California coastal dune inventory
Data files
Jun 09, 2026 version files 14.91 MB
-
Dataset_S2.tar.gz
14.90 MB
-
README.md
15.42 KB
Jun 09, 2026 version files 14.86 MB
-
Data_set_S2.tar.gz
14.84 MB
-
README.md
17.48 KB
Abstract
This dataset includes information on the historical (mid-to-late 1800s) and existing (2016) extent of coastal sand dunes in California, as well as the main drivers of loss and change over time. Information on the location and elevation of key dune features (i.e., toe and crest) recorded at 30-m intervals along the entire California coast is also provided.
Dataset DOI: 10.5061/dryad.hqbzkh1w7
Description of the data
This dataset includes information on the historical (mid-to-late 1800s) and existing (2016) extent of coastal sand dunes in California, as well as the main drivers of loss and change over time. The location and elevation of key dune features (i.e., toe and crest) recorded at 30-m intervals along the entire California coast is also included.
High-resolution scans of historical topographic maps or 'T-sheets' produced between 1845 and 1934 were used to generate regional (i.e., southern, central, and northern California) inventories of historical dunes and related coastal ecosystems. North American Datum (NAD) 1927 graticules spaced at one-minute intervals were used to digitally align the T-sheets before they were projected to NAD 1983 (2011) using ArcGIS Pro software (version 3.2.2). Different habitat types were identified and mapped using manual, heads-up digitization. To generate regional inventories of contemporary dunes, multispectral aerial imagery (NAIP 4-Band Imagery) and elevation data (2016 USGS West Coast El-Nino LiDAR DEM) were used in combination with a machine learning tool to map the location and elevation (North American Vertical Datum 1988) of the foredune toe and crest at 30-m intervals along the coast. Dune toe features (n = 10,000) were used to define the seaward boundary of individual dune systems. In contrast, the inland dune boundary was determined using aerial imagery, NDVI data (NAIP 4-Band Imagery), and elevation data downloaded from the NOAA Data Access Viewer website. Heads-up digitization was used to create line vectors of the inland boundary. These vectors were then connected to the dune toe features to create polygons of present-day dune extent and related features in ArcGIS Pro. Difference maps showing areas of dune loss and/or change over time were created by comparing the historical (1800s) and existing (2016) dune inventories, with specific drivers of change determined for each site using satellite imagery and historical archives (e.g., aerial photographs, maps, environmental reports).
Files and variables
Title of Dataset: Data_set_S2.tar.gz
Author(s): Timothy I. Baxter, Ian J. Walker, Jenifer E. Dugan, David M. Hubbard, Laura Engeman, Kyle A. Emery, Sean Vitousek, Karina K. Johnston, Andrea J. Pickart, Sarah Smith, Dakota R. Fee, Dan Willett, Jenna Wisniewski.
Corresponding Author Email Address: timothy.baxter@prm.ox.ac.uk
Corresponding Author Institutional Affiliation: University of California Santa Barbara and the Pitt Rivers Museum (University of Oxford)
Description: Shapefiles (.SHP, .CPG, .DBF, .PRJ, .SBN, .SBX, .SHX) showing the historical (mid-to-late 1800s) and existing (2016) extent of coastal sand dunes in California, as well as the main drivers of loss and change over time. The present-day (2016) location and elevation of key dune features (i.e., toe and crest) is also shown.
Files/Folders in Dataset: Four folders are included in the dataset: 'Dune_2016_extent', 'Dune_historical_extent', 'Dune_difference', and 'Dune_toe_and_crest'. Each folder contains data related to southern California (SoCal), central California (CenCal), and northern California (NoCal).
Title of Folder: Dune_2016_extent
Title of File: SoCal_2016_extent
Description: Extent of coastal sand dunes in southern California in 2016.
