DIS4ME DIS4ME Homepage | DESERTLINKS Homepage | © DESERTLINKS 2004
English-EN | Español-ES | Italiano-I | Ελληνικά-GR | Portuguese-PT
Desertification Indicator System for Mediterranean Europe

 

The main issues associated with Mediterranean desertification

Return to the introduction

 

Overgrazing
Lead author: Vasilios P. Papanastasis <vpapan@for.auth.gr>
With contributions from: Maria José Roxo and Pedro Cortesao Casimiro <mj.roxo@iol.pt>

This report draws on research results from projects other than DESERTLINKS.

  • Land Use systems in the Mediterranean Mountains and Marginal Lands 1995-1998 (contract no. AIR-3-CT-93-2426). This project assessed the effect of EU policies for livestock husbandry on local land use systems and the impact of grazing practices on Mediterranean mountains and marginal lands. Most of the data included in this section is derived from this project.
  • GeoRange 2001-2004 (contract no. EVK2-CT2000-00091). Like DESERTLINKS this was a project in the area of land degradation/desertification in the Framework V Environment and Sustainable Development Programme. GeoRange was set up by experts in rangeland ecology and management, ecosystem conservation and restoration, remote sensing and spatial information systems. With the direct involvement of responsible land managers, it aimed at the definition of optimised management strategies for multi-functional rangelands. For more information on GeoRange click here.

g Description of reasons leading to overgrazing and why it is an issue in the context of desertification
g Examples of reasons for overgrazing in European Mediterranean areas
g Portugal
g Greece
g Overview of how the indicators inter-relate
References

g Description of reasons leading to overgrazing and why it is an issue in the context of desertification
Author: Vasilios P. Papanastasis <vpapan@for.auth.gr>

Livestock grazing is an old practice in Mediterranean Europe. It dates back to the Neolithic period when the first domesticated sheep and goats arrived in the region. Since then livestock husbandry has become a dominant human activity supporting civilizations and shaping Mediterranean ecosystems and landscapes.

In discussing the factors underlying deforestation in the Mediterranean countries, Thirwood (1981) considers grazing by domestic animals among the major causes, with goats singled out for their predilection of woody forage. Also, Tsoumis (1985) considers that grazing had a major contribution to deforestation, more than agricultural clearances, with goats being the main catastrophic agents. The same views are shared by Tomaselli (1977), who in addition points out that grazing can prevent the evolution of maquis and garrigue to high forests. The convictions against goats were so strong in the middle of the past century that several countries had to take decisive measures to reduce their numbers or even eliminate them completely by subsidizing their slaughter (FAO, 1964). As a result of these views, the "Ruined Landscape" theory about Mediterranean Europe has been developed (Grove and Rackham, 2001). As a matter of fact, the blame on livestock for destroying the environment is confounded with the mismanagement applied for which the sole responsibility is with men and not the animals (Papanastasis, 1986).

Rangelands of the Mediterranean region include grasslands, also known as pastures, as well as "woody" rangelands, namely dwarf shrublands (e.g. phrygana, batha, tomillares), shrublands (e.g. garrigue, maquis, matorral) and open forests (less than 40% tree canopy), also known as silvopastoral systems. According to Le Houerou (1981), these rangelands are grazed by 270 million sheep-equivalents, which include horses, mules, donkeys, cattle, camels, pigs, sheep and goats. The latter two kinds of animals are the dominant group making 75% of the entire population. All these animals graze on about 830,000 km² of rangelands corresponding to a stocking rate of about 2.2 sheep-equivalents per hectare. If we consider that the grazing capacity of Mediterranean rangelands is no more than 1 sheep-equivalent/ha/year on the average, we may conclude that these rangelands are seriously overgrazed. However, the grazing pressure is not evenly distributed all over the Mediterranean rangelands and it is certainly higher in the south than in the north Mediterranean. In Mediterranean Europe, it is also unevenly distributed with areas being highly overgrazed (e.g. lowlands, around villages) as well as undergrazed (e.g. remote areas).

