arXiv subjects

Mark S. Bartlett

Publications and source records attributed to Mark S. Bartlett.

5 recordsLinked to original sources

arXiv2026

Physically-based dimensionless features for pluvial flood mapping with machine learning

Rapid delineation of flash flood extents is critical to mobilize emergency resources and to manage evacuations, thereby saving lives and property. Machine learning (ML) approaches enable rapid flood delineation with reduced computational demand compared to conventional high-resolution, 2D flood models. However, existing ML approaches are limited by a lack of generalization to never-before-seen conditions. Here, we propose a framework to improve ML model generalization based on dimensionless, multi-scale features that capture the similarity of the flooding process across regions. The dimensionless features are constrained with the Buckingham $Π$ theorem and used with a logistic regression model for a probabilistic determination of flood risk. The features were calculated at different scales by varying accumulation thresholds for stream delineation. The modeled flood maps compared well with the results of 2D hydraulic models that are the basis of the Federal Emergency Management Agency (FEMA) flood hazard maps. Dimensionless features outperformed dimensional features, with some of the largest gains (in the AUC) occurring when the model was trained in one region and tested in another. Dimensionless and multi-scale features in ML flood modeling have the potential to improve generalization, enabling mapping in unmapped areas and across a broader spectrum of landscapes, climates, and events.

physics.geo-ph↗

arXiv2026

Extending the Joint Probability Method to Compound Flooding: Transition Zone Delineation, Flood Depth Attribution, and Design Event Selection

Quantifying the frequency of compound flood depths is a fundamental challenge in low-gradient coastal watersheds, where flood hazards arise from the nonlinear interaction of storm surge, rainfall, and riverine flooding. Existing approaches often characterize either the joint occurrence of flood drivers or the flood response for prescribed events, but they do not derive the long-term frequency distribution of compound flood depths from probabilistic descriptions of rainfall and antecedent hydrologic conditions. Traditionally, coastal flood frequency has been quantified using the Joint Probability Method (JPM), which represents storm surge probabilistically. Although recent studies have incorporated rainfall into JPM-based analyses, rainfall is treated as a deterministic function of JPM storm characteristics rather than as a conditional probability distribution. Here, we extend the JPM by coupling its event-scale stochastic description of storm characteristics with probabilistic rainfall realizations and stochastic antecedent hydrologic conditions, thereby enabling propagation of these stochastic processes through the flood response to derive the compound flood-depth distribution. The framework provides a statistical basis for delineating compound flood transition zones, probabilistically attributing flood depths to hydrologic and coastal processes, and selecting response-based design storms for specified annual exceedance probabilities (or return periods). Application to the Lake Maurepas basin, Louisiana, shows that the statistically defined compound flood transition zone is more than twice the area identified from event-based analyses and that compound interactions increase flood depths by up to 0.7 m. This extended JPM establishes a probabilistic foundation for compound flood hazard assessment and response-based design.

physics.geo-ph↗

arXiv2025

Stochastic ecohydrological perspective on semi-distributed rainfall-runoff dynamics

Quantifying watershed process variability consistently with climate change and ecohydrological dynamics remains a central challenge in hydrology. Stochastic ecohydrology characterizes hydrologic variability through probability distributions that link climate, hydrology, and ecology. However, these approaches are often limited to small spatial scales (e.g., point or plot level) or focus on specific fluxes (e.g., streamflow), without accounting for the entire water balance at the basin scale. While semi-distributed models account for spatial heterogeneity and upscaled hydrologic fluxes, they lack the analytical simplicity of stochastic ecohydrology or the SCS-CN method and, perhaps more importantly, do not integrate the effects of past random variability in hydroclimatic conditions. This hinders an efficient characterization of hydrological statistics at the watershed scale. To overcome these limitations, we merge stochastic ecohydrology, the spatial upscaling of semi-distributed modeling, and the SCS-CN rainfall-runoff partitioning. The resulting unified model analytically characterizes watershed ecohydrological and hydrological statistics using probability density functions (PDFs) that are functions of climate and watershed attributes -- something unattainable with the Monte Carlo methods of traditional stochastic hydrology. Calibrated across 81 watersheds in Florida and southern Louisiana, the model PDFs precisely capture the long-term average water balance and runoff variance, as well as the runoff quantiles with a median normalized Nash-Sutcliffe (NNSE) efficiency of 0.95. These results also advance the SCS-CN method by providing an analytical PDF for the Curve Number (CN), explicitly linked to climate variables, baseflow, and the long-term water balance partitioning described by the Budyko curve.

physics.geo-ph↗

arXiv2019

Unified representation of the C3, C4, and CAM photosynthetic pathways with the Photo3 model

Recently, interest in crassulacean acid metabolism (CAM) photosynthesis has risen and new, physiologically based CAM models have emerged. These models show promise, yet unlike the more widely used physiological models of C3 and C4 photosynthesis, their complexity has thus far inhibited their adoption in the general community. Indeed, most efforts to assess the potential of CAM still rely on empirically based environmental productivity indices, which makes uniform comparisons between CAM and non-CAM species difficult. In order to represent C3, C4, and CAM photosynthesis in a consistent, physiologically based manner, we introduce the Photo3 model. This work builds on a common photosynthetic and hydraulic core and adds additional components to depict the circadian rhythm of CAM photosynthesis and the carbon-concentrating mechanism of C4 photosynthesis. This allows consistent comparisons of the three photosynthetic types for the first time. It also allows the representation of intermediate C3-CAM behavior through the adjustment of a single model parameter. Model simulations of *Opuntia ficus-indica* (CAM), *Sorghum bicolor* (C4), and *Triticum aestivum* (C3) capture the diurnal behavior of each species as well as the cumulative effects of long-term water limitation. The results show potential for use in understanding CAM productivity, ecology, and climate feedbacks and in evaluating the tradeoffs between C3, C4, and CAM photosynthesis.

q-bio.QM↗

arXiv2018

State dependent jump processes: Itô-Stratonovich interpretations, potential, and transient solutions

The abrupt changes that are ubiquitous in physical and natural systems are often well characterized by shot noise with a state dependent recurrence frequency and jump amplitude. For such state dependent behavior, we derive the transition probability for both the Itô and Stratonovich jump interpretations, and subsequently use the transition probability to pose a master equation for the jump process. For exponentially distributed inputs, we present a novel class of transient solutions, as well as a generic steady state solution in terms of a potential function and the Pope-Ching formula. These new results allow us to describe state dependent jumps in a double well potential for steady state particle dynamics, as well as transient salinity dynamics forced by state dependent jumps. Both examples showcase a stochastic description that is more general than the limiting case of Brownian motion to which the jump process defaults in the limit of infinitely frequent and small jumps. Accordingly, our analysis may be used to explore a continuum of stochastic behavior from infrequent, large jumps to frequent, small jumps approaching a diffusion process.

cond-mat.stat-mech↗