Methodology for assessing desertification risk in areas cultivated with oaks
Author: Constantinos Kosmas <lsos2kok@aua.gr>

g Indicator selection

A methodology has been developed for defining land desertification risk in oak forested areas by using simple indicators related to soil, climate, vegetation, socio-economic, and management characteristics. The list of indicators were defined based on: (a) the existing experience obtained in other desertification projects such as MEDALUS I, II, and II, MEDACTION, etc., (b) contacting farmers and land planners and defining a preferable list of indicators, and (c) organizing focus group workshops for finalising the list of possible indicators. In the present study 33 indicators were used which can be easily measured or made available. Most of the used indicators are related to local conditions (farm level) as shown in the Example Survey Form below.

Each indicator was described according to classes, which were defined using existing classification systems such as the geo-referenced soil data base, or existing research data.

Example Survey Form

5 top

g Data collection

Data were collected from the island of Lesvos, and Achaia (N. Peloponnesus) in 101 field sites. The field sites were located on topographic maps in grids of 400 meters by 400 meters. Some were selected by applying a systematic sampling design (Webster, 1977), but others were selected by travelling through the area until a farmer was found on his land. The location of each field site was accurately defined by using a GPS (Magellan, 500 DX).

The indicators were assessed using the following methods:

• Land ownership, family size, farm size, number of parcels, size of the study field site, age of the farmer, parallel employment (if any), application of fertilizers and available subsidies were defined in collaboration with the land user.
• Soil textural classes of particles <2 mm of the non-consolidated parent material, or the parent material, were estimated using the USDA system of soil texture designation. The parent material was defined using the geological map of the study area. The average soil depth was measured in auger holes or in soil cuts. The slope gradient was described using topographic maps of the appropriate scale. The following dominant slope classes were distinguished: <6%, 6-18%, 18-35%, and >35%. The rock fragments cover (>6 mm) in the soil surface were defined according to the percentage cover in three classes: >40%, 15-40%, and <15%. The drainage conditions were defined on the basis of the depth of hydromorphic features such as iron or manganese mottles or gray colors, and depth of the groundwater table. The following drainage classes were distinguished:
  Very well to well drained soils (soils with any Fe or Mn mottles or gray colors at some depth greater than 100 cm from the soil surface. The soil is not wet enough near the soil surface or the soil does not remain wet during the growing period of the plants. Water is removed from the soil rapidly.) Moderately well to somewhat poorly drained soils (Fe, Mn or gray mottles are present in the soil, at some depth between 30 and 100 cm from the soil surface. The soil is wet enough near the soil surface or the soil remain wet during the early growing period of the plants. Water is removed from the soil slowly.) Poorly to very poorly drained soils (Mottles of Fe and Mn are present in the upper 30 cm of the soil, or gray colors of reducing conditions are present. A permanent water table usually exist at a depth greater than 75 cm. In some of these soils the ground water may reach to the surface during the wet period of the year. Water is removed from the soil so slowly that the soils are wet at shallow depth for long periods.)
  
• Land terracing was determined by the extent to which a field site was covered by terraces (i.e. ratio of the terraced area to the total area) and the degree to which they were protected from collapsing (i.e. ratio of protected terraces to total terraces).
• Storage of water runoff was defined in terms of land management for reducing surface water runoff and increasing the soil infiltration rate. For example, the presence of adequate shrubby or annual vegetation cover, construction of terraces, shallow ploughing of the soil, concentrating runoff water in small ponds and retarding runoff, keeping plant residuals or rock fragments on the soil surface.
• Controlled grazing included the following actions: (i) selection of a sustainable number of grazing animals, (ii) fencing of grazing land and grazing alternatively meaning shifting the animals from field to field in order to allow regeneration of the grass, (iii) avoidance of grazing when soils are very wet, and (iv) fire protection of land. Whether the land was overgrazed or not was defined after assessing the stock carrying capacity of the area and comparing with the actual number of animals grazing the area. The sustainable stocking rate (SSR), expressed in animals per hectares, was calculated from the following equation: SSR=X*P*F/R, where: R is the required annual biomass intake per animal (sheep or goat 187.5 kg animal-1 year-1), X is the fraction including grazing efficiency and correction for biomass not produced during the latest growing season (grazed: 0.5, non-grazed 0.25 year-1), P is the averaged palatable biomass after dry season (kg ha-1), F is the average fraction of the soil surface covered by annual plant species.
  