In discussing the reasons for overgrazing in Mediterranean rangelands, Le Houerou (1981) considers the main reason to be the Mediterranean climate itself, and more specifically the mild and rainy winters which allow grazing animals to stay outdoors not only in the summer - as happens in the temperate areas - but also in the winter, resulting in nearly year-long grazing periods. As a second reason he considers the socio-economic conditions which provide a social status to farmers having large flocks, or force farmers to raise large flocks in order to make a living. However, there are additional, very important, reasons. One such reason is the marginality of grazing areas in the Mediterranean region made up of hilly and mountainous lands with relatively shallow and rocky soils and steep slopes resulting in low productivity and grazing capacity. In dry and semi-dry areas, this problem is further aggravated by the low and erratic precipitation. Another important reason is land tenure. A large proportion of rangelands in the Mediterranean region including southern Europe are state or municipality and community-owned areas that are communally grazed by livestock of the local people (e.g. the island of Crete - Papanastasis, 1993). Under this communal system, grazing management is difficult or impossible and most often leads to overgrazing (Papanastasis, 1988). A final reason is the EU policies applied before the Agenda 2000 was implemented that subsidised the number of animals, thus encouraging farmers in Mediterranean EU countries to increase their flocks in order to receive higher levels of subsidy (Dubost, 1998; Pulina et al., 1998).

Overgrazing by goats in a communal shrubland in north-eastern
Greece (photo by V. Papanastasis)

Grazing has multiple effects on natural ecosystems. Animals defoliate vegetation and consequently affect plant growth, plant vigour, plant reproduction, species composition, plant cover and biomass, thus resulting in bare soil. Grazing animals also trample the soil thus reducing bulk density and infiltration rates and increasing overland flow. If the slopes are steep and the soils erodible then soil erosion may result leading to desertification. This can happen, however, only when overgrazing is applied on a continuous basis, namely when too many animals are trying to feed on a limited supply of forage (Dregne, 1978).

Accelerated erosion in a rangeland overgrazed by cattle in north-western Greece (photo by V. Papanastasis).

Overgrazing has a negative effect on plant diversity. Although several individual plant species are adapted to intensive grazing or seem to be favoured due to the reduction of competition (Bergmeier, 1998; Egli, 1991; Grove and Rackham, 2001; Seligman and Perevolotsky, 1994), the overall impact of overgrazing is negative, particularly in grasslands (Koukoura et al. 1998; Koutsidou and Margaris, 1998; Papanastasis, 1985; Papanastasis et al. 2002). On the contrary, moderate grazing has a beneficial effect on plant diversity (Montalvo et al., 1993; Naveh and Whittaker, 1979; Noy-Meir, 1998; Puerto et al., 1990), but undergrazing or no grazing at all may also produce negative effects (Peco et al., 1998). Undergrazed or ungrazed rangelands present the problems of abandoned lands being invaded by woody species, which increase the fire risk and result in devastating wildfires.

In woody rangelands, the impact of overgrazing may not be adverse to plant diversity because the woody species can play a buffering role. This is the case with the phryganic communities, in which phryganic species, being unpalatable to animals themselves, can protect herbaceous species from overgrazing under their canopy. This buffering role however is eliminated if overgrazing is combined with wildfires (Papanastasis et al., 2002).

Wildfires set by shepherds to control undesirable growth of vegetation is a common practice in several parts of Mediterranean Europe (e.g. Corsica, Sardinia, Crete, western Greece). Although Mediterranean vegetation is well adapted to fire and usually grows back after burning, it can be destroyed if burning is combined with overgrazing. Several studies have shown that the combination of wildfires and overgrazing are the main cause of rangeland degradation and desertification in the Mediterranean Europe (Arianoutsou-Faraggitaki, 1985; Aru, 1986; Margaris and Koutsidou, 1998; Pantis and Mardiris, 1992; Pantis and Margaris 1988; Papanastasis, 1977; Papanastasis et al. 1990; Vokou et al., 1986).

In conclusion, overgrazing is an issue of desertification in Mediterranean Europe particularly in those areas where it is combined with pastoral wildfires

5 top


g Examples of reasons for overgrazing in European Mediterranean areas

g Lower Inner Alentejo, Portugal
Authors: Maria José Roxo and Pedro Cortesao Casimiro <mj.roxo@iol.pt>

The importance of cattle breeding in the area of the municipality of Mértola is well documented. Royal laws exist, dating from the medieval period, protecting the activity. In the past, sheep and pigs extensively grazed natural pastures, in a balanced use of local resources. Vast areas of the Mértola municipality provided spring grazing for flocks brought in from other regions of the country (for example the Algarve and higher Alentejo).

