Close Window

Analysis Of Land Cover Change in Eastern Para, Brazil Based On Four Dates Of Landsat Data From 1984 To 2002

Eric A Davidson, Woods Hole Research Center, (Presenting)
Arlete S. de Almeida, Museu Paraense Emílio Goeldi,
Tom A. Stone, Woods Hole Research Center,
Ima C. Vieira, Museu Paraense Emílio Goeldi,

The objective of this research is to spectrally characterize the vegetation and the land use of the region and to evaluate the current trends. The goal is to understand the change dynamics in an area that is highly fragmented and where most of the area is successional pasture in various stages of regrowth or reformation, perennial and permanent agriculture and secondary or successional forests. There are few large fragments of mature forest (floresta ombrofila densa) remaining. The study covers a period of 18 years and uses data from Landsat/MSS and TM from 1984, 1994, 1999 and 2002. All data were geometrically coregistered and were converted from digital numbers (DNs) to reflectivity. Over the 18 yr. time period, landcover types showing the largest declines include mature forest (floresta ombrofila densa) and selectively logged forest (mata explorada). Land covers showing the largest increases include intermediate aged secondary forest (capoeira intermediária), young secondary forest (capoeira jovem), and shrubby pasture (pasto sujo). Additionally, the amount of exposed soil has approximately doubled, indicating an increase in cultivated land. Although the initial wave of deforestation occurred several decades ago in this region, the process of transforming the landscape from high-biomass forests to low-biomass fields of cultivation, pastures, and young secondary forests has continued during the last two decades. Transition matrices for these land cover are being developed for four points in time and at random locations. The matrices should reveal typical cover change pathways. Additionally, we will examine the validity of classification signatures across the larger region.

Science Theme:  LC (Land Use and Land Cover Change)

Presentation Type:  Poster

Abstract ID: 107

Close Window