• The reclamation of affected soils (if any) such as poorly drained, salinized or acidified was described. Management techniques for reclaiming the soils were: (i) construction of channel network, (ii) application of excess of good water quality for leaching of soluble salts, (iii) application of lime for reducing soil acidity. The efficacy of land reclamation was rated as adequate, moderate, low, very low and none.
  
• Cultivation of plant species of low water requirements is an effective measure that might combat further the land degradation. The various vegetation types were classified in three categories in relation to water requirements such as high, moderate and low. For example cereals were graded as high water requirements plant species, while olive and pine trees as low water requirements.
  
• Increase in soil organic matter content for protecting soil crusting were often related to the crop residue management. Crop rotations involving grasses and legumes have long been recognized in the study areas for increasing soil aggregation and maintaining higher organic matter contents than continuous growing of row crops. The incorporation of animal manure has been also considered as a management technique for increasing soil organic matter content in the study areas.
  
• Soil erosion control measures have been already mentioned above. Some other measures undertaken in the study areas were contour farming, stabilization structures, grassed waterways, and fallow. Contour farming has been recently applied in extensive areas cultivated with olives. Following the contour farming, each furrow acts as a reservoir to receive and retain the runoff water. Stabilization structures along waterways consisted of reinforced concrete or monolithic reinforced concrete such as drop spillways, drop inlets, as well as temporary structures made by rocks, logs, brush, woven wire and other nondurable materials to dissipate the energy of running water and stabilize the soil in cuts from landslides. The effectiveness of the existing soil erosion control measures were rated as adequate, moderate, low, very low and none defined in a self explanatory way.
  
• Soil water conservation techniques, important for the study areas, included mulching, weed control, management of soil surface for maximum water vapour adsorption, tillage and covering soil surface by rock fragments. Enhancement of water vapour adsorption was achieved by: (a) reducing the density of the growing vegetation and increasing the soil-atmosphere interface, (b) using surface mulches such as rock fragments or plant residues partially covering the soil surface, and (c) ploughing the soil for increasing macro-porosity (Kosmas et al., 2001a). The existing techniques on soil water conservation, if any, were recorded for each study field site.
  
• The effectiveness of the policies on environmental protection depends on the degree of enforcement, while they are rated based on their degree of effectiveness. Hence, the information collected on the existing policies depended and their implementation /enforcement of the policy under consideration. For example, in the case of terracing protection policy, a relevant piece of information was the ratio of protected terraces to existing terraces in the study field site.
  

5 top

g Identification of ESAs and degree of soil erosion

The type of environmentally sensitive areas (ESAs) and the degree of soil erosion were estimated in each field site, relative to desertification risk. Four general types of environmentally sensitive areas (ESAs) to desertification have been distinguished based on the stage of land degradation (Kosmas et al., 1999):

• Critical ESAs: areas already highly degraded through past misuse, presenting a threat to the environment of the surrounding areas, i.e. badly eroded areas subject to high run-off and sediment loss. This may cause appreciable flooding downstream and reservoir sedimentation. Critical areas are subdivided in three sub-types C3, C2, and C1, in a decreasing stage of land desertification.
• Fragile ESAs: areas in which any change in the delicate balance between natural and human activity is likely to bring about desertification. For example, the impact of predicted climate change due to greenhouse effect is likely to enhance reduction in the biological potential due to drought causing areas to lose their vegetation cover, be subject to greater erosion, and finally shift to a critical ESA. A land use change (such as a shift towards cereals cultivation,) on sensitive soils might produce immediate increase in run-off and erosion, and perhaps pesticide and fertilizer pollution downstream. This type of ESA is subdivided in three sub-types F3, F2, and F1 in a decreasing stage of land desertification.
• Potential ESAs: areas threatened by desertification under significant climate change, if a particular combination of land use is implemented or where offsite impacts will produce severe problems elsewhere (for example pesticide transfer to downslope or downstream areas under variable land use or socio-economic conditions). This would also include abandoned land which is not properly managed. These ESAs are in a less severely desertified stage than fragile ESAs, for which nevertheless planning is necessary.
  
• Non Threatened ESAs: areas with deep to very deep soils, nearly flat, well drained, coarse-textured or finer soils, under semi-arid or wetter climatic conditions, independently of vegetation, are considered as being non-threatened by desertification.
  

