Evaluating uncertainties in latewood tree ring reconstructions of tropical cyclone precipitation using statistically/dynamically downscaled storms
Abstract
The short observational record of tropical cyclones (TCs) limits scientific understanding of how certain TC properties (e.g., frequency) respond to climate forcing. Paleohurricane records use natural archives (tree rings, coarse grained sediment) to reconstruct TC properties (rainfall, frequency) over the past few hundreds to thousands of years. These records have the potential to bolster TC statistics and subsequent understanding of climatic controls on TCs. Unfortunately, differing sensitivities and sampling biases in the various paleohurricane proxies restrict our ability to compile these records into regional or basin-scale TC estimates. In particular, no work has been done combining estimates of direct TC frequency (from sediment proxies) with TC-related climate properties (e.g., rainfall from tree cores). Here we explore the relationship between TC rainfall and TC frequency as captured by tree ring reconstructions using synthetic TCs. We generate a 1000-member ensemble of pseudo tree ring records of latewood width of longleaf pine (Pinus palustris, Mill.) from southern Mississippi using a Poisson process-based random draw from a large set of statistically/dynamically downscaled storms forced with Max Planck Institute (MPI-ESM-P) boundary conditions for the past millennium. Pseudo records convert synthetic TC rainfall into latewood width using a previously published statistical calibration and seasonal sensitivity (a forward, or proxy system modeling approach). We find that our individual pseudo latewood width records can reproduce low frequency variability found in proxy records. Storms passing to the west of our proxy site with slower translational velocities are more likely to produce dramatic peaks in rainfall, increasing latewood width. Comparing our pseudo tree ring records with pseudo sediment records generated for the Gulf Coast indicates promise in combining proxies sensitive to TC rainfall with proxies sensitive to storm overwash. Our findings lend support for multi-proxy efforts to reconstruct past TC frequency, and to bolster our understanding of how future climate change may impact storm statistics using data from the past as an analog for the future.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2022
- Bibcode:
- 2022AGUFMPP35D1003W