ND-01 Abstract

Land Cover Conversion in Amazonia, the Role of Environment and Substrate Composition in Modifying Soil Nutrient Cycling and Forest Regeneration

Paulo Barreto,  (SA-PI)
Dar A. Roberts, University of California (US-PI)
Joao Vianei Soares, INPE (SA-PI)

We propose an integrated remote sensing and field-based examination of the causes and biogeochemical consequences of regional land-cover change within the Arc of Deforestation in the southern Amazon Basin. Research components include field observations, regional synthesis, and multiresolution studies, with a major focus on remote sensing.  We will refine the standardized methods we developed during LBA-ECO Phase I for consistent mapping of regional land-cover conversion through time. The techniques will be expanded to include diverse environmental variation, and new methods will be developed to map forest degraded by logging, fire, and fragmentation.  These remote sensing products will be combined with ancillary data sets and field measurements to evaluate the relative importance of environmental and human controls on patterns of land-cover change, and the consequences of land-cover change for vegetation growth and succession and for soil and stream biogeochemistry.  Our previous research in Rond˘nia shows that variations in topography and rock type generate significant gradients in natural soil fertility and stream nutrient concentrations across the state.  Roads and urban networks, superimposed on this natural variation, constrain or accelerate land-cover conversion and have important consequences for human impacts on stream biogeochemistry.

Our questions represent diverse academic disciplines, but all aim to combine remote sensing with field measurements to develop an integrated understanding of regional patterns in land-cover change and its biogeochemical consequences.  In addition to producing a variety of remote sensing products to monitor land-cover change, we will address the following questions:  How can remote sensing methods be combined with field measurements to better quantify changes in vegetation community structure due to selective logging, and how extensive are those changes?  What are the topographic and geologic determinants of soil mineralogy and fertility across the state of Rond˘nia, and how does this natural variation in soil type constrain or promote land-cover conversion?  Finally, how do vegetation conversion and the evolving human settlement system affect terrestrial and aquatic biogeochemistry and over what scales do these drivers operate?

Our research approach can be divided into three intersecting phases: 1) mapping and monitoring a suite of land-cover changes; 2) evaluation of the natural and human drivers of forest conversion; and 3) measurement of the impacts of those changes on forest structure in selectively logged areas and on soil and stream biogeochemistry in areas converted to pasture.  The research will be conducted in the Brazilian states of Rond˘nia and Mato Grosso, which show significant intra- and interstate variations in deforestation rates, selective logging intensity, urban development, and environmental gradients.  An important aspect of the project is a dedication to exchange research methods and techniques in the LBA/Amazon research community, most intensively among Imazon, INPE, and UCSB.  Institutional exchanges will provide an opportunity for intensive training, data sharing and research collaboration among U.S. and Brazilian participants.