Hurricanes Katia, Irma, and Jose from west to east on 8 September 2017 (Photo credit: NASA/NOAA GOES Project).

U.S. Coastal Research Program Advancing the understanding of storm processes and impacts By Nicole Elko, Casey Dietrich, Mary Cialone, Hilary Stockdon, Matt W. Bilskie, Brandon Boyd, Bianca Charbonneau, Dan Cox, Kendra Dresback, Steve Elgar, Amanda Lewis, Patrick Limber, Joe Long, Chris Massey, Talea Mayo, Kathryn McIntosh, Norberto Nadal-Caraballo, Britt Raubenheimer, Tori Tomiczek, and Anna Wargula

n 2017, Hurricanes Harvey, Irma, and To determine needed advancements eXperiment), anticipated in 2020-2021, Maria caused more than $200 billion in storm forecasting, the U.S. Coastal were presented to encourage continued dollars of damage in the United States, Research Program (USCRP) hosted a engagement of coastal researchers across asI well as the incalculable cost of the loss Storm Processes and Impacts workshop disciplines. of life and mental trauma associated for coastal stakeholders 16-18 April This paper represents a synthesis of the with these disasters (Sullivan 2017). In 2018, in St. Petersburg, Florida. The at- forecasting challenges, research needs, a changing climate, sea level rise and the tendees included local coastal managers, infrastructure improvements, and com- potential for increasing tropical cyclone emergency managers, state and regional munication challenges presented and intensity can result in even more devastat- agencies, federal agency scientists and discussed during the workshop. Each ing damages (IPCC 2013; Knutson et al. engineers, academics, and private indus- section includes a table of prioritized 2010). Therefore, engineers, community try scientists and engineers. Workshop challenges or needs based on the work- planners, and coastal residents need ac- objectives were to synthesize present shop participants’ feedback. curate, timely, and accessible forecasting capabilities for modeling storm processes of storm processes and their impact on and forecasting impacts and to prioritize Three main research goals to address coastal communities to bolster national advancements. In addition, the workshop forecasting challenges and advance storm resilience and reduce risk to life and provided an opportunity to bridge the ap- impact predictions were identified: (1) property during these events. However, parent gap between the research of coastal expand the understanding and repre- along with uncertainties in understand- scientists and engineers and the infor- sentation of dynamic feedbacks between ing and modeling of storm processes, mation being distributed publicly and multiple time/length scales and processes there are complex challenges associated to emergency managers before, during, (Represent Dynamic Feedbacks); (2) with determining and meeting the needs and after storm events. Finally, plans for improve forecasting methodology and of end users who rely on these forecasts a large-scale, extreme-event field experi- communicate inherent model uncer- for emergency management decisions. ment, DUNEX (During Nearshore Event tainty (Improve Forecasting Methods); Shore & Beach  Vol. 87, No. 1  Winter 2019 Page 37 Table 1. about the timing of evacuation orders Prioritized forecasting challenges (Table 2). 1 — Integrate multiple hazards/processes over temporal & spatial scales The specific research needs (Table 2) 2 — Valuate accuracy of meteorological conditions were grouped into the following research 3 — Determine & propagate uncertainty themes or goals which are discussed in 4 — Predict timeline and level of impact the sections below. n Represent the dynamic feedbacks and (3) quantify the role of nature-based Furthermore, forecasting the impacts among storm-driven hydro- and mor- and engineered shorelines in mitigating of these hazards requires accurate me- phodynamics (2, 3, 4, 9); storm effects (Assess Hazard Mitigation). teorological conditions that may not be To advance these goals, improvements available. Storm track errors increase 30- n Improve forecasting methods to in observations and modeling (Infra- 40 nautical miles per day prior to landfall, better account for uncertainties (1, 5, 7, structure) are needed. A final challenge and errors in intensity forecasts increase 12); and involves translating these processes, for three days, then level off (hurricanes. n forecasts, and uncertainties to the end gov/surge). Determining and propagating Assess the hazard mitigation capa- user (Communication). this uncertainty into storm models and bilities of natural, nature-based, and built then into forecasts poses a significant features (6, 8, 10, 11). FORECASTING CHALLENGES challenge, and thus it is difficult to predict Forecasting challenges prioritized by REPRESENT the timeline and impact levels on the workshop participants are given in Table DYNAMIC FEEDBACKS temporal scales required by emergency 1 and discussed below. Workshop participants identified managers who are often expected to begin research needs related to the dynamic Extreme storms involve multiple making decisions up to one week before feedback between physical (hydrody- hazards driven by a variety of coastal storm impact. namic) processes and the system (mor- processes that operate on different tem- Forecasting at adequate spatial reso- phologic) response during storm events. poral and spatial scales. Storm surge, lution also presents a challenge. There Four specific objectives identified were to: dangerous surf, strong and damaging is a discrepancy in the resolution of 1) better understand sediment transport winds, tornadoes, and extreme rainfall national-scale models like the Sea, Lake, to improve our ability to simulate mor- can all contribute to the overall storm and Overland Surges from Hurricanes phologic change, 2) integrate relevant hazard. However, each of these hazards (SLOSH) model, which have a typical grid physics into numerical models, couple the may impact a coastal community at a dif- cell resolution of 500 to 1,000 m in coastal models, and provide a feedback mecha- ferent time during a storm, and the state regions, as compared to the parcel- to nism between models, 3) include the ef- of the art of mapping and modeling these street-level resolution required for evacu- fects of vegetation and sedimentation on multiple hazards has not yet advanced to ation zone assignments (Kerr et al. 2013). beach and dune response and recovery, provide coupled forecasts about water- and 4) incorporate rainfall runoff into related hazards. Emergency managers RESEARCH NEEDS the hydrodynamic and morphodynamic may be utilizing discrete products to Workshop participants prioritized a modeling capabilities. To represent assess the impacts of large rainfall events number of specific research needs, rang- these dynamic feedbacks during storms, and storm surge impacts. The integration ing from fundamental science questions research will require the integration of of multiple hazards and processes over (e.g. better understanding sediment multiple processes over a range of spatial the course of an extreme event remains a transport to improve morphologic and temporal scales as discussed in the significant forecasting challenge. change models) to practical questions forecasting challenges section above. Table 2. The coupling between hydrodynam- Prioritized research challenges ics and morphodynamics on a variety of spatial and temporal scales is needed 1 — Quantify spatial/temporal resolution and accuracy required for users to understand and predict the full range 2 — Improve models to integrate relevant processes of coastal impacts from storms and the 3 — Expand understanding of sediment transport – model improvements in recovery of beaches that occurs between morphology individual storms events. For instance, 4 — Determine feedbacks between physical processes feedbacks (two-way coupling) between 5 — Determine inputs for accurate coastal hazard forecasts in probabilistic ocean waves, wave-driven currents, and models geomorphology may result in offshore 6 — Study mitigation solutions - how to mitigate hazards movement of sandbars during storms 7 — Learn from model error to improve process understanding and models (Thornton and Humiston 1996, Gallagher 8 — Quantify the value of nature and nature-based features et al. 1998) and onshore migration of 9 — Investigate beach and dune recovery the bar during small to moderate wave 10 — Understand variability of impacts depending on shoreline type – cliff, conditions (Hoefel and Elgar 2003). marsh, mangrove, ice Similar coupling between oceanographic 11 — Assess impacts to structures (hydro, load, energy, surge) and geomorphologic processes affect the 12 — Assess available statistical methods to best use limited model results movement of other features in the beach profile such as the shoreline, berm, and

Page 38 Shore & Beach  Vol. 87, No. 1  Winter 2019 dune. Alongshore variations in sandbar suggesting that the transport of sediment likely are important during extreme depth and location can affect the maxi- from the dune is not linearly related to events when dune and bluff erosion can mum elevation of ocean water levels and the maximum water level elevations dur- be severe and strong rip currents (and breaking wave heights and accompanying ing a storm (Long et al. 2014). Instead, undertow) may carry sediments into deep wave runup on the beach (Raubenheimer process-based models are necessary to water. The net gain or loss of material to et al. 2001, Cohn and Ruggiero 2016), account for the timing, coupling of, and the outer surf zone during storms may be which in turn influences the potential for feedbacks between inner-shelf, surf zone, the determining factor for net shoreline dune erosion (Overbeck et al. 2017) and and beach processes (Roelvink et al. 2009, movement, and the exchange processes the quantity of sediment transported back Palmsten and Holman 2012, Callaghan may be closely coupled. Additionally, the to the surf zone. Sediment transported and Wainwright 2013, Dissanayake et al. large changes observed in erosional pat- from the dune to the surf zone can change 2014, Splinter et al. 2014). terns over small alongshore spatial scales the sandbar geometry, resulting in less (O(100-1,000 m)) are not modeled well revious observations and model (negative feedback) or more (positive) (de Winter et al. 2015), likely owing to simulations suggest alongshore- dune erosion. Although the offshore the simplifications and parameterizations variable dune erosion during surf zone transport owing to mean flows used in the models. Pwave collision events may be related to may be understood reasonably well, the inhomogeneous along-coast inner-shelf Feedbacks between hydrodynamic, processes affecting onshore transport bathymetry (shoals), spatial changes in morphologic, hydrogeologic, and Aeo- (including the recovery of beaches after the orientation of the coast relative to lian processes also may contribute to the storms) are less certain. Studies have the (alongshore homogeneous) incident observed alongshore inhomogeneity. shown that onshore transport may be wave direction, timing of the largest Periodic dune erosion hotspots may have driven by flow skewness (Trowbridge and waves with respect to high tide, the initial positive feedbacks with the growth of Young 1989), flow accelerations under location of the dune toe and the initial erosive beach cusps and rip currents that asymmetric waves (Hoefel and Elgar dune topography and/or saturation (dune carry sediments offshore (Castelle et al. 2003), and boundary layer processes shape) (Bender and Dean 2003, Schupp 2017). During beach recovery, wider cusp (Henderson et al. 2004), and is sensi- et al. 2006, Claudino-Sales et al. 2008, horns may be eroded by Aeolian sediment tive to near-bed turbulence and pick up Houser et al. 2008, Palmsten and Holman, transport, whereas wind-blown sand is functions, inter-granular stresses and 2011, Dissanyake et al. 2014, de Winter captured in dune scarps and low sections fluid-granular interactions (Hsu and Liu et al. 2015, Safak et al. 2017, Splinter et of the beach. Process-based models may 2004), stratification owing to the sus- al. 2018). However, parameters in the be able to simulate many of these feed- pended sediment (Falchetti et al. 2010), models used to investigate these processes back mechanisms but because these pro- and infiltration and exfiltration through are usually calibrated and evaluated by cesses occur on different temporal scales the seafloor (Turner and Masselink 1989, comparison of simulated with observed (hours to decades) a single model is not Nielsen et al. 2001, Chardón-Maldonado topographic and bathymetric changes capable of simulating all of the relevant et al. 2016). Wave-resolving models are between pre- and post-storm surveys processes. Moreover, the simulations are required to accurately simulate these (McCall et al. 2010, Splinter and Palm- sensitive to user-selected hydrodynamic transport processes but are not always sten 2012, Dissanyake et al. 2014, Harter and morphodynamic parameters (McCall feasible to implement when considering and Figlus 2017) and the processes that et al. 2010, Splinter and Palmsten 2012, the broad spatial extent of coastline that occur during the storm may not be well- Dissanyake et al. 2014, Harter and Figlus can be impacted by some extreme storms represented. In addition, in some cases 2017), which often are not well known (O(1,000km)). the available pre-storm topography has and calibrated values from one area may Storm-induced, large-scale (O(10,000 not been updated for years before the not be directly applicable to another. In m)) subaerial sand erosion patterns can event, which has been shown to impact particular, understanding the effects of be reproduced fairly well with some ex- model skill (e.g. Lindemer et al. 2010). infiltration and exfiltration through the isting models (0.35 < Brier skill score < The lack of model validation for these seafloor, sediment dynamics during and 0.74, McCall et al. 2010, Harter and Figlus storm processes is, in large part, because after dune slumping, Aeolian transport 2017, Overbeck et al. 2017). Empirical there are few observations of the near- (including effects of rain, groundwater, and semi-empirical models for dune shore (surf, beach, dune) bathymetry and vegetation), and overland flows need erosion are efficient and have reasonable and hydrodynamics before, during, and to be improved. skill representing observations (Stockdon after extreme storms, and thus the im- Integrating relevant physics into nu- et al. 2007, Long et al. 2014, Overbeck et portance of alongshore-variable surf zone merical models, coupling models, and al. 2017), but some methods forecast the bathymetry and waves and water levels in providing a feedback mechanism between type of beach response (e.g. Sallenger the geomorphological feedback process models is another challenging area of re- 2000, Stockdon et al. 2007) instead of is uncertain. search. Existing modeling systems have quantifying coastal change. In addition, Interactions between tidal, wind- limitations, both for long-term coastal empirical and semi-empirical methods driven, and wave-driven currents may management and for real-time forecast- cannot adequately represent the dynamic amplify forces on the beach and increase ing. For long-term management via the feedback between processes that occurs transport of sediment (Mulligan et al. development of flood risk maps by FEMA during storms. For instance, higher total 2008). Exchange of sediments between and its contractors, erosion is considered water levels may not always result in high- the subaerial beach and surf zone, and after the larger-scale waves and flooding er magnitudes of dune elevation change between the shoreline and inner shelf have been predicted. A coastal engineer

