How useful is the subfossil record for interpreting pre-human ecosystems and current extinctions?

Over the last 50,000 years, as humans spread out of Africa and around the globe, they have been implicated in a series of extinction events coinciding with their arrival in different geographic regions. The extinctions of large-bodied animals on each of the world's continents towards the end of the last Ice Age were followed by extinctions on different islands, as human colonists reached more and more isolated systems. Island extinctions during history and recent prehistory have been severe - for example, almost 40% of all historical mammal extinctions have occurred in the West Indies, and this region has altogether lost around 100 native mammals since humans first arrived there over 5,000 years ago. We are able to recognise these different extinction events in the recent past from fossil or subfossil material that shows relatively recent survival of now-extinct species. These records have often been used at face value to calculate the total number of extinctions caused by human activity throughout history and prehistory, to assess how much damage we have caused to global biodiversity and to predict how many more extinctions are likely in the future. However, the fossil record does not represent an objective source of information. Not only is it far from complete, with animals only being preserved under unusual circumstances, but it also contains many different biases: certain kinds of organisms in certain geographic regions are more likely than others to be fossilised. This means that what we think we know about past human-caused extinctions needs to be reassessed. Can we describe these biases further, and is it actually possible to use the fossil record to reconstruct the faunas and ecosystems that were present before humans spread around the globe? My research will attempt to answer this key question by focusing on a specific geographic region, and then comparing this region to similar systems in other parts of the world. I have already conducted research into extinct West Indian mammals, and recently described a new extinct rodent genus, showing that there is still much to learn about the area. I intend to determine the ecological niches occupied by known extinct West Indian mammals by morphological and isotopic analysis. Comparisons with the abundance and diversity of similar animals alive in other parts of the world will indicate the likely former abundances of the West Indian species. Preservation biases in the West Indian fossil record can then be examined by assessing whether the predicted abundances correspond with the number of specimens actually present in different museum collections. Similar comparisons with modern island ecosystems can also suggest whether we can identify any West Indian islands for which undescribed extinct mammal species probably still await discovery. These biases will also be studied by researching habitat preferences of Hispaniola's surviving mammals, and seeing whether subfossil deposits representing different environments show the expected abundances suggested by the modern study. This improved understanding of West Indian mammal ecology will then show whether it is also possible to identify ecological interactions, such as fruit dispersal, that these animals may have had with their environment, and possible effects that their extinction would have had on West Indian ecosystem structure. This improved understanding of the region's pre-human ecology can then suggest appropriate methods of restoring ecosystems to their original equilibrium, an especially important goal for ecologically degraded regions such as Haiti. Finally, the West Indian fossil record will be compared with what is known about recently extinct mammals and birds from other islands elsewhere across the world, to see whether biases vary in different regions. These studies will together allow us to determine the full extent of prehistoric species loss, and ultimately use the past to understand ongoing extinctions.