Projected Coordinate System: NAD 1983 (2011) UTM Zone 11N
Geographic Coordinate System: NAD 1983 (2011)
Attributes:
- FID = Identification number
- Shape = Polygon ZM
- Shape_Area = Habitat extent in m2
- Area_km = Habitat extent in km2
- Habitat = Habitat type:
- Dune = Coastal sand dunes
- Dune-Grassland transition = Transitional environment between dunes and grassland
- Dune-Wetland transition = Transitional environment between dunes and wetland
- Infrastructure = Developed land
Title of File: CenCal_2016_extent
Description: Extent of coastal sand dunes in central California in 2016.
Projected Coordinate System: NAD 1983 (2011) UTM Zone 10N
Geographic Coordinate System: NAD 1983 (2011)
Attributes:
- FID = Identification number
- Shape = Polygon ZM
- Shape_Area = Habitat extent in m2
- Area_km = Habitat extent in km2
- Habitat = Habitat type:
- Dune = Coastal sand dunes
- Dune (Mined) = Sand mining site
- Infrastructure = Developed land
- Water = Bodies of water
Title of File: NoCal_2016_extent
Description: Extent of coastal sand dunes in northern California in 2016.
Projected Coordinate System: NAD 1983 (2011) UTM Zone 10N
Geographic Coordinate System: NAD 1983 (2011)
Attributes:
- FID = Identification number
- Shape = Polygon ZM
- Shape_Area = Habitat extent in m2
- Area_km = Habitat extent in km2
- Habitat = Habitat type:
- Dune = Coastal sand dunes
- Beach-Dune transition = Transitional environment between beach and dune ecosystems
- Dune-Bluff transition = Transitional environment between dunes and coastal bluffs
- Dune-Grassland transition = Transitional environment between dunes and grassland
- Dune-Wetland transition = Transitional environment between dunes and wetland
- Grassland-Woodland = Grassland or woodland habitat
- Infrastructure = Developed land
- Water = Bodies of water
Title of Folder: Dune_historical_extent
Title of File: SoCal_historical_extent
Description: Extent of coastal sand dunes in southern California in mid-to-late 1800s.
Projected Coordinate System: NAD 1983 (2011) UTM Zone 11N
Geographic Coordinate System: NAD 1983 (2011)
Attributes:
- FID = Identification number
- Shape = Polygon ZM
- Shape_Area = Habitat extent in m2
- Area_km = Habitat extent in km2
- T_sheet = Name of topographic map ('T-sheet') or other historical resource used to map habitat extent
- Habitat = Habitat type:
- Dune = Active coastal sand dunes
- Dune (vegetated) = Stable, vegetated dunes
- Dune (relict) = Coastal dunes that formed and stabilized in the Pleistocene Period
- Beach-Dune transition = Transitional environment between beach and dune ecosystems
- Dune-Wetland transition = Transitional environment between dunes and wetland
- Estuary-Beach transition = Transitional environment between estuarine and beach environments that may have supported ephemeral dune ecosystems
- Water = Bodies of water
Title of File: CenCal_historical_extent
Description: Extent of coastal sand dunes in central California in mid-to-late 1800s.
Projected Coordinate System: NAD 1983 (2011) UTM Zone 10N
Geographic Coordinate System: NAD 1983 (2011)
Attributes:
- FID = Identification number
- Shape = Polygon ZM
- Shape_Area = Habitat extent in m2
- Area_km = Habitat extent in km2
- T_sheet = Name of topographic map ('T-sheet') or other historical resource used to map habitat extent
- Habitat = Habitat type:
- Dune = Active coastal sand dunes
- Dune (vegetated) = Stable, vegetated dunes
- Dune (relict) = Coastal dunes that formed and stabilized in the Pleistocene Period
- Beach-Dune transition = Transitional environment between beach and dune ecosystems
- Dune-Bluff transition = Transitional environment between dunes and coastal bluffs
- Estuary-Beach transition = Transitional environment between estuarine and beach environments that may have supported ephemeral dune ecosystems
- Water = Bodies of water
- Infrastructure = Developed land
Title of File: NoCal_historical_extent
Description: Extent of coastal sand dunes in northern California in mid-to-late 1800s.
