Estimating Changing Marshland Habitat and Conservation Potential for Diamondback Terrapins (Malaclemys terrapin) in New Jersey under Climate Change

Abstract

The diamondback terrapin (Malaclemys terrapin), a brackish water turtle species native to the eastern United States, is under “special concern” in the state of New Jersey, due to decreasing habitat from development and changing climatic conditions. Diamondback terrapins reside in saline marshes and coastal wetlands and nest in sandy substrate, primarily beaches and dunes, in June and July. New Jersey is vulnerable to sea level rise, leaving diamondback terrapin habitats and nesting areas at risk of inundation under future climate scenarios, and, as the most densely populated state, subject to continual development pressures on potentially conservable land. Changing sea levels and climatic conditions will cause accretion and migration of marshes into open grassy land, yielding new potential terrapin habitats, though changing temperatures could affect the availability of male-producing nesting sites and impact potential nesting patterns. This study spatially modeled lost, gained, and changed habitat and nesting areas under sea level rise scenarios for 2050 and 2100 in New Jersey and quantified these by municipality to offer insights into potential conservable land that may mitigate these changes for the vulnerable species. The results indicate an overall decrease in potential habitat coupled with a decrease in both overall and male-producing nesting ranges.

1. Introduction

The diamondback terrapin, Malaclemys terrapin, is a brackish water turtle species that resides in saline marshes and estuaries along the east coast of the United States [1]. In New Jersey, USA, this includes the area along the coast of the Atlantic Ocean and the Delaware Bay. It is often considered a salt marsh keystone species, as its predation of the periwinkle snail, Littoraria irrorata, regulates the extensive grazing of marsh grasses that could result in barren mudflats susceptible to erosion [1]. In June and July, female diamondback terrapins emerge from their marsh habitats in search of nesting sites to lay their eggs, with a preference for areas of loamy, loose sand such as beaches and dunes, that have the ideal water conductivity and ionic composition to support eggs and avoid desiccation [2]. It is estimated that this species has a maximum nesting range of 1300 m [3]. Hatchlings in northern regions tend to utilize their nesting habitat to over-winter in the fall for emergence in the spring, indicating that beaches are prioritized throughout the year [2].

The diamondback terrapin was listed in 2016 as a species of “special concern” by the New Jersey Department of Environmental Protection, due to its susceptibility to land development, habitat inundation from sea level rise, road mortalities, and bycatch from crab pots [4]. As of 2019, the IUN listed the species as “vulnerable” from an estimated over 35% decline in population across their range in response to these threats [3,4]. Estimates in other areas align with these results, with a mark–recapture study in South Carolina finding a nearly 75% reduction in population size since 1989 [5]. Little is known about the numeric population status of the species within New Jersey; however, studies on road mortalities and crab pot drowning suggest that this state is no exception [6,7].

Global warming is exacerbated by increases in radiative forcing from human-driven increases in carbon dioxide in the atmosphere, which further increases global air and sea surface temperatures. These increases result in a rise in sea level, which varies across space [8,9]. New Jersey is especially vulnerable to sea level rise because of its sinking land from ice sheet retreat, with historic records from 1911 to 2019 showing a rise of 0.45 m compared with the global 0.19 m [5]. Marshes in the state have seen the effects of this, with a predicted 11.9% of coastal wetlands in Barnegat Bay lost between 1972 and 2012, and an erosion rate of −0.5 m/year from 1930 to 2013 [10,11]. Beaches along the coast have had similar reactions, with the average erosion rate from the 1960s being −0.9 m/year in the northern region and −2.2 m/year in the southern region [12,13]. Warming is expected to increase in the coming decades, even under emission mitigation [14], which leaves diamondback terrapins at risk of losing habitat and nesting areas along their range with continuing sea level rise. Marshes can respond to increasing sea levels in a few ways: becoming completely inundated by water, accreting with deposited sediment and growing vertically, or migrating inland into open, undeveloped areas [6]. This suggests that rising sea levels could increase the amount of marsh in the state. Similar coastal species across the United States are expected to see variations in gains and losses of marsh habitat in future scenarios due to these phenomena [15]. To project the changes in sea level, the New Jersey DEP uses the range between 17% and 83% likelihood scenarios for sea level rise as their “likely range” [4]. Under moderate emissions, this leaves a range of 0.9–2.1 feet [0.27–0.64 m] of sea level rise by 2050 and 2.0–5.1 feet [0.61–1.55 m] by 2100 [5].

