Home

Research Projects:

Significance of biotic and climatic reconstruction in tropical areas

Estimating past climate from fossil leaves

Environments of the first Americans

Curriculum vitae

Contact information

Environmental Science Program

 




Satellite view of Africa, showing plant densities.


SIGNIFICANCE OF BIOTIC AND
CLIMATIC RECONSTRUCTION
IN TROPICAL AFRICA

The African continent is home to unique faunal and floral communities that developed over the last 65 million years through a combination of isolated evolution and the impact of immigrants from other continents. Today, tropical Africa's savanna regions are known for their diverse mammalian communities while much wetter areas support species-rich tropical rain forests. Yet, little is known about when these important ecosystems, which contribute significantly to global biodiversity, originated, what ancestral plant and animal species occupied their precursors, or in what patterns past communities were distributed across the tropical region at times in the distant past. Furthermore, ancient climates that would have been the primary determinant of past communities are poorly understood for this part of the world for much of the Cenozoic.

Documenting past climate for equatorial Africa has practical applications important to our understanding of modern and future climate. First, in order to understand current climate patterns, they must be put into a long-term perspective by exploring climates of the past. Secondly, quantitative reconstruction of climate provides test cases for computer-driven climate models designed to represent the world's climate system today and in the future, when atmospheric concentration of carbon dioxide is predicted to be 2 or 3 times current levels. Climate modelers attempt to recreate, through computer simulations based on changes to the starting (modern) conditions, past climates based on interpretation of fossils or isotopic data. If modelers can simulate past climates that are consistent with fossil and isotope data, then they can assume that their models are good representations of reality. However, to test models in that way, they need an ample number of data points for past time periods of interest, and data from low latitudes are greatly needed.