Projected Coordinate System: NAD 1983 (2011) UTM Zone 10N
Geographic Coordinate System: NAD 1983 (2011)
Attributes:
- FID = Identification number
- Shape = Polygon ZM
- Shape_Area = Habitat extent in m2
- Area_km = Habitat extent in km2
- T_sheet = Name of topographic map ('T-sheet') or other historical resource used to map habitat extent
- Habitat = Habitat type:
- Dune = Active coastal sand dunes
- Dune (vegetated) = Stable, vegetated dunes
- Beach-Dune transition = Transitional environment between beach and dune ecosystems
- Dune-Wetland transition = Transitional environment between dunes and wetland
- Estuary-Beach transition = Transitional environment between estuarine and beach environments that may have supported ephemeral dune ecosystems
- Grassland-Woodland = Grassland or woodland habitat
- Water = Bodies of water.
- Infrastructure = Developed land
Title of Folder: Dune_difference
Title of File: SoCal_difference
Description: The main drivers of dune ecosystem loss and change in southern California from the mid-to-late 1800s to 2016.
Projected Coordinate System: NAD 1983 (2011) UTM Zone 11N
Geographic Coordinate System: NAD 1983 (2011)
Attributes:
- FID = Identification number
- Shape = Polygon ZM
- Area_km = Habitat extent in km2
- Difference = Patterns and drivers of change
- Growth (Dune) = Dune growth since the mid-to-late 1800s
- No Change (Dune) = Areas of dune that have not changed in areal extent since the mid-to-late 1800s
- Loss (Dune) -> Beach or Erosion = Dune loss due to coastal erosion since the mid-to-late 1800s
- Loss (Dune) -> Infrastructure = Dune loss due to infrastructure development or conversion to agricultural land since the mid-to-late 1800s
- Loss (Dune) -> Habitat change = Dune loss due to the stabilization of dune ecosystems or transformation into wetland, grassland, or woodland environments since the mid-to-late 1800s
- Loss (Dune) -> Water = Dune loss due to riverine erosion or development of waterways
- Infrastructure = Infrastructure not built on historical dune ecosystems
Title of File: CenCal_difference
Description: The main drivers of dune ecosystem loss and change in central California from the mid-to-late 1800s to 2016.
Projected Coordinate System: NAD 1983 (2011) UTM Zone 10N
Geographic Coordinate System: NAD 1983 (2011)
Attributes:
- FID = Identification number
- Shape = Polygon ZM
- Area_km = Habitat extent in km2
- Difference = Patterns and drivers of change
- Growth (Dune) = Dune growth since the mid-to-late 1800s
- No Change (Dune) = Areas of dune that have not changed in areal extent since the mid-to-late 1800s
- No Change (Dune relict) = Areas of relict dune that have not changed in areal extent since the mid-to-late 1800s
- Loss (Dune) -> Beach or Erosion = Dune loss due to coastal erosion since the mid-to-late 1800s
- Loss (Dune) -> Infrastructure = Dune loss due to infrastructure development or conversion to agricultural land since the mid-to-late 1800s
- Loss (Dune) -> Habitat change = Dune loss due to the stabilization of dune ecosystems or transformation into wetland, grassland, or woodland environments since the mid-to-late 1800s
- Loss (Dune) -> Water = Dune loss due to riverine erosion or development of waterways
- Infrastructure = Infrastructure not built on historical dune ecosystems
Title of File: NoCal_difference
Description: The main drivers of dune ecosystem loss and change in northern California from the mid-to-late 1800s to 2016.
