A common farming tool is the setting of controlled fires. This is the most cost effective way to both clear a field of weeds and to fertilize the same field. While it is financially cost effective the misuse of fires in the Amazon can be very costly. During the dry season damages from forest fires used by farmers and ranchers accounts for about one-hundred million dollars in damages. The damage is not due to fire directly but also due to the smoke produced by the fires. The smoke causes car crashes, asthma attacks and airport closures. These factors led to the development of laws making forest fires illegal. Despite these laws there are concerns over developing industry in the Amazon, and the potential for this industry to light fires. Within this study they created a conceptual model for the occurrence of fire. The model was based on the premise that fires were dependent on land use, the number of potential ignitions and, environmental factors such as climate and, vegetation. They also set up equations for various probabilities such as accidental fires, like those caused by lightning strikes, or forest fires caused by a fire escaping the containment zone. When creating their GIS model they used data collected by from multiple digital maps about the legal amazon. They then used 5x5km cells to create a raster environment in order to analyze the data. They then used ones and zeros to represent whether a fire occurred in a specific area or not. In their analysis they also took economic factor into consideration. The two major factors they took into consideration were the production of beef, and the production of soybeans, two of the area’s most valuable commodities. Finally their model included quite a few control variables such as deforestation, and rainfall. They found that the occurrence of fires had a high correlation with the yield of the two main economic factors. They came to the conclusion that controlled fires in accordance with environmental laws may be the best of both worlds, causing high economic gains, while still being coupled with safe practices.
Arima, E. Y., Simmons, C. S., Walker, R. T., & Cochrane, M. A. (2007). Fire In The Brazilian Amazon: A Spatially Explicit Model For Policy Impact Analysis*. Journal of Regional Science, 47(3), 541-567.