Coastal development is a major threat to terrestrial and marine organisms that rely on these dynamic ecosystems for survival (Crain et al., 2009; Dugan et al., 2011; Patrick et al., 2016; Dugan et al., 2018). As the human population encroaches upon the coastline for recreational, economic, cultural, and aesthetic benefits, pressures on these ecosystems are tested (Kittinger and Ayers, 2010; Dugan et al., 2011; Gittman et al., 2016). Natural beaches accrete and erode, changing the shoreline over time (Hanley et al., 2014); however, construction of buildings, roads, and other infrastructure near the coast hampers these natural processes. Coastal squeeze, the degradation or complete loss of coastline as a result of anthropogenic structures restricting shoreline fluctuations, is compounded by sea level rise (Doody, 2004; Fish et al., 2008; Dugan et al., 2011). Coastal development can also have indirect impacts on the environment, including increasing pollution and accumulation of marine debris alongside developed areas (Nordstrom, 1994; Leite et al., 2014). These pressures are ever apparent in Florida, USA, whereby millions of tourists and residents are attracted annually to the nearly 1,300 km of sandy beaches (FDEP, 2021). Current estimates suggest that 83% of Florida’s coastline is critically eroding due to anthropogenic and/or natural impacts that have led to loss of beach to a degree that “upland development, recreational interests, wildlife habitat, or important cultural resources are threatened or lost (FDEP, 2021).”

Florida’s beaches are also a globally important nesting ground for threatened loggerhead sea turtles (Caretta caretta) and a regionally important area for endangered nesting leatherback (Dermochelys coriacea) and threatened green (Chelonia mydas) turtles (Meylan et al., 1995; Stewart et al., 2014; Ceriani et al., 2019). Unfortunately, 94% of Florida’s sea turtle nesting beaches are exposed to cumulative coastal modifications, and the number of permitted alterations continues to grow (Nelson Sella and Fuentes, 2019). Previous studies have evaluated the impacts of various facets of coastal development on sea turtles, including beach nourishment, artificial lighting, and armoring structures (Grain et al., 1995; Salmon et al., 1995; Butler, 1998; Brock et al., 2009; Lopez et al., 2015; Hirsch et al., 2019). In a survey of sea turtle experts across four continents, it was found that beach armoring was perceived as the largest threat to sea turtle nesting beaches in terms of coastal development, which ranked higher than artificial lighting, special events, beach cleaning, and beach sand placement activities (Nelson Sella et al. 2019). Beach armoring (e.g., vertical seawalls, sloped revetments, wooden walls, sandbag/geotextile containers systems) is a termed used to define manmade structures that are erected parallel to the shoreline with the intent of protecting inland development from coastal erosion and storm surge (Griggs, 2005; Dugan et al., 2011). These structures are used in place of a natural dune system, which typically functions to prevent beach erosion through sand accumulation via wind and wave action in the vegetated dune and provides storm protection by dissipating waves (Hanley et al., 2014). Unfortunately, beach armoring, most notably seawalls, increases longshore current intensity, limits how much energy can be converted into sand displacement when waves crash into the dune and flatten the beach, and sharpens and promotes erosion due to the offshore profile cut-off (Pilkey Jr. et al., 1984; Hall and Pilkey, 1991); therefore, these structures threaten the sandy beaches that sea turtles rely on to complete their reproductive cycle.

In addition to habitat loss, barriers that limit access to suitable sandy beach habitat can deter nesting sea turtles from completing the egg-laying process. When a female emerges, but abandons her nesting attempt, she will often return later that night or on subsequent nights to nest (Davis and Whiting, 1977; Talbert et al., 1980; Limpus, 1985; Miller, 1997). This results in additional energy expended by the female (Prange and Jackson, 1976; Mosier, 1998; Burns et al., 2016). Some females may still nest despite encountering an obstacle (i.e., obstruction); however, they may deposit their eggs in sub-optimal habitat. Nesting loggerheads that encountered a portable wall barrier nested just as frequently as those that were given access to the entire beach but laid their nests significantly closer to the waterline (Witherington et al., 2011). Similarly, leatherbacks that encountered steep escarpments due to beach erosion had more abandoned nesting attempts and were more likely to deposit their eggs in high-risk areas (Rivas et al., 2016). Depositing eggs closer to the waterline can increase tidal inundation, negatively impact developing embryos through suffocation, reduce hatching success, alter hatchling fitness, or lead to nest washout (Whitmore and Dutton, 1985; McGehee, 1990; Milton et al., 1994; Martin, 1996; Foley et al., 2006; Caut et al., 2010; Pike et al., 2015; Erb et al., 2018; Fleming et al., 2020; Limpus et al., 2020). Furthermore, if there is less available habitat for nests to be laid, there may be increased incidence of predation, microbial infection, or destruction of nests by other nesting females on high-density nesting beaches (Mazaris et al., 2009).