Shore & Beach  Vol. 87, No. 1  Winter 2019 Page 39 uses a one-dimensional beach transect erosion from storm surge and waves by al. 2017). Vegetation makes dunes serve to manually update the primary frontal acting as surface cover thereby creating as both habitat and flood protection fea- dune as necessary (Bellomo et al. 1999, drag, reducing flow velocities, overtop- tures, both functions of which provide FEMA 2003). The updated topography is ping, and overwash (Tanaka et al. 2009, invaluable ecosystem services as a result then coupled back to the overland waves Kobayashi et al. 2013, Anderson et al. of being able to withstand storm forces. and flooding, but in a semi-automated 2011, 2013). Storms have the potential to Many studies have focused on the way through the use of one-dimensional move large amounts of sediment causing beneficial reduction of storm waves and models for wave action landward of the erosion in some areas and deposition in surge that vegetation provide (Costanza et dune (FEMA 1989, Divoky 2007, FEMA others (Leatherman et al. 1977) creating al. 2008, Narayan et al. 2017) and the ef- 2007). For real-time forecasting, the a heterogenous dune environment vary- fect that such storms impose on marshes larger-scale wave and flooding models ing in morphological parameters such (Chabreck and Palmisano 1973, Stumpf (e.g. SLOSH or ADCIRC) are used with as height and slope (Houser et al. 2008). 1983, Nyman et al. 1995). However, quan- a fixed ground surface. The wave and Foredune stabilizing plants are capable tification of the feedbacks between waves water-level predictions can then be used of growing out of this deposition and and surge with sediment and vegetation to predict the erosion regime, such as surviving with as much as 1 m of burial is needed to fully forecast storm impacts overwash or inundation, along the coast (Maun 1994). and plan mitigation strategies. A broad (e.g. the USGS Coastal Change Hazards s previously mentioned, the approach including observational stud- Portal). None of these systems link the height of the dune relative to the ies, physical modeling, and numerical updated topography back to the larger- storm surge will affect the type modeling is required to better quantify scale waves and flooding. Aof impact that results, with dunes higher these feedbacks and, ultimately, the risk However, as topography erodes, waves than the surge experiencing swash or associated with storm impacts. More ob- and surge can progress further into collision and lower dunes experiencing servations of wave and surge interactions regions that were protected previously. overwash and inundation (Sallenger with mixed sediments (e.g. Roberts et al. For storms impacting Long Island, New 2000). The foredune is a defense line 2013) are necessary to improve our abil- York, overwash and breaching increased whereby any break in the crest will result ity to make risk-informed decisions. The the water levels in the back bay by as in water flowing into and potentially non-uniform nature of vegetated shore- much as 1 m (Canizares and Irish 2008). through the entire system. The initial line response requires an understanding Degradation of the Chandeleur Islands in dune profile morphology affects uprush of variation in vegetation structure, Louisiana can increase the wave heights and overtopping potential (Silva et al. cohesive sediment dynamics processes, in coastal marshes by 1 m to 4 m and the 2016, Figlus et al. 2011). In general, erosive properties of vegetation, and mass surge levels near New Orleans by 0.5 m vegetation affects dune morphology by failing (Priestas and Fagherazzi 2011). As (Wamsley et al. 2009). Erosion on Bolivar its ability to trap sediment and stabi- green engineering solutions are increas- Peninsula could increase the surge volume lize the dune (Hesp 1989, Murray et al. ingly applied in coastal protection, how in Galveston Bay by 50% to 60% (Rego 2008), which creates differences in ero- nature-based features such as dunes and and Li 2010). However, these studies were sion potential because plant response wetlands affect waves and surge need to conducted with flood models using fixed to wave action and overwash is species- be quantified (Gedan et al. 2011). Storms ground elevations. Even when erosion dependent (Charbonneau et al. 2017). can be important events for import of was simulated in a separate, process-based A more complete understanding of the sediment to coastal marshes (e.g. Cahoon model (e.g. XBeach), the dune crests were effects of waves and surge on sediment et al. 1995; Moskalski and Sommerfield lowered in the larger-scale circulation and vegetation requires a variety of data 2013) so the effects of such features on model before its simulation (Canizares types (Moore 2000, Delgado-Fernandez storm-induced sedimentation is equally and Irish 2008). Thus, these studies are et al. 2009), such as observational studies important for long-term coastal protec- overly conservative, in both magnitude and modeling (Larson et al. 2004, Mull tion. Pre-event data is requisite for deter- and timing of the altered surge. There is and Ruggiero 2014), both physical and mining the baseline in order to quantify a critical need to invest in model coupling numerical (Larson et al. 2004) to better storm effects. Remote sensing can provide to integrate the processes of storm-driven determine morphologic change and ul- necessary data on elevation and plant hydro- and morphodynamics over mul- timately categorize risk associated with species and density but may need to be tiple scales, and to investigate questions storm impacts. Pre-event data, which is ground-truthed to determine specific about how the erosion of beaches and often lacking, is required to establish a characteristics (Klemas 2013). Ultimately, dunes can affect the magnitude, extent, baseline allowing for the determination the valuation of storm risk reduction and timing of flooding. of how and why local and broad topo- from natural and nature-based features graphic changes occurred (Charbonneau (e.g. Narayan et al. 2017) is needed to Including the effects of vegetation et al. 2017). Remote sensing can provide justify future investment. and sedimentation on beach and dune necessary data on dune elevation, plant response and recovery is another portion Incorporating rainfall runoff into the species, and plant density (Timm et al. of the storm processes feedback loop. hydrodynamic and morphodynamic 2014, Brodie and Spore 2015), but these Vegetation causes wave attenuation, thus modeling dynamic feedback loop is an- data should be supplemented with physi- exposing the landmass to lower wave other need highlighted by the 2017 and cal sampling to determine more specific forces and thereby reducing sediment 2018 hurricane seasons. Flooding from details on the dune vegetation condition loss to erosion on vegetated dunes during rainfall-runoff in the riverine and coastal and verify mapping results (Simmons et storms. In addition, vegetation reduces areas during tropical cyclones can be as

Page 40 Shore & Beach  Vol. 87, No. 1  Winter 2019 devastating as the effects of storm surge frequency of data exchanges; 3) methods quire the representation of a continuous and/or wind, depending on the location, to implement structures in the riverine domain (i.e. deep ocean, nearshore, and drainage system, and intensity of the rain- areas (e.g. dams, bridges) into the model coastal floodplain) as discrete elements fall event. Flooding from both precipita- domain; 4) methods to address urban ar- (Westerink et al. 1994, Hagen et al. 2004, tion and storm surge is an increasing risk eas with these coupled modeling system, Westerink et al. 2008). The level of spatial for most of the coastal areas within the such as inclusion of a hydraulic model and discretization is directly related to model U.S. (Wahl et al. 2015). During the 2017 fluid-structure interactions; 5) lateral in- resolution. Coarse resolution results in and 2018 hurricane seasons, Hurricanes flows from the upland areas to the coastal lower computational cost and allows Harvey, Irma, and Florence produced sig- zones. Lastly, uncertainty quantification for probabilistic guidance from suites of nificant flooding due to precipitation in and sensitivity analysis need to be con- simulations, but sacrifices the model’s both the riverine and coastal areas, along sidered for different forecast inputs (e.g. accuracy in both its representation of with flooding caused by storm surge. As initial river stages, precipitation rates, the natural landscape (i.e. topography, mentioned in Blake and Zelinsky (2018), storm track, storm size, and intensity) to bathymetry, and surface roughness) and Hurricane Harvey produced the most formulate ensemble forecast. its mathematical solution. Finer model precipitation of any tropical cyclone resolution will result in higher compu- IMPROVING METEOROLOGIC within the recorded rainfall records of tational cost, but it allows the model to AND SURGE FORECASTING the United States. The rainfall associated better represent the natural landscape METHODS with this storm caused flooding within and yields more detailed and accurate During the last few decades, com- the city of Houston, TX, and regions of hydrodynamic calculations, especially putational models to forecast hurricane southeastern Texas, and in many cases in the nearshore and across the coastal winds and storm surge have improved exceeded the 100-year rainfall event for floodplain (Hagen et al. 2006, Bilskie and significantly and led to reduced uncer- these areas. Furthermore, Hurricane Irma Hagen 2013, Bilskie et al. 2015). tainty. However, there are remaining caused flooding in the Jacksonville, FL, challenges for current technologies Operational storm surge modeling area from the combined effect of rainfall- when accurate and rapid predictions systems have been developed with differ- runoff and storm surge (Cangialosi et al. are needed by emergency management ent approaches for resolution and accu- 2018). In the Jacksonville area, precipita- professionals. The top research need, as racy, and thus with different implications tion from the storm lead to peak riverine identified by the workshop participants, for end users. The Tropical Cyclone Storm flows occurring at the same time as peak was to “quantify spatial and temporal Surge Probabilities (P-Surge, http://slosh. storm surge in the St. Johns River, which resolution and accuracy for end users,” nws.noaa.gov/psurge2.0), a SLOSH- lead to historic flooding. Other tropical which motivates research in two key ar- based (Jelesnianski et al. 1992) product cyclones (e.g. Tropical Storm Allison, eas. First, storm surge models and other developed by the National Weather Hurricane Floyd) have also been associ- forecasting technologies can provide Service (NWS) in collaboration with the ated with flooding due to rainfall-runoff. rapid predictions with higher uncertainty NHC, is an example of a technology that Thus, it is evident that to capture these or time-consuming predictions with yields rapid but highly uncertain storm combined effects it is necessary to have higher accuracy. The National Hurricane surge predictions. SLOSH must be fast, a predictive dynamic modeling capabil- Center (NHC) uses the former approach, and so it is used with coarse represen- ity. To capture both the upland riverine by providing probabilistic guidance based tation of storm processes and coastal flows caused by the rainfall-runoff along on hundreds of relatively-coarse simula- geography, and then hundreds of SLOSH with the coastal flooding, several coupled tions, whereas academic researchers have simulations are combined statistically to modeling systems have been developed or pursued the latter approach, by providing provide a probabilistic forecast for each are under development to either utilize deterministic guidance based on a small advisory. P-Surge provides 10% to 50% a hydrologic model (e.g. Tromble et al. number of relatively-detailed simulations. exceedance peak storm surge elevations 2011, Dresback et al. 2013), hydrologic There is a need to close this gap between for preparedness and general emergency and hydraulic model (e.g. Ray et al. 2011, probabilistic and deterministic predic- management decision-making. The Christian et al. 2015, Torres et al. 2015) tions. Second, these models rely on input widespread of the 10% to 50% exceedance or employ USGS gauge information data that are highly uncertain; storm fore- interval, often encompassing 10 or more (e.g. Flowerdew et al. 2010, Warner et al. casts may have errors in parameters like feet, and lack of storm surge hydrograph 2010). In the case of coupling with the track, size, and intensity, while flooding estimation are of limited usefulness for USGS gauge information, a real-time models may have errors in bathymetric the operation of flood gates, pumps, and forecast can only extend outward on the and topographic elevations, surface and other components of flood-protection order of hours due to the lack of future bottom roughness, and other processes. systems. flows from the gauge information. There is a need to account for and com- The use of high-resolution, high- Modeling the compound flooding municate these uncertainties to end users. fidelity models is necessary when more from rainfall-runoff and storm surge, To balance timeliness and accuracy accurate storm surge predictions are requires several research areas to be in forecasts of storm-driven hazards, re- required. The ADvanced CIRCulation explored and pursued, include determin- searchers need to close the gap between (ADCIRC) model can be implemented ing: 1) the boundary location to couple a high-resolution, deterministic guid- as the ADCIRC Surge Guidance System hydrologic model with a coastal hydrody- ance and large-ensemble, probabilistic (ASGS) in high-performing computing namic model; 2) the coupling (one-way guidance. Models for the prediction of environments to perform individual or two-way) between models and the coastal flooding and related impacts re- simulations and provide high-fidelity

