English-EN | Español-ES | Italiano-I | Ελληνικά-GR | Portuguese-PT
Desertification Indicator System for Mediterranean Europe

1. Definition



Brief definition

The process of transforming large, contiguous forest patches into one or more non contiguous smaller patches, surrounded by disturbed areas.

Unit of measure

% of territory involved

Spatial scale


Temporal scale


2. Position within the logical framework DPSIR

Type of Indicator

Driving Force/State

3. Target and political pertinence


This indicator measures forest fragmentation. This refers to a landscape-level process in which a specific habitat is progressively sub-divided into smaller, geometrically more complex, and more isolated fragments, as a result of both natural and human activities.

Importance with respect to desertification

The primary cause of forest fragmentation is suburban development, which causes many of the remaining large parcels of forests to be converted for anthropogenic uses, most notably as residential development.

The fragmentation of forests by agricultural clearing has been exacerbated by land taken for urban development and infrastructures, as well as by large-scale and repeated fires in Mediterranean regions. In addition, human activity has utilized most of the remaining forest. In the late 1990s this trend was partly reversed by afforestation and spontaneous regrowth on former agricultural land.

Increasingly, previously extensive, continuous tracts of forest are being reduced to widely dispersed patches of remnant forest vegetation by logging and road-building, but few measures of the effects of roads on forest fragmentation are available. Fragmentation affects animal populations in a variety of ways, including decreased species diversity and lower densities of some animal species in the resulting smaller patches.

Fragmentation is also responsible for bringing about an alteration in the microclimate within and surrounding remnant patches of vegetation. This can result in altered physical fluxes eg. radiation, wind and water.

An important consequence of forest fragmentation is the increase in forest edges, along which there are strong microclimatic gradients, or edge effects. These include variations in light, temperature, soil moisture content and wind turbulence. It has been suggested that a 500-1,000 m (0.5 - 1.0 km) buffer zone is needed to accommodate all edge effects. Edge effects increase dramatically with fragmentation. Importantly, as fragmentation and deforestation continues, the proportion of the remaining total area of ancient forest affected by fragmentation increases.

Habitat fragmentation has become a problem of general concern. The fragmentation concept encompasses three aspects:
(1) reduction of the total area,
(2) scattering of residual formations in a mosaic in which the remaining individual elements are disjointed and
(3) increasing of total edge, therefore enhanced exchanges with the surrounding parts of territory.

The latter is strongly related to fire because of a larger and larger perimeter with increasing fragmentation, therefore easier passage for running fires. In a very fragmented forest scenario (Leone et al. 2001) it has been observed that edge values of 83 m ha-1 are present.

Fragmentation therefore involves changes in landscape composition, structure and function at many scales, and occurs on a backdrop of a natural patch mosaic, created by changing landforms and natural disturbances.

Defined in such a way, the fragmentation concept appears to be scale-free: it addresses no particular scale because the phenomena can occur at all scales. Also the definition addresses no particular kind of habitat because the fragmentation can occur in most types of habitat.

International Conventions and agreements

The UNCCD recognised the particular conditions of the Mediterranean affecting desertification processes, including the extensive forest coverage losses due to different causes (deforestation, frequent wildfires, intensive grazing etc) (Convention text as of September 1994 and as of September 2001)

Secondary objectives of the indicator

This indicator represents the impact due to reduction or loss of forest surface. Information about forest fragmentation can help in addressing political measures to recognize the loss of cover vegetation in a specific area and to organize efficient control and indirectly reduce desertification.

4. Methodological description and basic definitions

Definitions and basic concepts

Fragmentation describes vegetated land cover shrinkage, scattering and disjointing. These conditions induce the transformation of a continuous cover into a "porous" mosaic. Fragmentation involves changes in landscape composition, structure, and function at many scales and occurs on a backdrop of a natural patch mosaic, created by changing landforms and natural disturbances.

Benchmarks Indication of the values/ranges of value

Pf<0.4 = low incidence
0.4 < Pf < =0.6 = moderate incidence
Pf>0.6 and Pf-Pff>0 = high incidence
Pf > 0.6 and Pf = Pff = very high incidence

Methods of measurement

There are numerous measures of forest fragmentation and forest connectivity currently described in the literature. These include average forest patch size, percent interior forest, mean forest patch density, number of forest patches, interpatch distance, forest patchiness, forest contiguity, forest continuity, and proportion of forest in the largest forest patch (Vogelmann, 1995; Trani and Giles, 1999; Wickham et al. 1999). Several of these measures were assessed in a study conducted by Trani and Giles (1999) to determine how change in forest loss influenced their results.

