2013 NORTHEASTERN NATURALIST 20(2):255–274 Spatial Ecology of Fiddler Crabs, Uca pugnax, in Southern New England Salt Marsh Landscapes: Potential Habitat Expansion in Relation to Salt Marsh Change Yi Chuan Luk1,2 and Roman N. Zajac1,* Abstract - The spatial distribution of the fiddler crab Uca pugnax (Atlantic Marsh Fiddler Crab) in relation to salt marsh patch structure was investigated along the central Connecticut coast of Long Island Sound. Salt marsh landscape structure at the study sites exhibit characteristics consistent with changes noted in other systems along the US Atlantic coast over the last several decades, including significant seaward erosion, encroachment of low-marsh plants into high marsh, changing composition of high-marsh plant patch structure, and marsh dieback and drowning. Our objective was to determine whether the spatial patterns of U. pugnax inhabiting these systems differed from those previously reported for southern New England in light of these characteristics. Densities of crab burrows were highest in low-marsh patches of Spartina alterniflora (Atlantic Smooth Cordgrass) and unvegetated muds along tidal creek banks and mosquito ditches. Seaward-eroding low-marsh areas were generally devoid of live crabs and burrows. Crabburrow densities varied across the complex patch mosaics in high-marsh areas. Burrow densities were generally low in the extensive short S. alterniflora patches that comprised much of the high-marsh area at several sites. However, high burrow densities, equivalent to low-marsh densities, were found in certain high-marsh patch types and upland transition zones. These included patches of Spartina patens (Marsh Hay Cordgrass), Distichils Spicata (Desert Salt Grass), and mixes of these, and particularly in S. patens patches wholly or partly comprised of hummocks of vegetation surrounded by bare sediment. At several sites, burrow densities were high in upland transition zone patches of Phragmites australis (Common Reed). As such, crab-burrow distributions were highly variable at local, within-marsh system spatial scales. Live U. pugnax were found regularly in all patch types on all marshes. Our results indicate a much broader distribution of U. pugnax at relatively high densities across southern New England marsh landscapes than previously reported. This finding may represent a case of habitat expansion in response to salt marsh change, likely due to sea-level rise and other factors, creating high-marsh habitats in a variety of patch types that can support resident populations of fiddler crabs. Such an expansion of a dominant salt marsh species, which can significantly affect ecosystem dynamics, may potentially increase the complexity of current salt marsh change patterns and dynamics along southern New England coastlines. Introduction New England salt marsh landscapes have been typified by a general zonation of low-marsh, high-marsh, and upland-transition plant patches determined by tidal range and duration, plant tolerances to inundation, and related physical 1Graduate Program in Environmental Science, Department of Biology and Environmental Science, University of New Haven, 300 Boston Post Road, West Haven, CT 06516. 2Current address - University of Maine Learning Center at Bryant Pond, PO Box 188, Bryant Pond, ME 04219. *Corresponding author; rzajac@newhaven.edu. 256 Northeastern Naturalist Vol. 20, No. 2 and chemical conditions (e.g., salinity), and biotic interactions among the plants and animals inhabiting the marsh (Bertness 1985, 1991, 1992; Bertness and Ellison 1987; Niering and Warren 1980). Local disturbances generate open patches within these zones that go through successional changes, adding to the spatial variation of plant community structure. Other features such as tidal creeks, pools, and mosquito ditches in many marshes add to overall marsh landscape structure. It has long been appreciated that salt marshes are dynamic and changing systems (Miller and Egler 1950, Orson 1999), and have historically been impacted by human activities (Gedan et al. 2009, Kirwan et al. 2011). However, over the last several decades it has become increasingly evident that marshes in New England, along the US coast, and indeed globally, are exhibiting a suite of physical and ecosystem alterations. These include marsh erosion and loss, changing plant community structure, and marsh plant dieback, and they appear to be shifting these ecosystems into different states or causing their conversion to tidal-flat habitats (e.g., Adam 2002, Baily and Pearson 2007, Castillo et al. 2000, Donnelly and Bertness 2001, Gedan et al. 2011, Hartig et al. 2002, Holdredge et al. 2009, Miller et al. 2001, Smith 2009, Smith et al. 2012, Tiner et al. 2006, Warren and Niering, 1993). A major question that arises is how are resident and non-resident fauna that inhabit and/or utilize salt marsh ecosystems responding to changes occurring in many of these systems? Although many studies have focused on salt marsh change