Methodology & Data: Atlas of Forest Landscape Restoration Opportunity
Source information for each data layer:
1. Current forest coverage
The map of current forests shows the global extent of forest coverage, including categories of forest density. It is based on a combination of two satellite-derived products: 1) a global forest map derived from MODIS 250m data for the period 2000 to 2009 (South Dakota State University, 2011, unpublished dataset), which was used to map the general extent of forests independent of canopy density; and 2) a tree canopy density map derived from the MODIS vegetation continuous fields (VCF) data (Hansen et al., 2003), which was used to separate classes of tree density (e.g. closed, open, and woodland).
Cite as: Peter Potapov, Matthew Hansen, and Svetlana Turubanova. 2011. Global forest extent derived using MODIS 250m data. University of Maryland.
2. Potential forest coverage
The map of potential forests represents an estimate of where forests would grow under current climate conditions and without human influence. The main source of data for defining potential forest extent is the terrestrial ecoregions of the world (Olson et al., 2001). Each ecoregion was classified as belonging to one of four categories: dense forests, open forests, woodlands, or non-forest, depending on its description (including current and potential vegetation) and its proportion of different forest types, with additional input from the following datasets: current forest extent (see above); bioclimatic zoning and original forest cover extent (FAO, 1999; Bryant et al., 1997; Zomer et al., 2008); and a forest distribution map produced by modeling based on global climate variables and elevation.
Cite as: Peter Potapov, Lars Laestadius, and Susan Minnemeyer. 2011. Global map of potential forest cover. World Resources Institute: Washington, DC. Online at www.wri.org/forest-restoration-atlas.
3. Forest condition
A comparison of the maps of current and potential forests makes it possible to identify forest condition, including areas of historical forest loss and degradation. There are four basic categories for forest condition:
- Intact: No forest conversion or degradation has taken place;
- Fragmented/managed: Natural forests and woodlands that have experienced some level of timber extraction (e.g., selectively logged forests or secondary forests) or are managed as plantations.
- Degraded: A reduction in the volume, tree canopy cover and biodiversity of forested areas;
- Deforested: Formerly forested areas that have been converted to other non-forest land uses.
Cite as: Peter Potapov, Lars Laestadius, and Susan Minnemeyer. 2011. Global map of forest condition. World Resources Institute: Washington, DC. Online at www.wri.org/forest-restoration-atlas.
4. Restoration opportunities
Data on forest condition and current land use were used to derive the map of opportunities for restoration on degraded lands. The land-use data sets include population density, urbanized or industrial areas, and cropland distribution. Areas with high population density or those occupied by intensively managed croplands were considered as having no or low forest restoration potential. Areas with scattered cropland areas, pastures, agroforestry and all types of forest plantations were considered as providing promising opportunities for restoration. Deforested and degraded forest lands were divided into four categories, resulting in a map of restoration opportunity areas and other former forest lands:
- Wide-scale restoration: Less than 10 people per square kilometer and potential to support closed forest.
- Mosaic restoration: Moderate human pressure (between 10 and 100 people per square km).
- Remote restoration: Very low human pressure (density of less than one person per square km within a 500-km radius).
- Forests without restoration needs: Intact forests.
Other former forest lands:
- Agricultural land: Croplands with intensive usage for food production (Pittman, et al., 2010).
- Recent tropical deforestation: Loss of humid tropical forest between 2000 and 2005 (Hansen, et al., 2008).
- Urban areas: Densely populated and industrialized areas (LandScan, 2005).
Cite as: Peter Potapov, Lars Laestadius, and Susan Minnemeyer. 2011. Global map of forest landscape restoration opportunities. World Resources Institute: Washington, DC. Online at www.wri.org/forest-restoration-atlas.
