Dynamic patterns of deforestation in the Brazilian Amazon in relation to the value of ecosystem services- Solution
Abstract: The deforestation drivers sector models basic social, demographic and economic processes which are believed to be significant factors in Brazilian Amazonia deforestation. The land use/cover sector shows how different groups clear land, how the transition rates are dealt with and also the calculation of biomass amounts in the land stocks. This journal uses a Stella model to show the different land use has degraded the ecosystem in Brazilian Amazonia. The model includes a deforestation drivers sector, a land use/cover sector, ecosystems services sector, as well as ecosystem valuation as a sub-model.
Calculations are made in the ecosystem valuation sector which focuses on four primary ecosystem services that are provided for by an intact Amazonia region, and which contribute to people, including climate regulation, erosion control, nutrient cycling and species diversity. The calculation is made with the change in these values according to the land use practices. And finally, according to these findings, a discussion is made to show how an explicit monetary valuation of ecosystem services could create positive incentives for land stewardship and conservation. In all, this model used a dynamic way to describe and display the processes that degrade ecosystems in Brazilian Amazon.
The Amazon has been the focus of the attention of the naturalists, scientists and explorers all over the globe. It represents about 40 percent of the world’s remaining rainforests and holds by far the largest intact section of diverse humid flora and fauna (1997). To many people the Amazon has become the typical figurative last existed of a major wild, natural environment against the encroachment of civilization. Undoubtedly the Amazon has captured the imaginations of millions; but the future of this region should not be left to imagination, but rather to studied analyses based on the facts as we can best ascertain them. There has been outstanding improvement over the past decades in conducting hard, scientific studies of the ecology, biology, and economics of the Amazon rainforest. Nevertheless the region is still the subject of many popular myths.
The STELLA Model
The Systems Thinking for Education and Research is a great instrument in studying deforestation in the Brazilian Amazon. The STELLA model is often use in education and research and often assisted by computers and other programs. The issue and trends in the deforestation in the Brazilian Amazon can be explored using STELLA model through the following points (2000):
- How does the climate affect the environment in Brazilian Amazon?
- What are the effects of the deforestation of the Brazilian Amazon to the environment in Brazil?
- What are the effects of the deforestation of the Brazilian Amazon to the natural resources of Brazil?
The STELLA model is a great help in promoting dynamic learning styles by using different diagrams, charts and animations in discovering the different variables especially when it comes to environmental policy.
In this journal, the STELLA model diagrams were used. The schematic diagram showing the climate regulation is a very effective method of illustrating some of the basic concepts of systems thinking (2000). Deforestation is an inherently spatial phenomenon and more and more researchers have recognized the need to be spatially explicit when modeling deforestation. The schematic diagram makes it possible to analyze deforestation and its causes at pixel level (for example, sample points at 1 km intervals). Typical variables in the diagram include land-use type, distance to road, distance to market, and soil quality. However, other important variables do not easily lend themselves to geo-referencing and are therefore often ignored in these models.
The processes of growth and expansion in the Amazon are clearly complex and in addition they are evolving over time as the role of government policy changes and the frontier advances (1990). A good structural (i.e. a “theory-based”) model of these processes requires a comprehensive theoretical knowledge of the possible important relationships; leaving out any one of them could considerably prejudice the results if the processes under study are correlated with the omitted variables. However, identifying all the dynamic and spatial interactions and feedback relationships that could be expected to play an important role in the evolution of these processes. However, there are still several limitations associated with causal loops. First and foremost, they may specify relationships and predict behaviors, but they frequently do not describe the operations of systems. Without operational specification, it is impossible to understand an interactive, dynamic system.
Using the random reduction method in analyzing the erosion control ensures that the results are a function only of the initial set of possible explanatory variables and the chosen functional form of the regression, not of any pre-existing researcher bias about which variables should remain in the analysis. First, the levels are highly trending in most of our variables and thus a model in levels would be highly susceptible to spurious correlation (1997). Second, by taking the variables of the erosion control effectively eliminate the municipality-specific “fixed effects” from the analysis. In other words, all the time-invariant municipality characteristics that influence, say, the level of cattle herd, are present in both the start and the end period, so the change between periods cancels them out. In this it is effectively control for an enormous host of unobserved time-invariant variables that could bias the analysis if omitted from consideration. Controlling for the fixed effects also controls for the overall average level of spatial correlation in the data, thus cutting down considerably the scope for spatial correlation becoming a problem in the estimation.
In the frontier environment of the Amazon’s nutrient cycling can be extremely important in determining economic activity. In a standard regression analysis it is assumed that all the observations are independent, but this is clearly not the case in the Amazon; it matters a lot what is going on in neighboring municipalities (1984). Using similar data in Stella model, finds that the spatial distribution of both roads and people are very important for the modeling of deforestation processes in the Amazon. In fact a whole category of empirical models of deforestation focus on geographical location as the primary determinant of many economic activities. The STELLA model is based on adaptations of the location structure of different economic activities is described as functions of the distance.
The Stella model in this journal also presented information on species diversity, land quality, road networks, population densities, and agricultural The Brazilian Amazon will soon be possible to start modeling on a scale that is useful for sensible location of roads, forest reserves, logging and mining concessions, and for other planning purposes (1990). The Stella model has also put emphasis to the average coefficient estimate and what the spread of the coefficient estimates (highest and lowest value) was across all models in which that variable appeared. The variables that made it into the final model frequently “robust” and the spread of coefficient estimates is relatively tight or spread out another indication of how reliable the mean coefficient value is.
However, the STELLA model and System thinking have some limitations. In this journal there are some invariant municipality-specific characteristics could still play a role if they are important determinants of the growth of the variables, not just the levels (i.e. “fixed effects” in growth rates). This could be the case if there is a direct effect, but also if they are correlated with more specific spatial correlation that remains in the data. Nevertheless the model include a number of invariant variables including a full set of state dummy variables, measures of the original natural vegetation and soil type of each municipality, distance to state and federal market, length of navigable river, variables on the average monthly temperatures, and a dummy variable for high rainfall (2001). These variables will enter only if important for the growth of the dependent variable; many variables may be important for the level but not the growth rate and so do not appear in the final models