Diamondback terrapins are also vulnerable to changes in sex ratios due to increasing air and surface temperatures. Incubation temperature alters the sex determination of diamondback terrapins, with temperatures above 84.2 °F [29.0 °C], producing female turtles, and below producing males [7]. Studies on sea turtle nests at approximately 30 cm show a linear relationship between changes in air temperature and nest temperature, with a slope of 1.15 [8,9]. Diamondback terrapins have much shallower nesting depths, averaging 15 cm below the surface, and do not seem to consider temperature changes when selecting their depths [10]. With diamondback terrapin nests close to the surface, it can be estimated that increases in nest temperature will closely resemble increases in air temperature. New Jersey is expected to experience, under moderate emissions, an increase of 3.5 °F [1.9 °C] in air temperature by 2050, and an increase of 7.8 °F [4.3 °C] by 2100 [11]. This temperature increase has the potential to skew the sex ratio of nesting areas toward females.

A final threat to diamondback terrapins is coastal development. This includes road mortalities, particularly while traveling to and from nests [3], and habitat loss from marshes and beaches being built on [12]. The state has continuously seen declines in adequate salt marsh habitat, as analysis of land use and land cover changes from 2007 to 2015 cover saw a loss of 22.7 ha of coastal wetlands to urban land and 165.1 ha to a disturbed wetlands status [16,17]. New Jersey, however, has various areas of preserved land that cannot be developed, including lands under the Green Acres Program that protects open recreation, natural, and historic spaces, the Natural Areas System that protects land remaining in its natural state, and tidelands overseen by the Tidelands Resource Council in the NJDEP [18]. These areas, among others, are considered protected under the US Gap Analysis Project inventory of protected areas [19]. Terrapin habitat and nesting areas that fall within these lands can be considered protected, while those outside these boundaries may be at a higher risk of coastal development.

The purpose of this study is to estimate the extent of alteration in diamondback terrapin habitat, nesting area, and sex-based nesting area ratios under sea level rise and increased air temperature. The methods include an assessment of changes in land type, as well as potential habitat gained with marsh migration and changes in nesting range. Four scenarios of sea level rise: 2050 and 2100 at 83% and 17% likelihood scenarios are evaluated, as well as temperature increases in these years under moderate emissions, to determine the amount of potential habitat that is under conservation at the state and municipality levels. The results from this study may be used to contribute to the local vulnerability status of the species and to plan for areas of designated future habitat by federal and private conservation groups.

2. Materials and Methods

This study was conducted in the state of New Jersey, on the mid-Atlantic eastern coast of the United States. Positioned along the Atlantic Ocean and Delaware River and Bay, New Jersey’s border is nearly entirely defined by water bodies, except for the northeastern border with New York, and it is central to the range of the diamondback terrapin, which stretches from the Texas–Mexico border to Cape Cod, Massachusetts.

Data for this project included remotely sensed imagery and digital elevation models, municipal boundaries, protected areas, and land cover types of the state of New Jersey, and modeled marsh migration and accretion data.

Remotely sensed data included three rows along the same path (WRS path 14, rows 31–33) from a single date, 4 July 2022 of Landsat 8 Operational Land Imager (OLI) and Thermal Infrared Sensor (TIRS) imagery. This date was selected due to the rare synoptic conditions of nearly 0% cloud cover across the state, with temperatures comparable to the historic average for that date (72.6 °F/22.6 °C at the Atlantic City Airport Station, compared with a historic average of 76.3 °F/24.6 °C). Data were downloaded as collection 2, level 2 imagery, converted to reflectance values and degrees Kelvin per the Science Product Guide [20]. Data were further processed to mosaic into a continuous image and converted to indices and surface reflectance in degrees Fahrenheit using Terrset software version 19.0.8 [21].

Data regarding the boundaries of municipalities, protected open spaces, and protected tidelands were downloaded from the New Jersey Geographic Information Network (NJGIN), with sources from the New Jersey Office of GIS (NJOGIS), and the New Jersey Department of Environmental Protection (NJDEP) Bureau of GIS [22]. Additionally, the LiDAR-derived digital elevation model was also acquired through the NJGIN. The most recent official land cover data for New Jersey are from 2015, which served as the baseline of the methods described in Section 2.1.

Additional data used in this project included protected areas listed by the US Gap Analysis Project [19], soil type polygons from the USDA, and modeled marsh migration and accretion data from Duke University [23].

All spatial analyses were conducted using ArcGIS Pro version 3.1 [24].