5. Human pressure
A map of land-use intensity (human pressure) was used to assess opportunities for restoration of degraded lands as well as classify degraded lands according to suitability for different types of restoration. Several separately-mapped land-use classes were combined to make the land use intensity map, including population density, built-up areas, pasturelands, croplands and cultivated areas. The resulting data were divided into the following three categories of human pressure:
- High: Lands with high population density (more than 100 persons per square km), croplands, and urban areas. These lands offer opportunities for protective restoration only (e.g., buffering waterways near croplands; erosion prevention on steep slopes; and storm water runoff mitigation).
- Moderate: Lands with a rural population density between 10 and 100 persons per square km. These lands offer opportunities for mosaic restoration.
- Low: lands with a rural population density of less than 10 persons per square km. These lands offer opportunities for wide-scale restoration.
Cite as: Peter Potapov, Lars Laestadius, and Susan Minnemeyer. 2011. Global map of human pressure on the world's forests. World Resources Institute: Washington, DC. Online at www.wri.org/forest-restoration-atlas.
6. Bonn Challenge
At the invitation of the German Government and IUCN, the Bonn Challenge was established at a ministerial roundtable in September 2011 and calls for the restoration of 150 million hectares of deforested and degraded lands worldwide by 2020. The map marks the general locations where countries, regional organizations, businesses, and other entities have pledged to restore forests toward meeting the Bonn Challenge. Learn more, here.
Methodology Behind the Atlas of Forest Landscape Restoration Opportunities
In creating the maps contained in this atlas, the team sought to answer the following questions:
- How much of the world’s forest has been lost?
- Is it realistic to restore forests at a globally meaningful scale?
- How big are the restoration opportunities and where are they located?
We approached the analysis into two steps. First, we would map how much forest has been lost. Then, we would map the opportunities for forests and trees to return.
How much of the world’s forest has been lost?
To answer the question of how much forest has been lost, we first had to answer the question how much forest would there be if humans hadn’t changed the landscape? An initial thought was to create a map of historical forest extent, but that wouldn't work, since the climate has changed a lot in the ten thousand years since the last ice age and the beginning of the agricultural revolution. We would have to start by mapping where forests would be in today’s climate if it weren't for human influence.
To make this map of potential forest coverage, we combined data on climate, soils, and elevation and maps of current and historical forest extent (FAO, 1999; Bryant et al., 1997; Zomer et al., 2008). Then we derived the composition and density of these potential forests from a map of the world’s terrestrial ecoregions (Olson et al., 2001).
Having established this baseline, we proceeded to map current forest coverage. We used data from MODIS, a satellite-based sensor, to map forest location and tree canopy density (Hansen et al., 2003), which we classified into three types:
Closed forest (canopy density greater than 45 percent);
Open forest (canopy density between 25 and 45 percent); and
Woodlands (sparse, savanna type forest; canopy density between 10 and 25 percent).
By contrasting the maps of potential and current forest coverage, we can identify areas that have been deforested and replaced by other land uses. It is interesting to click back and forth between these two maps in the atlas. The deforested areas of the world are very large and comprise many areas that are seldom thought of as being deforested, given that forests have not existed there for hundreds if not thousands of years. These areas include major portions of China, India, Russia, Europe, United States, and Brazil.
But the human impact on the world’s forests is not limited to deforestation. Some forests have diminished without disappearing completely. These degraded forests are difficult to define and even more difficult to map, as the contrast between what they could be and what they are is not quite as clear. When considering how to map these degraded forests, we decided to define them as forests that exist in a state below their potential density. The reasons for degradation can be because of sustained human influence, such as crop or timber production, grazing, or settlements, but also because of fire or pests.
Many forest areas have also been fragmented by roads or affected by logging without causing a sustained loss in density. These fragmented/managed forests are essentially “working forests” and comprise the remainder of the world’s forests, other than deforested, degraded, and intact forest areas.
The resulting forest condition map shows the current situation in all parts of the world where forests could grow naturally. The map shows four types of forest condition:
Intact: Large blocks of virgin forest (> 50,000 hectares);
Fragmented/managed: Natural forests and woodlands that are fragmented by roads and/or managed for timber production;
Degraded: Forests with reduced density;
Deforested: Formerly forested areas.
