Spatial analysis of communal grazing resources and their utilisation by sheep in the highlands of Mexico
Estrada, Ernesto González
This thesis is concerned with analysing the interactions between smallholder sheep farming systems and the sustained maintenance of communal grazing resources in the temperate region of Mexico. The study was carried out in the parish of Coajomulco, which is located in the mountainous region south of Mexico City. Sheep production is a traditional agricultural activity in the region and extensive grazing is practised in the parish's communal forest. In 1988 the area was decreed as protected by the Ministry of Environment, and agricultural practices were highly regulated. Although sheep flocks could still have access to the forest, the establishment of an exclusion zone resulted in the inability of sheep farmers to make use of 48 % of the communal grazing area. Thus, the objectives of this thesis were concerned with finding a way of enhancing the development of the local smallholder sheep farming whilst concomitantly protecting the forest ecology. The ultimate objective of this thesis was to develop a spatial optimisation model for the grazing management of the communal land. This model produced the optimal distribution of flocks in time and space according to the characteristics of both the grazing resources and the sheep population. Prior to the development of the optimisation model, it was necessary to characterise the basic elements that affected the supply and demand of forage. Thus, under a farming systems research framework, the plant and animal elements of the farming system were characterised. The managerial and biological influences that defined the sheep grazing patterns were investigated and their resulting effects discussed. Participatory techniques were included as the core of the characterisation methodology. Findings derived from the characterisation were utilised to assist in the development of a geographical information system (GIS) and the application of biological simulation models. Two models, one that simulated flock dynamics and another that simulated sheep performance, were used. Subsequently, a two-way link was established between the simulation models, the GIS and the optimisation model.