Estuarine environments worldwide are among the most threatened habitats due to increased disturbances resulting from coastal infrastructure and rising population densities. Boating activity is a primary disturbance, as it induces biological stress and morphological changes along the coastline. This high-energy environment that boat wakes create has resulted in loss of surrounding oyster reefs and salt marsh vegetation, ultimately leading to shoreline and habitat erosion. Here, we characterize the boat wake climate in the Intracoastal Waterway, assess the bathymetry in this heavily trafficked area, and anticipate the effects of experimental living shorelines (natural breakwall and oyster restoration structures) on facilitating sediment deposition and slowing vegetation retreat. Field observations indicate that boat wakes suspend nearshore sediment and can reach heights greater than 40 cm. A numerical stability model of the breakwalls suggests that the optimal porosity is field-specific. The desired porosity for minimizing lateral displacement is 0.50, while it is 0.18 for maximum energy dissipation, which indicates a need to further investigate this complex problem. These findings demonstrate that boat wakes significantly and regularly disturb estuarine shorelines and may be altering their bathymetry as well as suggest that the design of the breakwalls may be optimized to best counteract this pervasive disturbance.

Decreasing frequency of freeze events due to climate change is enabling the poleward range expansion of mangroves. As these tropical trees expand poleward, they are replacing herbaceous saltmarsh vegetation. Mangroves and saltmarsh vegetation are ecosystem engineers that are typically viewed as having similar ecosystem functions. However, few studies have investigated whether predation regimes, community structure, and ecosystem functions are shifting at the saltmarsh-mangrove ecotone. In this study, we manipulated predator access to marsh and mangrove creekside habitats to test their role in mediating vegetation and invertebrate structure and stability in a two-year experiment. We also conducted a survey to evaluate how shifting vegetation is modifying structural complexity, invertebrate communities, and ecosystem functioning at the ecotone. Excluding larger (> 2 cm diameter) predators did not affect vegetation or invertebrate structure or stability in either saltmarsh or mangrove habitats. The survey revealed that the two habitat types consistently differ in structural metrics, including vegetation height, inter-stem distance, and density, yet they support similar invertebrate and algal communities, soil properties, and predation rates. We conclude that although mangrove range expansion immediately modifies habitat structural properties, it is not altering larger predator consumptive effects, community stability, community composition, or some other ecosystem functions and properties at the ecotone.

Foundation species provide habitat to other organisms and enhance ecosystem functions, such as nutrient cycling, carbon storage and sequestration, and erosion control. We focus on freshwater wetlands because these ecosystems are often characterized by foundation species; eutrophication and other environmental changes may cause the loss of some of these species, thus severely damaging wetland ecosystems. To better understand how wetland primary producer foundation species support other species and ecosystem functions across environmental gradients, we reviewed ~150 studies in subtropical, boreal, and temperate freshwater wetlands. We look at how the relative dominance of conspicuous and well-documented species (i.e., sawgrass, benthic diatoms and cyanobacteria, Sphagnum mosses, and bald cypress) and the foundational roles they play interact with hydrology, nutrient availability, and exposure to fire and salinity in representative wetlands. Based on the evidence analyzed, we argue that the foundation species concept should be more broadly applied to include organisms that regulate ecosystems at different spatial scales, notably the microscopic benthic algae that critically support associated communities and mediate freshwater wetlands’ ecosystem functioning. We give recommendations on how further research efforts can be prioritized to best inform the conservation of foundation species and of the freshwater wetlands they support.

Tidal creeks play a vital role in influencing geospatial evolution and marsh ecological communities in coastal landscapes. However, evaluating the geospatial characteristics of numerous creeks across a site and understanding their ecological relationships pose significant challenges due to the labor-intensive nature of manual delineation from imagery. Traditional methods rely on manual annotation in GIS interfaces, which is slow and tedious. This study explores the application of Attention-based Dense U-Net (ADU-Net), a deep learning image segmentation model, for automatically classifying creek pixels in high-resolution (0.5 m) orthorectified aerial imagery in coastal Georgia, USA. We observed that ADU-Net achieved an outstanding F1 score of $0.98$ in identifying creek pixels, demonstrating its ability in tidal creek mapping. The study highlights the potential of deep learning models for automated tidal creek mapping, opening avenues for future investigations into the role of creeks in marshes’ response to environmental changes.

Transformers are widely used deep learning architectures. Existing transformers are mostly designed for sequences (texts or time series), images or videos, and graphs. This paper proposes a novel transformer model for massive (up to a million) point samples in continuous space. Such data are ubiquitous in environment sciences (e.g., sensor observations), numerical simulations (e.g., particle-laden flow, astrophysics), and location-based services (e.g., POIs and trajectories). However, designing a transformer for massive spatial points is non-trivial due to several challenges, including implicit long-range and multi-scale dependency on irregular points in continuous space, a non-uniform point distribution, the potential high computational costs of calculating all-pair attention across massive points, and the risks of over-confident predictions due to varying point density. To address these challenges, we propose a new hierarchical spatial transformer model, which includes multi-resolution representation learning within a quad-tree hierarchy and efficient spatial attention via coarse approximation. We also design an uncertainty quantification branch to estimate prediction confidence related to input feature noise and point sparsity. We provide a theoretical analysis of computational time complexity and memory costs. Extensive experiments on both real-world and synthetic datasets show that our method outperforms multiple baselines in prediction accuracy and our model can scale up to one million points on one NVIDIA A100 GPU. The code is available at \url{https://github.com/spatialdatasciencegroup/HST}.