Based on previous experience, farmers are reluctant to attend stakeholder meetings involving management authorities and are more comfortable to be reached individually, therefore the farmer’s representative attended to speak for other shepherds/farmers. Hence, although over 20 stakeholders were invited, only 7 individuals attended (Figure 1). Despite this situation, CASCADE study site researchers considered that as the farmers’ representative is very well informed on current affairs due to his previous position as head of the pastoralists’ union of Heraklion, and long-term involvement of he and his family in the livestock farming profession, the shepherds’ views were adequately represented.

In the first part of the meeting, participants were shown the CASCADE video in order to demonstrate that problems in drylands are common among CASCADE’s study sites. Stakeholders could especially relate to and comment about the overgrazing problems described in Cyprus. Afterwards, TUC researchers gave a short presentation on the CASCADE results for Crete regarding (a) stress gradient experiments, (b) drought stress experiments, (c) CASCADE grazing principles, and (d) cost-benefit scenarios. Stakeholders were urged to interrupt and discuss any points they wanted to raise during the presentation.

Stakeholders’ perceptions of grazing principles in Crete

For descriptions of the principles discussed here, see »Guidelines for land managers: the overgrazing context_EN.

Every participant in the workshop agreed with criterion 1 "Keep a minimum of 30-40 % soil cover", and criterion 2 "rotate grazing areas and control the amount of animals" of principle 1 “Reduction of vegetation increases soil erosion, leading to less fertile soil and less productive pastures”. Criterion 1 was seen as feasible by all the participants. However, keeping animals in stables (criterion 3: use stall feeding, especially during the dry season), was considered feasible by non-pastoralists but unrealistic by pastoralists. This is due to current stable installations, as they are rudimentary, and stable construction costs are high.

Regarding principle 2: “Integrating trees and pastures has ecological and socio-economic benefits”, everyone considered protecting existing trees and planting fruit and fodder trees an effective measure. However, opinions were divided regarding how realistic it is for the stakeholders to apply this. Silvo-pastoralism and market diversification are already applied by a few more educated and open-minded farmers who took advantage of financial instruments to plant trees on their land or invest in agro-tourism. To some extent, land tenure was perceived as a barrier: small land owners may not be able to apply such instruments due to the high costs that this implies, and the lack of access to subsidies. Also, small producers are using their products only for subsistence.

Stakeholders agreed with the protection of wildlife and criteria in principles 3.1 and 3.2 “Protect ecosystem floral and faunal diversity, avoid killing predators”, and stated they already avoid killing predators such as snakes. Measure 3.3 “Protect trees against rats” was not relevant for the area.

Regarding 3.4 “Install fences and traps” and 3.5 “Provide nest boxes for birds of prey”, fencing was not considered realistic, due to the nature of the landscape, while trapping can only be managed at the administration level if such a need arises, otherwise it is illegal. Researchers mentioned that the use of nesting boxes could be appropriate, but administrators and farmers did not consider it necessary, as indigenous bird populations are not at risk.

Stakeholders agreed with principles 4.1 and 4.4: “Plan resting periods for pastures”, and “Increase health and productivity of individual animals instead of increasing the size of the herds” respectively. However, principle 4.4 was not considered feasible by any stakeholder. Regarding animal types and herd composition from principles 4.2 and 4.3 “Selectively remove unwanted species, while keeping some for soil protection if necessary” and “Diversify animal types”, stakeholders agreed with selectively removing unwanted species, while keeping some for soil protection if necessary. They also agreed with animal diversification, although extensive planning for rotational grazing was considered realistic by non-pastoralists but unrealistic by pastoralists. Animal diversification beyond sheep and goats was not considered feasible by any of the stakeholders. Stakeholders considered that including goats in the herd endangered forested areas, as their efficiency in grazing threaten the viability of the vegetation and limit regeneration. Furthermore, goats can also eat the bark of the trees, leaving them susceptible to diseases. Therefore, diversifying grazing including goats was perceived as a problem.

