Salt Marsh Ditches, Version 3.0, North Atlantic U.S. Coast

The data included are salt marsh ditches, a salt marsh ditch metric, and the data sources for the metric. The majority of salt marshes in the Northeast have been ditched, both to facilitate harvest of salt marsh hay and to control mosquitoes. Ditching changes the hydrology and flows of sediment and nutrients of marshes in ways that are not well understood, though ditched marshes may have altered invertebrate and shorebird communities, and may be less resilient to sea level rise. Marshes with intensive ditching (ca. 10 m spacing) appear to be most strongly affected.

The first dataset, “Salt Marsh Ditches (Lines), Version 3.0, North Atlantic U.S. Coast”, depicts North Atlantic salt marsh ditches at least 75 m in length. Results encompass about 64% of the coastal northeast (based on locations where high resolution LiDAR data were available, described in the dataset “Salt Marsh Ditch Sources and Status Map, Version 3.0, North Atlantic U.S. Coast”).

The second dataset, “Salt Marsh Ditch Metric, Version 3.1, North Atlantic U.S. Coast”, provides information about the potential magnitude of ditching, and impacts of ditching, in salt marshes. Values vary from 0 (no effect from salt marsh ditching) to 1 (severe effect). Salt marshes with a high ditching score are more likely to be degraded than those with a low score, all other factors being equal. Results encompass about 64% of the coastal northeast (based on locations where high resolution LiDAR data were available, described in the dataset “Salt Marsh Ditch Sources and Status, Version 3.0, North Atlantic U.S. Coast” below).

Using three kilometer by three kilometer tiles, the third dataset, “Salt Marsh Ditch Sources and Status, Version 3.0, North Atlantic U.S. Coast”, depicts the sources of the high resolution LiDAR-based elevation data used to create the dataset “Salt Marsh Ditch Metric, Version 3.1, North Atlantic U.S. Coast.” High resolution (one meter) LiDAR data were available for about 64% of the coastal northeast coast; “(No data)” is used to denote the remaining tiles where LiDAR were unavailable. For details on the dataset see the technical documentation: https://scholarworks.umass.edu/cgi/viewcontent.cgi?filename=0&article=1030&context=data&type=additional.

The data are part of a larger set of results developed by the Designing Sustainable Landscapes project https://scholarworks.umass.edu/designing_sustainable_landscapes/ led by Professor Kevin McGarigal of UMass Amherst. Funding for this project was provided by the North Atlantic Landscape Conservation Cooperative and Department of the Interior Project #24, Decision Support for Hurricane Sandy Restoration and Future Conservation to Increase Resiliency of Tidal Wetland Habitats and Species in the Face of Storms and Sea Level Rise.

Intended Uses

The data provide information about the potential magnitude of ditching, and impacts of ditching, in salt marshes. Salt marshes with a high ditching score are more likely to be degraded than those with a low score, all other factors being equal. Due to known uncertainties in the mapping process (summarized below) and in the uncertainties about the impacts of mapped ditches, the datasets may be best used as screening tools to summarize the extent and possible impacts of ditching on salt marsh function. They may also point users to locations that merit more in-depth, field-based investigation.

Description and Derivation

The derivation of the datasets “Salt Marsh Ditch Sources and Status,” “Salt Marsh Ditches,” and “Salt Marsh Ditch Metric” datasets are described in detailed technical documentation. https://scholarworks.umass.edu/cgi/viewcontent.cgi?filename=0&article=1030&context=data&type=additional Briefly, Digital Elevation Models (DEMs) at 1 m resolution derived from LiDAR were compiled from a number of state and federal sources described in the Salt Marsh Ditch Sources and Status dataset. A custom image analysis process was then used to identify local depressions in salt marshes that could be ditches. The results were next analyzed to tag long (>75 m), fairly straight depressions as ditches using “morphological skeletonizing” and “clockfacing” algorithms. These mapped ditches constitute the “Salt Marsh Ditches” dataset. Finally, the Salt Marsh Ditch Metric measured the intensity of ditches in the neighborhood of each salt marsh cell using a kernel estimator.

Known Issues and Uncertainties

As with any project carried out across such a large area, the salt marsh ditching project is subject to limitations. The results by themselves are not a prescription for on-the-ground action; users are encouraged to verify, with field visits and site-specific knowledge, the importance and impacts of salt marsh ditches identified in the datasets. Known issues and uncertainties include the following (see the technical documentation https://scholarworks.umass.edu/cgi/viewcontent.cgi?filename=0&article=1030&context=data&type=additional for a more complete discussion):

• Salt marsh ditches were mapped only where high resolution LiDAR were available, which consisted of 64% of the 3 km tiles containing salt marshes.

• The ditch identification process is believed to successfully map most ditches but it does not find all of them (balancing the trade-offs between correctly finding real ditches and minimizing incorrect mapping of non-ditches). Cases where ditches are not or may not be mapped include: a) ditches are shorter than 75 m (all are excluded); b) ditches are very narrow (approximately <1 m in width); c) ditches are not straight (deviate >5% from a straight line); d) ditches are very shallow; e) ditches occur in areas with “noisy” LiDAR (most common in the southern part of the region); and f) ditches occur in other areas with artifacts in LiDAR data or marsh mapping data.

• Some natural creek sections and other artifacts may be incorrectly identified as ditches, particularly when they are long and straight, but these situations are expected to be rare.