2.1. Determination of Current Diamondback Terrapin Habitat and Nesting Sites

New Jersey land use data from 2015 were used as a basis for current potential diamondback terrapin habitats, with ranges of interest and featured municipalities along the Delaware Bay and Atlantic Coastline highlighted in Figure 1. Areas classified with the field values of “phragmites dominate coastal wetlands”, “saline marsh (high marsh)”, “saline marsh (low marsh)”, and “tidal mud flat”, were selected and extracted. As the most recent data were released 8 years prior to this study, the method proposed by Al-Maliki and colleagues was reproduced to eliminate areas not suitable for marshes, masking the initial marsh land uses with areas derived from remotely sensed data in which marshes were either inundated, too dry, or lacked vegetation [25]. Metrics of the normalized difference vegetation index (NDVI), the normalized difference water index (NDWI), and the normalized difference moisture index (NDMI) were used for this calculation. Equations for these metrics follow below:

$$NDVI=\frac{NIR-Red}{NIR+Red}$$

(1)

$$NDWI=\frac{Green-NIR}{Green+NIR}$$

(2)

$$NDMI=\frac{NIR-SWIR}{NIR+SWIR}$$

(3)

For Landsat 8 imagery, NIR, Near Infrared, is band 5, red is band 4, green is band 2, and SWIR is band 6. These indices were thresholded such that calculated NDVI cells under 0.125 represented land with no vegetation, calculated NDWI cells above 0 represented land as fully underwater, and calculated NDMI cells below 0 represented land with no moisture. Each of these conditions resulted in areas where marsh landscapes were not viable under the conditions represented by the Landsat imagery from 4 July 2022, and all pixels fulfilling any of these criteria were merged into a single layer representing land not suitable for marshes. It is noted that all vegetation indices derived from remotely sensed data are subject to the momentary conditions in which their constituent imagery was captured and may represent an ephemeral state that differs from the preceding or succeeding conditions [26,27]. However, the consideration of this imagery vis-à-vis long-term conditions and the location and timing may mitigate the potential impact of any anomalous conditions and should not affect the future modeled scenarios. The original 2015 habitat layer was masked with an area overlapping with the combined unsuitable layer, creating an updated layer of suitable diamondback terrapin habitat for 2022 that served as a baseline for future scenarios.

2.2. Determination of Current Diamondback Terrapin Habitat and Nesting Sites

Nesting sites were classified using the USDA Soil Survey Geographic Data and selecting soil texture field values of “Sand” or “Sandy” and covers of “Barren Land” or “Shrub”, indicating beaches or dunes. This land was further reduced by clipping with a buffer of 1500 m from habitat, or the estimated nesting range with a slight overestimation for increased conservation purposes. Landsat 8 TIRS band 10 was converted into apparent land surface temperature and thresholded, with cells below 84.2 °F [29 °C] reclassified as male-producing areas and cells at 84.2 °F [29 °C] or higher reclassified as female-producing. Each classification was extracted as a layer and clipped to nesting sites to determine the nesting area with the potential to produce each sex, as highlighted in Figure 2.

2.3. Adjusting Potential Habitat Range by Sea Level Rise Changes and Marsh Migration/Accretion

Sea level rise projections from the 2020 New Jersey Scientific Report on Climate Change were used for 17% and 83% likelihood scenarios for 2050 and 2100 [25]. The digital elevation model was rescaled to 30 m/pixel to match the Landsat resolution and thresholded according to the inundated area under projected sea level rise to further mask the current habitat and nesting area. The previously generated accretion and marsh migration layers from Duke University were converted into vector polygons and merged with the remaining habitat to determine the total amount of possible habitat after each scenario. In order to best approximate potential changes in marsh habitat when exact date matches were unavailable, the closest later time period (2100 17%, 2107 83%, 2058 17%, and 2061 83%) for the accretion and migration layers were used. Potential nesting area within the range of the new habitat was redetermined with a buffer of 1500 m from habitat areas.

2.4. Determining Conservation and Results by Municipality

A data layer of protected land in New Jersey was created by uniting protected open spaces, protected tidelands, USGS GAP protected lands, and New Jersey-preserved natural areas layers. All boundaries were dissolved and intersected with the total marsh habitat and nesting layers to determine the area within the currently protected land cover, as shown in Figure 3 for marsh habitat. Area estimates were extracted by municipal boundaries to determine these statistics by municipality.

