Version classiqueVersion mobile
OpenEdition Books

La spatialisation de la biodiversité

Jean-Louis Guillaumet
Anne-Élisabeth Laques
Philippe Léna
et al.


Texte intégral

The spatialization of biodiversity for the sustainable management of territories

1Saying that biodiversity is a major issue for Amazonia and that it is endangered has become commonplace. Maintaining, improving or even restoring this biodiversity requires the drafting of public policies that have not yet been thought out and the development of tools for the analysis and management of areas that match these objectives. In this vast forest region – multifaceted as regards abiotic and biotic environments with numerous forms of traditional or non-traditional exploitation by man – distinction is made between the Amazonia of rivers where the populations use low-impact agroforestry systems, the Amazonia of roads where pioneer fronts are advancing, attracting settlers who focus on livestock, and finally the Amazonia of regions in which private and public stakeholders are trying to develop technical, economic and sociopolitical alternatives to the predatory exploitation of natural resources. The Municipio of Benjamin Constant, on the upper Amazon in Brazil, the land reform project of Benfica in the state of Para and the Municipio of the Uruará, in the colonization zone of the transamazonian road, have been held as representative of the three situations.

2The approach described proposes the integration of often scattered, disparate data to relate them to landscape units to map and display the key features of the spatial and temporal dynamics of biodiversity in order to relate them to socio-economic dynamics and public policies.

3A spatial continuum, the landscapes are segmented in two complementary levels of intelligibility – landscape type and landscape component – designed to serve as a common language for use between disciplines, terrains, images and evaluations of biodiversity. The first level is intended to be representative of a portion of space that is diversified but homogeneous as regards physiognomy, socio-economic use and the ecological functioning on which production is based. The second is the smallest elementary object perceived in the field and visible in satellite images. This may be a plant formation, a set of different formations or a reconstitution stage. The first part of the examination of the two levels starts with the identification of landscape types (the case of the Upper Solimões) or of components (Benfica). Revealing the former requires direct observation in the field, completed by the views provided by satellite images. Landscape types that could not be identified were not retained. This results in spatial segmentation into landscape units, making it possible to draw maps at the scale of territorial management units such as the assentamento, which are used in public policies. Analyses of biodiversity are performed at landscape component level. Here, it was tempting to use classic measurements of the richness and diversity of the flora but the authors preferred another method. Like scales for estimating wind speed, the intensity of earthquakes, risk of avalanche, a ‘complexity scale’ is proposed here for the classification of changes in biodiversity, with complexity being quantified by using simple, easily used indicators: diversity, continuity of the tree stratum, stratification and artificialization. Other indicators can be added according to specific features of the terrain and special indexes are used to describe how the local populations use the biodiversity of their environments. These are qualitative indexes when based on local perceptions and quantitative when they are more particularly of the quantitative ethnobotany type. They are applied to the diversity used and named, to category, to the value of use and to perception.

4The value of biodiversity for each landscape type takes into account the values of each of the landscape components according to the areas they occupy. Finally, application to all the landscape types identified in the territorial management unit concerned makes it possible to award a degree of diversity to all the analyzed landscapes of the study zone. The resulting maps plotted for landscape surveys on satellite images and the degree of biodiversity that can be attributed to them enable the rapid comparison of regions whose environments, forms of exploitation and rates of evolution are clearly different.

5Comparison of the duly dated maps means that the dynamics can be monitored: the maintaining, decrease or, more rarely, the increase in biodiversity. It is easy to move from the maps drafted in this way to graphs of the evolution of biodiversity in terms of the modification of the areas concerned by the changes. Public policies are one of the factors that affect fluctuations between local populations and their environment. They interfere with practically all the other factors, that can be assembled in several main categories: potential of the environment and land availability, cultural heritage, population growth, social and economic strategies and the social organization of stakeholders. As public policies cannot all be taken into account, criteria for choice must be drafted in function of an observation period governed by the availability of satellite images, with the preference going to those that have a direct, observable influence on the landscape units previously defined. In fact, the authors listed he public policies that have affected the territories studied over the last 15 to 30 years.

6Public policies may concern social inclusion, regional development and the development of farming and may include, among other features, landholding policies, the sale and processing of produce, loans and technical assistance and conservation of the environment. Finally, private development operations must also be taken into account.

7The types of indicator that can measure the impact of public policies on farming systems mainly focus on rates of change in the natural environment: the rate of annual deforestation and of the formation of productive capital, the annual rate of the abandoning of land to fallow and of the re-use of older cleared land. Other indicators such as initial capital, use of land, loans, etc. are used to refine and interpret the dynamics of landscape components.

8The maps of the evolution of the degree of biodiversity show two fairly different situations in time: in the Amazonia of roads, on the Benfica pioneer front, the changes in the landscape over the last 20 years reveal a sharp decrease in biodiversity, whereas in the Amazonia of rivers, in the Upper Solimões, the landscape has changed little during the same period of time and degrees of biodiversity are still high.

9The changes in or maintaining of biodiversity are to be related to the major development programs that have made a strong contribution to the choices of farming systems found in Amazonia insofar as their favour certain agricultural orientations at the expense of others (agroforestry, gathering, etc.). Government plans in eastern Amazonia were aimed at economic integration with the rest of the country thanks to incentive measures for colonising land and turning it into grazing facilities for meat production. In Benfica assentamento, the migrant livestock farmers have fairly different origins and traditions and so farming strategies are varied; however, their common feature is the generalization of cattle production in the medium term. The migrants who settle on pioneer fronts are strangers to the local environment and try to reproduce the production modes that they know; this generally leads to the destruction of forest for the installation of agrosystems that have variable but generally strong effects on the initial biodiversity. However, for reasons of difficult access and remoteness from major towns, the Brazilian government has never encouraged mass colonization in the western part of the forest area. In this better conserved environment, populations described as traditional, such as those of the Upper Solimões, have farming practices that use little forest area.

© IRD Éditions, 2009

Conditions d’utilisation :