relative to alterations in vegetation and geomorphologic processes, relatively few have addressed how salt marsh fauna may be responding (e.g., Rozas and Reed 1993), although many acknowledge that such changes will likely affect both resident and temporary fauna (e.g., Gedan et al. 2011, Greenberg et al. 2006, Warren and Niering 1993). We examined variation in the spatial distribution and abundance of Uca pugnax Smith (Atlantic Marsh Fiddler Crab) populations along the central Connecticut coast of Long Island Sound in salt marshes that exhibit characteristics consistent with salt marsh change. Uca pugnax is the most common species of fiddler crab in salt marshes along the southern New England coast. Our objectives were to determine the extent to which crab populations are restricted to certain portions of salt marsh landscapes, to assess if there is any indication that the crabs may be changing the range of habitats/patch types they occupy relative to previously reported distributions, and to determine whether the distributions found are a response to apparent habitat changes in salt marsh landscapes. Fiddler crabs are ubiquitous in salt marshes along the Atlantic coast of the US (Grimes et al. 1989, Teal 1958). Their northern limit is along the Massachusetts coast (Barnwell and Thurman 1984), and may be controlled by water-temperature effects on larval development as shown for Uca pugnax (Sanford et al. 2006). Uca spp. are integral components of salt marsh ecosystems via their behaviors and contributions to food-web dynamics (Bertness 1985, Krauter 1976, Montague 1980, Teal 1962). Numerous studies have focused on the behavioral, reproductive, and physiological ecology of these species (Crane 1975, Montague 1980, Nabout et al. 2010, Vernberg and Vernberg 1975), but relatively few studies have examined other aspects of their population ecology, including to what extent abundances and spatial distributions vary among and within marsh 2013 Y.C. Luk and R.N. Zajac 257 systems in a geographic area, and what factors may contribute to any differences. Furthermore, most population studies of North American Uca have been conducted in salt marsh systems along the southern Atlantic and Gulf coasts of the US, and less is known about the population ecology of fiddler crabs inhabiting salt marshes along the northeast coast. The spatial distribution of fiddler crabs in salt marshes along the US Atlantic coast varies geographically, likely due to regional differences in salt marsh landscape geomorphology, spatial extent, and hydrologic conditions. Teal (1958) found that Uca pugnax had a broad spatial distribution on Georgia salt marshes, with highest abundances in medium Spartina alterniflora Loisel (Atlantic Smooth Cordgrass) levees and in short S. alterniflora low-marsh zones (see Gallagher et al. 1988 regarding the different ecomorphs of S. alterniflora). Lower numbers were found on unvegetated creek banks and high-marsh short S. alterniflora areas. No individuals were found on marsh edges where tall S. alterniflora grows. Teal (1958) suggested that the distribution of U. pugnax is likely determined primarily by its preference for saline conditions and vegetated sediments that are comprised primarily of mud. Uca pugnax’s preference for higher salinities has been shown in several other studies (e.g., Miller and Mauer 1973). Aspey (1978) reported that U. pugnax inhabits areas that are about 60 cm below the high-tide mark in Georgia marshes. Wolf et al. (1975) found similar abundances of U. pugnax in medium and short S. alterniflora zones on a Georgia marsh as did Cammen et al. (1984) on several North Carolina marshes. More northern salt marshes are not as spatially expansive as those along the southeastern US coast, and specific vegetation zones/patches tend to be narrower/ smaller relative to tidal gradients. The distribution of Uca pugnax appears to be limited to portions of creek banks and low-marsh areas dominated by tall S. alterniflora and narrow, low-marsh/high-marsh transition areas that can be comprised of a mix of plants including short S. alterniflora and Spartina patens (Aiton) Muhl. (Marsh Hay Cordgrass). Fiddler crabs have generally not been reported in any significant densities in patches dominated by S. patens and other grasses such as Distichlis spicata (L.) Greene (Desert Salt Grass). In New Jersey, Bergey and Weis (2008) found U. pugnax burrows limited to open mud flats and edges of lowmarsh S. alterniflora at a density of ≈158 burrows m-2. In a study conducted along the central coast of Connecticut, McCaffrey (1977) found that abundances of U. pugnax declined greatly over short distances from very high abundances (≈254 burrows m-2 ) on creek banks dominated by tall S. alterniflora to the transitional low-marsh/high-marsh area 2 m away (≈64 burrows m-2). Almost no burrows were found in the middle of the adjacent S. patens meadow (≈0–5 m-2). Similar distributions were reported on Cape Cod salt marshes (Jaramillo and Lunecke 