Description: Highlights • Barrier island beach and dunes can inherit morphology from framework geology. • The influence of buried paleo-channels is affected by their scale and orientation. • Non-linear interaction occurs between geology and smaller-scale coastal processes. • Framework geology is key to predicting barrier island response to sea level rise. • Fine-scale variability in dune height is greater landward of buried paleo-channels. Abstract Barrier island response and recovery to storms, and island transgression with relative sea level rise, can be influenced by the framework geology. The influence of framework geology on barrier island geomorphology has previously been examined in areas where the framework is rhythmic alongshore or consists of an isolated paleo-channel or headland. The purpose of this paper is to examine the influence of framework geology on beach and dune geomorphology at Padre Island National Seashore (PAIS), Texas, USA, where the framework geology is variable alongshore. Alongshore beach and dune morphometrics and offshore bathymetric profiles were extracted from a combined topography and bathymetry digital elevation model (DEM) using an automated approach along the ~100 km study area, and an electromagnetic induction (EMI) survey was used to map the subsurface framework geology. Wavelet decomposition, Global Wavelet (GW), and bicoherence analyses were used to test for spatial relationships between and within the extracted alongshore metrics. GW trendlines demonstrate that beach and dune morphometrics are structurally controlled. Hotspots in wavelet coherence plots between framework geology and alongshore island morphometrics indicate that the paleo-channels dissecting the island influence beach and dune morphology, with large dunes found in the area directly landward of the paleochannels. Bicoherence analysis of alongshore beach and dune morphometrics indicates that low-frequency oscillations due to framework geology interact with higher-frequency oscillations, with greater small-scale variability in the dune line directly landward of the paleo-channels. These results suggest that the paleo-channels of PAIS non-linearly influence beach and dune morphology, which in turn alters the response of the island to storms and sea level rise. It is argued that an understanding of the framework geology is key to predicting island response to sea level rise and framework geology needs to be included in barrier island models. This paper demonstrates that an irregular framework geology influences small-scale coastal processes, and creates interactions across scales that influence beach and dune morphology and affects barrier island response to storms and sea level rise.

Description: Barrier island transgression is influenced by the alongshore variation in beach and dune morphology, which determines the amount of sediment moved landward through wash-over. While several studies have demonstrated how variations in dune morphology affect island response to storms, the reasons for that variation and the implications for island management remain unclear. This paper builds on previous research by demonstrating that paleo-channels in the irregular framework geology can have a directional influence on alongshore beach and dune morphology. The influence of relict paleo-channels on beach and dune morphology on Padre Island National Seashore, Texas, was quantified by isolating the long-range dependence (LRD) parameter in autoregressive fractionally integrated moving average (ARFIMA) models, originally developed for stock market economic forecasting. ARFIMA models were fit across ∼250 unique spatial scales and a moving window approach was used to examine how LRD varied with computational scale and location along the island. The resulting LRD matrices were plotted by latitude to place the results in the context of previously identified variations in the framework geology. Results indicate that the LRD is not constant alongshore for all surface morphometrics. Many flares in the LRD plots correlate to relict infilled paleo-channels, indicating that the framework geology has a significant influence on the morphology of Padre Island National Seashore (PAIS). Barrier island surface morphology LRD is strongest at large paleo-channels and decreases to the north. The spatial patterns in LRD surface morphometrics and framework geology variations demonstrate that the influence of paleo-channels can be asymmetric (i.e., affecting beach–dune morphology preferentially in one direction alongshore) where the alongshore sediment transport gradient was unidirectional during island development. The asymmetric influence of framework geology on coastal morphology has long-term implications for coastal management activities because it dictates the long-term behavior of a barrier island. Coastal management projects should first seek to assess the framework geology and understand how it influences coastal processes in order to more effectively balance long-term natural variability with short-term societal pressure.

Description: Shorelines exhibit long-range dependence (LRD) and have been shown in some environments to be described in the wavenumber domain by a power law characteristic of scale-independence. Recent evidence suggests that the geomorphology of barrier islands can, however, exhibit scale dependence as a result of systematic variations of the underlying framework geology. The LRD of framework geology, which influences island geomorphology and its response to storms and sea level rise, has not been previously examined. Electromagnetic induction (EMI) surveys conducted along Padre Island National Seashore (PAIS), Texas, USA, reveal that the EMI apparent conductivity σa signal and, by inference, the framework geology exhibits LRD at scales up to 101 to 102 km. Our study demonstrates the utility of describing EMI σa and LiDAR spatial series by a fractional auto-regressive integrated moving average process that specifically models LRD. This method offers a robust and compact way for quantifying the geological variations along a barrier island shoreline using three parameters (p,d,q). We discuss how ARIMA (0,d,0) models that use a single parameter d provide a quantitative measure for determining free and forced barrier island evolutionary behavior across different scales. Statistical analyses at regional, intermediate, and local scales suggest that the geologic framework within an area of paleo-channels exhibits a first order control on dune height. The exchange of sediment amongst nearshore, beach and dune in areas outside this region are scale-independent, implying that barrier islands like PAIS exhibit a combination of free and forced behaviors that affect the response of the island to sea level rise.

