Centennial‐Scale Shifts in Storm Frequency Captured in Paleohurricane Records From The Bahamas Arise Predominantly From Random Variability

Abstract

, coastal communities are becoming more vulnerable to TC-induced inundation. Unfortunately, the models we use to predict TC activity in real time and into the future are typically validated using only the observational record. Specifically, with only 168 years of observations that suffer from biases (e.g.,

On the Interpretation of Natural Archives of Atlantic Tropical Cyclone Activity

“…influenced by a complex array of local-scale meteorological phenomena that, for all intents and purposes, occur by chance (Wallace et al, 2020). Consequently, we have to question the extent to which individual event-based reconstructions of storms (e.g.…”

“…While these regional-scale climatic phenomena are likely to have affected the passage of storms across the region, there are still too few (<10) high-resolution event-based records to attribute confidently the spatial and temporal variability of past hurricane activity to climatic change. In a landmark study, Wallace et al (2020) set out to assess the relative importance of stochastic (random) and deterministic (climatic) processes on the deposition of discreet hurricane-induced event layers within a blue hole sediment record on South Andros Island in The Bahamas (Figure 1f, Wallace et al 2019). They simulated 1000 virtual sediment records of hurricane activity for South Andros Island drawn randomly from a synthetic storm dataset derived from a deterministic (climate-driven) hurricane model, which was, in turn, constrained by the initial and boundary conditions of the Max Planck Institute Earth System Model (MPI-ESM).…”

“…By comparing the ensemble average values of 1,000 simulated sediment records from each of South Andros, Long Island and Caicos Island (all similarly constrained by the boundary conditions of the MPI‐ESM), Wallace et al. (2020) establish a virtual climate signal for each site. For those study sites located within 400 km of each other (South Andros/Long Island and Long Island/Caicos Island), the climate signals correlate significantly, albeit with moderate to low correlation coefficients ( r = 0.32 and r = 0.36, respectively).…”

“…However, when comparing the variability of the individual simulated records with that of the model's ensemble average, which represents the underlying climate signal, much of the centennial-scale variability was lost indicating that random processes, not climatic change, were responsible for the observed variability in the storm record. To test the cause of the centennial-scale variability in the simulated records, Wallace et al (2020) created a second set of virtual sediment records using a pure Poisson process (Figure 1a), which generates hurricane strikes in a random manner (i.e., without the influence of climate). Since these virtual records generated similar centennial-scale variability to that of the physical sediment records, the authors argued that the variability exhibited by the latter also likely occurred as a result of Nevertheless, Wallace et al (2020) suggest that there is some promise in combining individual event-based sediment records located within a few hundred kilometers of each other to capture a broader spatial range of tropical cyclone events and construct a regional composite record that teases out the underlying climate signal.…”

“…To test the cause of the centennial-scale variability in the simulated records, Wallace et al (2020) created a second set of virtual sediment records using a pure Poisson process (Figure 1a), which generates hurricane strikes in a random manner (i.e., without the influence of climate). Since these virtual records generated similar centennial-scale variability to that of the physical sediment records, the authors argued that the variability exhibited by the latter also likely occurred as a result of Nevertheless, Wallace et al (2020) suggest that there is some promise in combining individual event-based sediment records located within a few hundred kilometers of each other to capture a broader spatial range of tropical cyclone events and construct a regional composite record that teases out the underlying climate signal. By comparing the ensemble average values of 1,000 simulated sediment records from each of South Andros, Long Island and Caicos Island (all similarly constrained by the boundary conditions of the MPI-ESM), Wallace et al (2020) establish a virtual climate signal for each site.…”