Critical military training and testing on lands along the nation’s coastal and estuarine shorelines are increasingly placed at risk because of development pressures in surrounding areas, impairments due to other anthropogenic disturbances, and increasing requirements for compliance with environmental regulations. The U.S. Department of Defense (DoD) intends to enhance and sustain its training and testing assets and to optimize its stewardship of natural resources through the development and application of an ecosystem-based management approach on DoD installations. To accomplish this goal, particularly for installations in estuarine/coastal environments, the Strategic Environmental Research and Development Program (SERDP) launched the Defense Coastal/Estuarine Research Program (DCERP) as a minimum 10-year effort at Marine Corps Base Camp Lejeune (MCBCL) in North Carolina. The results of the first six years of the program (DCERP1) are presented here.

The overarching objectives of DCERP are to: (1) understand the effects of military training activities, infrastructure development, and other anthropogenic stressors, as well as natural disturbances, on the coastal ecosystems at MCBCL and other coastal military installations; (2) develop models, tools, and indicators to evaluate ecosystem health; and (3) recommend adaptive management strategies to sustain ecosystem natural resources within the context of an active military installation.

DCERP1 was implemented in two phases. Phase I of the program was a planning period that was conducted between November 2006 and June 2007 and resulted in the development of the DCERP Strategic Plan, the DCERP Baseline Monitoring Plan, and the DCERP Research Plan, which collectively serve as the foundation for DCERP activities at MCBCL. Implementation of these plans (Phase II) began in July 2007 and resulted in the establishment of more than 350 monitoring and research sites, completion of 13 research projects, and development of the Data Information and Management System (DIMS). DIMS currently archives DCERP1 monitoring and research data and provides a standard data format that optimizes data storage and retrieval for integrated analysis, allowing for exchange of information among the various DCERP1 partners and other interested researchers and stakeholders.

During Phase I, the DCERP1 Team developed the overall approach that was implemented during Phase II and used to meet the program’s objectives. This approach started with identifying ecosystem processes and stressors and developing an overarching conceptual model for DCERP1 that included four ecological modules: the Aquatic/Estuarine Module, the Coastal Wetlands Module, the Coastal Barrier Module, and the Terrestrial Module. Because the atmosphere has an overarching influence on all four ecosystem modules, it was treated as a fifth ecosystem module (i.e., the Atmospheric Module). After developing the overall conceptual model (and conceptual models for individual modules), the DCERP1 Team identified knowledge gaps in the models, worked with the installation staff to identify the needs of MCBCL management, and then determined potential research questions to address these basic knowledge gaps and management needs.

Because DCERP1 was designed to be an adaptive program, the DCERP Baseline Monitoring Plan was developed to gather environmental data to address MCBCL’s management concerns and support the 13 research projects. The monitoring activities established important baseline conditions in each of the modules against which changes in ecosystem processes from both military training activities, other non-military stressors, and natural phenomena (i.e., extreme events including hurricanes, droughts) could be measured. The monitoring activities also provided data at the temporal frequencies and spatial extents needed to assess variability in the environmental parameters of importance and were used to inform and validate ecosystem models. Results from the research projects fed back into the adaptive monitoring efforts so that changes in sampling frequency, spatial scale of sampling locations, or parameters to be sampled could be adapted as necessary. Results were used to identify ecosystem indicators and develop associated threshold values, tools, or design models that address MCBCL’s management needs. This information was communicated to MCBCL personnel to assist them in making decisions about what type of management actions could be taken to mitigate the effects of military-related activities on the ecosystems.

Key Scientific Findings

The first objective of DCERP1 was to provide basic scientific information needed to help understand the physical, chemical, and biological processes associated with the coastal ecosystems of MCBCL. The main findings associated with these processes included the hydrodynamics of the New River Estuary (NRE) and the adjacent coastal system, sediment transport among all four MCBCL ecosystems (i.e., terrestrial lands, streams and estuary, coastal wetlands, and coastal barrier island), and nutrient cycling (particularly nitrogen) within the aquatic/estuarine and marsh ecosystems, as well as the role of atmospheric deposition of nitrogen to the ecosystems.

The NRE is a semi-lagoonal estuary with a long history of water quality degradation resulting from loadings of nutrients, particularly nitrogen, resulting in phytoplankton blooms, extended periods of hypoxia or anoxia, and resultant fish kills. The results indicate that anthropogenic nutrient-driven eutrophication and resulting algal bloom dynamics are controlled by climatically driven hydrologic variability in the NRE. When the water residence time within the estuary is too short to allow for nutrient assimilation by the phytoplankton, then bloom development is constrained. Freshwater discharge is of critical importance from both ecological function and ecosystem “health” perspectives because it controls nitrogen inputs and rates of nitrogen cycling.