3. Results

3.1. Change in Terrapin Habitat across the State

All current habitat ranges of the diamondback terrapin decrease with rising sea levels, though these losses are slightly mitigated in the first 2050 scenario by marsh accretion. Changes in habitat show an overall decrease in area with sea level rise, except for the 2050, 83% scenario, as listed in

Table 1. Changes in marsh habitat area in acres under each scenario of sea level rise across New Jersey.

Scenario	SLR ft [m]	Habitat Lost Acres [ha]	Percent Habitat Lost	Habitat Remaining Acres [ha]	Gained Marsh Area Acres [ha]	Percent Habitat Gained	Total Habitat Area Acres [ha]	Protected Habitat Area Acres [ha]	Percent Habitat Protected
Current	0.00 [0.00]	0.0 [0.0]	0.00%	187,563.4 [75,905.9]	000.0 [000.0]	0.0%	187,563.4 [75,905.9]	122,684.1 [49,649.6]	65.4%
2050 (83% confidence)	0.90 [0.27]	28,204.4 [11,414.2]	15.0%	159,359.1 [64,491.7]	43,066.5 [17,428.8]	23.0%	199,415.5 [80,702.3]	124,880.1 [50,538.28]	62.6%
2100 (83% confidence)	2.00 [0.61]	101,339.5 [41,011.5]	54.0%	86,223.9 [34,894.3]	54,034.7 [21,867.5]	28.8%	137,486.9 [55,640.2]	79,941.7 [50,538.28]	58.1%
2050 (17% confidence)	2.10 [0.64]	111,061.8 [44,946.1]	59.2%	76,501.6 [30,959.8]	50,864.1 [20,584.4]	27.1%	126,465.2 [51,179.8]	72,268.0 [29,246.46]	57.1%
2100 (17% confidence)	5.10 [1.55]	186,521.6 [75,484.3]	99.4%	1041.8 [421.6]	59,823.8 [24,210.4]	31.9%	60,850.4 [24,625.8]	30,402.3 [12,303.64]	50.0%

. Habitat area lost by 2050 ranges from 28,204.4 acres [11,414.2 ha] (15.04%) to 111,061.8 acres [44,946.1 ha] (59.21%) and by 2100, it ranged from 101,339.5 acres [41,011.5 ha] (54.03%) to 186,521.6 acres [75,484.3 ha] (99.44%), following the 83% and 17% confidence interval estimates, respectively. The addition of gained marsh shows an increase in total habitat for the 2050, 83% scenario and mitigates some of the loss from other scenarios. In the 2050, 83% scenario, there is a predicted gain in marsh habitat of 43,066.5 acres [17,428.8 ha] (23.0%), leading to a total area of 199,415.5 acres [80,702.3 ha], with 124,880.1 acres [50,538.28 ha] (62.62%) protected. For the 17% scenario, there is a predicted gain of 50,864.1 acres [20,584.4 ha] (27.1%), and a total area of 126,465.2 acres [51,179.8 ha], with 72,268.0 acres [29,246.46 ha] (57.14%) protected. In 2100, there is a predicted gain in marsh habitat of 54,034.7 acres [21,867.5 ha] (28.8%), leading to a total area of 137,486.9 acres [55,640.2 ha], with 79,941.7 acres [50,538.28 ha] (58.14%) protected under the 83% scenario. Under the 17% scenario, there is a predicted gain of 59,823.8 acres [24,210.4 ha] (31.9%) acres, with a total area of 60,850.4 acres [24,625.8 ha] and 30,402.3 acres [12,303.64 ha] (50,049.96%) protected. Changes to the original habitat and the net change with gains are listed in Table 1.

Similarly, available nesting areas overwhelmingly decrease under rising sea levels. By 2050, according to the 83% and 17% likelihood scenarios, the amount of original nesting area lost ranges from 1365.0 acres [552.4 ha] (8.8%) to 2343.9 acres [948.6 ha] (15.0%), and by 2100, it ranges from 2210.8 acres [894.7 ha] (14.2%) to 9224.0 acres [3732.9 ha] (59.1%), respectively. While there is an overall loss of nesting areas, the changes in the location of marshes influence the location of potential nesting areas within range, leading to varying changes in total nesting area. For that reason, under the 83% and 17% scenarios, the total available nesting area by 2050 is estimated to be 17,191.8 acres [6957.4 ha], with 9713.1 acres [3930.8 ha] (56.5%) protected, and 16,502.2 acres [6678.3 ha], with 9155.6 acres [3705.2 ha] (55.5%) protected, respectively. By 2100, the total available nesting area is estimated to be 16,225.9 acres [6566.5 ha], with 9101.8 acres [3683.4 ha] (56.1%) protected, and 8913.3 acres [3607.2 ha], with 3846.3 acres [1556.6 ha] (43.2%) protected, respectively. The ranges, including lost and new gains, are detailed in

Table 2. Changes in the nesting area in acres under each scenario of sea level rise overall in New Jersey.

