Project Outcomes Report for Collaborative Research: Examining Pyrotechnology and Ecosystem Change in the Archaeological Record

Award NSF BCS 2018896

This project investigated fire ecology and the potential timing and nature of fire use in early humans, using a unique approach that can only be applied in places where entire ancient landscapes are preserved. Rather than identifying individual places on ancient landscapes that may have been exposed to fires, we instead studied evidence of burning over hundreds of square kilometers. We focused this research in the Turkana Basin in Kenya, because this is one of the few localities in the world where volcanic ashes allow us to trace ancient landscapes. We pieced together multiple strands of evidence to identify changes in the fire regime during the time when theoretical studies have suggested that humans were harnessing fire.

The research team from George Washington University studied 3 major time horizons between 2.0 and 1.4 million years ago. This time frame covers the period when a species of our ancestors (Homo erectus) is first found in East Africa. This species had a brain and body size that was much larger than previous human ancestors. It has been hypothesized that this increase in size was spurred on by the ability to extract greater nutrients from foods by cooking them. However, the archaeological record exhibits very scant evidence of burning at this time. Previous attempts at studying the origin of fire focused on evidence of specific fires at singular archaeological sites. We have changed this perspective by studying the ecosystems in the past that may have been impacted by changes in the fire regime.  

The international team of researchers used various different methods to study change sin fire regime across this 600,000 year time frame. One of the most important advances was the development of an agent-based model that simulated the production of evidence of burning under different climatic contexts. An important result of this work showed that during timeframes with increased floodplains (as lakes recede) burnable vegetation increases which makes fires more widespread and the evidence of burning more extensive.

In addition to this modeling, we embarked on an ambitious sampling strategy that identified horizons that are contemporaneous across large areas. We used the extensive volcanic ash deposits in the Koobi Fora Formation, which allow us to determine if layers are contemporary. We collected samples that allow us to measure the levels of burning using the presence of charcoals. We also measured the distribution of vegetation by using phytoliths, which are fossils of plant cells and let us reconstruct past vegetation. This sampling allowed us to reconstruct the ecosystems in the areas around archaeological sites. The results of this sampling allowed us to understand the places that early humans had spent time in over 1 million years ago. In addition, we can now determine that the ecosystems around archaeological sites were substantially impacted by changes in the fire regime. We believe that these changes are the result of ancient humans creating fire in the past. This indicates that humans have been modifying the ecosystems they live in for over 1 million years.

These findings, which show, for the first time, that humans most likely were able to use fire to modify their surroundings and modify food sources, were complemented by a series of innovative geochemical analyses. The collaborative grant at Harvard University is exploring the use of biomarkers that can be specifically linked to ancient combustion. Importantly these biomarkers can provide some guide to the degree of overall burning intensity. These may be able to distinguish between fires that are naturally occurring because of lightning strikes and those that are the result of controlled fires. We have sampled over large parts of the ancient landscape to identify these biomakrers. Our collaborative team is comparing this data to that seen in long term records of the entire Turkana Basin. The differences between local signals and regional signals should reflect the impact that humans have had on local systems.

This project also resulted in significant impacts to several communities. Our research team has worked closely with REU students who are working on our projects in Kenya. These students are from underrepresented groups (Latinx, Native American) and received some of their very first insights into independent research as part of our program. We established a tiered mentoring systems for research interns, several of whom presented their research results are conferences. In addition, we helped develop the exhibitions associated with a new local museum in Ileret (the closest village to our field site). This will have a lasting impact on this community.

This research project made advances toward understanding when pyrotechnology started to shaped the evolutionary trajectory of Homo erectus and perhaps other Pleistocene hominins.

Last Modified: 12/09/2024
Modified by: David Braun