Degraded and deforested lands, in a biophysical sense, present areas of opportunity for returning forests and trees to the landscape. Biophysical potential, however, is not the same as actual restoration potential. This led us to the second step in the analysis: where can forest and trees realistically be returned to the landscape?
Where can forests and trees be restored?
The actual potential to bring back forests and trees is obviously much smaller than the total area of degraded and deforested land shown on the forest condition map. A large portion of these lands is used for food production, while cities, settlements and a host of other human uses also require land. A realistic map of restoration opportunities would have to exclude such areas.
To do this, we created a map of human pressure by combining urban areas, areas with dense rural population (> 100 person/km2), cultivated areas, and other intensively used areas (Pittman, et al., 2010; LandScan, 2005; GLC2000, Sanderson, et al., 2002). We then eliminated these areas from further consideration, assuming, for the purpose of this analysis, that they are incompatible with returning forests to the landscape.
After applying these constraints, the remaining area is the world map of restoration opportunity areas.
But opportunities to bring back what, exactly? We decided to distinguish two types of restoration opportunity landscapes with regard to the most likely type of restoration:
Wide-scale restoration aims to restore closed forests to the landscape. This type of restoration is more likely in deforested or degraded landscapes with low population density (< 10 people/km2) that are also areas where closed forests formerly dominated the landscape.
Mosaic restoration integrates trees into mixed-use landscapes, such as agricultural lands and settlements, where trees can support people through improved water quality, increased soil fertility, and other ecosystem services. This type of restoration is more likely in deforested or degraded forest landscapes with moderate population density (10 - 100 people/km2).
Remote restoration is reserved for deforested or degraded forest lands that are completely unpopulated and located far away from human settlements, such as northern Canada and Siberia. The reduced density of forests in these areas is likely due to fire and pests, and their remoteness makes them a more costly and lower-priority restoration opportunity.
The restoration opportunity map shows the result of this analysis and suggests several interesting conclusions regarding restoration opportunity globally:
Restoration opportunity landscapes are widespread. More than two billion hectares worldwide offer opportunities for either wide-scale or mosaic-type restoration—an area larger than South America.
Restoration opportunities exist on all continents.
Restoration opportunities are not tied to ongoing deforestation. Most of the opportunity landscapes were deforested or degraded a long time ago and far away from current deforestation hotspots.
Mosaic restoration is the greatest opportunity. A full three quarters of the total opportunity area (1.5 billion hectares) is in this category.
The Way Forward
Knowing the global extent and location of restoration opportunity landscapes is just the starting point, however. At a resolution of 1 km2 (where one pixel is the combined area of 140 soccer fields), these maps provide the big picture of opportunity and show the locations worthy of a closer look. National and local-level analyses to identify the specific types of restoration needed in specific locations are necessary next steps. WRI, in partnership with IUCN, is already working on these detailed assessments. Work in Brazil, Rwanda, Guatemala, Mexico, and Ghana is underway and with locations on the horizon. WRI and IUCN’s Restoration Opportunity Assessment Methodology provides an overview of how we are approaching these assessments.
In 2011, the government of Germany and the Global Partnership on Forest Landscape Restoration established the Bonn Challenge: a call to action for governments and civil society organizations to restore 150 million hectares of deforested and degraded lands worldwide by 2020. The Bonn Challenge map identifies the countries and organizations that have made pledges toward achieving this goal to date, with more pledges being made each year. Keep checking the map to see which new countries have joined the Bonn Challenge and are making forest landscape restoration a priority.
- WRI and IUCN's Restoration Opportunities Assessment Methodology (ROAM) Handbook
- L. Laestadius, S. Maginnis, S. Minnemeyer, P. Potapov, C. Saint-Laurent and N. Sizer, Mapping opportunities for forest landscape restoration, Unasylva 238, 62, 47 (2011/2).
Projects that include this Resource
Global Restoration Initiative
WRI is partnering with governments, businesses, and communities around the world to restore millions of hectares of deforested and degraded land.Part of Forests