Search

The effect of livestock grazing on the under-story vegetation in a Mediterranean Pinus nigra forest pasture in the Pyrenees was studied during 6 years, by comparing changes in herbaceous and shrub vegetation under cattle grazing conditions or in absence of livestock. Cattle grazed the pasture each year at a moderate stocking rate of 0.2 LU/ha, for 3 months in spring and 3 months in the autumn. Four 10 m × 10 m exclosures were set up in different locations in the range. Each year, at the end of the grazing season, herbage height and biomass and shrub number, volume and biomass were measured inside and outside the exclosures. Herbage quality was measured at the end of the study. After 6 years, herbage biomass was much greater inside the exclosures (1730 kg DM/ha versus 679 kg DM/ha, p

The ‘Sierra de Guara’ Natural Park (81,491 ha, Huesca, Spain) is a protected Mediterranean mountain area dominated by shrub and forest pastures. Traditional agriculture, mainly extensive grazing systems, has decreased in the last decades; concurrently, invasion of shrub vegetation, landscape changes and higher risk of forest fires have been observed. A study, which started in 2000, was carried out with two broad objectives: at the farm level, to analyse the farming systems and evaluate management strategies; at the regional level, to give useful information to conservation authorities for better decision-making. An integrated approach with different spatial scales and methods of analysis was used. First, a survey covering all farms that utilized the Park was carried out and livestock farming systems were characterized in terms of grazing management, technical and socio-economic factors. Second, six representative areas were selected to evaluate, depending on livestock utilization, grass and shrub vegetation dynamics (biomass, green/dead ratio). Third, vegetation and livestock data were analysed using a Geographic Information System to identify constraining factors and areas of intervention. Key imbalances were identified that can threaten the sustainability of the Park: low continuity of farming families; intensification of the management system; degradation of grazing resources; and concentration of grazing areas. A number of management recommendations are raised.

The management of natural resources in Protected Natural Areas needs to bring together the interests of many different stakeholders. For this purpose, multi-agent simulation systems (MAS) can be used to develop virtual environments in which the consequences of different management strategies on the ecosystem can be tested. In this context, a prototype of a MAS model was built for Sierra de Guara Natural Park (Spain, 80,739 ha), a typical Mediterranean mountain area where agricultural abandonment has occurred progressively in the last decades. In order to determine the evolution of vegetation types in time and space, information on current distribution of vegetation cover was used to set up the initial state of the model, which could change due to a ‘densification’ function defined from previous studies and that was dependent on management. According to the theory of states and transitions in vegetation dynamics, once a threshold was reached, a particular vegetation unit could change to a denser vegetation type. Three different management scenarios were evaluated: abandonment of agriculture and livestock production, with occurrence or absence of natural fires, and continuity of current farming practices (on 53% of the total Park area). Simulations were carried out until an equilibrium state was reached (60 to 80 years depending on scenarios). Agricultural abandonment led to a rapid densification of vegetation, with lower diversity at the equilibrium situation (higher proportion of forest cover). This trend was more irregular when natural fire was included in the model, and more grassland and shrub pastures were observed at equilibrium. Current farming practices still led to some changes in vegetation, but were mostly concentrated in non-used areas; vegetation cover diversity at equilibrium was higher in this scenario. The utility of this approach as a decision support tool, even at its preliminary stage, was demonstrated.