Search

The world's coastlines are spatially highly variable, coupled-human-natural systems that comprise a nested hierarchy of component landforms, ecosystems, and human interventions, each interacting over a range of space and time scales. Understanding and predicting coastline dynamics necessitates frequent observation from imaging sensors on remote sensing platforms. Machine Learning models that carry out supervised (i.e., human-guided) pixel-based classification, or image segmentation, have transformative applications in spatio-temporal mapping of dynamic environments, including transient coastal landforms, sediments, habitats, waterbodies, and water flows. However, these models require large and well-documented training and testing datasets consisting of labeled imagery. We describe "Coast Train," a multi-labeler dataset of orthomosaic and satellite images of coastal environments and corresponding labels. These data include imagery that are diverse in space and time, and contain 1.2 billion labeled pixels, representing over 3.6 million hectares. We use a human-in-the-loop tool especially designed for rapid and reproducible Earth surface image segmentation. Our approach permits image labeling by multiple labelers, in turn enabling quantification of pixel-level agreement over individual and collections of images., (© 2023. This is a U.S. Government work and not under copyright protection in the US; foreign copyright protection may apply.)

Coastal wetlands are not only among the world's most valued ecosystems but also among the most threatened by high greenhouse gas emissions that lead to accelerated sea level rise. There is intense debate regarding the extent to which landward migration of wetlands might compensate for seaward wetland losses. By integrating data from 166 estuaries across the conterminous United States, we show that landward migration of coastal wetlands will transform coastlines but not counter seaward losses. Two-thirds of potential migration is expected to occur at the expense of coastal freshwater wetlands, while the remaining one-third is expected to occur at the expense of valuable uplands, including croplands, forests, pastures, and grasslands. Our analyses underscore the need to better prepare for coastal transformations and net wetland loss due to rising seas.

Coastal wetlands are among the most productive and carbon-rich ecosystems on Earth. Long-term carbon storage in coastal wetlands occurs primarily belowground as soil organic matter (SOM). In addition to serving as a carbon sink, SOM influences wetland ecosystem structure, function, and stability. To anticipate and mitigate the effects of climate change, there is a need to advance understanding of environmental controls on wetland SOM. Here, we investigated the influence of four soil formation factors: climate, biota, parent materials, and topography. Along the northern Gulf of Mexico, we collected wetland plant and soil data across elevation and zonation gradients within 10 estuaries that span broad temperature and precipitation gradients. Our results highlight the importance of climate-plant controls and indicate that the influence of elevation is scale and location dependent. Coastal wetland plants are sensitive to climate change; small changes in temperature or precipitation can transform coastal wetland plant communities. Across the region, SOM was greatest in mangrove forests and in salt marshes dominated by graminoid plants. SOM was lower in salt flats that lacked vascular plants and in salt marshes dominated by succulent plants. We quantified strong relationships between precipitation, salinity, plant productivity, and SOM. Low precipitation leads to high salinity, which limits plant productivity and appears to constrain SOM accumulation. Our analyses use data from the Gulf of Mexico, but our results can be related to coastal wetlands across the globe and provide a foundation for predicting the ecological effects of future reductions in precipitation and freshwater availability. Coastal wetlands provide many ecosystem services that are SOM dependent and highly vulnerable to climate change. Collectively, our results indicate that future changes in SOM and plant productivity, regulated by cascading effects of precipitation on freshwater availability and salinity, could impact wetland stability and affect the supply of some wetland ecosystem services., (Published 2018. This article is a U.S. Government work and is in the public domain in the USA.)

Coastal wetland responses to sea-level rise are greatly influenced by biogeomorphic processes that affect wetland surface elevation. Small changes in elevation relative to sea level can lead to comparatively large changes in ecosystem structure, function, and stability. The surface elevation table-marker horizon (SET-MH) approach is being used globally to quantify the relative contributions of processes affecting wetland elevation change. Historically, SET-MH measurements have been obtained at local scales to address site-specific research questions. However, in the face of accelerated sea-level rise, there is an increasing need for elevation change network data that can be incorporated into regional ecological models and vulnerability assessments. In particular, there is a need for long-term, high-temporal resolution data that are strategically distributed across ecologically-relevant abiotic gradients. Here, we quantify the distribution of SET-MH stations along the northern Gulf of Mexico coast (USA) across political boundaries (states), wetland habitats, and ecologically-relevant abiotic gradients (i.e., gradients in temperature, precipitation, elevation, and relative sea-level rise). Our analyses identify areas with high SET-MH station densities as well as areas with notable gaps. Salt marshes, intermediate elevations, and colder areas with high rainfall have a high number of stations, while salt flat ecosystems, certain elevation zones, the mangrove-marsh ecotone, and hypersaline coastal areas with low rainfall have fewer stations. Due to rapid rates of wetland loss and relative sea-level rise, the state of Louisiana has the most extensive SET-MH station network in the region, and we provide several recent examples where data from Louisiana's network have been used to assess and compare wetland vulnerability to sea-level rise. Our findings represent the first attempt to examine spatial gaps in SET-MH coverage across abiotic gradients. Our analyses can be used to transform a broadly disseminated and unplanned collection of SET-MH stations into a coordinated and strategic regional network. This regional network would provide data for predicting and preparing for the responses of coastal wetlands to accelerated sea-level rise and other aspects of global change.

Due to their position at the land-sea interface, coastal wetlands are vulnerable to many aspects of climate change. However, climate change vulnerability assessments for coastal wetlands generally focus solely on sea-level rise without considering the effects of other facets of climate change. Across the globe and in all ecosystems, macroclimatic drivers (e.g., temperature and rainfall regimes) greatly influence ecosystem structure and function. Macroclimatic drivers have been the focus of climate change-related threat evaluations for terrestrial ecosystems, but largely ignored for coastal wetlands. In some coastal wetlands, changing macroclimatic conditions are expected to result in foundation plant species replacement, which would affect the supply of certain ecosystem goods and services and could affect ecosystem resilience. As examples, we highlight several ecological transition zones where small changes in macroclimatic conditions would result in comparatively large changes in coastal wetland ecosystem structure and function. Our intent in this communication is not to minimize the importance of sea-level rise. Rather, our overarching aim is to illustrate the need to also consider macroclimatic drivers within vulnerability assessments for coastal wetlands., (Published 2015. This article is a U.S. Government work and is in the public domain in the USA.)