Scenario	SLR ft [m]	Nesting Area Lost Acres [ha]	Percent Nesting Area Lost	Total Nesting Area Acres [ha]	Protected Nesting Area Acres [ha]	Percent Nesting Area Protected
Current	0.00 [0.00]	0.0 [0.0]	0	18,192.6 [7362.4]	10,500.1 [4249.3]	57.7%
2050 (83% confidence)	0.90 [0.27]	1365.0 [552.4]	8.8%	17,191.8 [6957.4]	9713.1 [3930.8]	56.5%
2100 (83% confidence)	2.00 [0.61]	2210.8 [894.7]	14.2%	16,225.9 [6566.5]	9101.8 [3683.4]	56.1%
2050 (17% confidence)	2.10 [0.64]	2343.9 [948.6]	15.0%	16,502.2 [6678.3]	9155.6 [3705.2]	55.5%
2100 (17% confidence)	5.10 [1.55]	9224.0 [3732.9]	59.1%	8913.3 [3607.2]	3846.3 [1556.6]	43.2%

.

3.2. Changes in Nesting Area Sex Ratios across the State

Under the 83% and 17% likelihood scenarios, the percentage of nesting area available to produce males by 2050 ranges from 1.5%, with 55.6% of the area protected, to 1.3%, with 57.0% of the area protected, respectively. By 2100, the percentage of nesting area available to produce males ranges from 0.4%, with 59.6% of the area protected, to 0.3%, with 55.7% of the area protected. The results including the changes in temperature show a strong skew toward female-producing nesting areas, as highlighted in

Table 3. Changes in the sex ratios of the nesting area in acres under each scenario of sea level rise and temperature increase across New Jersey.

Scenario	SLR ft [m]	Temperature Increase °F [°C]	Female Nesting Area Acres [ha]	Male Nesting Area Acres [ha]	Percent Female Nesting Area	Percent Male Nesting Area	Protected Female Nesting Area Acres [ha]	Protected Male Nesting Area Acres [ha]	Percent Female Nesting Area Protected	Percent Male Nesting Area Protected
Current	0.00 [0.00]	0.0 [0.0]	16,611.8 [6722.7]	1580.8 [639.7]	91.3%	8.7%	9488.4 [3839.9]	1011.7 [409.4]	57.1%	64.0%
2050 (83% confidence)	0.90 [0.27]	3.5 [1.9]	16,932.7 [6852.6]	259.0 [104.8]	98.5%	1.5%	9569.1 [3872.6]	144.0 [58.3]	56.5%	55.6%
2100 (83% confidence)	2.00 [0.61]	7.75 [4.3]	16,154.2 [6537.5]	71.7 [29.0]	99.6%	0.4%	9059.1 [3666.2]	42.7 [17.3]	56.1%	59.6%
2050 (17% confidence)	2.10 [0.64]	3.5 [1.9]	16,292.8 [6593.6]	209.4 [84.7]	98.7%	1.3%	9036.3 [3656.9]	119.3 [48.3]	55.5%	57.0%
2100 (17% confidence)	5.10 [1.55]	7.75 [4.3]	8882.8 [3594.8]	30.5 [12.3]	99.7%	0.3%	3829.3 [1549.7]	17.0 [6.9]	43.1%	55.7%

.

3.3. Results of Habitat Change and Conservation Potential by Municipality

The conservation potential for the state varies widely, as New Jersey operates under “home rule”, in which each of the 564 municipalities determines its own zoning and land use. Under current conditions, 186 municipalities have potential habitat and 72 have potential nesting areas. Of those municipalities, an average of 42.5% of the habitat is protected, and an average of 46.1% of the nesting area is protected. By 2050 under the 83% and 17% scenarios, 209 to 219 municipalities have habitat areas, with an average of 42.5% to 41.0% of it protected, and 74 to 77 municipalities have nesting areas, with 49.2% to 50.5% of it protected, respectively. By 2100 under the 83% and 17% scenarios, 211 to 207 municipalities have habitat areas, with an average of 41.8% to 40.8% of it protected and 74 to 69 municipalities have nesting areas, with an average of 49.3% to 45.5% of it protected, respectively. Future sea level rise scenarios alter the number of municipalities with habitat, as well as the amount protected, as listed in

Table 4. Changes in habitat and nesting area in acres at the municipality level.