Sheep, Serra de Mértola, Mértola (photo by Maria Roxo, Pedro Casimiro)

Now, incentives over many decades to increase cereal production have led to a significant decrease of natural pasture areas. This has resulted in the remaining areas for grazing cattle becoming more intensively used, and consequently increasingly degraded.

In addition CAP incentives for sheep, pig and cattle production have led to a significant increase in numbers of animals per hectare and new methods of husbandry. Previously shepherds grazed their cattle over extensive areas, but now they stay in the same place supported by the installation of an infrastructure of fences, stables and small dams to provide drinking water. This causes a more prolonged effect on the soil from trampling and pasture exhaustion.

Cattle, Serra Mértola, Mértola (photo by Maria Roxo and Pedro Casimiro)

All these factors have contributed to more intensive and widespread land degradation in the area. It is important that attitudes towards the activity are changed and that husbandry systems are adopted that are more suited to the specific soil and climatic conditions of the area.

5 top

g Psilorites mountain, Crete (Greece)
Author: Vasilios P. Papanastasis <vpapan@for.auth.gr>

Psilorites mountain has a maximum altitude of 2,456 m a.s.l. It is located at the middle of Crete with an area of about 500 km², mostly lying above 600 m a.s.l. It is permanently inhabited by about 18,000 people living in 20 village communities. The dominant bedrock is undivided limestone and the soils are red. The climate is sub-humid Mediterranean with wet and mild winters that become cold in high altitudes. Vegetation is complex; it includes mostly evergreen but also a few deciduous woody species as well as phryganic species, which dominate the mountain. Land tenure is also complex; agricultural lands are privately owned but forests and rangelands are state owned and the right to graze them belongs to the local people.

General view of Psilorites mountain (photo by V. Papanastasis).

As far as the soils are concerned, they were found to be very shallow (<15 cm) or shallow (15-30 cm) on most of the mountain, with about 40% of the surface area occupied by rocky outcrops (Pendarakis, 1994). They get deeper in karstic dolines, which however are limited in area, and found towards the foot of the mountain.

Livestock husbandry is a traditional activity that has shaped Psilorites mountain since the Neolithic period (Lyrintzis and Papanastasis, 1995). In the last few decades, however, the number of grazing animals have been dramatically increased, mainly due to national initiatives,and, especially since 1981 (when Greece became a member of the European Union), due to EU subsidies. Based on National Statistical Data, the numbers of sheep were increased by 529% and of goats by 279%, from 1961 until 1991. These high numbers in 1991 resulted in an average stocking rate of 4.6 sheep-equivalents/ha/year (Menjli, 1994), which is at least four times higher than the grazing capacity of rangelands (Papanastasis et al., 1990), suggesting overgrazing.

Evolution of livestock numbers between 1961 and 1991 in the Psilorites mountain of Crete (Data from National Statistical Service).

In order to study the impact of overgrazing on Psilorites mountain, air photographs taken in 1961 and 1989 were processed and the major land use/cover types were identified and recorded for the two periods. The results showed an increase of sparse and medium density shrublands (mainly phryganic ecosystems) at the expense of the dense ones, as well as of forests. This evolution indicates degradation, since fewer shrubs suggest less woody cover to protect the soil and sustain productivity. The impact of overgrazing was more severe when it was combined with pastoral wildfires.

Land use/cover changes on the Psilorites mountain of Crete between 1961 (top) and 1989 (below) (Bankov, 1998).
Desertification caused by the combination of pastoral wildfires
and overgrazing on Psilorites mountain, Crete (photo by V. Papanastasis).

5 top


g Overview of how the indicators inter-relate
Author: Vasilios P. Papanastasis <vpapan@for.auth.gr>

Overgrazing is management oriented but as a process it is affected by several physical and socio-economic factors as well. Assessment indicators may relate to management, vegetation and abiotic conditions as well as to social, economic and political ones.

