The Global Centre for Risk and Innovation (GCRI)
Forests hold most of the world's terrestrial species, store carbon, regulate water and sustain rural livelihoods in every tropical, temperate and boreal region. Their health depends not only on how much forest remains but on how it is arranged: large intact blocks behave differently from fragments separated by farms and roads. This page sets out how forest loss, degradation and fragmentation are measured, how they propagate through landscapes, and what the evidence means for those who govern land use and supply chains.
Forests and landscapes · Wetland complex · Accounting year 5
| Water level regime | 0.7 | 0.7 | 1.0 |
| Native vegetation cover | 0.5 | 0.4 | 1.0 |
| Invasive species pressure | 0.3 | 0.3 | 0.0 |
| Connectivity | 0.6 | 0.5 | 1.0 |
Fig.Ecosystem condition account. Design wireframe with placeholder names and illustrative values, not a record or a live system.
Deforestation is the conversion of forest to another land use, while forest degradation is the loss of a forest's structure, function or species without conversion. Both are measured from satellites and forest inventories, and fragmentation into smaller patches changes how the remaining forest behaves. Commodity supply chains and finance link distant buyers to forest loss.
Mountain ranges, deserts, wetlands, glaciers and reefs, and how remote each division is from large cities, shown for the Hindu Kush Himalaya as an example geography for this subject.
Forest health and landscape fragmentation. The dossier below sets out how the risk develops, what is measured, where the data come from, how it is modelled and who decides.
Clearing converts forest to other uses; degradation removes trees, structure and species while the land stays classed as forest. Selective logging, fire and fuelwood extraction often precede clearance, so degradation is an early signal of conversion.
Roads and clearings break forests into smaller patches with more edge. Edges are hotter, drier and windier, raising tree mortality and fire risk, and they isolate species that cannot cross open land.
Drought weakens trees, pests and pathogens spread through stressed stands, and dead fuel feeds larger fires. In some regions repeated fire can shift forest toward savanna or shrubland that does not return on its own.
Most tropical clearance is driven by agriculture, including cattle, soy, oil palm, cocoa and wood fibre. Demand travels through traders, processors and lenders, so forest loss is a supply-chain and portfolio risk as well as a local one.
Each term as this page uses it. Where a standard-setting body defines the term, the wording follows that body.
| Term | Definition |
|---|---|
| Tree cover loss | Removal or death of tree canopy detected by satellite, which includes harvest and fire and is not the same as deforestation. |
| Forest fragmentation | The breaking of continuous forest into smaller, more isolated patches with more edge. |
| Intact forest landscape | A large, unbroken expanse of natural forest ecosystems with no remotely detected signs of significant human activity. |
| Edge effect | Changes in microclimate, species and fire risk near the boundary between forest and cleared land. |
Each indicator carries its unit, its cadence, the kind of source behind it and the evidence label a value would carry. No values are shown: a programme supplies them under its own authority.
| Indicator | Unit | Cadence | Source type | Label |
|---|---|---|---|---|
| Tree cover loss | hectares per year | annual | Landsat time-series classification | OBSERVED |
| Forest area | hectares and percent of land area | five-yearly | National forest inventories reported to FAO | OFFICIAL |
| Forest disturbance alerts | alerts per square kilometre per week | weekly | Optical and radar change detection | OBSERVED |
| Above-ground biomass | tonnes per hectare | periodic | Lidar, radar and inventory plots | ESTIMATED |
| Effective mesh size | square kilometres | periodic | Landscape metrics on forest maps | ESTIMATED |
| Burned area | hectares per month | monthly | Satellite burned-area products | OBSERVED |
| Forest loss by driver | hectares per year by driver class | annual | Driver classification of mapped loss | MODELLED |
The topic as a computational graph: named sources feed the methods, an evidence review gates the outputs, and the decisions stay with the authorities that hold them.
Fig. 02Nodes carry their type, two metric rows and a state; connectors show what feeds what; dashed frames name the owners; the timeline is the cadence. Design of the pipeline, not a record of a run.
Named public datasets and institutional records, each with its licence class. Sources by agreement are used only under their terms and never republished.
| Source | What it gives | Licence class |
|---|---|---|
| Global Forest Watch and Global Forest Change (University of Maryland) | Annual tree cover loss and gain mapped from Landsat | Free and open, attribution |
| FAO Global Forest Resources Assessment | Country-reported forest area, characteristics and management | Free and open, attribution |
| GLAD and RADD forest disturbance alerts | Near-real-time disturbance alerts from optical and radar imagery | Free and open, attribution |
| GEDI lidar and ESA Climate Change Initiative biomass | Forest structure and above-ground biomass estimates | Public domain; free and open |
| Intact Forest Landscapes | Mapped large unfragmented forest areas and their reduction over time | Free and open, attribution |
| Copernicus Sentinel-1 and Sentinel-2 | Radar and optical imagery for monitoring cloud-prone forests | Free and open, attribution |
| National forest inventories | Plot measurements of species, size, condition and growth | Terms per agency |
Biodiversity Nexus prepares evidence. The decisions below are taken by the authorities named, under their own rules, and never by the platform.
| Who decides | The question | Evidence this page serves |
|---|---|---|
| Forestry and land-use ministries | Where to enforce, recognise tenure or restore, and how to report forest reference levels | Loss, degradation and driver maps with stated accuracy; inventory data |
| Competent authorities under deforestation-free supply rules | Whether a commodity plot was cleared or degraded after the regulatory cut-off date | Plot-level loss history with alert dates and map accuracy |
| Commodity buyers and lenders | Which sourcing regions and clients carry deforestation and conversion risk | Loss by driver and jurisdiction; fragmentation and fire trends |
| Indigenous and community forest governance bodies | How to monitor and defend their territories against incursion | Alerts and loss maps for their lands, held under their own data terms |
Forest and landscape ecologists work on questions that cross scales from the tree to the region.
Clearing removes habitat outright, but degradation through selective logging, grazing, fire and road building often affects larger areas and is harder to see. Fragmentation splits what remains into smaller patches with more edge. Edges are hotter, drier and windier than interiors, and they let in invasive species, hunters and fire. Small patches support smaller populations that are more prone to local extinction, and isolated patches are less likely to be recolonised.
Climate change adds stress from within. Hotter droughts raise tree mortality, warmer winters let some pests survive and spread, and longer dry seasons lengthen fire seasons. In some regions these pressures combine with fragmentation to create a risk that forest gives way to more open vegetation, with consequences for regional rainfall, water supply and carbon storage.
Optical and radar imagery map clearing and canopy loss consistently over time. Degradation beneath the canopy and in dry forests is harder to detect.
Patch size, edge density and connectivity are computed with geospatial information systems. Metrics depend on map resolution and class definitions.
Permanent plots measure growth, mortality and species. They are the reference for satellite estimates but sparse in many regions.
Circuit and least-cost models predict movement routes. They rest on assumed resistance values that are rarely validated.
The current service level is R0: information and enquiries. An enquiry receives a reference and names the next step; applications, payments, compute allocations and registrations open only when their team, terms and review capacity exist.