All stakeholders agreed with principle 5 “Controlled grazing reduces risk of fires” the most. This is because they saw that controlling grazing can reduce the risk of fires by reducing the fuel load. Installation of fire breaks and the reduction of bush cover in order to allow grazing and reduce fire risks are already being applied. However, they did not agree with reducing grazing during the dry season, and perceived that grazing should be increased when dry matter increases. Reducing grazing during the wet season makes more sense for stakeholders. Excluding grazing for at least 4 months during the wet season was suggested. This mimics the traditional transhumance pattern of moving to lower grounds during the winter season both for shelter and to allow vegetation to grow before it is grazed. They considered that grazing should be allowed during the dry season when biomass has grown otherwise it would only provide fuel for fires.

All stakeholders agreed with the effectiveness of the measures in principle 6 “Actions after a fire or drought”. Burned lands are typically included in a reforestation zone plan. Two years was considered the minimum resting period by farmers; 5-6 years was perceived as adequate.

Stakeholders’ perceptions of findings from the scenario analysis in Crete

Stakeholder discussions of findings from the scenario analysis in Crete can be summarised as follows. For degraded sites, participants considered that increasing vegetation cover requires extensive financial resources and the benefit is only environmental. Furthermore, small interventions make little difference. In this sense, the vegetation trends and financial benefits/losses described in the models were considered realistic. In non-degraded/restored sites there is high potential for sustainable management. Excluding grazing during the wet season can be profitable. In order to successfully apply this though, fodder needs to be provided.

The assumption made in management scenarios implying annual decisions on livestock destocking (keeping animals in a stable rather than selling them) or restocking (moving animals from stable to pasture rather than buying new animals) is problematic for degraded initial conditions. Currently stable installations are rudimentary at best and therefore the cost of destocking should include stable construction. Even the traditional model of transhumance (moving animals to higher ground in the summer and lower grounds in the winter) is currently unfeasible due to labour costs and land fragmentation. Therefore, stakeholders consider that partial animal exclusion in traditional/degraded lands is largely unrealistic. Partial exclusion was however considered realistic in non-degraded or restored lands when lower animal densities are also assumed. Total exclusion is considered unrealistic in all cases.

Contrary to the reactions towards the results of the grazing model, the stakeholders showed some disagreement with the grazing guidelines and recommendations of the grazing principles. As mentioned before, they did not agree with reducing grazing during the dry season. On the contrary, stakeholders considered that grazing should be increased when fresh matter decreases and dry matter increases as vegetation dries out. They felt that reducing grazing during the wet season makes more sense, also in accordance with the “resting in wet season” scenario. Excluding grazing for at least 4 months (wet season) was considered appropriate.

In responding to question 3 (in sites where vegetation was successfully restored, how severely was the vegetation degraded (% cover) when restoration started?), stakeholders could not answer using a simple percentage. This is because the perception of degradation as a function of vegetation cover was not considered entirely realistic. Some participants noted that some sites have been successfully restored after having 0% vegetation cover. Several such examples were discussed and these observations also agree with findings from the CASCADE stress gradient experiments (see »Simulated pressures and ecosystem responses). According to the stress gradient results, degraded sites may be more fertile, with higher amounts of soil organic carbon (SOC) and N, probably as a result of the higher amounts of manure left during grazing. The limiting factor in these cases is soil depth which gradually erodes when vegetation cover is low. Vegetation cover is in some cases a good proxy for degradation but does not depict the permanent loss of other land resources such as soil.

How many animals per hectare were grazed prior to degradation on these sites (=cause of degradation)?

The perception of the number of animals considered as overgrazing varied, and stakeholders were reluctant to set a clear threshold for overgrazing. In their view it is a complex situation that depends on grazing strategies (constant or rotational grazing). E.C. Reg. 1782/2003 suggests livestock density has to be maintained at 1.4 head/ha (Hadjigeorgiou, 2011). The limit for Natura 2000 is 0.2 to 3 LU/ha. Stricter regulations for single areas are possible (Dimopoulos et al., 2006) and indeed necessary (Papanastasis et al., 2002). Current average density is 1-1.4 head/ha. Nevertheless, as much as 4-8 heads/ha can be considered sustainable if rotational grazing is applied. If rotational grazing is not applied over 2 heads/ha may be considered overgrazing.

Note: For full references to papers quoted in this article see

» References