Overgrazing is caused when the number of animals carried in a rangeland are more than its grazing capacity, suggesting these extra animals could be from a few to too many. As a result, the number of animals grazing in a rangeland or the grazing intensity, expressed as stocking rate, is a very important indicator of rangeland degradation (Papanastasis, 1998; 2000). Due to the different way that the various grazing animals collect the forage, their impact on vegetation is different (Rook et al., 2004) and therefore the kind of animal species is very important in the overgrazing process. Also, the system with which animals graze in rangelands is important, too. For example, continuous grazing on a year-long basis is more adverse to species composition than a seasonal or rotational grazing system (Sternberg et al., 2000). It has been already mentioned that combining overgrazing with wildfires can be more detrimental than either of the two processes alone. The distribution of available infrastructure may decide whether a rangeland will be evenly used or not; normally the animals tend to graze more intensively near the watering points and animal sheds than away from them (Ghossoub, 2003). Finally, overgrazing will be avoided if alternative feed resources are available and therefore the production system applied is affecting the impact of grazing animals on rangelands (Papanastasis, 1990).

Increased land degradation around a watering point on Dia islet (Crete) (photo by V. Papanastasis).

For vegetation, important indicators are the amount of biomass produced or left at the end of the grazing period, the plant cover (Papanastasis et al., 2003) and the species composition. Usually, when overgrazing has occurred, the pasture is filled with weeds, which are undesirable plant species to animals. Soil and climatic variables such as soil depth, slope gradient, parent material, soil erosion, rainfall (amount and distribution), and temperature are all affecting vegetation and therefore the number of animals that can be grazed on a rangeland. There is a direct relationship between the soil depth (Papanastasis, 1994) or the amount of surface rocks (Alexandris et al., 1997) and herbage production in grasslands. Also, herbage production is very much affected by rainfall and air temperature (Papanastasis, 1982).

Considering socio-economic indicators, they may include EU subsidies for animal numbers, the local traditions as far as the flock size is concerned, the land tenure (private, state or communal rangelands), the alternative income that the farmers can have apart form raising livestock and the laws that dictate the use of rangelands by the farmers. All these indicators interrelate with each other and with several physical indicators (e.g. productivity of rangelands) as well as management factors (e.g. number and kind of animals, grazing system, etc).

Overgrazing is related ro other issues, particularly land abandonment and deforestation.