Shore & Beach  Vol. 87, No. 1  Winter 2019 Page 41 predictions. Individual forecast simu- with weak eye structure, establishing the assimilation methods, or Kalman filters, lations can take from a half hour to a position of the vortex can be a challenge. operate in two main steps, a forecast step few hours to complete, depending on Of particular concern, however, are the and an analysis step. In the forecast step, the capabilities of the computing en- uncertainties in wind radii, which are an initial estimate of the model state (e.g. vironment. Alternatively, surrogate or often the basis for the estimation of Rmax. modeled water elevations) is forecasted meta-models (Jia et al. 2016; Taflanidis Recent improvements in the estimation to some later time using the numerical et al. 2017, Nadal-Caraballo et al. 2018) of Rmax include the development of asym- forecast model (e.g. a storm surge model). can be trained on existing high-fidelity metric vortex models. When using these In the analysis step, observed data (i.e. ob-

ADCIRC databases, and subsequently vortex models, Rmax can be estimated by served water elevations) is used to update executed to predict storm surge in a few quadrant typically by fitting a velocity the forecasted model state. The model seconds while conserving the fidelity of profile model to match the velocity and state generally does not lie in the same the underlying ADCIRC database. These distance of the highest isotach, as estimat- space as the observed data (e.g. data may meta-models can also be used to simulate ed by NHC forecasters. Other approaches be observed at a subset of the locations ensembles of thousands of hurricanes in include estimating Rmax by averaging the at which it is modeled). Thus, the model probabilistic frameworks and account wind profiles, and thus Rmax, associated state is first projected into the observation for the uncertainties in meteorological with all available isotachs. Moreover, space through an observation operator. forcing. Rmaxestimated by different reanalysis ef- The residual between the observed data forts have often produced inconsistent and the projection of the forecasted state To reduce and communicate uncer- and conflictive results, with estimates off is then computed, weighted, and added to tainties in both model inputs and results, by up to a factor of two (Levinson et al. the forecasted state to form the updated researchers need to better quantify these 2010). Given the sensitivity of storm surge state, or the analysis. The weight is known uncertainties and how they propagate predictions to meteorological forcing, as the Kalman gain, which is defined in a through the forecast models. Workshop improvements to the methods for estima- way that minimizes the uncertainty of the participants identified a key research tion of hurricane forcing parameters are update. The Kalman gain is also used to need related to “inputs for accurate warranted. It is expected for forecasters produce an estimate of the error variance coastal hazard forecasts in probabilistic to progressively shrink the uncertainty of the update from that of the forecast. models.” Storm surge forecasts are depen- bands associated with hurricane intensity, dent on meteorological forcing, and are These methods allow data to be as- position and projected trajectory, but the thus subject to the inherent uncertainties similated sequentially, as data become estimation of parameters that not are in hurricane forecasts. Regardless of the available, making them particularly directly measured or observed, such as approach (i.e. deterministic or probabi- advantageous for real-time forecasting. R , requires further investigation. listic), storm surge model simulations max This is a direct approach to reducing un- rely on cyclone vortex models that are hese uncertainties in atmospheric certainty in modeled data, as simulated constructed with a limited set of hur- forecasts must then be translated storm surge heights are explicitly adjusted ricane parameters, such as landfalling or into uncertainties in storm surge toward observed values in an optimal bypassing location (x ), central pressure Tforecasts, leading to research needs in way. The methods also produce an es- 0 deficit ∆( ρ), radius of maximum winds “assessing available statistical methods timate of uncertainty in the improved (Rmax), translational speed (Vf), and head- to best use limited model results” and storm surge forecast. In meteorological ing direction (θ). Torn and Snyder (2012) “learning from model error to improve forecasting, pressure data are often used and Landsea and Franklin (2013) esti- process understanding and models.” In to improve weather models of precipita- mated the uncertainties for wind speed, recent years, research efforts have at- tion (Kalnay 2003). The same methods central pressure, position, and gale, storm tempted to better resolve uncertain physi- can be implemented for storm surge and hurricane wind radii. The uncertainty cal processes that contribute to errors in forecasting. Measurements of water eleva- associated with the cyclone position, for storm surge forecasts (e.g. surface wind tions and currents can be used to better example, can be expected to range from 9 field representation, bottom friction, estimate storm surge model output, i.e. km to 65 km depending on the intensity wind-driven waves, and rainfall). How- water elevations and currents, as well as and organization of wind circulation. ever, understanding and incorporating their uncertainties (Butler et al. 2012). Wind speed uncertainty can vary from 9 these physical processes into storm surge Additionally, they can be used to estimate km to 34 km/h; likewise, central pressure modeling remains a significant challenge. uncertain model parameters, such as bot- uncertainty can vary and from 1.5 hPa It is necessary that all of these uncertain- tom friction (Mayo et al. 2014). to 15 hPa. The gale (63 km/h), storm (93 ties (i.e. those resulting from uncertain Ongoing research efforts should focus km/h) and hurricane (119 km/h) wind physical processes and meteorological on closing the gap between probabilistic radii uncertainties range from to 9 km forcing) as well as their impacts be quan- and deterministic guidance, but with to 111 km. tified and accounted for effective storm emphases on meeting needs of and on surge forecasting. Meteorologists and other scientists communicating uncertainties to end are improving forecasting models and Data assimilation methods offer an users. Meta-model databases will be reducing uncertainty through ensembles approach to this type of consideration expanded to include simulations of and data assimilation techniques. The and quantification of uncertainty. The storms with new combinations of pa- NHC forecast track errors have trended methods combine observed data with rameters and to improve representation down during the past 40 or so years (NHC numerical model output to improve the of the coastal zone with new geospatial 2017). For unorganized tropical cyclones accuracy of modeled data. Statistical data data and higher model resolution. The

Page 42 Shore & Beach  Vol. 87, No. 1  Winter 2019 development of new numerical algo- to degrade the structural integrity of sea dunes can be eroded on the seaward face rithms and implementation of the latest cliffs (Trenhaile 1987; Sunamura 1992), causing it to narrow, or the dune may be computing technologies will reduce cost ultimately leading to sudden landslides overwashed and inundated when water and wall-clock time of high-resolution that threaten cliff-top infrastructure. In levels rise above the maximum dune models (Dietrich et al. 2011, Dietrich many places, cliffs erode primarily during height (Sallenger 2000, Long et al. 2014). et al. 2013), thus leading to faster and storm events, when elevated water levels Under these conditions, the elevation of more reliable predictions. Instead of ever- (i.e. surge and set-up) allow large waves the dune can be reduced significantly or increasing model resolution, sub-mesh- to impact the cliff with high frequency breached completely and sand can be scale processes can be parameterized in and intensity (e.g. Earlie et al. 2015). For transported landward and away from the coarse-resolution models, and geospatial example, sea cliffs near San Francisco, active beach system. As a result, when techniques can be used to downscale California retreated by up to 14 m dur- dunes become lower or narrower, the and extrapolate the flooding guidance ing the high-energy 1997-1998 El Niño vulnerability of that section of coastline to critical infrastructure. Increasing data winter season (Sallenger et al. 2002). to the next storm increases. The time scale availability will allow for increased use of Although storm-driven coastal cliff ero- over which dunes re-grow (O(decades)) data assimilation in real-time storm surge sion occurs globally and often accounts is much longer than the storm time scales predictions. Research should continue to for the majority of observed cliff erosion, (O(hours)) that cause them to lower and improve both accuracy and efficiency of the process can be particularly difficult can be related to the frequency of storms model forecasts. to measure in-situ during extreme storm that impact a particular region (Houser wave conditions — limiting our overall and Hamilton 2009, Houser et al. 2015). ASSESSING knowledge of cliff evolution. HAZARD MITIGATION revious studies have used pre- and Coastal communities are continually Recent advances in low-cost remote post-storm observations of dunes challenged to adapt to the impact caused sensing technology, including Structure- to characterize and quantify how by storms and to prepare for future from-Motion (SfM) topographic data Psand dunes respond to storm forcing impacts caused by sea level change and collection (Ruzic et al. 2013; Warrick et (e.g. McCall et al. 2010, Lindemer et al. increased storm activity. Essential to al. 2016) and time lapse surf zone pho- 2010, Sherwood et al. 2014). Using these this adaptation is the understanding of tography (e.g., Holman et al. 2013), as observations, process-based (Roelvink et the hydrodynamic forces impacting the well as the growing use of seismometers al. 2009, Harter and Figlus 2017), statisti- beach, coastal dune system, and engi- to record the geomorphic response of cal (Plant and Stockton 2012, Wahl et al. neered structures, improving models of cliffs to wave impacts in different envi- 2016), and empirical models (Long et al. storm processes and coastal response, and ronments (Adams et al. 2005; Dickson 2014) of dune evolution have been devel- ultimately refining our ability to forecast and Pentney 2012; Norman et al. 2013; oped and tested. While these models have impacts to future events. Along the U.S. Earlie et al. 2015; Young et al. 2016), are been shown to be skillful at reproducing coastline, beaches, dunes, cliffs, wetlands, expanding our understanding of coastal dune evolution for individual storms and and, often, engineered structures provide cliff erosion. But our understanding specific locations, they have not been fully a primary line of protection to landward could be expanded further by systematic evaluated for a wide range of conditions. infrastructure, habitat, and populations pre-storm “rapid response” deployments In addition, some models that are capable from elevated water levels caused by – involving cost and instrument sharing of predicting part of the storm-induced extreme tides, surge and wave runup. between coastal scientists – to directly dune response (e.g. dune erosion; Larson Understanding the variability of the pro- monitor storm impacts on cliffs. Addi- et al. 2004, Palmsten and Holman 2012) cesses and response of different types of tional research needs to improve forecasts are not able to capture the full range of shorelines, both natural and engineered, of storm impacts to cliffs include, but are processes, such as collision, overwash, will better inform decisions for addressing not limited to: (1) modeling and quantify- and inundation. Due to the difficulty of risk on the coast and mitigating hazards. ing wave transformation across shore in measuring coastal processes and change different morphological settings to better during a storm event, only a few studies, Large stretches of the U.S. Pacific and understand the energy impacting cliffs; typically based on laboratory observa- Great Lakes coastlines are backed by (2) relationship between rock strength tions, have assessed the accuracy of these cliffs and bluffs. Coastal cliff erosion and and cliff behavior, such as microseismic models in predicting time-dependent progressive landward retreat is caused fatigue, rapid retreat, sudden failure, dune evolution (Figlus et al. 2011, McCall by a complex and unique combination over a range of time scales; and (3) de- et al. 2010, Palmsten and Splinter 2016). of subaerial and marine factors. These velopment of automated approaches for Novel observations of the oceanographic include direct and indirect wave impacts observing cliff position and retreat from forcing, including wave run-up, and dune (Sunamura 1977; Emery and Kuhn 1982), high resolution imagery. response during storms in a field setting rainfall and groundwater percolation are needed to increase understand- (Young et al. 2009; Young 2015), per- Even along the highly studied dune- ing about how dunes erode (slumping, mafrost thaw in Arctic environments, backed shorelines, models are not able saturation [Palmsten and Holman 2011], especially near remote native communi- to capture spatial variability in storm overwash, etc.), while longer-term studies ties (Barnhart et al. 2014; Erikson et al. response or time evolution of the dune are needed to better quantify recovery 2015), biochemical and biophysical rock structure accurately or quickly enough processes and associated time scales. erosion (Coombes et al. 2017), and, oc- to meaningfully inform decisions about With these data, models of dune erosion casionally, seismic shaking (Hapke and mitigation. When water levels during and recovery can be combined, enhanc- Richmond 2002). These processes act storms reach these protective features,