The model of Riitters et al. (2000), developed to assess forest fragmentation at the global level, generates categories that describe the type of forest fragmentation condition that exists for a given forest pixel. Two values that characterize a forest pixel at the centre of a moving window are necessary. The first is the proportion of forest, or Pf. In a window where the centre pixel is forest, Pf is the number of pixels in the window that are forest divided by the number of pixels in the window that are not forest.

pF = (forest pixels in window)/(pixels in window that are not forest)

The second measure is the connectivity of the forest, or Pff. Pff measures connectivity using pixel pairs. It is the number of pixel pairs where both pixels are forest divided by the number of pixel pairs with at least one forest pixel.

Pff = (pixel pairs that are both forest)/((pixel pairs that are both forest)+(pixel pairs with one forest))

Pf and Pff are used to distinguish between six types of forest pixels:
Interior forest (Pf = 1) All of the pixels surrounding the centre pixel are forest.
Patch forest (Pf<0.4) Pixel is part of a forest patch on a non-forest background, such as a small wooded lot within an urban region.
Transitional forest (0.4 < Pf < 0.6) About half of the cells in the surrounding area are forested and the centre forest pixel may appear to be part of a patch, edge, or perforation depending on the local forest pattern
Edge forest (Pf>0.6 and Pf-Pff<0) Most of the pixels in the surrounding area are forested, but the centre pixel appears to be part of the outside edge of forest, such as would occur along the boundary of a large urban area, or agricultural field.
Perforated forest (Pf>0.6 and Pf-Pff>0) Most of the pixels in the surrounding area are forested, but the centre pixel appears to be part of the inside edge of a forest patch, such as would occur if a small clearing was made within a patch of forest.
Undetermined forest (Pf > 0.6 and Pf = Pff) Most of the pixels in the surrounding area are forested, but this centre forest pixel could not be classified as a type of fragmentation in the surrounding area.

Limits of the indicator

One limit of the indicator is the difficulty in evaluating the protective cover effects of the mosaic.

Linkages with other indicators

Deforested area, Area of matorral, Biodiversity conservation, Forest productivity, Forest management quality

5. Evaluation of data needs and availability

Data required to calculate the indicator

E.O. data, land-cover maps, forest maps.

Data sources

Land cover maps, forest maps, aerial photos, digital land cover maps classified from remote-sensing data.

Availability of data from national and international sources

Data on forest surface can be provided by the Eurostat and National Statistic Organizations. Satellite data too can be used to measure current forest cover.

6. Institutions that have participated in developing the indicator

Main institutions responsible

University of Basilicata, Italy

Other contributing organizations


7. Additional information


Forman, R.T.T. 1995. Land mosaics: the ecology of landscape and regions. Cambridge University Press, Cambridge.

Riitters, K., J. Wickham, R. O'Neill, B. Jones, and E. Smith. 2000. Global-scale patterns of forest fragmentation. Conservation Ecology 4(2):1-28. URL: http://www.consecol.org/vol4/iss2/art3

Trani, M.K. and R.H. Giles, Jr. 1999. An analysis of deforestation: metrics used to describe pattern change. Forest Ecology and Management. 114:459-470.

Vogelmann, J.E. 1995. Assessment of forest fragmentation in southern New England using remote sensing and geographic information system technology. Conservation Biology 9(2):439-449.

Wickham, J.D., K.B. Jones, K.H. Riitters, T.G. Wade and R.V. O'Neill. 1999. Transitions in forest fragmentation:implications for restoration opportunities at regional scales. Landscape Ecology 14: 137-145.

Leone V., Lombardi V., Lovreglio R., Mininni M., Semerari P.,2002 Inventario dei boschi dell'Alta Murgia. Primi risultati in "Studi per il Piano d'area dell'Alta Murgia" a cura del DAU, Dipartimento di Architettura e Urbanistica, Politecnico di Bari

Other references


Contacts Name and address

Prof. Agostino Ferrara
University of Basilicata
Via dell'Ateneo Lucano
85100 Potenza, Italy
e-mail: ferrara@unibas.it