Biodiversity implications of forest disturbance and related landscape dynamics in the Brazilian Amazon

Mark Alan Cochrane, South Dakota State University/GIScCE, mark.cochrane@sdstate.edu (Presenter)
Jos Barlow, Lancaster University, jos.barlow@lancaster.ac.uk
Carlos Souza, IMAZON, souzajr@imazon.org.br
David Roy, South Dakota State University/GIScCE, david.roy@sdstate.edu
Eugenio Arima, University of Texas - Austin, arima@austin.utexas.edu
Izaya Numata, South Dakota State University/GIScCE, izaya.numata@sdstate.edu
Christopher Barber, South Dakota State University/GIScCE, christopher.barber@sdstate.edu
Luiz Mestre, South Dakota State University/GIScCE, luiz.mestre@sdstate.edu
Sanath Sathyachandran, South Dakota State University/GIScCE, sanath.kumar@sdstate.edu
Rafael Andrade, South Dakota State University/GIScCE, rafael.andrade@sdstate.edu
Juliana Silveira, Universidade Federal de Lavras, silveira.juli@gmail.com

Biodiversity threats are not solely from deforestation. Human land use fragments remaining forests into smaller and smaller pieces. Forest fragments are altered habitats that change community composition and structure. It has not generally been understood that an equal or greater percentage of standing Amazonian forests have been degraded by fire, selective logging and fragmentation than have been deforested. Biodiversity implications of ongoing forest degradation and its relation to human landuse have not been quantified or placed in the spatial context necessary for predicting future landcover and hence biodiversity dynamics.

The fundamental hypothesis underlying this research is that biodiversity levels of Amazonian forests are strongly related to two competing factors: forest disturbance and time since last disturbance.

Using Landsat, we have produced the first complete map of forest disturbance for the Brazilian Amazon, for 10 years (2000-2009). The extents and locations of forest fragmentation, forest degradation (selective logging and forest fires), as well as deforestation, have been determined. We conducted the first fine-scale analysis of Brazil’s extensive Amazonian protected area network (1.8 million km2) and have quantitatively estimated conservation effectiveness in light of changing human development pressures in the surrounding landscape. By integrating MODIS fire products and Landsat analyses, we are examining fire occurrence in the Amazon by type, location and timing. The extensive investigation of ecological community responses (based on 4 taxa - understory birds, dung beetles, trees and ants) to disturbance frequency and post-disturbance recovery is providing key insights into the state of Amazonian biodiversity within millions of hectares of standing, but damaged, forests. We are currently constructing a model incorporating economic, land cover and physical-geographic factors affecting fire ignition and spread probability, which will enhance policy analyses, conservation planning and ability to forecast biodiversity impacts of disturbance for the Brazilian Amazon.

Presentation Type:  Poster

Session:  Science in Support of Decision Making   (Wed 10:00 AM)

Associated Project(s): 

• Cochrane, Mark: Applied remote sensing for conservation monitoring ...details
• Cochrane, Mark: Biodiversity Implications of Forest Disturbance and Related Landscape Dynamics in the Brazilian Amazon ...details
• Roy, David: Fire Type Classification in the Brazilian Legal Amazon ...details

Poster Location ID: 130