The degree of soil erosion was assessed qualitatively in the field. It was characterised, according to: (i) the presence or not of the A-horizon, (ii) the existence and percentage of eroded spots, (iii) the degree of exposure of the parent material on the soil surface, and (iv) the presence of erosional gullies. In each study field site, three transects of thirty meters each were chosen perpendicular to the contour lines, and the length of the eroded areas and of the rock outcrops were measured. Eroded areas were defined spots in which current erosion features could be easily distinguished, such as washing out or deposition of soil materials (Kosmas et al., 2000). The degree of erosion was then assessed by the ratio of the total length of the eroded areas plus rock outcrops to the total length of the transect, expressed as a percentage(Kosmas et al., 2000). Five classes of erosion were used, very severe, severe, moderate, slight and no erosion.

An empirical approach was adopted to define desertification risk based on the degree of soil erosion and the type of ESA. The type of ESA describes the existing condition of land degradation caused by various processes acting previously. In sloping land, where this study was conducted, the main process of land degradation and desertification was soil erosion. Four categories of desertification risk were distinguished, high, moderate, low and none and they were associated with ESA status as follows.

5 top

g Analysis and results

The various classes of the indicators were rated according to the importance to desertification risk and indices were assigned to each class. For example:

The analysis was conducted by using the statistical package STATISTICA (1999 edition). A forward stepwise multiple regression was applied, with desertification risk being the dependent variable and all the indicators as independent variables. An algorithm was derived from the analysis relating the most important indicators (which are shown in the diagram).

Thus:

DR=(9.11)-(0.44*land ownership)+(0.09*number of animals)-(0.50*plant cover)-(0.83*depth)+
(0.39*slope gradient)-(1.81*rainfall)+(1.09*slope exposure)+
(0.13*presence of terraces)+(0.20*controlled grazing)-(0.65*forest fire protection)+(0.77*policy enforcement)

Desertification risk was classified according to the following range in value of DR:

• No risk DR<1.49
• Low risk 1.50<DR<2.49
  
• Moderate risk 2.50<DR<5.49
  
• High risk DR>5.50

The indicators related to desertification risk for oak-forested areas were concerned with soils and topography, climate, vegetation, and land management. The most important indicators related to soils and topography were soil depth, slope gradient and slope exposure. As soil depth decreased, growth of oak trees was reduced, erosion rates were higher and desertification risk increased. As it is expected the greater the slope gradient the higher desertification risk as soil erosion increased. Also oak-forested areas located in north-facing slopes were under lower desertification risk than areas with south-facing slopes. Desertification risk decreased as annual rainfall increased. Plant cover was the most important indicator from the vegetation characteristics affecting desertification risk. As plant cover increased desertification risk decreased. A number of indicators relating to land management were important. In several cases, oak forested areas are privately owned and a great impact is expected by human actions in those areas. The most important indicators were fire protection measures, presence of terraces, number of animals grazing the land, controlled grazing, land ownership, and policy enforcement. Fire protection measures were highly variable in the study areas depending on the type of land ownership. In case that land was private and grazed by animals, farmers used to put fire to favour the growth of annual grass. Therefore, desertification risk was higher in those cases than in land controlled by the public. In some cases, terraces have been constructed in those areas highly reducing soil erosion and desertification risk. Desertification risk increased as the number of animals per hectare increased, favouring higher soil erosion rates. Of course desertification risk decreased in cases that controlled grazing was applied. As in the case of pasture land, land ownership had negatively affected desertification risk in the oak forested areas. Finally, as in the other land use types, policy enforcement was a major issue in protecting the land from desertification.

5 top

g References

• FAO-UNESCO, 1989. Soil Map of the World, revised legend. World Resources Report 60, FAO. Rome. Reprinted as technical paper 20, ISRIC, Wageningen.
• Finke, P., Hatwich, R., Dudal, R., Ibanez, J., Jamagne, M., King, D., Montanarella, L., and Yassoglpu, N., 1998. Georeferenced soil data base for Europe, Manual of procedures. European Soil Bureau Scientific Committee. EUR 18092 EN, 170 p
  
• Kosmas, C., Kirkby, M. and Geeson, N. 1999. Manual on: Key indicators of desertification and mapping environmentally sensitive areas to desertification. European Commission, Energy, Environment and Sustainable Development, EUR 18882, 87 p.
  
• Kosmas, C., Danalatos, N.G, and Gerontidis, St., 2000. The effect of land parameters on vegetation performance and degree of erosion under Mediterranean conditions. Catena, 40:3-17.
  
• Kosmas, C., Marathianou, M. Gerontidis, St., Detsis, V., Tsara, M. and Poesen, J., 2001. Parameters affecting water vapor adsorption by the soil under semi-arid climatic conditions. Agricultural Water Management J. 36:157-168.
  
• Webster, R. 1977. Quantitative and numerical methods in soil classification and survey. Clarendon Press, Oxford, p. 255.

5 top