Scenario	SLR ft [m]	Mun. with Habitat	Mun. with Nesting Area	Avg. Percent Habitat Protected	Avg Percent Nest Area Protected	Mun. with Male Nesting Area
Current	0.00 [0.00]	186	72	42.5%	46.1%	41
2050 (83% confidence)	0.90 [0.27]	209	74	42.5%	49.2%	29
2100 (83% confidence)	2.00 [0.61]	211	74	41.8%	49.3%	17
2050 (17% confidence)	2.10 [0.64]	219	77	41.0%	50.5%	28
2100 (17% confidence)	5.10 [1.55]	207	69	40.8%	45.5%	14

. These numbers also vary geographically, with municipalities along the east coast often having a higher percentage of marsh habitat protected than those in the west, as seen in Figure 4.

Across the state and scenarios, Middle Township, Maurice River Township, and Galloway Township have some of the most distinct changes, as highlighted in

Table 5. List of the changes in habitat and nesting areas for the municipalities with the largest changes.

Scenario	SLR ft [m]	Mun. with the Most Habitat	Percent Habitat Protected	Mun. with the Most Habitat Loss	Mun. with the Most Nesting Area	Percent Nesting Area Protected	Mun with the Most Nesting Area Loss	Mun. with the Most Male Nesting Area
Current	0.00 [0.00]	Middle Township	70.1%	-	Maurice River Township	78.8%	-	Maurice River Township
2050 (83%)	0.90 [0.27]	Middle Township	71.9%	Galloway Township	Maurice River Township	78.8%	Maurice River Township	Brigantine City
2100 (83%)	2.00 [0.61]	Middle Township	73.5%	Galloway Township	Maurice River Township	79.3%	Maurice River Township	Brigantine City
2050 (17%)	2.10 [0.64]	Middle Township	73.5%	Galloway Township	Maurice River Township	79.4%	Maurice River Township	Brigantine City
2100 (17%)	5.10 [1.55]	Downe Township	51.8%	Middle Township	Berkeley Township	28.9%	Maurice River Township	Brigantine City

and Figure 5. The municipality with the highest potential amount of terrapin habitat is Middle Township in Cape May County at 15,495.2 acres [6270.8 ha], with 70.1% of it being protected. The municipality with the current maximum amount of nesting area is Maurice River Township in Cumberland County at 5539.4 acres [2241.8 ha], with 78.8% protected. The municipality with the most male nesting area is also Maurice River Township in Cape May County at 366.5 acres [148.3 ha], 79.2% of which is protected.

In both the 83% and 17% scenarios by 2050, Middle Township remains with the most habitat area, ranging from 16,190.9 acres [6552.4 ha] to 11,927.2 acres [4826.9 ha], and Galloway Township loses the most habitat area, with 2701.2 acres [1093.2 ha] and 12,848.1 acres [5199.6 ha], respectively. Maurice River Township continues to have the most nesting area, with 4840.0 acres [1958.7 ha] and 4405.8 acres [1783.0 ha], while also losing the most nesting area of 699.4 acres [283.0 ha] and 1133.6 acres [458.8 ha], respectively. In both scenarios, Brigantine City has the most male nesting area, with 56.4 acres [22.8 ha] and 48.1 acres [19.5 ha], respectively.

By 2100, Middle Township continues to have the most habitat area in the 83% scenario, with 13,145.5 acres [5319.9 ha], while Downe Township emerges with the most in the 17% scenario, with 4680.6 acres [1894.2 ha]. Galloway Township loses the most habitat area in the 83% scenario, with 11,939.2 acres [4831.7 ha], whereas Middle Township loses the most in the 17% scenario, with 15,466.0 acres [6259.0 ha]. Maurice River Township again has the most nesting area in the 83% scenario, with 4480.4 acres [1813.2 ha], but Berkeley Township has the most in the 17% scenario, with 900.5 acres [364.4 ha]. Maurice River Township loses the most nesting area in both the 83% and 17% scenarios, with 1059.0 acres [428.6 ha] and 5507.7 acres [2228.9 ha], with Brigantine City having the most male nesting area, with 24.4 acres [9.9 ha] and 13.9 acres [5.6 ha], respectively.