Shore & Beach  Vol. 87, No. 1  Winter 2019 Page 43 ing their power and usefulness, providing fect dune response to storm, as recently (2016) and FEMA (2011) do not explicitly stakeholders with tools and actionable observed during Hurricane Sandy where include wave height in equations predict- information to quantify vulnerability to erosion varied as a function of the species ing the horizontal breaking wave-induced future storm events as well as to evaluate dominating foredunes (Charbonneau et force, and guidance is limited for esti- viable restoration options of dune features al. 2016). Plants in coastal dune habitats mating the vertical wave uplift forces on (e.g. Mickey et al. 2017). can thus be considered ecosystem engi- elevated residential near-coast structures. neers as they initiate, build, and stabilize As mentioned in the dynamics feed- Some experimental work has mea- dunes. The biophysical feedback loop back section above, nature-based fea- sured these horizontal and vertical forces between vegetation and dune, or ecology tures, such as sand dunes, have been on structures in idealized settings. For and geology, has only recently begun to identified as key elements on resilient example, Bea et al. (1999) proposed a be explored (Stallins 2006; Murray et al. coasts. A more complete understanding method for estimating the total force on 2008). Understanding the role of vegeta- of the processes that control their growth a bridge deck, and Bradner et al. (2011) tion for dune building and stabilization is and stability is necessary for evaluating similarly measured wave-induced forces important for understanding topographic options for use in mitigating future storm and pressures on a large-scale model of heterogeneity and storm response which impacts. A major control on foredune a bridge superstructure. More recently, feedback on beach management. growth and stability is vegetation (Bryant in a large-scale wave flume experiment et al. 2017). Species-specific morphology, here beaches and dunes have examining forces on an idealized elevated defined by form and structural charac- not been able to recover natu- coastal structure, Park et al. (2017) found teristics, has been shown to affect dune rally, engineered features are that the vertical force was on the same system geomorphology in a biophysi- sometimesW constructed to protect against order of magnitude as the horizontal force cal feedback loop (Murray et al. 2008; future storm events. To determine struc- for some elevation-wave height combina- Zarnetske et al. 2010; Duran and Moore tural response to storms, multiple pro- tions, reinforcing the need for additional 2013; Fei et al. 2014). Plant shoots catch cesses (hazards) associated with coastal research on vertical wave forces on coastal aeolian sand but the efficiency of capture flooding — such as storm surge flood- structures. Additional work is required varies depending on species morphology, ing, wave action, flow velocity — must to accurately characterize a structure’s establishment, survival ability, and den- be integrated. A current challenge is to response to wave impacts and to iden- sity (Hesp 1989; Zarnetske et al. 2010). understand the environmental load asso- tify scaling effects from laboratory to Differences in plant community struc- ciated with storms and the impacts these prototype structures and conditions. To ture, transport potential, and sediment loads have on coastal structures. Previ- bridge this knowledge gap, datasets pro- supply along a coast (Houser and Mathew ous work has made use of observational, viding full-scale measurements of wave- 2011) create heterogeneous environments theoretical, physical, and computational induced forces and structural response that lead to variability in the topography models to investigate relationships be- in a realistic environment are required, of the dune system (Hilton et al. 2006; tween storm characteristics (storm surge as are careful laboratory investigations Hacker et al. 2011). Shoots create drag elevations, significant wave heights, wind of scaling effects and more generalized and surface cover, thereby creating a bio- velocities), structural characteristics (el- wave conditions. These data may inform shield that retards wind and wave erosion evation of lowest horizontal member, date computational fluid dynamics models, and increases dune stability (Tanaka et al. of construction, building archetype) and which can, in turn, inform community 2009). The above-ground biomass also damage in coastal communities. Fragil- stakeholders about potential structural serves to disrupt run-up, thus reducing ity models, which predict the likelihood vulnerabilities and mitigation alterna- erosion (Bryant et al. 2018). of exceeding a given damage threshold tives. Community robustness as a whole, based on a hazard intensity measure, have depends on the impact on individual Plant roots stabilize otherwise un- made strides in evaluating storm impacts features, both natural and engineered stable sand particles (Maun 2009) by de- on structures. For example, Kennedy et coastal features, as well as successive and veloping complicated root networks that al. (2011) assessed building damage on integrated impacts. A remaining research bind sand, providing structural integrity the Bolivar Peninsula, TX, after Hurri- challenge for adequately understanding (Forster and Nicolson 1981; Tanaka et cane Ike (2008); using these data and an coastal response and mitigation will be al. 2009; Sigren et al. 2014). Laboratory ADCIRC+SWAN hindcast of the storm, to link these processes and develop the tests have shown that the belowground Tomiczek et al. (2014) derived empirical timeline of impacts and damage. biomass (plant root network) provides fragility models predicting the probabil- sediment stability and thus, reduces ero- INFRASTRUCTURE NEEDS ity of elevated, wood-framed structures sion (Bryant et al. 2018). Many foredune The prior sections identified obser- on the peninsula experiencing collapse plants invest in symbiotic relationships vational and modeling needs to better limit state failure. Environmental vari- with fungi that produce belowground represent dynamic feedbacks, improve ables (e.g. significant wave height, storm hyphal networks that both directly and forecasting methods to better account surge elevation) showed more skill in indirectly reduce soil erosion (Mardhiah for uncertainties, and assess the hazard predicting failure than standards-based et al. 2016). Similarly, microorganisms, mitigation capabilities of natural, nature- force computations, suggesting that de- like bacteria, living on roots increase based, and built features. Here, we elabo- sign equations must be refined to better sand aggregation thereby positively af- rate on needed observations, modeling ad- represent physical processes associated fecting stability (Forster and Nicolson vancements, and community efforts. Table with wave impacts. Design guidance from 1981). Plants vary in their investment in 3 summarizes the infrastructure needs the American Society of Civil Engineers roots versus shoots, which will then af- prioritized by workshop participants.

Page 44 Shore & Beach  Vol. 87, No. 1  Winter 2019 Better understanding storm processes Table 3. and impacts will require improved ob- Prioritized infrastructure needs servations of physical processes and re- 1 — Community of Practice: Communication; sharing instruments; sharing sponses during storm events. Researchers need real-time data before, during, and of model results, data, and knowledge after storm events. These data may be col- 2 — Focused, comprehensive process-based during storm observations lected with new arrays of instrumentation (rapid response) that have greater spatial resolution than is 3 — Data: presently available or with high spatial- Available data sets in common language, central repository and temporal-resolution remote sensing Real time data with adequate spatial coverage data. This will help update the initial Updated initial conditions (topo, bathy, waves) physical conditions, such as bathymetry, Observations during storms topography, and nearshore oceanographic High resolution remote sensing data conditions for more accurate model input. 4 — Instrumentation of built environment for rapid response (smart struc- It may also lead to modeling advance- tures) ments, for example predicting rapid 5 — Integration, support, & coordination for field studies morphologic change during storm events. 6 — Community involvement (citizen science) Collecting accurate source data, such 7 — Reliable fast predictions? New algorithms? Models? HPC? as georeferenced aerial photographs, 8 — Novel instrumentation platforms for better observations of sediment & satellite imagery, and lidar surveys, and morpho-dynamics subsequently digitizing topographic 9 — Pre-event funding pool features is a time-intensive process, es- pecially for long stretches of coastline. environment are required, as are careful provide a community of practice to not Recent research has focused on develop- laboratory investigations of scaling effects only serve data, but to also address a ing automated feature position extrac- and more generalized wave conditions. number of other infrastructure needs. tion routines from topographic Lidar These data may inform computational Virtual and in-person meetings would data (Palaseanu-Lovejoy et al. 2016), but fluid dynamics models, which can, in provide regular interactions of like- such routines do not exist for imagery. turn, inform community stakeholders minded researchers to share open source For beaches, Luijendijk et al. (2018) about potential structural vulnerabilities models, model results, data, instrumen- used automated shoreline detection and and mitigation alternatives. Smart struc- tation, challenges, best practices, lessons leveraged the wealth of publicly available, tures that can support instrumentation learned, and knowledge. high-temporal-resolution satellite data and the associated power and data storage COMMUNICATION CHALLENGES on the Google Earth Engine to calculate requirements to measure storm impacts The discussion above focuses on sci- global rates of shoreline change. Ideally, on the built environment are required. Fi- entific and engineering challenges that a similar approach could be developed nally, the research community is presently will be addressed by coastal researchers, for cliffs, allowing easy (and centralized) lacking the integration, coordination, and ideally in partnership with national and access to short-term (weekly) time series support of multidisciplinary field studies regional government forecasters. The of cliff positions, which can then be used related to this type of work. The field stud- following discussion focuses on commu- to monitor and predict storm impacts. ies should include stakeholder engage- nicating this research to local forecasters ment and citizen science opportunities. Improving forecasts of sediment and emergency managers. Table 4 sum- transport and morphological changes Once datasets of before, during, and marizes the communication challenges during storms requires simultaneous after storm events have been collected, prioritized by workshop participants. observations of nearshore hydrodynam- researchers would benefit from a mecha- For purposes of this paper, the role of ics, sediment fluxes, and bathymetric and nism to host a central repository that direct communication with the public is topographic changes and improved mod- makes the data available in a common assumed to be filled by the local emer- els for coupled hydrodynamic, meteoro- language. Such a mechanism (an online gency and coastal managers. They are logic, and sediment processes, including platform or other virtual hub) would predictions of coastal inundation. Due to the difficulty of measuring coastal pro- Table 4. cesses and change during a storm event, Prioritized communication challenges novel observations of the oceanographic 1 — Communicate and manage uncertainty (temporal/spatial variability) forcing, including wave runup, and dune 2 — Translate numerical output to relatable user products response during storms in a field setting 3 — Educate (surge, impacts, post-storm conditions) are needed to increase understanding 4 — Communicate economic benefits of mitigation about how coastal features evolve. 5 — Engage social scientists In addition to morphologic obser- 6 — Tailor messaging vations, datasets providing full-scale 7 — Share best practices/data (see Infrastructure: Community of Practice) measurements of wave-induced forces 8 — Consolidate and unify tools and structural response in a realistic 9 — Help underserved communities access and deliver information