In contrast to the NRE, which receives inputs of nutrients (particularly nitrogen from the New River watershed), the marshes of the lower NRE and Intracoastal Waterway (ICW) are, by comparison, nitrogen starved. Shallow groundwater from the uplands contains almost no nitrates, and much of the source inputs from the New River watershed to the upper estuary are completely assimilated by phytoplankton in the upper estuary. The intertidal salt marshes of MCBCL were found to be overwhelmingly large sinks for nitrogen, and nitrogen sink strength was dominated by nitrogen burial during sediment accretion (80 to 90%) and by denitrification (10 to 20%).

Wind- and wave-driven hydrodynamic movement were responsible for both NRE shoreline erosion and sediment resuspension in this shallow estuary. Transport of sediments from erosion of sediment banks was found to provide at least half of the sediment required by the lower NRE and ICW marshes to keep pace with sea level rise (SLR) through accretion. Although wind and wave energy were most important as erosional forces in the estuary, within the confined channel of the ICW, boat wake energy became a more important factor affecting shoreline erosion processes. Wave energy from wind and boat wakes, compounded by the effects of routine dredging of the ICW, are the main erosional processes responsible for doubling the width of the ICW channel over the past 70 years. In addition, the very presence of the ICW traps the landward transport of sand from overwash events and aeolian transport across the barrier island, depriving marshes to the west of the ICW of this sand subsidy. As a result, salt marshes on the eastern side of the ICW were found to have a higher elevation above mean sea level than the marshes on the western (landward) side of the ICW. This finding has implications for the sustainability of the marshes landward of the ICW at MCBCL and also in other coastal areas along the ICW’s extent.

A comparison of washover extent across the barrier island suggests that the primary forcing mechanism generating overwash processes has been tropical storm activity. Transgressive barriers exemplified by the southwestern portion of Onslow Island may be overwashed more frequently as sea level rises in the future, producing further erosion of dunes and creating more washover fans. Studies of the two most substantial washover deposits generated by Hurricane Irene (in 2011) found that these areas had not been overwashed in the past 70 years, and thus were not just reoccurrences of overwash of previously breached dunes as occurred at several other sites towards the middle of Onslow Island. This suggests that overwash occurs along the barrier both in new areas and those that have previously overwashed.

Longleaf pine (Pinus palustris) forest habitat at many southeastern installations are managed to promote habitat quality for the endangered red-cockaded woodpeckers (RCWs; Picoides borealis). A basic question of concern was whether forest management practices to improve habitat for RCWs were beneficial or detrimental to other avian species. Research findings from MCBCL suggest that these practices in general appear to benefit the avian community as a whole. This is reflected in measures of community composition. Both species richness (the number of species) and species diversity, which takes into account the relative abundance of the species present, increased as RCW habitat quality increased. This was especially true for another open pine habitat species of concern, Bachmann’s sparrow (Peucaea aestivalis).

The relationship between vegetation and avian communities was assessed on 45 pine plots and was found to be highly correlated. Both vegetative composition and avifaunal communities were compositionally different among longleaf pine, loblolly pine (Pinus taeda), and high-pocosin sites. The overlap between the two communities suggests that the composition of avifaunal communities is correlated with differences in understory vegetative composition that can emerge in the different mature pine stands. Future efforts aimed at recovering avifaunal species may depend upon the recovery of the understory plant communities.

Models, Tools, and Indicators to Assess Ecosystem Health

The second objective of DCERP was the development of ecosystem models, decision-support tools, and environmental indicators that could be used to evaluate ecosystem health. For example, the Estuarine Simulation Model (ESM) predicted hydrodynamic exchanges in the NRE and included components such as eelgrass (Zostera marina), the most abundant species of submerged aquatic vegetation in the NRE, and total suspended solids (TSS). The ESM was used to examine the changes in point and non-point source inputs on estuarine water quality variables and ecosystem processes. The Marsh Equilibrium Model (MEM) was applied to MCBCL to forecast changes in the relative elevation of the marsh surface and biomass response to different rates of SLR possible in a climate changed future. Model simulations of different SLR scenarios showed that the marsh vegetation survived 100 years only when the SLR was less than 60 cm (24 in); otherwise, the marsh vegetation rapidly declined. This model can be transferred to other locations, but requires local accretion rates, standing biomass, TSS, and water level data. Finally, the Run-up and Overwash Model (ROM) was used to predict where overwash would occur on Onslow Beach. This model correctly identified the four overwash areas resulting from the passage of Hurricane Irene (in 2011) illustrating its use in forecasting vulnerable areas.