Projected Coordinate System: NAD 1983 (2011) UTM Zone 10N
Geographic Coordinate System: NAD 1983 (2011)
Attributes:
- FID = Identification number
- Shape = Polygon ZM
- Area_km = Habitat extent in km2
- Difference = Patterns and drivers of change
- Growth (Dune) = Dune growth since the mid-to-late 1800s
- No Change (Dune) = Areas of dune that have not changed in areal extent since the mid-to-late 1800s
- Loss (Dune) -> Beach or Erosion = Dune loss due to coastal erosion since the mid-to-late 1800s
- Loss (Dune) -> Infrastructure = Dune loss due to infrastructure development or conversion to agricultural land since the mid-to-late 1800s
- Loss (Dune) -> Habitat change = Dune loss due to the stabilization of dune ecosystems or transformation into wetland, grassland, or woodland environments since the mid-to-late 1800s
- Loss (Dune) -> Water = Dune loss due to riverine erosion or development of waterways
- Infrastructure = Infrastructure not built on historical dune ecosystems
Title of Folder: Dune_toe_and_crest
Title of File: SoCal_Dune_crest_2016
File Type: Point Shapefile
Description: The location and elevation of the foredune crest recorded at 30 m intervals along the coastline of southern California.
Projected Coordinate System: NAD 1983 (2011) UTM Zone 11N
Geographic Coordinate System: NAD 1983 (2011)
Attributes:
- FID = Identification number
- Shape = Point ZM
- lat = Latitude
- lon = Longitude
- easting = Easting
- northing = Northing
- elevation = North American Vertical Datum 1988
- BLdistance = Distance (m) along the coastline from the US-Mexico border
Title of File: CenCal_Dune_crest_2016
Description: The location and elevation of the foredune crest recorded at 30 m intervals along the coastline of central California.
Projected Coordinate System: NAD 1983 (2011) UTM Zone 10N
Geographic Coordinate System: NAD 1983 (2011)
Attributes:
- FID = Identification number
- Shape = Point ZM
- lat = Latitude
- lon = Longitude
- easting = Easting
- northing = Northing
- elevation = North American Vertical Datum 1988
- BLdistance = Distance (m) along the coastline from the US-Mexico border
Title of File: NoCal_Dune_crest_2016
Description: The location and elevation of the foredune crest recorded at 30 m intervals along the coastline of northern California.
Projected Coordinate System: NAD 1983 (2011) UTM Zone 10N
Geographic Coordinate System: NAD 1983 (2011)
Attributes:
- FID = Identification number
- Shape = Point ZM
- lat = Latitude
- lon = Longitude
- easting = Easting
- northing = Northing
- elevation = North American Vertical Datum 1988
- BLdistance = Distance (m) along the coastline from the US-Mexico border
Title of File: SoCal_Dune_toe_2016
Description: The location and elevation of the foredune toe recorded at 30 m intervals along the coastline of southern California.
Projected Coordinate System: NAD 1983 (2011) UTM Zone 11N
Geographic Coordinate System: NAD 1983 (2011)
Attributes:
- FID = Identification number
- Shape = Point ZM
- lat = Latitude
- lon = Longitude
- easting = Easting
- northing = Northing
- elevation = North American Vertical Datum 1988
- BLdistance = Distance (m) along the coastline from the US-Mexico border
Title of File: CenCal_Dune_toe_2016
Description: The location and elevation of the foredune toe recorded at 30 m intervals along the coastline of central California.
Projected Coordinate System: NAD 1983 (2011) UTM Zone 10N
Geographic Coordinate System: NAD 1983 (2011)
Attributes:
- FID = Identification number
- Shape = Point ZM
- lat = Latitude
- lon = Longitude
- easting = Easting
- northing = Northing
- elevation = North American Vertical Datum 1988
- BLdistance = Distance (m) along the coastline from the US-Mexico border
Title of File: NoCal_Dune_toe_2016
Description: The location and elevation of the foredune toe recorded at 30 m intervals along the coastline of northern California.
Projected Coordinate System: NAD 1983 (2011) UTM Zone 10N
Geographic Coordinate System: NAD 1983 (2011)
Attributes:
- FID = Identification number
- Shape = Point ZM
- lat = Latitude
- lon = Longitude
- easting = Easting
- northing = Northing
- elevation = North American Vertical Datum 1988
- BLdistance = Distance (m) along the coastline from the US-Mexico border
Code/software
Shapefiles can be accessed using a range of software, including ArcGIS Pro, QGIS, MATLAB, and Python.