While coastal armoring is considered a key threat to the preservation of threatened and endangered sea turtles as it may directly impact nesting behavior and reproductive success, few studies aim to quantify these impacts (Horrocks and Scott, 1991; Mosier, 1998; Mosier and Witherington, 2002; Rizkalla and Savage, 2011). Mosier (1998) found that loggerheads emerged from the ocean less often and were less likely to nest in front of a seawall; however, the study was limited temporally (across 25 d of nesting season. The only other in-depth study on coastal armoring impacts also concluded that loggerheads nest less frequently in front of seawalls and nests that were deposited were more susceptible to being washed out during storm events (Rizkalla and Savage, 2011). Both studies focused solely on loggerheads, whose nesting behaviors vary from that of other species (Dodd, 1988). Therefore, the goal of this study was to use a historical dataset to evaluate the impacts of coastal armoring on different species of sea turtle. Because Florida is a hotspot for coastal erosion, beach armoring, and sea turtle nesting (Fuentes et al., 2016), the objectives of this study were to (1) examine nest density, nesting success (i.e., the proportion of successfully laid nests compared to the total number of crawls), washout rates, and hatching and emergence success at hard-armoring sites in comparison to a control area and (2) characterize the impacts of obstructions encountered by leatherback, loggerhead, and green sea turtles nesting in northern Palm Beach County, Florida USA.

Previous studies examining the impacts of hard-armoring structures have almost exclusively focused on loggerheads (Mosier, 1998; Mosier and Witherington, 2002; Rizkalla and Savage, 2011). Therefore, the information presented here improves upon our understanding of the effects of coastal armoring to include different species of sea turtles at a high-density nesting beach. While the general nesting process is similar for leatherback, loggerhead, and green turtles, these three species do exhibit some variation in their nesting behaviors (Miller, 1997) that likely led to some of the varying results found in this study. Leatherbacks tend to lay their nests on the open beach (Whitmore and Dutton, 1985; Nordmoe et al., 2004), are less sensitive to disturbances than other sea turtle species, and typically complete the nesting process if they emerge onto the beach (Carr and Ogren, 1959; Hughes et al., 1967; Tucker, 1988; Meylan et al., 1995; Reina et al., 2002). Nest abandonment in leatherbacks likely has a higher energetic cost than smaller sea turtle species; therefore, unsuccessful nesting attempts could have been selected against throughout their evolutionary history to conserve energy. Loggerheads usually lay their nests between the open beach and the supra-littoral vegetation, abandon nesting attempts when disturbed by lighting or movement, and have nesting success rates of ~50% (Dodd, 1988). Green turtles also nest successfully ~50% of the time but tend to lay their nests on the upper portion of the nesting beach, in or near vegetation, and are particularly sensitive to movement, artificial light, and vibrations (Hendrickson, 1958; Bustard and Greenham, 1969; Weishampel et al., 2003; Witherington et al., 2006).

Leatherback nesting success was not significantly impacted by the hard-armoring structures; however, nest density was significantly higher at the hard-armoring structures in comparison to the control site. These results reflect high nesting success rates of leatherbacks and their tendency to nest despite facing disturbances (Carr and Ogren, 1959; Hughes et al., 1967; Tucker, 1988; Meylan et al., 1995; Reina et al., 2002). Numerous obstacles to nesting besides hard-armoring structures are present on Tequesta and Jupiter Inlet Colony Beaches; therefore, we were also interested in evaluating how nesting success was impacted by various other obstructions to help elucidate if the physical barrier may be the primary reason for nest abandonment. We found that leatherback nesting success was not significantly different for turtles that physically encountered obstructions. Our findings support previous literature that leatherbacks are likely to nest when emerging onto the beach, regardless of encountering obstructions. There were no differences in washout rates or hatching success and emergence success between the two sites for leatherbacks – possibly a result of timing of nesting, as leatherbacks nest earlier in the season and may therefore be exposed to less storm activity and subsequent tidal overwash (Stewart and Johnson, 2006; Dewald and Pike, 2014). Leatherback nesting was similar between the control and hard-armoring sites, which suggests that seawalls may not hinder the nesting process for this species. The results presented here should be interpreted cautiously as sample sizes for leatherbacks were low (1–19 nests/year across both sites) and similar analyses on higher-density nesting beaches are warranted.

