Research Overview

Stochastic dynamics of barrier islands

The elevation of barrier islands follows from the competion of sand accretion forming the coastal dunes and sand erosion driven by high-water events overtopping the beach. In collaboration with PhD student T. Rinaldo, Prof. I. Rodriguez-Iturbe, our co-chaired PhD student K.A. Ramakrishnan, and postdoc Dr. B. Schaffer (Princeton), we found high-water erosional events, which also control nuisance flooding, can be modeled as a marked Poisson process with exponentially distributed marks and relatively constant distribution parameters (PNAS, 2021a). The finding of a marked Poisson process allowed us to write a relatively simple master equation for the stochastic dynamics of barrier island elevation (PNAS, 2021b). This equation has an analytical solution for the transient distribution with two dimensionless control parameters, relating the average intensity and frequency of erosional events to the maximum dune height and dune formation time. Depending on the control parameters, the solution converges toward a high-elevation barrier, a low-elevation barrier, or a mixed, bimodal, state. We find the average after-storm recovery time—a relaxation time characterizing barrier’s resiliency to storm impacts—changes rapidly with the control parameters, suggesting a tipping point in barrier response to external drivers.

Active research topics include:

Validation of the stochastic model prediction (T. Rinaldo, K.A. Ramakrishnan).
Using the physical model to uncover the controls on barrier resilience, the role of barrier width and to predict barrier respose to sea level rise (T. Rinaldo).

Barrier islands eco-morphodynamics

We also investigate the stochastic dynamics of barrier islands using a physical model that explicitely resolve vegetation growth, sediment transport and dune eco-morphodynamics (PNAS, 2013; Nature Clim. Change, 2015). This allow us to extend the phase space of barrier dynamics to include the effects of plant colonization, which tends to slowdown after-storm recovery and thus increase barrier vulnerability (K.A. Ramakrishnan).

Active research topics include:

Effects of plant stress on barrier dynamics (K.A. Ramakrishnan).
Role of water-driven erosion/accretion in barrier island formation and drowning (A. Yousefi).

CNN-based image segmentation for coastal research

Together with my PhD student B. Kang, we applied Convolutional Neural Networks (CNN)-based image segmentation to analyze high resolution coastal imagery as part of the investigation of after-storm beach and dune recovery. We studied the stochastic properties of run-up flooding events and confirmed they can be modeled as a Poisson process with exponentially distributed marks. We are currently investigating the factors controlling aeolian sand transport on beaches and the wet-to-dry transition behind coastal recovery.

Water and planetary ripples and dunes

In collaboration with B. Andreotti (ENS, Paris) and P. Claudin (ESPCI, Paris Tech), I expanded my previous research on aeolian dunes and dune fields (e.g. PNAS, 2009 & Earth Surf. Proc. Land., 2010), to investigate the origin of subaqueous ripples and dunes using a continuous description of sediment transport modulated by the surface shear stress. Numerical simulations reproduced for the first time the scaling of both types of bedforms and predicted the transition between them as function of the transport relaxation length and the grain Reynolds number, thus opening the door to a unified theory of bedforms that also explains Martian and Venusian bedforms (Nature Geoscience, 2019).

Sediment transport modeling

I investigate the characteristics of bedload and saltation using two-phase numerical simulations based on a discrete element method for particles coupled to a continuum Reynolds averaged description of hydrodynamics (Aeolian Res., 2011 & Phys. Fluids, 2012). The direct numerical simulations of grains interacting with a wind flow reproduced the growth of aeolian ripples and led to a new formation mechanism, involving ‘resonant’ grain trajectories (PNAS, 2014). In collaboration with T. Pähtz (Zhejiang University, China), we use transport simulations to develop a novel analytical model for the transport cessation threshold (JGR: Earth Surface, 2018; Review of Geophysics, 2020) and a new rheology unifying dense and dilute granular regimes, usually described by either a viscoplastic rheology or kinetic theory, respectively (Phys. Rev. Lett., 2019). We also use the balance of fluctuation energy within the transport layer to unify aeolian and fluvial sediment transport (Phys. Rev. Lett., 2020).

Eco-morphodynamics of vegetated dunes (parabolic and coastal dunes)

I developed an ecomorphodynamic model of coastal vegetated dunes (Coastal Dune Model, doi:10.5281/zenodo.847746) and identified important equilibrium states and control parameters characterizing the response of the coastal dune ecosystem to environmental stressors. The main control parameters are: (i) the ‘mobility index’ quantifying the competition between vegetation growth and sand erosion/accretion that triggers a transition from mobile to stable dunes and thus exerts a primary control on the stability of coastal dune ecosystems (Phys. Rev. Lett., 2006); (ii) the ‘vegetation limit’, a length quantifying the interaction between ‘dune-building’ grasses and the shoreline that controls the maximum size of coastal dunes and explains a well-known empirical correlation between dune size and beach type, which has direct implications for coastal vulnerability to storms (PNAS, 2013); and (iii) the ‘vulnerability index’ quantifying the competition between vegetation recovery, storm erosion and sea level rise that induces a bistable response of barriers islands, where dunes may not recover after an overwash and islands can be trapped in a perpetual state of vulnerability (Nature Clim. Change, 2015). We are currently improving the latter result using a better description of the stochastic water-driven erosion.

Eco-morphodynamics of salt marshes

I also worked on a simplified process-based large-scale model of salt marsh dynamics that fills the gap between complex physics-based models at smaller scales and phenomenological point-models typically used at large scales. This model was used to better understand the current and future resilience of marshes in Virginia and Massachusetts (J. Env. Economics and Management, 2019; Limnol. and Oceanography, 2020). Furthermore, our finding of a scale-invariant sediment redistribution over a marsh platform led to a self-similar mechanism for runaway marsh fragmentation, triggered at a critical value of the rate of sea level rise, characterized by a power-law distribution of the size of marsh ponds and the eventual drowning of the marsh platform (One Earth, 2021). Model predictions were later confirmed using remote sensing data from Blackwater, MD (Front. Mar. Sci, 2021).

Google Sites

Report abuse