| dc.description.abstract |
<p>Human and natural disturbance affect the Amazon basin at several spatial and temporal
scales. In this thesis, I used satellite-detected hot pixels to examine patterns of
human-caused disturbance and protected areas in the Brazilian Amazon from 1996-2006.
Deforestation fires, as measured by hot pixels, declined exponentially with increasing
distance from roads. Fewer deforestation fires occurred within protected areas than
outside and this difference was greatest near roads. However, even within reserves,
more deforestation fires occurred in regions with high human impact than in those
with lower impact. El Niño-related droughts affected deforestation fires most outside
of reserves and near roads. There was no significant difference in fire occurrence
among inhabited and uninhabited reserve types. </p><p>Within this context of disturbance
in the Brazilian Amazon basin, I examined relatively undisturbed savanna-like `campina'
ecosystems. I reviewed the literature on campinas and discussed their variation and
their significance for beta diversity. As one of two case studies, I assessed spatio-temporal
patterns of disturbance (fire and blowdowns), and vegetation change from 1987 to 2007
in campinas in the central Brazilian Amazon using Landsat imagery. In 2001 images,
an increase in open areas corresponded with significantly more visible signs of disturbance,
likely precipitated by the 1997-98 El Niño. Bird community data indicated a trend
of more generalist/savanna species in more frequently disturbed campinas. </p><p>As
the second case study, I used daily 500 m resolution MODIS reflectance data to assess
seasonal and inter-annual flooding in ~33,000 km2 of campinas in the Negro river basin.
Flooding cycles of these wetland campinas critically influence regional ecosystem
processes. Flooded areas ranged from 15,000 km2 at the end of the rainy season (August-Oct)
to little, if any, open water in the driest times (Jan-Mar). Predictable seasonal
flood pulses occurred, but also displayed high inter-annual variability. This variability
was weakly correlated with the Multivariate El Niño Southern Oscillation Index (MEI).
</p><p>Campina ecosystems are an important, but largely overlooked, component of the
biodiversity of the Amazon basin. My research shows that climate, particularly ENSO-associated
droughts, strongly affects campinas even in remote areas, just as it increases fire
frequencies in more populated regions of the Amazon.</p>
|
|
