Land Change, Urban Expansion, and Gold Mining in Latin America: A Land System Science Approach
Álvarez Berríos, Nora L.
Aide, Mitch (Consejero)
Latin American and the Caribbean region (LAC) is undergoing dynamic land changes due to agricultural expansion, fast-growing urban areas, and intensive mining production. These land changes are accelerating, often intensified by global drivers such as the increasing demand for commodities (e.g. gold), international migration, and climate change. My dissertation applies an interdisciplinary approach framed within Land System Science to study understudied land change in LAC and the global and local drivers of change. Specifically, I combined remote sensing analyses with socio-economic information to investigate land change related to: 1) urban expansion in Colombia, Ecuador, Peru, and Bolivia; 2) woody vegetation and agricultural changes in the Greater Antilles in the Caribbean, and 3) gold mining deforestation in the Tropical Moist Forest of South America. At a local scale, I conducted groundwork using acoustic recorders to examine the effect of gold mining activities on avian and amphibian populations in a small-scale mining event in Peru. In the Central Andean region, urban expansion was examined between 1992 and 2009 using night-time lights satellite images combined with census data. Overall, countries with longer urban growth histories (Colombia and Peru) underwent lower urban population growth and reduced rates of urban expansion, whereas countries in intermediate stages of urban growth (Ecuador and Bolivia) expanded faster in response to urban population growth. In the Greater Antilles between 2001 and 2010, analyses of MODIS images in combination with a literature review showed that land change was dominated by woody vegetation regrowth and agricultural abandonment. This region underwent a net woody vegetation gain of 801 km2 and mixed-woody/plantations increased by 1,482 km2, while agriculture/herbaceous vegetation declined by 1,498 km2. Most of the increase in woody vegetation was attributed to the decline of sugarcane cultivation and the expansion of an invasive shrub on abandoned fields in Cuba. MODIS data was also used to investigate gold mining expansion in South America from 2001 to 2013 by analysing the loss of forest in ~1600 potential and active gold mining sites. Aproximately 1680 km2 of tropical moist forest was lost in these mining sites between 2001 and 2013. Deforestation was significantly higher during the 2007–2013 period, and this was associated with the increase in global demand for gold after the international financial crisis. More than 90% of the deforestation occurred in four major hotspots: Guianan moist forest ecoregion (41%), Southwest Amazon moist forest ecoregion (28%), Tapajós–Xingú moist forest ecoregion (11%), and Magdalena Valley montane forest and Magdalena–Urabá moist forest ecoregions (9%). This increase in mining activities throughout the tropics is likely to significantly alter wildlife communities in the region. I performed a rapid assessment on potential biodiversity changes occurring in a small-scale mining community within the Southwest Amazon moist forest ecoregion in Peru, where my study provided evidence of mining activities altering the composition of bird and amphibian communities at a location within the buffer zone of the Tambopata National Reserve. Amphibians prevailed in the active and abandoned mining land-use types, while birds were much more sensitive to mining activities. The differences in the composition of bird species across the land-use types (i.e. active, abandoned, forest) were correlated with habitat disturbance and anthropogenic noise, suggesting that some birds are fleeing active and abandoned mining areas. In conclusion, my research indicated that global factors are major drivers shaping local changes in LAC, with multiple effects across spatial scales. Monitoring and modeling drivers at different scales are therefore essential to understand the major components of land change and to help design conservation efforts based on up-to-date information. The methods and scientific approaches of this dissertation can be applied to monitor other regions undergoing rapid land change, particularly due to urban and gold mining expansion.