Shore & Beach  Vol. 87, No. 1  Winter 2019 Page 45 typically employed by a coastal county In order for managers to adopt new mod- www.fema.gov/media-library/assets/ or municipality and may not have been els and tools, they need consolidated, uni- videos/153336), which describes how trained in coastal science or engineering, fied tools and user products that translate improved building codes have helped to but have the experience with, and ac- complex numerical model results. reduce storm impacts on coastal homes cess to, successful messaging techniques over time, resulting in more resilient esearchers aim to offer higher tailored to their local community. Local coastal communities. resolution models with less uncer- managers are best positioned to increase tainty. Local area forecasters are SUMMARY public risk salience through local, tailored Rchallenged to provide forecast guidance Storm processes and impacts research messaging strategies to reduce the psy- and descriptions of impacts that are com- needs include chological distance to risk (e.g. Lorenzoni prehensive enough to be useful, yet do not et al. 2007; O’Neill and Nicholson-Cole A) Represent the dynamic feedbacks imply greater precision than is justified. 2009). When coastal researchers com- among storm-driven hydro- and mor- Local emergency managers sometimes municate with local managers and/or phodynamics. The research objectives are have to make evacuation decisions based the public, interactions are better with a to: 1) better understand sediment trans- on high forecast uncertainty. In these two-way flow of information rather than port to improve our ability to simulate cases, managers need to understand a one-way flow from scientist to stake- morphologic change, 2) integrate relevant the worst-case scenario to manage risk. holder (NASEM 2018). physics into numerical models, couple the Thus, the challenge may not in reducing models, and provide a feedback mecha- Today’s local coastal and emergency uncertainty but in communicating un- nism between models, 3) include the ef- managers are experienced, knowledge- certainty to users. As one manager stated, fects of vegetation and sedimentation on able, and capable of utilizing fairly com- “hurricane evacuation decision making is beach and dune response and recovery, plex model results to make risk-based as much art as it is science, so the more and 4) incorporate rainfall runoff into decisions. However, during an emergency practitioners understand the tools and the hydrodynamic and morphodynamic event, they are extremely limited by time. the process, the better the art.” modeling capabilities. Quantification of Unlike weather models, which peri- It is important for local emergency the feedback between hydrologic, hydrau- odically update using the latest observa- managers to understand and have infor- lic, sediment transport, morphologic, and tions, local emergency managers cannot mation to educate residents about surge, vegetation conditions is needed to fully iteratively change or update evacuation impacts, and post-storm conditions. Rap- forecast storm impacts and plan mitiga- orders. Evacuation modifications result in paport (2014) identified storm surge and tion strategies. a loss of public trust and apathy. In addi- rising water to be the deadliest component tion to the changing storm characteristics B) Improve forecasting methods to of hurricanes, with water responsible for and the local geomorphology and urban better account for uncertainties. The as many as 90% of all storm-related deaths setting, managers must navigate multi- research goal is to close the gap between and storm surge responsible for half of layered bureaucratic decision-making high-resolution, deterministic guid- those. To increase public awareness of the processes and sometimes complex social ance and large-ensemble, probabilistic risk from water-related deaths, creative and economic factors within their com- guidance. This research will require the communication methods are being em- munities. development of better descriptions of ployed around the nation. For example, in coastal regions and faster models for The preceding sections have described Pinellas County, FL, property owners re- storm-driven hazards, as well as a better differences in probabilistic and determin- ceive information about their evacuation understanding of uncertainties in model istic models to predict storm impacts. zone on monthly electric bills and on an- inputs and how they affect the model Many tools have been developed for nual property tax statements. Educational results. emergency managers to utilize the output signage has been installed at schools and from these different models; however, parks to help residents visualize potential C) Assess the hazard mitigation ca- managers appear to be overwhelmed by storm surge elevations. pabilities of natural, nature-based, and the number of models, tools, and infor- built features. The research goal is to help It is also important for local build- mation related to coastal risk assessment. coastal communities adapt to the impact ing officials to have information about There may be limitations on the useful- caused by storms and to prepare for future the economic benefits of mitigation to ness of these products to support decision impacts caused by sea level change and provide to their coastal residents. Local making. Managers acknowledge they end increased storm activity. This involves officials can communicate structural up relying on products that are intuitive an understanding of the hydrodynamic resilience/vulnerability to low-income and easy to use, rather than potentially forces impacting the beach, coastal dune and socially vulnerable populations in more sophisticated but complicated tools system, and engineered structures, im- order to improve mitigation decisions (NASEM 2018). Emergency managers proving models of storm processes and and ensure life-safety during extreme tend to use nationally approved tools – for coastal response, and ultimately refining events. An education tool is FEMA’s example, the National Hurricane Pro- our ability to forecast impacts to future “Evolution of Mitigation” video (https:// gram’s HURREVAC and NHC’s SLOSH. events.

Page 46 Shore & Beach  Vol. 87, No. 1  Winter 2019 Infrastructure needs include advance- Meeting participants ments to provide coastal data before, • Peter Adams (University of Florida) • Andrew Kennedy (University of Notre during, and after storms at high spatial • Aileen Aponte (University of Puerto Rico Dame) and temporal resolution. Communica- [UPR] Rio Piedras Campus) • Brian LaMarre (NOAA/National Weather tion challenges include addressing the • Maritza Barreto (UPR Rico Rio Piedras Service, Tampa Bay Area/Ruskin FL) Campus) • Say-Chong Lee (JACOBS) disconnect between the research needs to • Tanya Beck (U.S. Army Corps of • Kelly Legault (USACE SAJ) reduce the uncertainty and increase the Engineers Engineer Research and • Jeff Lillycrop (USACE ERDC) resolution of model output and the emer- Development Center [USACE ERDC]) • Linda Lillycrop (USACE ERDC) gency managers’ need to make evacuation • Matthew Bilskie (Louisiana State • Patrick Limber (USGS) University, Center for Coastal Resiliency) • Christina Lindemer (FEMA) decisions based on an intuitive and easy- • Justin Birchler (U.S. Geological Survey • Yi Liu (Virginia Tech) to-use model with high certainty several [USGS]) • Joseph Long (USGS) days to a week before landfall. • John Bishop (Pinellas County FL Coastal • Bonnie Ludka (Scripps Institution of Management) Oceanography, University of California As discussed in the infrastructure • Sally Bishop (Pinellas County FL San Diego) needs section, the coastal research com- Emergency Management) • Rick Luettich (University of North Carolina • Amber Boulding (St. Petersburg FL Office at Chapel Hill) munity should develop a storm processes of Emergency Management) • Arjen Luijendijk (Deltares & TU Delft) and impacts community of practice to • Brandon Boyd (USACE ERDC Coastal • Victor Malagon Santos (University of foster collaboration and provide a plat- Hydraulics Lab [CHL]) Central Florida) • Leighann Brandt (BOEM) • Jennifer Marlon (Yale University) form to share data, models, instrumenta- • Jenna Brown (USGS) • Chris Massey (USACE ERDC CHL) tion, and knowledge. Emergency manag- • Nahir Cabrera (UPR Rico Rio Piedras • Talea Mayo (University of Central Florida) ers and local administrators should be Campus) • Rangley Mickey (USGS) included in the discussions to inform • Libby Carnahan (Florida Sea Grant • Jon Miller (Stevens Institute of University of Florida/IFAS Extension) Technology) research investments and to help deter- • Q. Jim Chen (Northeastern University) • Jennifer Miselis (USGS-SPCMSC) mine whether research advancements are • Jun Cheng (University of South Florida, • Karen Morgan (USGS) useful to stakeholders. Coastal Research Laboratory) • Ryan Mulligan (Queens University) • Mary Cialone (USACE ERDC) • Muthu Narayanswamy (Michael Baker • Willie Colon (UPR Rico Rio Piedras International) Campus) • Timothy Nelson (USGS) • Drew Condon (USACE, Jacksonville • Maitane Olabarrieta (University of Florida) District) • Jacquelyn Overbeck (Alaska Division of • Matthew Conlin (University of Florida) Geological & Geophysical Surveys) • Dan Cox (Oregon State University) • Meg Palmsten (U.S. Naval Research Lab) • Donald Cresitello (USACE) • Davina Passeri (USGS) • Jane Darby (Mayor, Town of Edisto Beach • Kevian Perez-Valentin (UPR Rico Rio SC) Piedras) • Elizabeth Diaz (UPR) • Jenna Phillips (Taylor Engineering) • Casey Dietrich (North Carolina State • Nathaniel Plant (USGS) University) • Jack Puleo (University of Delaware) • Kara Doran (USGS St. Petersburg) • Britt Raubenheimer (WHOI) • Steve Elgar (Woods Hole Oceanographic • Don Resio (University of North Florida) Institution [WHOI]) • Christopher Sherwood (USGS) • Nicole Elko (American Shore & Beach • Hilary Stockdon (USGS) Preservation Association [ASBPA]) • Andrew Sussman (Florida Division of • Li Erikson (USGS) Emergency Management) • Jason Fleming (Seahorse Coastal • Steven Traynum (Coastal Science & Consulting) Engineering) • Marshall Flynn (Tampa Bay Regional • Mathieu Vallee (University of South Planning Council) Florida, Coastal Research Laboratory) • Diane Foster (University of New • Thomas Wahl (University of Central Hampshire) Florida) • Grover Fugate (Rhode Island Coastal • Ping Wang (University of South Florida, Resources Management Council) Coastal Research Laboratory) • Kate Gooderham (ASBPA) • Anna Wargula (U.S. Naval Academy) • Angelos Hannides (Coastal Carolina • Jeff Waters (USACE ERDC CHL) University) • Robert Weaver (Florida Institute of • Cheryl Hapke (USGS) Technology) • Tom Herrington (Monmouth University, • Zachary Westfall (University of South Urban Coast Institute) Florida, Coastal Research Laboratory) • Caitlin Hoch (USACE) • Greg Wilson (Oregon State University) • Laurie Hogan (NOAA/National Weather • Jim Wilson (Pinellas County FL Fort Service, Eastern Region HQ) DeSoto Park) • Claire Jeuken (Deltares USA) • Han Byul Woo (University of Florida) • Tori Tomiczek (U.S. Naval Academy) • Jennifer Wozencraft (USACE-ERDC-CHL- • Felix Jose (Florida Gulf Coast University) JALBTCX)