Access information
Data were derived from the following sources:
- Dataset: 2016 USGS West Coast El-Nino Lidar DEM (WA, OR, CA)
- Citation: Office for Coastal Management, 2025: 2016 USGS West Coast El-Nino Lidar DEM (WA, OR, CA)
- Link: https://www.fisheries.noaa.gov/inport/item/48383
- Data Use Constraints: None. However, users should be aware that temporal changes may have occurred since this data set was collected and that some parts of these data may no longer represent actual surface conditions. This data were produced for the USGS according to specific project requirements. This information is provided "as is". Further documentation of this data can be obtained by contacting: U.S. Geological Survey, PO Box 25046, MS 510, Denver, CO 80225. Telephone (303) 202-4419.
- Dataset: 2016 Coastal California NAIP 4-Band 8 Bit Imagery
- Citation: OCM Partners, 2025: 2016 Coastal California NAIP 4-Band 8 Bit Imagery
- Link: https://www.fisheries.noaa.gov/inport/item/53106
- Data Use Constraints: None.
- Dataset: NOAA Index of Non-Georeferenced Manuscripts.
- Link: https://geodesy.noaa.gov/pub/Shoreline/T-Sheet%20Raster%20Images/nongeomanu_index.html
- Data Use Constraints: None.
Mapping Historical Dune Extent
High-resolution scans of T-sheets (n = 140) produced between 1845 and 1934 were obtained from the NOAA Index of Non-Georeferenced Manuscripts and the National Archives Records Administration (NARA). We generated a historical inventory of coastal sand dunes through a systematic process of georectification, vectorization, interpretation, and classification.
North American Datum (NAD) 1927 graticules spaced at one-minute intervals were used to digitally align the T-sheets, with an average of 11 control points used to georeference each map. The average (mean) root-mean-squared-error (RMSE) derived from these control points was 000049 ± 0.000030 decimal degrees (mean ± SD). The T-sheets were then reprojected to NAD 1983 (2011) using ArcGIS Pro software (version 3.2.2). To reduce the potential for major positional errors, a representative sample of T-sheets (n = 66) from across California were also examined against distinct corresponding features (e.g., road intersections, rocky outcrops, buildings) identified in contemporary aerial imagery (NAIP 4-Band Imagery) downloaded from the NOAA Data Access Viewer website (coast.noaa.gov/dataviewer). Based on five control points per T-Sheet, the average RMSE for these maps was 8.16 ± 3.42 m (mean ± SD).
In cases where T-sheets were unavailable, damaged, or insufficiently detailed, the historical location and extent of coastal dunes was determined and/or reviewed using additional resources. This includes early aerial photographs obtained from the University of California Santa Barbara (UCSB) Library, oblique aerial images and terrestrial photographs obtained from public libraries and local historical societies, regional-scale maps of dune extent, historical topographic maps produced by the US Coast Guard and U.S. Geological Survey (USGS), and geological maps produced by the USGS, California Geological Survey, and Dibblee Geology Center. While many of these resources were obtained in formats containing geospatial information (e.g., GeoTIFF), several maps and photos required georectification using latitude/longitude graticules or fixed, physical features identifiable in contemporary imagery.
Once the historical maps and photographs were georeferenced, vectors of different habitat types were created using manual (heads-up) digitization. In cases where T-sheets overlapped, earlier maps were relied on to resolve differences, except when later maps were more detailed. Although a full suite of coastal environments is shown in T-sheets, we focused on mapping coastal dunes and related features. In total, over 1,500 polygons were mapped, interpreted, and classified into 11 different habitat types. This includes active dunes, stable (vegetated) dunes, and relict dunes (i.e., those formed over 11,700 years ago during the Pleistocene), wetland environments (e.g., swales), grassland and/or forested areas, and infrastructure (e.g., buildings and roads) as well as ‘transitional’ environments which, due to their location, environmental setting, and proximity to sediment source, may have supported ephemeral dune systems. These transition zones include spits and barriers close to river mouths (‘Estuary–Beach’ environments), beaches supporting incipient dunes without a clearly defined foredune (‘Beach–Dune’ environments) or at the base of coastal bluffs (‘Dune–Bluff’ environments), and ephemeral wetlands located within or adjacent to dune fields (‘Dune–Wetland’ environments).
