Menu Close

Characterizing and Modeling Patterns of Deforestation and Agricultural Extensification in the Ecuadorian Amazon

Citation

Walsh, Stephen J.; Messina, Joseph P.; Crews-Meyer, Kelley A.; Bilsborrow, Richard E.; & Pan, William K. Y. (2002). Characterizing and Modeling Patterns of Deforestation and Agricultural Extensification in the Ecuadorian Amazon.. Walsh, Stephen J. & Crews-Meyer, Kelley A. (Eds.) (pp. 187-214). Boston: Kluwer Academic Publishers.

Abstract

We examine human-environment interactions in the Oriente region of the Ecuadorian Amazon through a Geographic Information Science (GISc) perspective. A remote sensing time-series is used to represent LULC (land use/land cover) dynamics, a GIS to assess resource endowments at local and regional settings, a longitudinal household survey to measure socioeconomic conditions and changes over time at the farm or household finca level, and a community-level survey administered to community leaders, farmers, teachers, women, and health workers in places ranging from tiny communities to the largest city, Lago Agrio to measure infrastructural linkages between households and their communities. Here, we (a) describe the multi-thematic, spatially explicit database assembled to address deforestation and agricultural extensification; (b) indicate changes in finca demographic and land use characteristics reported in the 1990 and 1999 household surveys; (c) use a hybrid digital classification approach for repeatable LULC characterizations for selected Landsat Thematic Mapper (TM) and Multispectral Scanner (MSS) images; (d) apply pattern metrics to classified satellite data to assess the composition and spatial organization of the landscape through trajectories of landscape structure; and (e) use a cellular automata (CA) approach to simulate LULC patterns for antecedent and future periods, given historical and current LULC patterns, as defined through the satellite time-series and hypotheses about the importance of geographic access to and spatial diffusion from the region’s central city.
Preliminary analyses of the longitudinal household surveys indicate the importance of land subdivisions since 1990 in sample fincas located in close proximity to Lago Agrio and other main towns in the region, and as a consequence, a reduction in the mean size of family land holdings. Analyses of the pattern metrics, calculated from the classified satellite time-series, suggest that the Oriente is undergoing a conversion of LULC types from high-density forest to agriculture as well as to low and medium density forest through secondary plant succession. These changes result from (a) initial deforestation by migrant households settling along roads built by petroleum companies in the early 1970s; (b) further clearing of forests to support subsistence agriculture and increasingly commercial cultivation; and (c) secondary plant succession of agricultural lands related to declining soil fertility and pasture formation. Simulation models indicate the important role of the central city, Lago Agrio, and of road networks that connect farms to community services and off-farm employment. CA models indicate that, over time, urbanization will become more significant and land fragmentation through deforestation and agricultural extensification more concentrated in areas currently remote from Lago Agrio. Areas surrounding Lago will become more mature, with a mix of crops, pasture, and residential/commercial development, while high-density forest areas, previously concentrated in large tracts will be converted to heterogeneous low to small medium density forest tracts, on fincas intermixed with agriculture and pasture.

URL

http://dx.doi.org/10.1007/978-1-4615-0985-1_9

Reference Type

Book Section

Year Published

2002

Author(s)

Walsh, Stephen J.
Messina, Joseph P.
Crews-Meyer, Kelley A.
Bilsborrow, Richard E.
Pan, William K. Y.