Shore & Beach  Vol. 87, No. 1  Winter 2019 Page 47 REFERENCES within the advanced circulation (ADCIRC) ricane protection.” Ambio, 37, 241-8. Adams, P. N., Storlazzi, C. D., and R. S. Anderson, modeling framework for hurricane storm De Winter, R.C., Gongriep, F., and B.G. Ruessink, 2005. “Nearshore wave‐induced cyclical flex- surge forecasting.” Monthly Weather Review, 2015. “Observations and modeling of along- ing of sea cliffs.” J. Geophysical Res.: Earth 140(7), 2215-2231. shore variability in dune erosion at Egmond Surface, 110(F2). Cahoon D.R., Reed, D.J., Day Jr. J., Steyer, G. D. G., aan Zee, the Netherlands.” Coastal Eng., 99, Anderson, M.E., Smith, J.M., Bryant, D.B., and Boumans, R., Lynch, J. C., McNally ,D., Latif, 167-175. R.G.W. McComas, 2013. “Laboratory studies N., and J.W. Day, 1995. “The influence of Delgado-Fernandez, I., Davidson-Arnott, R., and of wave attenuation through artificial and Hurricane Andrew on sediment distribution J. Ollerhead, 2009. “Application of a remote real vegetation.” ERDC TR-13-11, U.S. Army in Louisiana coastal marshes.” J. Coastal Res., sensing technique to the study of coastal Engineer Research and Development Center, Special Issue, 280-294. dunes.” J. Coastal Res., 255, 1160-1167. Vicksburg, MS. Callaghan, D.P., and D. Wainwright, 2013. “The Dickson, M.E.. and R. Pentney, 2012. “Micro- Anderson, M.E., Smith, J.M., and S.K. McKay, 2011. impact of various methods of wave transfers seismic measurements of cliff motion under “Wave dissipation by vegetation.” CHETN- from deep water to nearshore when deter- wave impact and implications for the devel- I-82, U.S. Army Engineering Research and mining extreme beach erosion.” Coastal Eng., opment of near-horizontal shore platforms.” Development Center, Vicksburg, MS. 74, 50-58. Geomorphology, 151, 27-38. American Society of Civil Engineers (ASCE), Cangialosi, J.P., Latto, A.S., and R. Berg, 2018. Dietrich, J.C., Dawson, C.N., Proft, J.M., Howard, 2016. “Minimum Design Loads for Build- “National Hurricane Center Tropical Cyclone M.T., Wells, G., Fleming, J.G., Luettich Jr., ings and Other Structures.” ASCE/SEI 7-16. Report: Hurricane Irma (AL112017).” Techni- R.A., Westerink, J.J., Cobell, Z., Vitse, H., Washington, DC. cal Report, available at https://www.nhc.noaa. Lander, H., Blanton, B.O., Szpilka, C.M., and Barnhart, K. R., Anderson, R. S., Overeem, I., gov/data/tcr/AL112017_Irma.pdf. J.H. Atkinson, 2013. “Real-time forecasting Wobus, C., Clow, G. D., and F. E.Urban, 2014. Canizares, R., and J.L. Irish, 2008. “Simulation of and visualization of hurricane waves and “Modeling erosion of ice‐rich permafrost storm-induced barrier island morphodynam- storm surge using SWAN+ADCIRC and bluffs along the Alaskan Beaufort Sea coast.” ics and flooding.” Coastal Eng., 55, 1089-1101. FigureGen.” Computational Challenges in the J. Geophysical Res.: Earth Surface, 119(5), Castelle, B., Bujan, S., Ferreira, S., and G. Dodet, Geosciences. C. Dawson and M. Gerritsen, 1155-1179. 2017. “Foredune morphological changes and Springer New York. 156: 49-70. Bea, R.G., Xu, T., Stear, J., and R. Ramos, 1999. beach recovery from the extreme 2013/2014 Dietrich, J.C., Zijlema, M., Westerink, J.J., Holthui- “Wave forces on decks of offshore platforms.” winter at a high-energy sandy coast,” Marine jsen, L.H., Dawson, C., Luettich Jr., R.A., Jen- J. Waterway, Port, Coastal, and Ocean Eng., Geology, 385, 41-55. sen, R.E., Smith, J.M., Stelling, G.S., and G.W. 125, 136-144. Chabreck, R., and A. Palmisano, 1973. “The ef- Stone, 2011. “Modeling hurricane waves and Bellomo, D., Pajak, M.J., and J. Sparks, 1999. fects of Hurricane Camille on the marshes storm surge using integrally-coupled, scalable “Coastal flood hazards and the National Flood of the Mississippi River delta.” Ecology, 54, computations.” Coastal Eng., 58: 45-65. Insurance Program.” J. Coastal Res., 28, 21-26. 1118-1123. Dissanayake, P., Brown, J., and H. Karunarathna, Bender, C.J., and R.G. Dean, 2003. “Wave trans- Charbonneau, B.R., Wnek, J.P., Langley, J.A., Lee, 2014. “Modelling storm-induced beach/dune formation by two-dimensional bathymetric G., and R.A. Balsamo, 2016. “Above vs. evolution: Sefton coast, Liverpool Bay, UK.” anomalies with sloped transitions. Coastal belowground plant biomass along a barrier Marine Geology, 357, 225-242. Eng., 50(1-2), 61-84. island: Implications for dune stabilization.” J. Divoky, D., 2007. “Supplementary WHAFIS docu- Bilskie, M.V., Coggin, D., Hagen, S.C., and S.C. Environmental Management, 182(C), 126-133. mentation: WHAFIS 4.0: A revision of FEMA’s Medeiros, 2015. “Terrain-driven unstructured Charbonneau, B.R., Wootton, L.S., Wnek, J.P., WHAFIS 3.0 program.” Technical report, mesh development through semi-automatic Langley, J.A., and M.A. Posner, 2017. “A spe- Federal Emergency Management Agency, vertical feature extraction.” Advances in Water cies effect on storm erosion: Invasive sedge Washington, DC. Resources, 86, Part A: 102-118. stabilized dunes more than native grass during Dresback, K.M., Fleming, J.G., Blanton, B.O., Kaiser, Bilskie, M. V. and S.C. Hagen, 2013. “Topographic Hurricane Sandy.” J. Applied Ecology, 54(5), C., Gourley, J.J., Tromble, E.M., Luettich, R.A., accuracy assessment of bare earth lidar- 1385-1394. Kolar, R.L., Hong, Y., Van Cooten, S., Vergara, derived unstructured meshes,” Advances in Chardón-Maldonado, P., Pintado-Patiño, J.C., and H.J., Flamig, Z.L., Lander, H.M., Kelleher, Water Resources, 52: 165-177. J.A. Puleo, 2016. “Advances in swash- zone K.E., and Durán, V.O. and L.J. Moore, 2013. Blake, E.S., and D.A. Zelinsky, 2018. “National research: Small-scale hydrodynamic and “Vegetation controls on the maximum size of Hurricane Center Tropical Cyclone Report: sediment transport processes.” Coastal Eng., coastal dunes.” PNAS, 110(43), 17217-17222. Hurricane Harvey (AL092017).” Technical 115, 8-25. Earlie, C.S., Young, A.P., Masselink, G., and P.E. Rus- Report, available at https://www.nhc.noaa. Christian, J., Fang, Z., Torres, J., Deitz, R., and P.B. sell, 2015. “Coastal cliff ground motions and gov/data/tcr/AL092017_Harvey.pdf Bedient, 2015. “Modeling the hydraulic ef- response to extreme storm waves.” Geophysical Bradner, C., Schumacher, T., Cox, D., and C. Hig- fectiveness of a proposed storm surge barrier Research Letters, 42(3), 847-854. gins, 2011. “Experimental setup for a large system for the Houston ship channel during Emery, K.O., and G.G. Kuhn, 1982. “Sea cliffs: their scale bridge superstructure model subjected to hurricane events.” ASCE Journal of Natural processes, profiles, and classification.” Geologi- waves.” J. Waterway, Port, Coastal, and Ocean Hazards Review, 16(1). cal Society of America Bulletin, 93(7), 644-654. Eng., 137 (1), 3-11. Claudino-Sales, V., Wang, P., and M.H. Horwitz, Erikson, L.H., McCall, R.T., van Rooijen, A., and Brodie, K.L., and N.J. Spore, 2015. “Foredune 2008. “Factors controlling the survival of B. Norris, 2015. Hindcast storm events in the classification and storm response: Auto- coastal dunes during multiple hurricane Bering Sea for the St. Lawrence Island and mated analysis of terrestrial LIDAR DEMS.” impacts in 2004 and 2005: Santa Rosa bar- Unalakleet regions, Alaska (No. 2015-1193). Presented at Proc. Coastal Sediments 2015, rier island, Florida.” Geomorphology, 95(3-4), U.S. Geological Survey. World Scientific. 295-315. Falchetti, S., Conley, D.C., Brocchini, M., and S. Bryant, D.B., Bryant, M.A., and A.S. Grzegowze- Cohn, N., and P. Ruggiero, 2016. “The influence Elgar, 2010. “Nearshore bar migration and wski, 2017. “Erosion of coastal foredunes: of seasonal to interannual nearshore profile sediment-induced buoyancy effects.” Conti- A review on the effect of dune vegetation.” variability on extreme water levels: Modeling nental Shelf Res., 30(2), 226-238. CHETN-I-94, U.S. Army Engineering Re- wave run-up on dissipative beaches.” Coastal Federal Emergency Management Agency (FEMA), search and Development Center, Vicksburg, Eng., 115, 79-92. 1989. Wave height analysis for flood insur- MS. Coombes, M.A., Viles, H.A., Naylor, L.A., and E.C. ance studies (technical documentation for Bryant, D.B., Bryant, M.A., Sharp, J., and G. Bell, La Marca, 2017. “Cool barnacles: Do com- WHAFIS program version 3.0), amended with 2018. “Erosion of vegetated dunes.” Proc. 7th mon biogenic structures enhance or retard software. Technical report, Washington, DC. International Conference on the Application of rates of deterioration of intertidal rocks and Federal Emergency Management Agency (FEMA), Physical Modelling in Coastal and Port Engi- concrete?” Science of the Total Environment, 2003. Guidelines and specifications for flood neering and Science (Coastlab18), Santander, 580, 1034-1045. hazard mapping partners, Appendix D: Spain, 22-26 May 2018. Costanza, R., Pérez-Maqueo, O., Martinez, M.L., Guidance for coastal flooding analyses and Butler, T., Altaf, M.U., Dawson, C.N., Hoteit, I., Sutton, P., Anderson, S. J., and K. Mulder, mapping. Technical report. Washington, DC. Luo, X., and T. Mayo, 2012. “Data assimilation 2008. “The value of coastal wetlands for hur- Federal Emergency Management Agency (FEMA),