Overall, the state shows a trend of the total habitat decreasing over the four scenarios, with a slight increase in the initial 2050, 83% scenario. By municipality, however, this trend varies widely, as shown in Figure 6. Southern, western, and mid-eastern municipalities gain habitat in the initial 2050, 83% scenario, and those in the southwest and mid-east continuously gain habitat until the maximum 2100, 17% scenario. Some in the west, especially in the 2050, 83% scenario, gain new habitat from nothing. Those in the southeast and northeast tend to continuously lose habitat, with some losing all terrapin habitat by the 2100, 17% scenario.

New Jersey loses nesting area across the sequence of all four scenarios, with a slight upturn in the 2050, 17% scenario. By municipality, most along the coast continuously lose nesting areas, with a few in the northeast gaining access to more nesting areas as habitat ranges migrate north. By the 2100, 17% scenario, most municipalities are either losing or have lost all previous nesting areas, as shown in Figure 7.

4. Discussion

While the severity and location of changes in diamondback terrapin habitat and nesting areas vary across New Jersey, the overall pattern indicates severe losses for the coastal marsh areas needed to support this species. Although municipalities along the Delaware Bay and Delaware River show initial increases in the 2050, 83% scenario, the state could see a detrimental loss of up to 99.4% of its original habitat by the 2100, 17% scenario. This later decrease is most prominent along the eastern Delaware Bay and east coast, which consists of their traditional range [1]. Marsh migration helps mitigate some of these losses, with a potential increase of up to 31.9% of the original habitat. In the municipalities with positive changes, the amount of habitat increase from marsh migration and accretion exceeds the losses. Most of this migration is along tributaries from the bays, such as around Great Bay, Maurice River, and the many smaller tributaries from Delaware Bay. Across the state, marshes are projected to move inland and gain consistently near the coast of the Delaware River in southwestern New Jersey and the central-eastern coast through the 2100, 17% sea level rise scenario. While this study is specific to New Jersey, other species are projected to see similar reductions and alterations in marsh habitat from sea level rise and marsh migration throughout the United States [15]. These methods of summarizing changes by municipality could be effective in suggesting and comparing the conservation potential for habitat across local governments.

The loss of habitat in the traditional range of diamondback terrapins may force them to migrate or be displaced toward the west and north, though the coast of the Delaware River is not thought to be a traditional habitat for diamondback terrapins [1]. The salinity of the river ranges from 6.4 ppt in the mouth of the bay to 0 ppt in western-central New Jersey [28]. Diamondback terrapins thrive best in salinity levels between 5 ppt and 10 ppt, where they can maintain proper hydration for growth and metabolic function [29]. The energetic cost required to live under freshwater conditions may result in a smaller body mass but has not been shown to adversely affect other aspects of terrapin health [30]. It is therefore possible for the diamondback terrapin habitat to be shifted west if the southern east coast habitat becomes submerged, but it may have physiological effects on terrapin growth. Concentrated migration north may be more beneficial, as the areas along the coast will have higher salinities from the Atlantic Ocean. Migration or displacement west could also lead to an invasive species event where diamondback terrapin intrusion causes the population decline or displacement of other species, as was the case for the red-eared slider (Trachemys scripta elegans) [31]. While the competitive ability of diamondback terrapins is unknown, an overlap in range would occur for native species such as the painted turtle (Chrysemys picta), whose similar diet and basking behavior could lead to competition for resources and space [32,33,34,35].

Another issue with this change in marsh movement is the lack of nesting areas along Delaware Bay and the Delaware River when compared with the southeastern coast. Maurice River Township has the most nesting area along Delaware Bay and loses substantial habitat in comparison with the western municipalities, which have minimal nesting areas. The prime nesting areas of dunes and beaches of the southeastern coast tend to lose habitat, with municipalities more inland gaining habitat. This could negatively impact diamondback terrapin nesting habits, resulting in longer nesting ranges or using less suitable substrate. A possible solution for this issue is dredging. While less preferable than loamy sand, Wnek and colleagues found that diamondback terrapin eggs in dredged soil aged for one year had a 59.4% hatching success rate [2]. It could therefore be possible to artificially provide nesting substrate in these areas, gaining habitat with little natural nesting area. Dredging may, however, pose negative impacts on marine and riverine ecosystems, such as increases in dissolved oxygen and metal contamination [36], as well as sediment disturbances and increased contamination impacting benthic organisms [26] and fish [37]. Careful planning would be required to minimize the impact of dredging on other ecosystems. Therefore, central-eastern New Jersey may be a more suitable area for displacement compared with western New Jersey regarding nesting area availability, with some municipalities gaining nesting area up to the 2050, 17%, 2.1 ft rise scenario.

