Progress in modeling inherited age effects of charcoal on paleoseismic dates at Pallett Creek on the San Andreas fault
Glenn P. Biasi, Devin McPhillips, & Katherine M. ScharerSubmitted September 7, 2025, SCEC Contribution #14657, 2025 SCEC Annual Meeting Poster #TBD
Paleoearthquake dates at the Pallett Creek paleoseismic site depend on radiocarbon date estimates of charcoal fragments embedded in stratigraphic layers of interest. Dates from detrital samples have an inherited age, the time between fixing the carbon in the sample and the date of deposition in the layer in which it is found. Scharer et al. (2024) investigated the question of inherited age by sampling two modern fires that affected paleoseismic sites on the San Andreas fault on the northern side of the San Gabriel Mountains. At both sites the distribution of inbuilt dates spanned over 400 years, with a median date offset of 86 years for charcoal in stream deposits in the Pallett Creek watershed. The distribution of ages in the late Holocene Pallett Creek deposit is very similar, suggesting one can use the distribution of modern fire dates to remove bias from inheritance in the deposits. We call the distribution of modern dates a "fire curve."
A common practice in paleoseismic event dating is to use just the youngest ages in the layers in the age model, but this does not address inheritance. Paleoearthquake dates derived from this process are almost certainly artificially precise and the true date does not have to be within the uncertainties from radiocarbon and layer ordering. If we model a set of samples as drawn at random from the fire curve, then the broad, 400 year uncertainty of the fire adds to the radiocarbon measurement uncertainty, ballooning the uncertainties on each paleoearthquake date. We investigate applications of order statistics to address the inherited age bias while restraining the uncertainties of the fire curve. Order statistics give expectations for offsets among samples drawn at random from a pdf like the distribution of modern fire dates. We first test the probability of sets of sample dates as being random picks from the fire curve. Many layers fit well as samples from the fire curve, but some show less dispersion and would be 'overcorrected' by the fire curve. Notably, most layer distributions are poorly fit by the fire curve since ca. 1200 AD; around this time we document a drop in the sedimentation rate and increase in charcoal flux. These observations suggest a change in geomorphic and fire conditions require a different fire curve above this layer. We discuss progress in understanding these relationships and their implications for paleoseismic event dates at Pallett Creek.
Key Words
paleoseismic methods
Citation
Biasi, G. P., McPhillips, D., & Scharer, K. M. (2025, 09). Progress in modeling inherited age effects of charcoal on paleoseismic dates at Pallett Creek on the San Andreas fault. Poster Presentation at 2025 SCEC Annual Meeting.
Related Projects & Working Groups
Earthquake Geology