Page 48 Shore & Beach  Vol. 87, No. 1  Winter 2019 2007. Coastal hazard analysis mapping sediment transport and sandbar migration,” Knutson, T.R., McBride, J.L., Chan, J., Emanuel, K., program (CHAMP 2.0). Technical report. Science, 299(5614), 1885-1887. Holland, G., Landsea, C., Held, I., Kossin, J.P., Washington, DC. Holman, R., Plant, N., and T. Holland, 2013. “Bathy: Srivastava, A.K., and M. Sugi, 2010. “Tropical Federal Emergency Management Agency (FEMA), A robust algorithm for estimating nearshore cyclones and climate change.” Nature Geosci- 2011. Coastal Construction Manual. FEMA bathymetry.” J. Geophysical Res.: Oceans, ence, 3, 157–163, https://www.nature.com/ P-55. Washington, D.C. 118(5), 2595-2609. articles/ngeo779, accessed 14 November 2018. Fei, S., Phillips, J., and M. Shouse, 2014. “Biogeo- Houser, C., and S. Hamilton, 2009. “Sensitivity of Kobayashi, N., Gralher, C., and K. Do, 2013. “Ef- morphic impacts of invasive species.” Annual post‐hurricane beach and dune recovery to fects of woody plants on dune erosion and Review of Ecology, Evolution, and Systematics, event frequency.” Earth Surface Processes and overwash.” J. Waterway, Port, Coastal, and 45, 69-87. Landforms, 34(5), pp.613-628. Ocean Eng., 139(6), 466-472. Figlus, J., Kobayashi, N., ASCE, M., Gralher, C., and Houser, C., Hapke, C., and S. Hamilton, 2008. “Con- Landsea, C.W., and J.L. Franklin, 2013. “Atlantic V. Iranzo, 2011. “Wave overtopping and over- trols on coastal dune morphology, shoreline hurricane database uncertainty and presen- wash of dunes.” J. of Waterway, Port, Coastal, erosion and barrier island response to extreme tation of a new database format.” Monthly and Ocean Eng., 137(1), 26-33. storms.” Geomorphology, 100(3-4), 223-240. Weather Review, 141(10), 3576-3592. Flowerdew, J., Horsburgh, K., Wilson, C., and K. Houser, C., and S. Mathew, 2011. “Alongshore Larson, M., Erikson, L., and H. Hanson, 2004. “An Mylne, 2010. “Development and evaluation variation in foredune height in response to analytical model to predict dune erosion of an ensemble forecasting system for coastal transport potential and sediment supply: due to wave impact.” Coastal Eng., 51(8-9), storm surge.” Q J R Meteorological Society, South Padre Island, Texas.” Geomorphology, 675-696. 136(651), 1444-1456. 125(1), 62-72. Leatherman, S.P., Williams, A. T., and J. S. Fisher, Forster, S.M., and T.H. Nicolson, 1981. “Aggregation Houser, C., Wernette, P., Rentschlar, E., Jones, H., 1977. “Overwash sedimentation associated of sand dune from a maritime embryo sand Hammond, B., and S. Trimble, 2015. “Post- with a large-scale northeaster.” Marine Geol- dune by microorganisms and higher plants.” storm beach and dune recovery: Implications ogy, 24(2), 109-121. Soil Biology and Biochemistry, 13, 199-203. for barrier island resilience.” Geomorphology, Levinson, D.H., Vickery, P.J., and D.T. Resio, 2010. Gallagher, E.L., Elgar, S., and R.T. Guza, 1998. “Ob- 234, 54-63. “A review of the climatological characteristics servations of sand bar evolution on a natural Hsu, T.J., and P.L.F. Liu, 2004. “Toward modeling of landfalling Gulf hurricanes for wind, wave, beach.” J. Geophysical Research: Oceans, turbulent suspension of sand in the near- and surge hazard estimation.” Ocean Eng., 103(C2), 3203-3215. shore.” J. Geophysical Res.: Oceans, 109(C6). 37(1), 13-25. Gedan, K.B., Kirwan, M.L., Wolanski, E., Barbier, IPCC, 2013. Climate Change 2013: The Physical Sci- Lindemer, C.A., Plant, N.G., Puleo, J.A., Thompson, E.B., and B.R. Silliman, 2011. “The present ence Basis. Contribution of Working Group I to D.M. and T.V. Wamsley, 2010. “Numerical and future role of coastal wetland vegetation the Fifth Assessment Report of the Intergovern- simulation of a low-lying barrier island’s mor- in protecting shorelines: Answering recent mental Panel on Climate Change (Stocker, T.F., phological response to Hurricane Katrina.” challenges to the paradigm.” Climate Change, D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Coastal Eng., 57(11-12), 985-995. 106, 7-29. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Long, J.W., de Bakker, A. and N.G. Plant, 2014. Hacker, S.D., Zarnetske, P., Seabloom, E., Rug- Midgley [eds.]). Cambridge University Press, “Scaling coastal dune elevation changes across giero, P., Mull, J., Gerrity, S., and C. Jones, Cambridge, United Kingdom and New York, storm‐impact regimes.” Geophysical Res. Let- 2011. “Subtle differences in two non-native NY, USA, 1,535 pp. ters, 41(8), 2899-2906. congeneric beach grasses significantly affect Jelesnianski, C.P., Chen, J., and W.A. Shaffer, 1992. Lorenzoni, I., Nicholson-Cole, S., and L. Whit- their colonization, spread, and impact.” Oikos, “SLOSH: Sea, Lake, and Overland Surges from marsh, 2007. “Barriers perceived to engaging 121(1), 138-148. Hurricanes.” U.S. Dept. of Commerce, Nation- with climate change among the UK public and Hagen, S.C., and D. M. Parrish, 2004. “Meshing al Oceanic and Atmospheric Administration, their policy implications.” Global Environmen- requirements for tidal modeling in the western National Weather Service, Silver Spring, MD. tal Change, 17(3):445-459. North Atlantic.” International J. Computa- Jia, G., Taflanidis, A.A., Nadal-Caraballo, N.C., Luijendijk, A., Hagenaars, G., Ranasinghe, R., Baart, tional Fluid Dynamics, 18(7), 585-595. Melby, J.A., Kennedy, A.B., and J.M. Smith, F., Donchyts, G., and S. Aarninkhof, 2018. Hagen, S.C., Zundel, A.K., and S. Kojima, 2006. 2016. “Surrogate modeling for storm surge “The state of the world’s beaches.”Scientific “Automatic, unstructured mesh generation prediction using an existing database of Reports, 8. for tidal calculations in a large domain.” In- synthetic storms and addressing output size Mardhiah, U., Caruso, T., Gurnell, A., and M.C. ternational J. Computational Fluid Dynamics, and time-dependence.” Natural Hazards, Rillig, 2016. “Arbuscular mycorrhizal fungal 20(8), 593-608. 81(2), 909-938. hyphae reduce soil erosion by surface water Hapke, C., and B. Richmond, 2002. “The impact of Kalnay, E., 2003. Atmospheric modeling, data as- flow in a greenhouse experiment.” Applied Soil climatic and seismic events on the short-term similation and predictability. Cambridge Ecology, 99, 137-140. evolution of seacliffs based on 3-D mapping: University Press. Maun, M.A., 1994. “Adaptations enhancing survival northern Monterey Bay, California.” Marine Kennedy, A., Rogers, S., Sallenger, A., Gravois, U., and establishment of seedlings on coastal Geology, 187(3-4), 259-278. Zachry, B., Dosa, M., and F. Zarama, 2011. dune systems.” Vegetatio, 111(1), 59-70. Harter, C., and J. Figlus, 2017. “Numerical modeling “Building destruction from wave and surge Maun, M.A., 2009. The Biology of Coastal Sand of the morphodynamic response of a low-ly- on the Bolivar Peninsula during Hurricane Dunes. New York, NY: Oxford University ing barrier island beach and foredune system Ike.” J. Waterway, Port, Coastal, and Ocean Press. inundated during Hurricane Ike using XBeach Eng.,137 (3), 132-141. Mayo, T., Butler, T., Dawson, C.N., and I. Hoteit, and CSHORE,” Coastal Eng., 120, 64-74. Kerr, P.C., Donahue, A.S., Westerink, J.J., Luettich 2014. “Data assimilation within the Advanced Henderson, S.M., Allen, J.S., and P.A. Newberger, Jr., R.A., Zheng, L.Y., Weisberg, R.H., Huang, Circulation (ADCIRC) modeling framework 2004. “Nearshore sandbar migration predicted Y., Wang, H.V., Teng, Y., Forrest, D.R., Roland, for the estimation of Manning’s friction coef- by an eddy‐diffusive boundary layer model.” J. A., Haase, A.T., Kramer, A.W., Taylor, A.A., ficient.” Ocean Modelling, 76, 43-58. Geophysical Res.: Oceans, 109(C6). Rhome, J.R., Feyen, J.C., Signell, R.P., Hanson, McCall, R.T., De Vries, J.V.T., Plant, N.G., Van Hesp, P.A., 1989. “A review of biological and J.L., Hope, M.E., Estes, R.M., Dominguez, Dongeren, A.R., Roelvink, J.A., Thompson, geomorphological processes involved in R.A., Dunbar, R.P., Semeraro, L.N., Wester- D.M., and A.J.H.M. Reniers, 2010. “Two- the initiation and development of incipient ink, H.J., Kennedy, A.B., Smith, J.M., Powell, dimensional time dependent hurricane foredunes.” Proc. Royal Society of Edinburgh. M.D., Cardone, V.J., and A.T. Cox, 2013. “U.S. overwash and erosion modeling at Santa Rosa Section B. Biological Sciences, 96B, 181-201. IOOS coastal and ocean modeling testbed: Island.” Coastal Eng., 57(7), 668-683. Hilton, M., Harvey, N., Hart, A., James, K., and C. Inter-model evaluation of tides, waves, and Mickey, R., Long, J.W., Thompson, D.M., Plant, Arbuckle, 2006. “The impact of exotic dune hurricane surge in the Gulf of Mexico.” J. N.G., and P.S. Dalyander, 2017. “Storm- grass species on foredune development in Geophysical Res.: Oceans, 118,10, 5129-5172, impact scenario XBeach model inputs and Australia and New Zealand: a case study of DOI 10.1002/jgrc.20376. results.” U.S. Geological Survey. Ammophila arenaria and Thinopyrum juncei- Klemas V., 2013. “Remote sensing of coastal wetland Moore, L.J., 2000. “Shoreline mapping techniques.” forme.” Australian Geographer, 37(3), 313-334. biomass: an overview.” J. Coastal Res., 29(5), J. Coastal Res., 16(1), 111-124. Hoefel, F., and S. Elgar, 2003. “Wave-induced 1016-1028. Moskalski S.M., and C.K. Sommerfield, 2013.