Almost all studies evaluating impacts of coastal armoring on sea turtles focus on loggerheads (Mosier, 1998; Mosier and Witherington, 2002; Rizkalla and Savage, 2011). We found that loggerhead nest density did not differ between the control and hard-armoring structures, which contradicts findings from loggerheads nesting in Indian River County, Florida (Rizkalla and Savage, 2011). These dissimilarities may be due to differences in beach width and hard-armoring type (e.g., rock revetment, vertical seawall, etc.). The present study evaluated impacts from a range of sloped revetments and vertical seawalls, whereas the Indian River County study only evaluated vertical seawalls (Mosier and Witherington, 2002), which may be perceived differently from gradual sloped structures. In the current study, nesting success was significantly lower at the hard-armoring structures, similar to loggerheads nesting at three study areas from Melbourne Beach to Wabasso Beach, Florida (Mosier, 1998). We also found that nest densities were similar between the two sites, indicating that more loggerheads attempted to nest at the hard-armoring site than at the control site. It is still unclear as to what the turtles react to in terms of beach-based structures, as loggerheads that were presented with a portable wall barrier nested just as successfully as those that were allowed to access the entire beach. Witherington et al. (2011) concluded that loggerheads may be reacting to visual topographical cues prior to interaction with a physical barrier.

Loggerheads from this study that physically encountered hard-armoring structures had significantly lower nesting success in comparison to those that encountered other obstructions or no obstructions, indicating a negative effect of hard-armoring structures on loggerhead nesting behavior. Surprisingly, nesting success was also 2% lower for turtles that did not encounter any obstructions in comparison to those that encountered other obstructions. The reason for this difference is unclear; however, nest site selection is poorly understood in sea turtles and is likely affected by a combination of factors including physical obstructions and barriers, elevation, temperature, moisture, and sand characteristics (Miller et al., 2003). Lastly, there were no differences in washout rate or hatching and emergence success at the hard-armoring structures compared to the control site for loggerheads. No data were available on whether large erosive events took place during the study period, so we are uncertain if washout events and hatching and emergence success may have been negatively impacted by erosion. These data were not available during our study period and this question requires further investigation and increased sample sizes.

In the present study, green turtles were seemingly most affected by the hard-armoring structures, showing significantly lower nest density and nesting success at the hard-armoring structures in comparison to the control site. Because green turtles prefer to use upper portions of the beach (Brock et al., 2009), they were more likely to interact with these structures. Overall, 32% of the green turtles emerging onto the beach physically interacted with the hard-armoring structures, compared to just 18% of loggerheads and 13% of leatherbacks. Additionally, green turtle nesting success was significantly lower for those turtles that encountered the concrete or rock seawall compared to those that encountered other obstructions or no obstructions; however, no difference in nesting success was observed between green turtles that encountered other obstructions (e.g., trees, debris, natural rock, escarpments, etc.) or no obstructions. This suggests that hard-armoring structures reduce nesting success in green turtles due to their preference to nest in vegetated areas high on the beach (Bustard and Greenham, 1969; Brock et al., 2009), features that are minimal or absent at the hard-armoring site. Abandoned nesting attempts expend limited energy reserves and can result in turtles laying nests in high-risk habitats (Mosier, 1998). However, we found no difference in washout rates or hatching and emergence success for green turtle nests at the hard-armoring site; therefore, while green turtles deposited fewer nests in this area, those that were laid successfully were not more susceptible to lower hatching or emergence success.

