Main Navigation

Stella Mosher

<<Return to main menu

High-resolution sedimentary charcoal records of fire illustrate linkages between fire, climate, and vegetation in the Fynbos Biome, South Africa


I am a Ph.D. candidate working with Drs. Mitchell Power (Geography) and Tyler Faith (Anthropology). I use sedimentary macro-charcoal to reconstruct burning activity over millennia, with a primary focus on South Africa's fire-adapted and highly biodiverse Fynbos Biome. In particular, I am interested in ecosystem change in response to climate, sea level, and human impacts over the past ~4,000 years, spanning the arrival of pastoralists ~2,000 years ago and European colonization beginning in the mid-1600s. I am also interested in methodological development and have worked on experimental burning studies of fynbos and common Utah vegetation to better understand how charcoal morphometrics can tell us about the types of vegetation that burned.


High-resolution sedimentary charcoal and pollen records provide insights into linkages among fire, vegetation, and climate in South Africa’s fire-adapted Fynbos Biome. A new record of fire from a fynbos-afrotemperate ecosystem, Eilandvlei, provides a nearly annual (~2-years/sample) resolution paleo-fire record. Research to-date has reconstructed fire from 4250 to 1900 cal BP, demonstrating the earliest part of this record was a time of frequent and high-magnitude fire events from 4250 to 3500 cal BP. A decline occurred by ~3400 cal BP, possibly linked to multi-decadal drought. Linking sedimentary charcoal and pollen evidence suggests increases in fynbos pollen taxa co-occurred with elevated charcoal influx, and that increased afrotemperate forest taxa co-occurred with reduced charcoal influx. This work demonstrates that fynbos fire regimes are more dynamic through time than previously considered, and points to the importance of long-term fire-vegetation-climate analysis to understand how climate and fire have shaped ecosystems over millennia.


My project is about reconstructing fire activity over the past ~4,000 years from two sites in South Africa's biodiverse and fire-adapted Cape Floristic Region (CFR), which includes the unique and highly endemic fynbos vegetation. Modern day management strategies typically use prescribed burning at 10-13 year intervals, but these conservation plans have relied on historical ecological baselines, and we do not know if these intervals are maintained in deeper time. I am reconstructing fire history over millennia at exceptionally high-resolution 2-year timesteps, and finding that there has been variability in fynbos fire return intervals, ranging between ~1 fire every 14 years to 1 fire every 50 years. I have also found that in the absence of fire, the fire-adverse afrotemperate forest benefits at the expense of the fynbos. My next steps are to identify changes in the fire history record that may co-occur with known changes in human activity in the region, and to link fire activity with relevant climate drivers.