Search for a command to run...
This dataset provides annual raster maps of ecosystem carbon stocks and fluxes following restoration within four National Park Service lands the Ackerson Meadow restoration project in Yosemite National Park, Captain John Smith Chesapeake National Historic Trail, Indiana Dunes National Park, and Cape Cod National Seashore. Data were produced using the Land Use and Carbon Simulator (LUCAS; Sleeter et al. 2022), which integrates a land cover change model with a carbon budget model. The model assesses the effect of land cover change (such as that through wetland restoration) on ecosystem carbon dynamics. Landscape sub-models were parameterized using data from the National Land Cover Dataset (NLCD) and run at 30-m resolution within the areas of interest. Upland forest classes in NLCD were further stratified into forest type groups using the Existing Vegetation Type Product (LANDFIRE, 2016). The Woody Wetland and Wetland classes from NLCD were further stratified into additional wetland classes (Palustrine Forested Wetland and Palustrine Emergent Wetland, respectively) using the NOAA Coastal Change Analysis Program (Office for Coastal Management, 2023). Using the Rangeland Condition Monitoring Assessment and Projection, the Shrubland and Grassland classes in NLCD were further stratified into additional grassland and shrubland classes (Rigge et al. 2021). LUCAS simulates the annual changes in the carbon pools that occur due to growth, turnover, decomposition, natural disturbances, and land management for multiple cover classes, including upland forests, forested wetlands, herbaceous wetlands, and agricultural lands among others (Sleeter et al., 2022). Updates to the wetland sub-models of LUCAS improved the representation of wetland-specific fluxes, including methane emissions, lateral flux of carbon into the aquatic pool, burial of carbon into the deep soil, and changes in decay dynamics. Wetland-specific carbon stocks and fluxes derived from USGS field studies (Stagg et al. 2017a, 2017b, 2017c, 2017d, 2017e; 2017f; Baustian et al. 2017, 2020a, 2020b, 2021a, 2021b; Krauss et al. 2018) were incorporated into the model. All analyses were conducted using the ST-Sim package (version 4.5.0) running under SyncroSim version 3.1.12 (ApexRMS, 2025). Park-specific Restoration Activities: Ackerson Meadow, Yosemite National Park: Ackerson Meadow is an area recently purchased and donated to Yosemite National Park. Prior to acquisition by the National Park Service, much of the existing wetland was drained through a gully network formed by timber harvesting, ranching, and farming activities. Restoration was completed in 2025. Herbaceous and woody plants were planted or transplanted as part of the restoration activities at Ackerson Meadow. We used a polygon of the restored area provided by Ecohydro (Ecohydro.org) in May 2025 to delineate the restoration footprint. To determine the land cover classes present within the analysis area, we used the land cover model in LUCAS and applied the following assumptions: (1) areas classified by LUCAS as forest were assumed to be restored to "Palustrine Forested Wetland," and (2) areas classified as shrubland or grassland were assumed to be restored to "Palustrine Emergent Wetland". Following restoration, the area of shrubland, grassland, and upland forest decreased, while the area of forested palustrine wetland and emergent palustrine wetland increased. Captain John Smith National Historic Trail: The historic site known as Werowocomoco was recently acquired by the National Park Service and is now managed by the Captain John Smith National Historic Trail. Prior to this acquisition, construction of a pond caused the formation of a sinkhole and loss of wetland habitat. Through streambed and wetland restoration, the pond will be removed and the wetland habitat restored in 2027. Because the pond was not included in the historical NLCD record, the pond was digitized using satellite imagery. We assumed that wetland restoration transitioned from "Water" to "Palustrine Emergent Wetland" in 2027. Following restoration, the area of water decreased while the area of palustrine emergent wetland increased. Indiana Dunes National Park: Restoration at Indiana Dunes National Park occurred in 2021 and included the removal of hybrid cattail (Typha spp.), common reed (Phragmites australis), and non-native shrubs in addition to the planting of native plants. There were no changes in land cover post-restoration because the species removed and those added were all classified within the same cover class ("Palustrine Emergent Wetland"). Cape Cod National Seashore: For the Cape Cod National Park data, we used an existing wetland type map from the National Park Service and a projected future wetland type map representing conditions under a scenario in which all tidal gates are opened to 10 feet in May 2025. We simulated wetland restoration to occur in 2040. Following restoration, the area of developed land, grassland, palustrine forested wetland, upland forest, and palustrine emergent wetland decreased while the area of water, unconsolidated shore, and estuarine emergent wetland increased. Scenarios: For all four parks, we simulated the carbon dynamics associated with two alternative scenarios of land cover: the first scenario assumed no restoration (i.e., land cover did not change over time due to restoration, “Without Wetland Restoration”), while the second scenario assumed land cover changed due to restoration according to the scheduled date of the planned/actual restoration activities (“With Wetland Restoration”). The effect of the restoration was then assumed to be the difference in carbon dynamics over time between these two scenarios. The simulation period for both scenarios was 2001 to 2075. To establish a starting value for carbon stocks, carbon was initialized in all land cover classes following Sleeter et al. (2022). To initialize carbon in the pond in Captain John Smith Chesapeake National Historic Trail, we assumed that the pond was originally forested wetland and converted to a pond in 2000. To initialize carbon in developed land in Cape Cod, we assumed that the developed land was originally palustrine forested wetland and was converted to developed land in 1900. In addition to simulating land cover changes due to restoration, we assessed the historical impact of land cover change on carbon dynamics from 2001 to 2021 in the analysis area prior to restoration. Historical land cover transitions can include agricultural expansion, agricultural contraction, fire, forest harvest, urbanization, and urban intensification. Carbon Metrics: The model output includes metrics of carbon stocks, fluxes, net fluxes, and cumulative metrics that can be used as indicators of restoration impact. The carbon sub-model accounts for 15 LUCAS ecosystem carbon stocks as well as all carbon fluxes across all stocks, including emissions of carbon dioxide (CO2) and methane (CH4) as well as lateral flux of dissolved carbon. The LUCAS model estimates the net flux of carbon to and from the atmosphere, accounting for all emissions (i.e. losses to the atmosphere and to the aquatic pool) and removals (i.e. gains in the above- and belowground pools) providing an estimate of the net ecosystem carbon balance (NECB; Chapin et al., 2006). If emissions are greater than removals, NECB values are negative, and the system is considered a carbon source. If emissions are less than removals, NECB values are positive, and the system is considered a carbon sink; also referred to as net carbon sequestration. This tool also aggregates carbon across the 15 LUCAS stocks into the live IPCC recommended carbon stocks (IPCC, 2006). Cumulative carbon metrics are used to quantify the impact of the restoration projects. Changes in carbon associated with restoration were estimated by subtracting the NECB values for the reference scenario (e.g., Without Wetland Restoration) from the restoration scenario (e.g., With Wetland Restoration). Positive changes in NECB represent a gain in net carbon sequestration associated with restoration.