Mapping Existing Dune Extent
High-resolution (0.5–1m) aerial imagery (2016 NAIP 4-Band Imagery) and oblique photographs from the California Coastal Records Project (www.californiacoastline.org) were used to conduct a general reconnaissance of the entire California coast and create an initial list of potential dune sites. At each location where dunes were identified, elevation data (2016 USGS West Coast El-Nino LiDAR DEM) were used in combination with HAML—Human Attuned Machine Learning Tool for Dune Annotation—to map the location and elevation (North American Vertical Datum 1988) of foredune toe and crest features at 30-m intervals along the coastline using a 1625-km-long baseline created for the 2024 USGS National Shoreline Change project.
HAML is a custom ArcGIS plugin developed by the U.S. Naval Research Laboratory that facilitates the quick and accurate identification and annotation of key dune features through the creation of elevation profiles derived from a user-provided digital elevation model (DEM). When first initialized, the tool uses algorithmic annotation to place features along shore-perpendicular profiles: the dune crest and toe are defined as the highest point of elevation and location of maximum change in downslope curvature, respectively. Users can adjust annotated features on newly generated profiles through a simple click-and-drag interface. Once the tool has accumulated a reasonable amount of training data (achieved after approximately 30 profiles), the machine-learning backend initiates, providing updated suggestions for crest and toe placement, while still allowing user edits.
The positional accuracy of features mapped using HAML depends on the horizontal and vertical accuracy of the DEM used to generate the elevation profiles and the amount of human input used verify and/or adjust the placement of features initially located by the tool. In this study, we used HAML to help identify the location and height of nearly 20,000 dune toe and crest features using elevation data (2016 USGS West Coast El-Nino LiDAR DEM) with a horizontal accuracy of 0.21 m, vertical accuracy of 0.12 m, and resolution of 0.5 m. The position of each feature was individually reviewed by an expert in coastal geomorphology on at least two occasions: once following its initial placement by HAML, and again when the entire dataset for a region had been created. In most cases, only small adjustments (i.e., 1–2 m) were necessary. However, on occasion, more substantial changes (i.e., 10–20 m) were required when features were incorrectly placed by HAML. Given the accuracy and resolution of the 2016 DEM used for selection and the two‐step review processes, we believe the toe and crest features mapped in this study can be considered accurate to approximately 1 m.
Following the review process, toe features were connected and used to define the seaward boundary of dune systems. In contrast, the inland dune boundary was manually identified using elevation data downloaded from the NOAAs Data Access Viewer website. For smaller dune systems, the inland boundary was determined using the same DEM used to identify toe and crest features (i.e., 2016 USGS West Coast ElNino LiDAR DEM). However, as the 2016 dataset only extends for approximately 500 m inland, alternative DEMS were used to determine the inland extent of larger dune systems. The horizontal and vertical accuracy of the DEMs used to map these larger systems ranged between 0.12–1.00 m and 0.05–0.20 m, respectively. However, in some locations with limited data availability, less accurate (i.e., > 2m), older DEMs were used (e.g., 2002–2003 IfSAR data for Southern California DEM).
Line vectors of the inland dune boundary were created using manual (heads‐up) digitization in ArcGIS Pro (version 3.2.2; Esri, Redlands, California, USA) based on changes in slope and elevation shown in the DEMs (i.e., at the base of the lee slope of the backdune). In some cases, high‐resolution aerial imagery (NAIP 4‐Band Im agery; EROS, 2023) and NDVI data were also used, especially in low‐lying areas where changes in land cover and/or plant communities provided a better indication of dune and non‐dune areas (i.e., dune to wetland eco systems). The accuracy of the inland dune boundary varies by region depending on the accuracy and resolution of the DEM referenced, quality and availability of supporting data sets (i.e., aerial images), and type of boundary assessed, with those defined by clear changes in elevation and/or land‐use (i.e., dune to infrastructure) generally considered more accurate (<2 m error) than those produced for gently sloping, transitional environments (i.e., dune to wetlands, 5–15 m error). Given these uncertainties, the inland dune boundary should be considered broadly accurate to approximately 5–20 m across the State.