5 top


g References

  • Alexandris, S., V.P. Papanastasis and K. Papadimitriou. 1997. Range sites in calcareous areas of western Macedonia. In: Proc. Sustainable Utilization of Rangelands and Pastures, Thessaloniki, pp. 78-90 (In Greek with English summary)
  • Arianoutsou-Faraggitaki, M. 1985. Desertification by overgrazing in Greece: The case of Lesvos island. Journal of Arid Environments, 9:237-242.
  • Aru, A. 1986. Aspects of desertification by overgrazing in Sardinia-Italy. In: Fantechi R. and N.S. Margaris (eds), Desertification in Europe. D. Reidel Publishing Co., Dordrecht, pp. 194-198.
  • Bankov, N. 1998. Dynamics of land cover/use changes in relation to socio-economic conditions on the Psilorites mountain of Crete, Greece. M.Sc. Thesis. Mediterranean Agronomic Institute, Chania, Crete, 136 pp.
  • Bergmeier, E. 1998. Are Cretan endemics threatened by grazing? In: Papanastasis, V.P and D. Peter (eds), Ecological Basis for Livestock Grazing in Mediterranean Ecosystems. European Commission, EUR 18308 N, Luxemburg, pp. 90-93.
  • Dregne, H.E. 1978. Desertification: man's abuse of land. J. Soil and Water Cons., 33:11-14.
  • Dubost, M. 1998. European policies and livestock grazing in the Mediterranean ecosystems. In:
  • Papanastasis, V.P. and D. Peter (eds), Ecological Basis for Livestock Grazing in Mediterranean Ecosystems. European Commission, EUR 1830N, Luxemburg, pp. 298-311.
  • Egli, B.R. 1991. The special flora, ecological and edaphic conditions of dolines in the mountains of Crete. Botanica Chronika 10:352-335.
  • FAO, 1964. Rapport sur les politiques de l' elevage de la chévre dans la région mediterraneenne et la Proche-Orient. PEAT/1929, Rome.
  • Ghossoub, R. 2003. Impact of piospheric points in Mediterranean rangelands. M.Sc. Thesis. Mediterranean Agronomic Institute, Chania, Crete, 126 pp.
  • Grove, A.T. and O. Rackhman. 2001. The Nature of Mediterranean Europe, An Ecological History. Yale University Press, London.
  • Koukoura, Z., C. Tsiouvaras, and V.P. Papanastasis. 1998. Long term effects of grazing on biodiversity of a Mediterranean grassland in northern Greece. In: Papanastasis, V.P. and D. Peter (eds),
  • Ecological Basis for Livestock Grazing in Mediterranean Ecosystems. European Commission, EUR 18308 N, Luxemburg, pp. 53-57.
  • Koutsidou, E. and N.S. Margaris. 1998. The regeneration of Mediterranean vegetation in degraded ecosystems as a result of grazing pressure exclusion: the case of Lesvos island. In: Papanastasis, V.P. and D. Peter (eds), Ecological Basis for Livestock Grazing in Mediterranean Ecosystems. European Commission, EUR 18308 N, Luxemburg, pp. 76-79.
  • Le Houerou, H.N. 1981. Impact of man and his animals on Mediterranean vegetation. In F.di Castri et al. (eds), Mediterranean Type Ecosystems, Elsevier Scientific Co., Amsterdam, p. 479-521.
  • Lyrintzis, G. and V.P. Papanastasis. 1995. Human activities and their impact on land degradation - Psilorites mountain in Crete: A historical perspective. Land Degradation and Rehabilitation, 6:79-93.
  • Margaris, N.S. and E. Koutsidou. 1998. Fires and overgrazing: the main driving forces of desertification in the Aegean islands. In: Papanastasis, V.P. and D. Peter (eds), Ecological Basis for Livestock Grazing in Mediterranean Ecosystems. European Commission, EUR 1830N, Luxemburg, pp. 167-170.
  • Menjli, 1994. Effects of pastoral activities on desertification of mountain Psilorites. M.Sc. Thesis. Mediterranean Agronomic Institute of Chania, Crete, 99 p.
  • Montalvo, J., M.A. Casado, C. Levassor, and F.D. Pineda. 1993. Species diversity patterns in Mediterranean grasslands. Journal of Vegetation Science, 4:213-222.
  • Naveh, Z. and R.H. Whittaker. 1979. Structural and floristic diversity of shrublands and woodlands in northern Israel and other Mediterranean regions. Vegetatio, 41:171-190.
  • Noy-Meir, I. 1998. Effects of grazing on Mediterranean grasslands: the community level. In: Papanastasis, V.P. and D. Peter (eds), Ecological Basis for Livestock Grazing in Mediterranean Ecosystems. European Commission, EUR 1830 N, Luxemburg, pp. 27-39.
  • Pantis, J. and N.S. Margaris. 1988. Can system dominated by asphodels be considered as semi-deserts? International Journal of Biometeorology, 32:87-91.
  • Pantis, J.P. and Th.A. Mardiris. 1992. The effects of grazing and fire on degradation processes of Mediterranean ecosystems. Israel Journal of Botany, 41:233-242.
  • Papanastasis, V.P. 1977. Fire ecology and management of phrygana communities in Greece, In: Mooney, H.A. and Conrad, C.E. (coors), Proc. Symp. Environmental Consequences Fire and Fuel Manage in Mediterranean-type Ecosystems. USDA Forest Service, Gen. Tech. Rep. WO-3, Washington D.C, p. 476-82.
  • Papanastasis, V.P. 1982. Production of grasslands in relation to air temperature and precipitation in northern Greece. Dassiki Erevna, 3, 111 pp. (In Greek wiht English summary).