Description: Barrier Islands represent some of the most dynamic and complex systems within the Critical Zone worldwide. Although coastal systems tend not to be recognized as Critical Zone environments, the evolution of Barrier Islands and the ecological functions they provide can be characterized in terms of a complex feedback among sediment supply (lithosphere), hydrology, the atmosphere, and ecology (biosphere). This represents an interesting departure from the traditional view of Barrier Island evolution (either regression or transgression) as a result of variations in sea level, sediment supply, and accommodation space. This chapter takes a Critical Zone approach to the response of Barrier Island evolution to sea-level rise and storm activity, explicitly recognizing the feedback among sediment supply, aeolian transport, disturbance regimes, vegetation development, and hydrology.

Description: Highlights • Driving on the beach affects beach-dune response to and recovery following a storm. • Non-driving areas show substantial evidence of recovery within 6 months post-storm. • Driving on the beach limits incipient foredune development and recovery. • Limiting low dune recovery can reinforce the effects of framework geology. Abstract Beach and dune morphology are spatially and temporally variable, changing over a broad range of scales simultaneously. Strong wind, waves, and storm surge from Hurricane Harvey substantially eroded the beach and dunes along the Texas coast, causing significant scarps and berms. This paper presents information about how anthropogenic activity, such as driving on the beach, affected the response and recover of a barrier island with regard to the post-storm resiliency along the Texas-Gulf of Mexico coast by comparing two adjacent 7 km stretches of coast: a driving section and a limited-access section. A collection of field photos, aerial imagery, and a September 2016 LiDAR-derived digital elevation model (DEM) dataset provide information on pre-storm morphology, while field photos, taken only 3 days after Hurricane Harvey made landfall, and a structure-from-motion (SfM)-derived DEM and imagery provide qualitative and quantitative information about the post-storm morphology. While beach and dune erosion in the non-driving section was restricted entirely to the beach and incipient dune system, the driving section exhibited complex patterns of erosion and deposition along the beach and the entire foredune profile was altered. Despite the hurricane making landfall north of the study site and closer to the non-driving section, beach-dune erosion and scarping was greatest in the southern section which is accessible to public vehicles. Results demonstrate that human activity affects the response and recovery of the beach-dune system along the Gulf side of the island by decreasing alongshore variability in erosion-deposition and limiting vegetative and geomorphic recovery.

Description: Groundwater resources in coastal regions are facing enormous pressure caused by population growth and climate change. Few studies have investigated whether offshore freshened groundwater systems are connected with terrestrial aquifers recharged by meteoric water, or paleo-groundwater systems that are no longer associated with terrestrial aquifers. Distinguishing between the two has important implications for potential extraction to alleviate water stress for many coastal communities, yet very little is known about these connections, mainly because it is difficult to acquire continuous subsurface information across the coastal transition zone. This study presents a first attempt to bridge this gap by combining three complementary near-surface electromagnetic methods to image groundwater pathways within braided alluvial gravels along the Canterbury coast, South Island, New Zealand. We show that collocated electromagnetic induction, ground penetrating radar, and transient electromagnetic measurements, which are sensitive to electrical contrasts between fresh (low conductivity) and saline (high conductivity) groundwater, adequately characterize hydrogeologic variations beneath a mixed sand gravel beach in close proximity to the Ashburton River mouth. The combined measurements – providing information at three different depths of investigation and resolution – show several conductive zones that are correlated with spatial variations in subsurface hydrogeology. We interpret the conductive zones as high permeability conduits corresponding to lenses of well-sorted gravels and secondary channel fill deposits within the braided river deposit architecture. The geophysical surveys provide the basis for a discharge model that fits our observations, namely that there is evidence of a multilayered system focusing groundwater flow through stacked high permeability gravel layers analogous to a subterranean river network. Coincident geophysical surveys in a region further offshore indicate the presence of a large, newly discovered freshened groundwater system, suggesting that the offshore system in the Canterbury Bight is connected with the terrestrial aquifer system.

