, NGOs, project developers, and others across agri-food value chains helping them to swiftly and accurately identify areas with the greatest potential for productive restoration. By highlighting these priority locations, the RAIZ Mapping Tool supports more effective allocation of investment capital, targeted policy interventions, and on-the-ground restoration actions, ultimately directing collaborative efforts toward sites where they can deliver the highest economic and environmental benefits. The current RAIZ mapping tool is a preliminary, exploratory analysis developed through a consultative process with technical partners and informed by their expertise. Through upcoming improvements, the platform will work in further harmonize methodologies across Conventions dealing with agricultural land and indicating areas that can yield investment.

Objective of the RAIZ mapping tool

In the initial phase leading up to the UNFCCC’s 30th Conference of the Parties (COP30) in Belém, Brazil, the mapping tool identifies degraded cropland and pastureland, using grassland and shrubland as proxies for pastureland. It presents a global heatmap that highlights regions with the highest concentrations of degraded pastureland and cropland. Building on this preliminary work, the focus following COP30 will shift to exploring and prototyping functionalities, such as mapping priority restoration areas based on socioeconomic and biophysical indicators to maximize impact and investment benefits.

Methodology

The platform is developed using a suite of global datasets to quantify and visualize the extent of cropland and pastureland degradation. It integrates multiple supporting layers that capture key biophysical and socio-economic attributes, providing a comprehensive spatial understanding of land conditions. The maps presented within the platform primarily indicate areas where land degradation is most prevalent, based on the underlying analytical parameters. The list of layers used to identify degraded lands are:

1. Global land Degradation, SDG 15.3.1 : The SDG 15.3.1 shows the status of land (from degraded to improved) at pixel level over the period 2001 to 2015. This dataset has been generated using the Trends.Earth model to assess Land Degradation according to the UN Metadata Guideline for SDG. In order to assess the area degraded, SDG Indicator 15.3.1 uses information from 3 sub-indicators which is Vegetation productivity, Land cover and Soil organic carbon. The layer is compiled by UNEP GRID Geneva, UNCCD is the custodian of the SDG 15.3.1.

2. Map of agreement on global cropland : Accurate and precise measurements of global cropland extent are needed for monitoring the sustainability of agriculture at all scales. Recent advancement in remote sensing and land cover mapping methods have greatly increased the ability to estimate cropland area distribution and trends. This map of agreement/disagreement was produced by consolidating information at pixel level from six high-resolutions maps (10-30m). The input datasets represent the most advanced land cover and thematic layers available today for cropland area extent circa 2020. Main concepts and input datasets are described in the pre-print "Measuring the world's cropland area.

3. Global Agro-Ecological Zoning V5 Grassland and Shrubland: The land cover classes (Grassland – class 3); Shrub-covered areas – class 5; Shrub/Herbaceous regularly flooded – class 6; and Lichen and mosses – class 8) from the Share of land cover class (Global - ~1 km) - GAEZ v5 were selected to represent the distribution of grassland and shrubland. The same methodology is applied in the forthcoming FAO report on the Status of the World’s Land and Water Resources for food and agriculture (SOLAW 2025). The RAIZ tool uses SOLAW’s approach where the combined share of grassland and shrubland within each cell is dominant, defined as having the highest proportional share and exceeding a 50% coverage threshold. The resulting layer of grassland and shrubland represents a proxy for the extent of permanent meadows and pastures (FAO, 2025).

The Global Land Degradation (SDG 15.3.1) and Global Cropland Agreement layers are used together to identify degraded cropland areas. Similarly, the SDG 15.3.1 and GAEZ layers are used to identify degraded pastureland areas. Heatmap was generated to highlight the density of degraded cropland and pastureland. Identifying high density is crucial because it signifies the continuity of degradation across the landscape, which translates into a much greater potential for executing efficient, large-scale restoration projects. It is important to note that while these heatmaps reveal the concentration and potential continuity of degraded territory, they do not currently indicate the severity of the degradation itself. The platform also provides several complementary data layers to allow users to assess other critical factors of interest.

4. Global Soil Organic Carbon: The Global Soil Organic Carbon Map (GSOCmap) by FAO allows the estimation of SOC stock from 0 to 30 cm. The GSOCmap represents the first ever global soil organic carbon assessment produced through a participatory approach in which countries developed their capacities and stepped up efforts to compile all the available soil information at national level. Values are in tonnes per hectare.

5. Population Distribution: The GHS-POP dataset represents the spatial distribution of the human population, expressed as the number of people per grid cell. Population data are derived from global census information harmonized by CIESIN for the Gridded Population of the World, version 4.11 (GPWv4.11). These data are disaggregated from census or administrative units to grid cells using the distribution, classification, and density of built-up areas mapped in the GHSL global layers.

6. Precipitation Anomaly: The CHIRPS precipitation anomaly is the difference between a specific period's rainfall and its historical average, calculated using the CHIRPS dataset.

7. Active Fire Density: Active Fires density based on MODIS Collection 6 Active Fire Product MCD14ML. The density of fires is reported as the count of fires per km2 occurred into pixels of 0.1 decimal degree (pixel dimension varies from 20 km2 at the Poles to 122 km2 at the Equator). Only location points described as "presumed vegetation fire" in the attributes are included in the frequency calculation.

Limitations of the tool

The tool relies on global datasets that often miss the complexity and diversity of agricultural landscapes, particularly in low- and middle-income countries. Limited contextual information and coarse spatial resolution reduce overall quality and accuracy. Because training data and methods are frequently biased toward Global North contexts, they don’t transfer well to local conditions or informal practices. Reliability is further undermined by scarce ground-truth data for training and validation. Additionally, understanding cropland extent, management, and seasonal dynamics requires high-resolution, locally calibrated data and context-specific knowledge. This tool serves as an indicative resource for identifying areas of degradation, but it should not be considered a substitute for on-the-ground assessment or validation.

The tool relies on global datasets that often miss the complexity and diversity of agricultural landscapes, particularly in low- and middle-income countries. Limited contextual information and coarse spatial resolution reduce overall quality and accuracy. Because training data and methods are frequently biased toward Global North contexts, they don’t transfer well to local conditions or informal practices. Reliability is further undermined by scarce ground-truth data for training and validation. Additionally, understanding cropland extent, management, and seasonal dynamics requires high-resolution, locally calibrated data and context-specific knowledge.
This tool serves as an indicative resource for identifying areas of degradation, but it should not be considered a substitute for on-the-ground assessment or validation.

References