CCE banner

Atmospheric CO2 increase driven by human activities is causing changes in Earth’s climate at an increasing rate and mitigating these trends is a high priority for international policy makers. The two major sources of anthropogenic CO2 are fossil fuel emissions and tropical deforestation. Fossil fuel emissions are fairly accurately known (+/- 10%) based on economic market statistics. Tropical deforestation makes up about 20% of the total anthropogenic carbon emissions into the atmosphere but this estimate is highly uncertain (+/- 70%). A number of global policy initiatives are focused on moderating tropical deforestation not only for reducing greenhouse gas emissions but also for maintaining important biodiversity and other ecosystem services provided by these systems (e.g. UN Framework Convention on Climate Change – Reducing Emissions from Deforestation and Degradation or REDD). The international community is crafting a carbon credit system in which developing countries can receive carbon credits for reducing deforestation, thereby mitigating the greenhouse gas burden in the atmosphere. To make this system works there needs to be a better understanding than we have now of current and future carbon flux consequences of various land management practices.

Nearly half of all global deforestation is occurring along the Amazonian agricultural frontier. Historically (since the inauguration of the Trans-amazon Highway in 1970) deforestation in this region was largely associated with slash and burn subsistence agriculture and small-scale cattle ranching. See Figure 1.








deforestation Figure 1. Undisturbed forests are mechanically knocked down using bulldozers and chains then initially burned (conditions shown in second picture) in the first stage. In the second stage, the remaining biomass is accumulated into piles and burned again. The final stage results in large tracts of land ready for agriculture.
learn more

Later large scale cattle ranching began to dominate land management practices and, more recently large scale mechanized agriculture for soybean export has expanded. Over the last decade 40% of the deforestation in Brazilian Amazonia took place in the Brazilian state of Mato Grosso where 90% of cropland expansion in Brazilian Amazonia has occurred. See Figure 2.

deforestation Figure 2. Study Area: Mata Grosso

A previously published NASA funded study by Morton et al. reported that in Mato Grosso about 20% of deforested area is used directly for crops, 72% for pastures and 8% cleared but not in production. The deforestation events are associated with fires that are used to clear the woody forest biomass in preparing the land for pasture or crop.

Deforestation typically follows a time trajectory of carbon emissions and biomass changes as indicated in Figure 3. When forests are cleared for pastures large tree trucks are removed or burned leaving significant amounts of carbon in stumps and underground wood. The extra costs of more complete removal of woody biomass are incurred in the case of transitions to mechanized agriculture in order to accommodate the operation of farm machinery.




land use

Figure 3. Schematic representation of hypothetical trajectories of biomass accumulation, fire emissions and net ecosystem production (NEP) for 3 types of land use transistions. Transitions that do not lead to regrowth of forests result in a long term net fux of carbon to the atmosphere. Note that compared to agriculture transition (cropland) the pasture transition produces less fire emissions and more respiration.

A pair of NASA funded studies recently published (DeFries et al,; van der Werf et al, 2008) estimated the carbon emissions for the most important types of deforestation transitions: when forests are converted to croplands, forests are converted to pastures and pastures are converted to cropland.

They found that converting a forested area to a cropland releases the most carbon to the atmosphere, while converting a equal sized pasture to cropland the least. The authors of this study suggest that by promoting croplands on lands that have already been cleared, policy makers can reduce the amount of deforestation and of carbon emissions into the atmosphere even in the case where some of the converted pasture is replaced through new deforestation.

The DeFries et al. study also shows that from 2000–2005 in Mato Grosso, 90% of the deforested area and of the carbon emissions are associated with >250,000 m2 (or >25 ha, Figure 4).




Figure 4.  Mean annual fraction of total deforestation events, deforestation area, and carbon emissions from all deforestation transitions from annual clearings less than 25 ha, 25 to 200 ha, and greater than 100 ha for 2001 to 2005.

Therefore, despite the high frequency of small scale deforestation events, policies aimed at reducing deforestation activities associated with large scale mechanized farming will have a greater impact on reducing potential deforestation carbon emissions.

The study also shows that polar orbiting moderate resolution satellite instruments (e.g. NASA’s MODIS instruments on the Terra and Aqua satellite platforms) are well suited for monitoring these large deforestation events and their subsequent fates because of their capability of high frequency observations (~daily). Multiple views minimize the viewing interference of clouds and aerosols and allow post disturbance monitoring of the surface that provides information about transition types. This high frequency of viewing comes at the cost of fine spatial scale resolution and these instruments are limited to detecting surface changes >60,000 m2 . Fine resolution instruments can observe disturbances at 1 to 100’s of m2 but pass over the same location at lower frequencies. The results of DeFries et al. showing that 90% of the deforestation area and emissions are associated with events on the scale of >250,000 m2 in Mato Grosso demonstrates the usefulness of moderate resolution polar orbiting satellite instruments for monitoring tropical deforestation.






DeFries RS, Morton DC, van der Werf GR, Giglio L, Collatz GJ, Randerson JT, Houghton RA, Kasibhatla PK, Shimabukuro Y, Fire-related carbon emissions from land use transitions in southern Amazonia.  Geophysical Research Letters 35, L22705, doi:10.1029/2008GL035689, 2008

van der Werf GR, Morton DC, DeFries RS, Giglio L, Randerson JT, Collatz GJ, Kasibhatla PS,  Estimates of fire emissions from an active deforestation region in the southern Amazon based on satellite data and biogeochemical modeling.  Biogeosciences 2009 (in press)

Morton DC, DeFries RS, Shimabukuro YE, Anderson LO, Arai E, del Bon Espirito-Santo F, Freitas R, Morisette J, Cropland expansion changes deforestation dynamics in the southern Brazilian Amazon. Proceedings of the National Academy of Sciences 103:  14637-14641