As previously discussed, there was a strong relationship discovered between hydrological flow of the New River and phytoplankton biomass production. Freshwater flow of 27 cubic meters per second (590 cubic feet per second) was identified as a tipping point for the estuary. At freshwater flows above this threshold, which occurs approximately 22 days a year, the water residence time in the NRE was too short for nutrient assimilation by phytoplankton in the water column; thereby restricting algal bloom development. Benthic chlorophyll a concentration was also found to be an excellent indicator of the effectiveness of the benthic microalgae to act as a nutrient filter. When chlorophyll a ranged from 70 to 83 mg m^-2 or above, benthic microalgal biomass increased, and nitrogen was sequestered from the water column. When conditions restricted photosynthesis (chlorophyll a ranged from less than 70 to 83 mg m^-2), the microalgae released nitrogen into the water column. In shallow estuaries such as the NRE, hydrologic changes can modify both phytoplankton primary production and affect benthic microalgal production, which modulates internal nutrient cycling.

Recommended Adaptive Management Strategies to Sustain Ecosystems

DCERP’s third objective was to recommend adaptive management strategies to sustain ecosystem natural resources within the context of an active military installation. These strategies are most applicable to the coastal wetlands, coastal barrier, and terrestrial ecosystems. More than 80% of the NRE shoreline is contained within the boundary of MCBCL. Only 19% of this shoreline has been hardened with revetments, sills, and seawalls. This constrained shoreline development has water quality benefits with respect to reduced nutrients and sediment runoff and allows ecosystem services (e.g., storm surge protection, fish and shellfish nursery areas) of the unhardened shorelines to be preserved. Although historic MCBCL practices of hardening NRE shoreline in high-energy areas are appropriate, marshes and sediment banks, which supply sediment vital for marsh accretion, will be needed in the future to help mitigate for rising sea level. MCBCL managers should consider whether additional shoreline hardening is needed, and if so, new shoreline hardening should be offset with marsh restoration in hardened areas where wave energy is low and marsh restoration efforts would be successful. These restoration efforts will promote sustainability by allowing the marshes to migrate landward as sea level rises and continue to provide shoreline ecosystem services not provided by hardened structures.

Marshes along the lower NRE and ICW are an integral part of amphibious training and are a conduit for moving amphibious vessels between mainland training areas and the barrier island. Reinforced splash points within the ICW have a lower shoreline change compared to unmodified splash points. MCBCL managers should consider strategies to reduce erosion rates and enhance sustainability of splash points for future training, including reinforcing splash points with concrete ramps, implementing marsh restoration, or diverting some military training activities from overused splash points to underused ones.

Similar to the ICW and coastal marshes, the barrier island provides essential beach for the Marines to conduct amphibious assault training. At current training levels (frequency and intensity), military training did not have a measurable impact on the landscape (e.g., sediment texture, topography, habitat) or biology of Onslow Beach. For example, the installation’s constraint of training to existing egress and ingress areas along the barrier island and restricting vehicular and pedestrian traffic on the dunes and backbarrier marshes also prevents impacts to the barrier. Overall, the studies found that natural processes (wind and wave actions) overshadow anthropogenic effects.

Forest management at MCBCL and other installations along the coastal, southeastern United States is directed toward returning hardwood-pine lands to open canopy longleaf pine-wiregrass communities. Installation restoration goals were better achieved by the combined use of understory/midstory thinning in pine-hardwood plots followed by prescribed burning. Used in combination, these practices removed more woody material and consumed more than three times the fuel of prescribed burning alone. Thinning during the growing season was more effective in reducing the understory/midstory woody plants than dormant season thinning or prescribed burning alone. If these differences persist, they would indicate a possible benefit of growing season thinning that could be implemented at other installations with longleaf pine restoration goals. In either case with or without thinning, continued suppression of woody growth requires regular prescribed burns. Additionally, the use of understory/midstory thinning appears to have a secondary benefit in reducing PM_2.5 emissions, thus providing improved smoke management for installations.

The recovery of RCWs at installations in the southeastern United States drives forestry management strategies. Forest management that specifically targets improving habitat conditions for the RCWs was found to result in habitat changes that benefit the biodiversity of terrestrial ecosystems in general and the total avian community specifically; therefore, this management practice should be encouraged. Installation managers also should maintain the availability of nesting substrate (e.g., live pines, pine snags, hardwood snags) for the wide variety of cavity-nesting avian species because that determines the strength of interactions among these species. Specifically, a shortage of dead or dying pine snags would likely result in negative impacts on RCWs due to the takeover of their cavities in live pines by other cavity nesting species.

The research conducted as part of DCERP1 has resulted in a greater understanding of MCBCL’s biologically diverse ecosystems and their interactions with military training activities. In addition, the research results provide an understanding of what on- and off-installation activities affect these ecosystems and what management actions could be implemented to best sustain MCBCL military training and testing resources. Through the research projects, the DCERP1 Team developed a series of indicators, models, and tools designed to benefit installation managers by providing support for environmental decision making. Knowledge gained from DCERP1 research will provide benefits to other military installations in coastal settings and to the scientific community and general public at large.