To create polygons of contemporary (2016) dune extent, the inland dune boundary and seaward dune boundary were connected in ArcGIS Pro. Intotal, over 1,400 polygons were generated. These polygons were then classified into 10 different habitat types (Figure S2 in Supporting Information S1). Most of these habitat types correspond directly to those identified in the historical inventory, including infrastructure, wetland, and transitional envi ronments (e.g., Dune–Bluff, Dune–Wetland, and Dune–Grassland transition zones). However, due to challenges associated with accurately differentiating between active and stable (vegetated) dunes using aerial imagery, these habitats were combined into a single “dune” category in the contemporary inventory. In general, habitat type was classified using aerial imagery from 2016 (i.e., 2016 NAIP 4‐Band Imagery; EROS, 2023) as this year was the most comprehensive in terms of data coverage, accuracy (i.e., 6 m), and resolution (i.e., 1 m). However, in a few isolated cases, imagery from 2002 to 2022 was also used in areas not covered by the 2016 data set, such as at Vandenberg Space Force Base.
Site visits (n = 29) conducted throughout 2024 and 2025 were used to visually assess the location of dune systems using fixed landmarks, such as roads and buildings. These surveys were particularly beneficial in helping to identify and record the presence/absence of smaller dune sites that may have been missed during our initial reconnaissance of the coastline using aerial imagery (e.g., dunes on Casper Headlands State Beach, Mendocino County). Feedback on the contemporary dune inventory was also gathered from local managers, practitioners, and members of the California Dune Science Network (www.resilientcoastlines.com) who were engaged through a series (n=10) of targeted workshops held across California. During these workshops, participants were provided with an opportunity to comment on our preliminary findings and suggest additional sites for analysis. Feedback from these workshops was then used, in combination with the site visits, to produce a more exhaustive list of dune systems across the State and gather site specific information on potential drivers of change.
Difference Maps
Difference maps showing areas of dune loss and/or change over time were created by comparing the historical (1800s) and contemporary (2016) dune inventories, with specific drivers of change determined by identifying the dominant land-use and/or habitat type shown in modern aerial imagery (USDA NAIP 4-Band). For example, infrastructure development was considered to be the main driver of change in areas where historical dunes had been replaced by residential buildings or roadways. In some cases, historical archives (i.e., aerial photographs, maps, and environmental reports) were also used to provide additional context.
Drivers of change were classified into the following categories: (a) loss through coastal erosion, (b) loss through river erosion or the redirection of waterways, (c) loss through processes that facilitated the transformation of historical dunes to wetland, grassland, or forested environments, (d) growth through the seaward progradation of foredunes, and (e) growth through the landward migration of dune systems due to aeolian transport. Changes in vegetation did not result in the classification of historical dunes as “lost” in isolation. Loss was only considered for areas deemed to be both ecologically and geomorphologically distinct from adjacent dune ecosystems. This includes, for example, wetland environments and low‐lying grassland and/or forested areas with little or no topography. As it was generally difficult to determine a single, causal mechanism that may have facilitated the transformation of historical dunes to wetland, grassland, or forested areas—whether through environmental, geomorphic, or anthropogenic drivers—these transitions where collectively classified as “habitat change” in the inventory. This contrasts to historical dunes lost through human development and land‐use change (e.g., industry, residential communities, transport links, agricultural land), that were more readily identifiable using high‐resolution (1 m) aerial imagery and land cover data (i.e., Impervious Cover) from the NOAA Coastal Change Analysis Program.
Changes after Jun 9, 2026: Minor grammatical and spelling errors in the README files have been rectified.