Shore & Beach  Vol. 87, No. 1  Winter 2019 Page 49 “Effects of northeaster storms on water level Young, 2016. “Automatic delineation of seacliff Schupp, C.A., McNinch, J.E., and J.H. List, 2006. and turbidity in a Delaware Bay subestuary.” limits using lidar-derived high-resolution “Nearshore shore-oblique bars, gravel out- J. Coastal Res., 291, 205-213. DEMs in southern California.” J. Coastal Res., crops, and their correlation to shoreline Mull, J., and P. Ruggiero, 2014. “Estimating storm- 76(sp1), 162-173. change.” Marine Geology, 233(1-4), 63-79. induced dune erosion and overtopping along Palmsten, M.L., and R.A. Holman, 2011. “Infiltra- Sherwood, C.R., Long, J.W., Dickhudt, P.J., Daly- U.S. west coast beaches.” J. Coastal Res., 298, tion and instability in dune erosion.” J. Geo- ander, P.S., Thompson, D.M., and N.G. Plant, 1173-1187. physical Res.: Oceans,116(C10). 2014. “Inundation of a barrier island (Chan- Mulligan, R.P., Hay, A.E., and A.J. Bowen, 2008. Palmsten, M.L., and R.A. Holman, 2012. “Labora- deleur Islands, Louisiana, USA) during a “Wave‐driven circulation in a coastal bay dur- tory investigation of dune erosion using stereo hurricane: Observed water‐level gradients and ing the landfall of a hurricane.” J. Geophysical video.” Coastal Eng., 60, 123-135. modeled seaward sand transport.” J. Geophysi- Res.: Oceans, 113(C5). Palmsten, M.L., and K.D. Splinter, 2016. “Observa- cal Res.: Earth Surface, 119(7), 1498-1515. Murray A.B., Knaapen, M.A., Tal, M., and M.L. Kir- tions and simulations of wave runup during a Sigren, J.M., Figlus, J., and A.R. Armitage, 2014. wan, 2008. “Biomorphodynamics: physical- laboratory dune erosion experiment.” Coastal “Coastal sand dunes and dune vegetation: biological feedbacks that shape landscapes.” Eng., 115, 58-66. https://doi.org/10.1016/j. restoration, erosion, and storm protection.” Water Resources Res., 44(11): W11301. coastaleng.2016.01.007. Shore & Beach, 82, 5-12. Nadal-Caraballo, N.C., Lewis, A.B., and F. Diop, Park, H., Tomiczek, T., Cox, D. T., van de Lindt, J.W., Silva, R., Martínez, M. L., Odériz, I., Mendoza, E., 2018. “StormSim-CHRP: Rapid prediction and P. Lomonaco, 2017. “Experimental mod- and R. A. Feagin, 2016. “Response of vegetated of coastal storm hazards.” U.S. Army Corps eling of horizontal and vertical wave forces on dune-beach systems to storm conditions.” of Engineers, Flood Risk Management (FRM) an elevated coastal structure.” Coastal Eng., Coastal Eng., 109, 53–62. Newsletter, January 2018, 11(1). 128, 58-74. Simmons, J.A., Harley, M.D., Marshall, L.A., and Narayan S., Beck, M. W., Wilson P., Thomas, C.J., Plant, N.G., and H.F. Stockdon, 2012. “Probabilistic I.T. Coastal, 2017. (n.d.). “Calibrating and Guerrero, A., Shepard, C. C., Reguero, B. G., prediction of barrier‐island response to hur- assessing uncertainty in coastal numerical Franco, G., Ingram, J. C. and D. Trespalacios, ricanes.” J. Geophysical Res.: Earth Surface, models.” Elsevier. 2017. “The value of coastal wetlands for flood 117(F3). Splinter, K.D., Carley, J.T., Golshani, A., and R. damage reduction in the northeastern USA.” Priestas, A.M., and S. Fagherazzi, 2011. “Morphol- Tomlinson, 2014. “A relationship to describe Scientific Reports, 7, 1-12. ogy and hydrodynamics of wave-cut gullies.” the cumulative impact of storm clusters on National Academies of Sciences, Engineering, Geomorphology, 131, 1-13. beach erosion.” Coastal Eng., 83, 49-55. and Medicine (NASEM), 2018. Understand- Rappaport E.N., 2014. “Fatalities in the United Splinter, K.D., and M.L. Palmsten, 2012. “Modeling ing the Long-Term Evolution of the Coupled States from Atlantic tropical cyclones: new dune response to an East Coast Low.” Marine Natural-Human Coastal System: The Future data and interpretation.” Bulletin of the Geology, 329, 46-57. of the U.S. Gulf Coast. Washington, DC.: American Meteorological Society, March 2014. Splinter, K.D., Kearney, E.T., and I.L. Turner, 2018. The National Academies Press. https://doi. https://doi.org/10.1175/BAMS-D-12-00074.1 “Drivers of alongshore variable dune erosion org/10.17226/25108. Raubenheimer, B., Guza, R.T., and S. Elgar, 2001. during a storm event: Observations and mod- National Hurricane Center (NHC), 2017. “An- “Field observations of wave‐driven setdown elling.” Coastal Eng., 131, 31-41. nual average official track errors for Atlantic and setup.” J. Geophysical Res.: Oceans, Stallins, J.A., 2006. “Geomorphology and ecology: basin tropical storms and hurricanes for 106(C3), 4629-4638. Unifying themes for complex systems in the period 1970-2017.” National Oceanic Ray, T., Stepinski, E., Sebastian, A., and P.B. Bedient, biogeomorphology.” Geomorphology, 77(3- and Atmospheric Administration, available 2011. “Dynamic modeling of storm surge and 4), 207-216. at https://www.nhc.noaa.gov/verification/ inland flooding in a Texas coastal floodplain.” Stockdon, H.F., Sallenger Jr., A.H., Holman, R.A., pdfs/1970-present_OFCL_ATL_annual_trk_ J. Hydraulic Eng., 137(10), 1103-1110. and P.A. Howd, 2007. “A simple model for errors_noTDs.pdf. Rego, J.L., and C. Li, 2010. “Storm surge propaga- the spatially-variable coastal response to Nemunaitis-Monroe, K.L., 2013. “Skill assessment tion in Galveston Bay during Hurricane Ike.” hurricanes.” Marine Geology, 238(1-4), 1-20. of a real-time forecast system utilizing a J. Marine Systems, 82, 265-279. Stumpf, R.P., 1983. “The process of sedimentation coupled hydrological and coastal hydrody- Roberts T.M., Wang, P., and J.A. Puleo, 2013. on the surface of a salt marsh.” Estuarine, namic model during Hurricane Irene (2011).” “Storm-driven cyclic beach morphodynamics Coastal and Shelf Science, 17, 495-508. Continental Shelf Res., 71, 78-94. of a mixed sand and gravel beach along the Sullivan, B.K., 2017. The most expensive U.S. hurri- Nielsen, P., Robert, S., Møller-Christiansen, B., and mid-Atlantic coast, USA.” Marine Geology, cane season ever: by the numbers.” Bloomberg, P. Oliva, 2001. “Infiltration effects on sediment 346, 403-421. 26 November 2017, retrieved 18 October mobility under waves.” Coastal Eng., 42(2), Roelvink, D., Reniers, A., Van Dongeren, A.P., de 2018. https://www.bloomberg.com/news/ 105-114. Vries, J.V.T., McCall, R., and J. Lescinski, 2009. articles/2017-11-26/the-most-expensive-u- Norman, E.C., Rosser, N.J., Brain, M.J., Petley, D.N., “Modelling storm impacts on beaches, dunes s-hurricane-season-ever-by-the-numbers and M. Lim, 2013. “Coastal cliff‐top ground and barrier islands.” Coastal Eng., 56(11-12), Sunamura, T., 1977. “A relationship between wave- motions as proxies for environmental pro- 1133-1152. induced cliff erosion and erosive force of cesses.” J. Geophysical Res.: Oceans, 118(12), Ružić, I., Marović, I., Benac, Č., and S. Ilić, 2014. waves.” J. of Geology, 85(5), 613-618. 6807-6823. “Coastal cliff geometry derived from struc- Sunamura, T., 1992. Geomorphology of rocky coasts Nyman, J., Crozier, C., and R. DeLaune, 1995. “Roles ture-from-motion photogrammetry at Stara (Vol. 3). John Wiley & Son Ltd. and patterns of hurricane sedimentation in an Baška, Krk Island, Croatia.” Geo-Marine Let- Taflanidis, A.A., Zhang, J., Nadal-Caraballo, N.C., estuarine marsh landscape.” Estuarine, Coastal ters, 34(6), 555-565. and J.A. Melby, 2017. “Advances in surrogate and Shelf Science, 40, 665-679. Safak, I., List, J.H., Warner, J.C., and N. Kumar, 2017. modeling for hurricane risk assessment: O’Neill, S., and S. Nicholson-Cole, 2009. “‘Fear “Observations and 3D hydrodynamics-based storm selection and climate change impact.” won’t do it’: Promoting positive engagement modeling of decadal-scale shoreline change In Proc. 12th International Conference on with climate change through visual and iconic along the Outer Banks, North Carolina.” Structural Safety & Reliability, 6-10 August, representations.” Science Communication, Coastal Eng., 120, 78-92. Vienna, Austria. 30(3):355-379. Sallenger Jr., A.H., 2000. “Storm impact scale for Tanaka, N., Nandasena, N.A.K., Jinadasa, K.B.S.N., Overbeck, J.R., Long, J.W., and H.F. Stockdon, 2017. barrier islands.” J. Coastal Res., 16(3), 890-895. Sasaki, Y., Tanimoto, K., and M.I.M. Mow- “Testing model parameters for wave‐induced Sallenger Jr., A.H., Krabill, W., Brock, J., Swift, R., jood, 2009. “Developing effective vegetation dune erosion using observations from Hur- Manizade, S., and H. Stockdon, 2002. “Sea- bioshield for tsunami protection.” Civil Eng. ricane Sandy.” Geophysical Res. Letters, 44(2), cliff erosion as a function of beach changes and Environmental Systems, 26(2), 163-180. 937-945. and extreme wave runup during the 1997- Thornton, E.B., and R.T. Humiston, 1996. “Bar/ Palaseanu-Lovejoy, M., Danielson, J., Thatcher, C., 1998 El Niño.” Marine Geology, 187(3-4), trough generation on a natural beach.” J. Foxgrover, A., Barnard, P., Brock, J., and A. 279-297. Geophysical Res., 101: 12097-12110.

Page 50 Shore & Beach  Vol. 87, No. 1  Winter 2019 Timm, B.C., Smith, S.M., and S.E. Greenspan, 2014. flow conditions.”J. Geophysical Res.: Oceans, Westerink, J.J., Luettich Jr, R.A., Feyen, J.C., “Remotely sensed mapping of Ammophila 94(C8), 10971-10991. Atkinson, J.H., Dawson, C., Roberts, H.J., spp. distribution and density at Cape Cod Na- Turner, I.L. and G. Masselink, 1998. “Swash infiltra- Powell, M.D., Dunion, J.P., Kubatko, E.J., and tional Seashore.” J. Coastal Res., 296, 862-867. tion‐exfiltration and sediment transport.”J. H. Pourtaheri, 2008. “A basin- to channel- Tomiczek, T., Kennedy, A.B., and S. Rogers, 2014. Geophysical Res.: Oceans, 103(C13), 30813- scale unstructured grid hurricane storm “Collapse limit state fragilities of wood framed 30824. surge model applied to southern Louisiana.” residences from storm surge and waves during Wahl, T., Jain, S., Bender, J., Meyers, S.D., and M.E. Monthly Weather Review, 136, 833-864. Hurricane Ike.” J. Waterway, Port, Coastal, and Luther, 2015. “Increasing risk of compound Westerink, J.J., Luettich Jr., R.A., and J.C. Muccino, Ocean Eng., 140 (1), 43-55. flooding from storm surge and rainfall for 1994. “Modeling tides in the western North Torn, R.D., and C. Snyder, 2012. “Uncertainty of major US cities.” Nature Climate Change, 5, Atlantic using unstructured graded grids.” tropical cyclone best-track information.” 1093-1098. Tellus, 46A, 178-199. Weather and Forecasting, 27(3), 715-729. Wahl, T., Plant, N.G. and J.W. Long, 2016. “Probabi- Young, A.P., Guza, R.T., Flick, R.E., O’Reilly, W.C., Torres, J.M., Bass, B., Irza, N., Fang, Z., Proft, J., listic assessment of erosion and flooding risk and R. Gutierrez, 2009. “Rain, waves, and Dawson, C.N., Kiani, M., and P. Bedient, 2015. in the northern Gulf of Mexico.” J. Geophysical short-term evolution of composite seacliffs “Characterizing the hydraulic interactions of Res.: Oceans, 121(5), 3029-3043. in southern California.” Marine Geology, hurricane storm surge and rainfall-runoff Wamsley, T.V., Cialone, M.A., Smith, J.M., Ebersole, 267(1-2), 1-7. for the Houston-Galveston region.” Coastal B.A., and A.S. Grzegorzewski, 2009. “Influ- Young, A.P., 2015. “Recent deep-seated coastal land- Eng., 106, 7-19. ence of landscape restoration and degrada- sliding at San Onofre State Beach, California.” Trenhaile, A.S., 1987. The geomorphology of rock tion on storm surge and waves in southern Geomorphology, 228, 200-212. coasts. Oxford University Press, USA. Louisiana.” J. Natural Hazards, 51, 207-224. Young, A.P., Guza, R.T., O’reilly, W.C., Burvingt, O., Tromble, E.M., Kolar, R.L., Dresback, K.M, Hong, Y., Warner, J., Armstrong, B., He, R., and J. Zambon, and R.E. Flick, 2016. “Observations of coastal Vieux, B.E., Luettich, R.A., Gourley, J.J., Kelle- 2010. “Development of a coupled ocean-atmo- cliff base waves, sand levels, and cliff top shak- her, K.E., and S. Van Cooten, 2011. “Aspects of sphere-wave-sediment (COAWST) modeling ing.” Earth Surface Processes and Landforms, coupled hydrologic-hydrodynamic modeling system.” Ocean Modeling, 35(3), 230-244. 41(11), 1564-1573. for coastal flood inundation.” In: Proc. 11th Warrick, J.A., Ritchie, A.C., Adelman, G., Adelman, Zarnetske, P.L., Hacker, S.D., Seabloom, E.W., Rug- International Conference on Estuarine and K., and P.W. Limber, 2016. “New techniques to giero, P., Killian, J.R., Maddux, T.B., and D. Coastal Modeling, eds. Spaulding, M.L., ASCE, measure cliff change from historical oblique Cox, 2010. “Biophysical feedback mediates Reston, VA., 724-743. aerial photographs and structure-from- effects of invasive grasses on coastal dune Trowbridge, J., and D. Young, 1989. “Sand trans- motion photogrammetry.” J. Coastal Res., s h ap e .” Ecology, 93(6), 1439-1450. port by unbroken water waves under sheet 33(1), 39-55.

Authors: Nicole Elko, P.O. Box 1451, Folly Beach, SC 29439 Casey Dietrich, 2501 Stinson Drive, Raleigh, NC 27607 Mary Cialone, 3909 Halls Ferry Road, Building 3200, Vicksburg, MS 39180 Hilary Stockdon, 12201 Sunrise Valley Drive, Reston, VA 20192 Matt W. Bilskie, 3255 Patrick F. Taylor Hall, Baton Rouge, LA 70803 Brandon Boyd, 3909 Halls Ferry Road, Building 3200, Vicksburg, MS 39180 Bianca Charbonneau, 433 S. University Avenue, 330 Leidy Labs, Philadelphia, PA 19147 Dan Cox, 101 Kearney Hall, Corvallis OR 97331 Kendra M. Dresback, 202 W. Boyd Street, CEC Room 334, Norman, OK 73019 Steve Elgar, WHOI, Mail Stop #11, 266 Woods Hole Road, Woods Hole, MA 02543 Amanda Lewis, 3909 Halls Ferry Road, Building 3200, Vicksburg, MS 39180 Patrick Limber, Department of Marine Science, P.O. Box 261954, Conway, SC 29528 Joe Long, 601 S. College Road, Wilmington, NC 28403 T. Chris Massey, 3909 Halls Ferry Road, Building 3200, Vicksburg, MS 39180 Talea Mayo, 12201 Research Parkway, Suite 501, Orlando, FL 32826 Kathryn McIntosh, 100 I St SE #406, Washington, DC 20003 Norberto Nadal-Caraballo, 3909 Halls Ferry Road, Building 3200, Vicksburg, MS 39180 Britt Raubenheimer, WHOI, Mail Stop #11, 266 Woods Hole Road, Woods Hole, MA 02543 Tori Tomiczek, 590 Holloway Road, Mail Stop 11D, Annapolis, MD 21402-5042 Anna Wargula, 590 Holloway Road, Mail Stop 11D, Annapolis, MD 21402-5042

Shore & Beach  Vol. 87, No. 1  Winter 2019 Page 51