Marsh migration is associated with terrapin habitat moving inland, which could place terrapins at a higher risk of development pressures and other human activity. Living in more developed land has the potential to lengthen nesting travel through more dangerous terrain [38] with the risk of exposure to roads [3]. The results indicate that as habitat ranges migrate, they often exceed the boundaries of protected areas, with a maximum of 15.5% of habitat being subtracted from the current 65.4% of habitat under protection. The percentage of protected nesting areas shows a similar decrease of up to 14.6% from the current 57.7%. This higher percentage of habitat and nesting area in unprotected land means future habitat could be in closer proximity to anthropogenic lands, such as roads, or at risk of further development. At the municipality level, those with higher percentages of habitat protection remain along the east coast and Delaware Bay. While these percentages are higher, the loss of overall habitat could be fueling these increases, therefore not proving beneficial to diamondback terrapin conservation. Most western municipalities have less than 25% of their habitat protected, which could be detrimental in the event of western migration. Municipalities along the central-eastern coast have a greater fraction of protected habitat and may be a better location for displacement in terms of conservation. Additionally, municipalities near the Delaware River should consider conserving more of their marshland in the event of migration. Downe Township, for example, is projected to have the most habitat area in the 2100, 17% likelihood scenario but will only have 51.8% of that land protected.

Sea level rise will negatively impact available nesting areas, which is estimated to decrease by up to 59.1% in the 2100, 17% rise scenario, again with a slight uptick in the 2050, 17%, scenario. It is also possible for the available nesting area to increase, as the movement of marsh habitat changes the location of nesting areas within range. Most of this loss will occur along their traditional habitat, with some municipalities losing all of their nesting areas. This could reduce the ability of diamondback terrapins to reach adequate nesting sites and settle for less suitable substrate, which would lower the chances of hatching success [2]. Furthermore, Berkeley Township has the greatest potential to conserve nesting area, with the most nesting area by municipality in the 2100, 17% likelihood scenario, but it will only have 28.9% of that land protected under current efforts.

Another concern is the loss of male-producing nesting areas resulting from temperature increases. The percentage of nesting areas able to produce males could be reduced to 0.3% under the 2100, 17% sea level rise scenario with a 7.75 °F [4.3 °C] temperature increase. For loggerhead sea turtles that lay approximately 100 eggs at a time [39] and have similar nesting substrates, it would require males to mate five times annually at a 95% to 5% female sex ratio to maintain a stable population [40]. Comparatively little is known about the mating frequency of male diamondback terrapins; however, the ability of males to sustain the population in combination with the loss of nesting area implies a dire scenario. A possible strategy for increasing the likelihood of male-producing incubation is to provide shading to reduce nest temperatures [2]. This could be achieved with the addition of dune and beach vegetation. Brigantine City and Maurice River Township are expected to have the most male-producing nesting area and overall nesting area, respectively, in the worst and hottest scenarios, and may be good candidates for implementing strategies to shade and cool nests.

5. Conclusions

This study projects the impacts of sea level rise and temperature increase on diamondback terrapin habitat and conservation in New Jersey by the years 2050 and 2100 under the 83% and 17% likelihood scenarios. The results suggest that New Jersey may lose up to 99.4% of its current habitat with sea level rise in the 2100, 17% likelihood scenario with 5.1 feet [1.55 m] of rise. Marsh migration and accretion offer some mitigation of these effects, with a gain of up to 31.9% of the current habitat in the same scenario. The amount of habitat protected is also projected to decrease down to 50.0% in the same scenario, potentially exposing habitat to development pressure. Changes in habitat show approximately continuous gains with increasing sea level rise in the southwestern and central-eastern municipalities up until the 2100, 17% scenario, lending to potential locations for turtle migration and displacement. The lack of suitable nesting areas and protected land in southwestern New Jersey, however, poses concerns for its habitability.

The available nesting area is projected to decrease as well, with a loss of up to 59.1% of New Jersey’s current nesting area. This loss is continuous for municipalities along the east coast with increasing sea level rise. Ratios of sex-producing nesting areas continuously skew heavily toward females with increasing temperature and sea level rise, with a minimum of 0.3% of all nesting areas being capable of producing males. With strong local governance of land use and development, there is considerable potential to mitigate the effects of sea level rise for diamondback terrapins with increased protection of their current and future habitat and nesting areas.