Understanding the Impacts of Agricultural Expansion on Biodiversity and Habitat Loss

In recent years, the expansion of agricultural lands into areas rich in biodiversity has led to a conservation dilemma between the need for food security for an expanding human population and the goal of conserving species and habitat to curb biodiversity loss. In this dissertation, I evaluate several different concerns about agricultural expansion from a conservation perspective. One of the central goals of conservation is the preservation of species and habitats within the context of anthropogenic threats. Agriculture has come to exemplify these threats to conservation both directly through habitat loss and indirectly through increased human/wildlife conflict, reduced in connectivity between intact areas and loss of ecosystem services in farming areas. Many of these effects can be observed through monitoring of land use change and populations of critical species. Still other effects will only be observed in the future, as agricultural areas continue to expand. In recognition of the importance of addressing these questions, there has been an increasing push by both agroecologists and conservation scientists to adopt increasingly interdisciplinary approaches in their research, focusing both on ways to minimize, mitigate and predict possible negative effects of agricultural production on biodiversity and the environment.

In Chapter 1, I examine the global impacts of a rapidly expanding commodity crop, oil palm, on deforestation and biodiversity. Here I address the deforestation associated with development of oil palm over the past 25 years in 20 different countries, discussing the implications for future deforestation and risk of biodiversity loss in the context of a changing climate. I conclude that the potential expansion of oil palm agriculture threatens many of the world’s most biodiverse places, but that the exact areas of highest conservation priority are dependent on the biodiversity criteria by which such areas are selected.

In Chapter 2, I build upon this analysis with a study of oil palm in the context of other agricultural development in Peru--a country found to have sharply increasing deforestation related to oil palm in Chapter 1. Here I show that oil palm is contributing to deforestation more than other crops. I also show how the spatial pattern of this impact differs from other crops, with larger and more clustered patches of deforestation. I expand on the analysis of areas at long-term of deforestation from oil palm from Chapter 1 by examining specific biophysical variables that show how oil palm is suitable in habitats not typically exploited for agriculture in this region. I also assess short-term risk of deforestation based on variables associated with human population and accessibility. Finally, I evaluate the effectiveness of protected areas and officially recognized indigenous areas in meeting the threat from oil palm across the different ecoregions of Peru’s Moist Tropical Forest biome.

In Chapter 3, I examine the impacts of agricultural activity on an iconic predator species, the cheetah (Acinonyx jubatus). This species is designated as threatened by the IUCN and plays an important role in the ecosystems in which it occurs. Currently, cheetah are facing loss of habitat and restrictions of connectivity from agriculture, especially livestock production. Farmers in areas where many cheetah occur are also a risk to the species through persecution and killing individuals. We found that cheetah are decreasing in number, supporting an argument to uplist the cheetah to endangered.

I did not only want to focus on the problems associated with agricultural production, but also consider a proposed solutions. Thus, in Chapter 4, I consider an option that could balance the needs for food security of human populations and habitat for species conservation: increased intensity of production on existing agricultural land, such as growing multiple crops per year. This suggestion is not without possible drawbacks or pitfalls, motivating the need to study multiple cropping systems and the consequences of their expansion. However, the study of cropping intensity over large geographical areas is complicated by the lack of high-quality maps of cropping intensity at such scales. I evaluate cropping intensity throughout South America using MODIS Enhanced Vegetation Index (EVI) data in Google’s Earth Engine over the period 2003-2015. I conclude that there is great potential for this approach to reduce habitat loss in South America, but there are also potential complications that could arise from its widespread adoption.