It is important to note that there are several differences between the control and hard-armoring sites in this study that could have confounded our results. The zones where the hard-armoring structures occur are more residentially developed and nesting behavior may also be influenced by artificial lighting, although Palm Beach County regulates beachfront lighting through their Sea Turtle Protection and Sand Preservation Ordinance requiring lights to be shielded and of long wavelengths (Unified Land Development Code Article 14, Chapter A). Additionally, fine-scale nest-site selection can be influenced by a variety of environmental cues including temperature, moisture content, salinity, and beach slope, in addition to beach width and obstructions (Wood and Bjorndal, 2000), all of which were not evaluated in this study. The control and hard-armoring sites were adjacent to each other, and so abiotic variations were likely minimal. However, because the sites were in proximity, turtles that emerge in one treatment may have ultimately ended up crawling into the other treatment area. A limitation of field-based studies is the inability to control for numerous factors and before-after-control-impact (BACI) designs (e.g., Hirsch et al., 2019) can help limit confounding factors by evaluating both control and study sites before and after the habitat alterations. A BACI design was not possible in the current study as a comprehensive sea turtle nesting dataset was not available prior to the installation of the hard-armoring structures. Future studies should also aim to determine the driving factors for sea turtle nesting behavior, in addition to comparing how different armoring structures (e.g., vertical seawalls, sloped revetments, geotextiles containers, etc.) impact nesting, since these influence approach to the nesting beach and nest site selection (Schroeder and Mosier, 2000).

Coastal armoring is increasing in frequency on sea turtle nesting beaches around the globe; therefore, the data presented here provide a better understanding into the impacts that these structures have on leatherback and green turtles, which have not previously been evaluated. Additionally, this study evaluated a longer temporal and spatial scale for loggerheads than previous studies (Mosier, 1998; Mosier and Witherington, 2002; Rizkalla and Savage, 2011). Identifiable changes in population size for sea turtles can often take decades due to the time it takes for sea turtles to reach reproductive maturity (Mazaris et al., 2017). Since 2007, we have seen increases in loggerhead nesting on Tequesta, Coral Cove, and Jupiter Inlet Colony beaches; however, Florida’s statewide nesting population of leatherbacks, loggerheads, and green turtles have shown population declines (NALWG, 2018), stable trends (Ceriani et al., 2019), and significant increases in nest numbers, respectively, although in recent years the nesting trends of green turtles have become less predictable (FWC, 2021). Green turtles were most impacted by coastal armoring, as they use the upper portions of the beach where these types of structures are likely to be installed, increasing the chance of physical interaction with the structure, and ultimately resulting in a reduction in nesting success. Decreases in nesting success can lead to individual turtles constructing sub-optimal egg chambers on subsequent nesting attempts or release of the eggs in the water (Mosier, 1998). Additionally, the impacts of the increased energy expenditure/exertion from multiple unsuccessful nesting attempts to sea turtle physiology are unknown and could possibly lead to metabolic alterations (Phillips et al., 2015; Innis et al., 2010), especially considering the unique physiologic state and decreased nutritional condition of nesting sea turtles (Plot et al., 2013; Stacy et al., 2013; Perrault et al., 2014; Perrault and Stacy, 2018; Page-Karjian et al., 2020), particularly those that are further along in nesting season (i.e., have laid more clutches). Because hard-armoring structures often lead to habitat loss, coastal managers should carefully consider impacts to nesting turtles and setback limits (i.e., how far away the structure is from the waterline) of these structures if they are to permit their installation; however, more natural methods (e.g., dune restoration) to beach armoring should first be considered.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The animal study was reviewed and approved by Florida Fish and Wildlife Conservation Commission on Marine Turtle Permit #154.

Conceptualization: SEH and JRP; methodology: SEH, JDR, SRH, and JRP; validation: SEH, JDR, and SRH; formal analysis: SEH and JRP; data curation: SEH, JDR, and SRH; writing – original draft preparation: SEH, MT, and JRP; writing – review and editing: SEH, MT, JDR, SRH, and JRP; funding acquisition: SEH and JRP. All authors have read and agreed to the published version of the manuscript.

Funding for this project was acquired, in part, from the Florida Department of Environmental Protection (#20-PBL, #21-PBL), Palm Beach County Environmental Resource Management, and Loggerhead Marinelife Center.

The authors wish to thank the staff and volunteers of Loggerhead Marinelife Center for assistance with data collection. We also thank Meghan Koperski of Florida Fish and Wildlife Conservation Commission for assistance with permitting.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.