Description: Gully formation has been associated to groundwater seepage in unconsolidated sand- to gravel-sizedsediments. Our understanding of gully evolution by groundwater seepage mostly relies on experiments and nu-merical simulations, and these rarely take into consideration contrasts in lithology and permeability. In addition,process-based observations and detailed instrumental analyses are rare. As a result, we have a poor understandingof the temporal scale of gully formation by groundwater seepage and the influence of geological heterogeneityon their formation. This is particularly the case for coastal gullies, where the role of groundwater in their for-mation and evolution has rarely been assessed. We address these knowledge gaps along the Canterbury coastof the South Island (New Zealand) by integrating field observations, luminescence dating, multi-temporal un-occupied aerial vehicle and satellite data, time domain electromagnetic data and slope stability modelling. Weshow that gully formation is a key process shaping the sandy gravel cliffs of the Canterbury coastline. It is anepisodic process associated to groundwater flow that occurs once every 227 d on average, when rainfall intensi-ties exceed 40 mm d−1. The majority of the gullies in a study area southeast (SE) of Ashburton have undergoneerosion, predominantly by elongation, during the last 11 years, with the most recent episode occurring 3 yearsago. Gullies longer than 200 m are relict features formed by higher groundwater flow and surface erosion〉2 kaago. Gullies can form at rates of up to 30 m d−1via two processes, namely the formation of alcoves and tunnelsby groundwater seepage, followed by retrogressive slope failure due to undermining and a decrease in shearstrength driven by excess pore pressure development. The location of gullies is determined by the occurrenceof hydraulically conductive zones, such as relict braided river channels and possibly tunnels, and of sand lensesexposed across sandy gravel cliffs. We also show that the gully planform shape is generally geometrically similarat consecutive stages of evolution. These outcomes will facilitate the reconstruction and prediction of a prevalenterosive process and overlooked geohazard along the Canterbury coastline.

Description: © The Author(s), 2021. This article is distributed under the terms of the Creative Commons Attribution License. The definitive version was published in Goldstein, E. B., Buscombe, D., Lazarus, E. D., Mohanty, S. D., Rafique, S. N., Anarde, K. A., Ashton, A. D., Beuzen, T., Castagno, K. A., Cohn, N., Conlin, M. P., Ellenson, A., Gillen, M., Hovenga, P. A., Over, J.-S. R., Palermo, R., Ratliff, K. M., Reeves, I. R. B., Sanborn, L. H., Straub, J. A., Taylor, L. A., Wallace E. J., Warrick, J., Wernette, P., Williams, H. E. Labeling poststorm coastal imagery for machine learning: measurement of interrater agreement. Earth and Space Science, 8(9), (2021): e2021EA001896, https://doi.org/10.1029/2021EA001896.

Description: Classifying images using supervised machine learning (ML) relies on labeled training data—classes or text descriptions, for example, associated with each image. Data-driven models are only as good as the data used for training, and this points to the importance of high-quality labeled data for developing a ML model that has predictive skill. Labeling data is typically a time-consuming, manual process. Here, we investigate the process of labeling data, with a specific focus on coastal aerial imagery captured in the wake of hurricanes that affected the Atlantic and Gulf Coasts of the United States. The imagery data set is a rich observational record of storm impacts and coastal change, but the imagery requires labeling to render that information accessible. We created an online interface that served labelers a stream of images and a fixed set of questions. A total of 1,600 images were labeled by at least two or as many as seven coastal scientists. We used the resulting data set to investigate interrater agreement: the extent to which labelers labeled each image similarly. Interrater agreement scores, assessed with percent agreement and Krippendorff's alpha, are higher when the questions posed to labelers are relatively simple, when the labelers are provided with a user manual, and when images are smaller. Experiments in interrater agreement point toward the benefit of multiple labelers for understanding the uncertainty in labeling data for machine learning research.

Description: The authors gratefully acknowledge support from the U.S. Geological Survey (G20AC00403 to EBG and SDM), NSF (1953412 to EBG and SDM; 1939954 to EBG), Microsoft AI for Earth (to EBG and SDM), The Leverhulme Trust (RPG-2018-282 to EDL and EBG), and an Early Career Research Fellowship from the Gulf Research Program of the National Academies of Sciences, Engineering, and Medicine (to EBG). U.S. Geological Survey researchers (DB, J-SRO, JW, and PW) were supported by the U.S. Geological Survey Coastal and Marine Hazards and Resources Program as part of the response and recovery efforts under congressional appropriations through the Additional Supplemental Appropriations for Disaster Relief Act, 2019 (Public Law 116-20; 133 Stat. 871).