  • Papanastasis, V.P. 1985. Impact of grazing on a grassland of the Lahanas region of Thessaloniki. Dassiki Erevna, VI(2): 155-168 (In Greek with English summary).
  • Papanastasis, V.P. 1986. Integrating goats into Mediterranean forests. Unasylva, 154(38):44-52.
  • Papanastasis, V.P. 1988. Livestock grazing legislation in Greece. In: Forestry Legislation. Forstwissenschaftliche Beitrage 6, Fachbereich Forstokoenomie und Forstpolitik. ETH. Zurich, pp.117-128.
  • Papanastasis, V. 1990. Pastoral and forage systems in Greece. 6th Meeting FAO Subnetwork on Mediterranean Pastures and Fodder Crops, Bari, Italy. pp. 177-180.
  • Papanastasis, V. 1993. Legal status of land tenure and use and its implication for open landscapes of western Crete. Landscape and Urban Planning, 24:273-277.
  • Papanastasis, V. 1994. Structure and production characteristics of limestone rangelands in western Macedonia, Greece. In: L. t' Mannetje and J. Frame, (eds in chief), Grassland and Society. Wageningen Press, The Netherlands, pp. 316-317.
  • Papanastasis, V.P. 1998. Grazing intensity as an index of degradation in semi-natural ecosystems: the case of Psilorites mountain in Crete, In: G. Enne, M. D' Angelo and C. Zanolla, (eds), Indicators for Assessing Desertification in the Mediterranean. Nucleo Ricerca Desertificatione. Universitá di Sassari. Italy, pp.146-158.
  • Papanastasis, V.P. 2000. Land degradation caused by overgrazing and wildfires and management strategies to prevent and mitigate their effects. In: Desertification in Europe: mitigation strategies, land-use planning G. Enne, Ch. Zanolla and D. Peter (eds). European Commission, EUR 19390, Luxembourg, pp. 187-198.
  • Papanastasis, V., S. Kyriakakis, and J. Ispikoudis. 1990. Forestry and grazing practices in Crete. In: Grove, A.T., O. Rackham and J. Moody (eds), Stability and Change in the Cretan Landscape. Petromaroula 1, Corpus Christi College, England, pp. 42-46.
  • Papanastasis, V.P., S. Kyriakakis and G. Kazakis. 2002. Plant diversity in relation to overgrazing and burning in mountain Mediterranean ecosystems. J. Mediterranean Ecology, (2-3):53-63.
  • Papanastasis, V.P., S. Kyriakakis, G. Kazakis, M. Abid and A. Doulis. 2003. Plant cover as a tool for monitoring desertification in mounting Mediterranean rangelands. Management of Environmental Quality: An International Journal, 14(1): 69-81.
  • Pulina, G., A. Cappio-Borlino, M. d'Angelo and A.H.D. Francesconi. 1998. Grazing in Mediterranean ecosystems: a complex approach as addressed in the EU MEDALUS project. In: Papanastasis, V.P. and D. Peter (eds), Ecological Basis for Livestock Grazing in Mediterranean Ecosystems. European Commission, EUR 1830N, Luxemburg, p. 173-188.
  • Peco, B., T. Espigares, and C. Levassor. 1998. Trends and fluctuations in species abundance and richness in Mediterranean pastures. Applied Vegetation Science, 1:21-28.
  • Pendarakis, K.E. 1994. Identification of basic soil parameters related with the desertification process in two sampling zones of Psilorites area. Mimeo. 42p.
  • Puerto, A., M. Rico, M.D. Matias and J.A. Garcia. 1990. Variation in structure and diversity in Mediterranean grasslands related to trophic status and grazing intensity. Journal of Vegetation Science, 1:445-452.
  • Rook, A.J., B. Dumont, J. Isselstein, K. Osoro, M.F. Wallis De Vries, G. Parente and J. Mills. 2004. Matching type of livestock to derived biodiversity outcomes in pastures-a review. Biological Conservation, 119:137-150.
  • Seligman, N.G. and A. Perevolotsky, 1994. Has intensive grazing by domestic ungulates degraded Mediterranean basin rangelands? In: Arianoutsou, M. and R.H. Groves (eds), Plant-Animal Interactions in Mediterranean - Type Ecosystems. Kluwer Academic Publ. The Hague, pp. 93-103.
  • Sternberg, M., M. Gutman, A. Perevolotsky, E.D. Ungar and J. Kigel. 2000. Vegetation response to grazing management in a Mediterranean herbaceous community: a functional group approach. Journal of Applied Ecology 37:224-237.
  • Thirgood, J.V. 1981. Man and the Mediterranean Forest. Academic Press, New York.
  • Tomaselli, R. 1977. Degradation of the Mediterranean maquis, In: Mediterranean Forests and Maquis. Ecology, Conservation and Management, MAB Technical Series 2. UNESCO, p. 32-72.
  • Tsoumis, G. 1985. The depletion of forests in the Mediterranean region- An historical review from the ancient times to present. Scientific Annals of the Department of Forestry and Natural Environment, Vol. KH (11):281-300.
  • Vokou, D., J. Diamantopoulos, Th.A. Mardiris and N.S. Margaris. 1986. Desertification in northern Greece: the case of Kella. In: Fantechi, R. and N.S. Margaris (eds), Desertification in Europe. D. Reidel Publishing Co. Dordrecht, pp. 155-160.

5 top