1988, Katz 1980, Krebs and Valiela 1978). Bertness and Miller (1984) and Bertness (1985) reported that most U. pugnax burrows in a protected salt marsh in Narragansett Bay, RI, were found at intermediate tidal heights at the transition from the marsh edge to the high-marsh (what they referred to as the marsh flat). Few burrows were found in the high-marsh areas dominated by S. patens and D. spicata, and these were primarily in bare areas. Low abundances in high-marsh areas appear to be due 258 Northeastern Naturalist Vol. 20, No. 2 to dense root mats of S. patens and other plants that retard their ability to burrow (Bertness and Miller 1984, Ringold 1979). Experimental work indicated that U. pugnax can inhabit high-marsh areas if substrate they can burrow into is available (Bertness and Miller 1984). Although the focus of some of the studies cited above was not specifically on the spatial distribution and abundance of Uca on northeast salt marshes, they do provide the only data available to assess general trends and spatial patterns across the region. Based on these data, U. pugnax appears to be primarily restricted to low-marsh and low/high-marsh transition areas in southern New England salt marshes. Fiddlers can occupy patch types in high-marsh patches when substrate conditions are amenable to burrowing. With respect to the types of salt marsh alterations noted above, we might predict that the limited spatial distribution of fiddler crabs on southern New England marshes may change depending on the nature of changes in salt marsh landscapes and any associated changes in habitats preferred by Uca pugnax. For example, severely eroding seaward areas of salt marshes may not be able to support Uca populations, whereas shifts in vegetation patterns due to sea-level rise, such as expanding areas of short Spartina alterniflora and/or decreasing areas of S. patens (Warren and Niering 1993), or shifts in biotic interactions that increase bare areas on the marsh (e.g., Holdrege et al. 2009), may increase the area of favorable habitat for U. pugnax, resulting in altered spatial distributions and population abundances across these marsh systems. Salt marshes in southern New England currently exhibit highly variable vegetation patch structure from one marsh to another (see Results), likely due to the complex set of factors that determine salt marsh landscape structure including both natural physical and ecological dynamics, and human impacts interacting in different ways to shape specific systems. Some have the typical zonation noted above, and others are better characterized as mosaics of different types of patches. In this study, we examined how the distribution of U. pugnax varies relative to such salt marsh mosaics. Methods Study sites This study was primarily conducted at three salt marsh systems along the central coast of Connecticut (Fig. 1; see also Supplemental File 1, available online at https://www.eaglehill.us/NENAonline/suppl-files/n20-2-1156-Zajac-s1, and, for BioOne subscribers, at http://dx.doi.org/10.1656/N1156.s1). They are referred to here as the Banca and Pleasant Point marshes in Branford and the Chaffinch Marsh in Guilford. The Banca and Pleasant Point marshes are separated into front and back marshes by an old trolley track that is now a part of the Branford Trolley Trail. The Chaffinch Marsh is located behind coastal dunes which are fronted by a small salt marsh exposed to Long Island Sound (LIS). Collectively, the field sites included: Banca Front (BF), Banca Back (BB), Pleasant Point Front (PF), Pleasant Point Back (PB), and Chaffinch Back (CB). The separation of front and back marshes for the study areas reflects an up-estuary tidal gradient and exposure to LIS, thereby providing a broader representation of the variation in salt marsh landscape structure and physical conditions at these study sites. The fronting mashes of Chaffinch 2013 Y.C. Luk and R.N. Zajac 259 were not included in the crab sampling as no Uca pugnax were ever found in this portion of the marsh over the course of the study. This marsh is undergoing significant erosion and has only small areas of low marsh with much of the remaining high-marsh areas elevated at approximately mean higher high water (MHHW). In addition to the main study sites, several other marshes were sampled to obtain additional information on the spatial distribution of U. pugnax on different marshes. These sites were only sampled once, and included East River, Guilford Marina, Hoadley Creek, and Shell Beach marshes (Fig. 1). The region of LIS where the study sites are located has semi-diurnal tides with a mean tidal range of 1.88 m (measured in New Haven, CT; NOAA 2007). The elevations of the low-marsh and high-marsh areas sampled ranged ≈0.1 m–0.7 m, and ≈0.7 m–1.0 m above mean Figure 1. Locations and aerial photographs of the primary study sites along the central Connecticut coast of Long Island Sound (LIS). Sites are designated as B = Banca, P = Pleasant Point, H = Hoadly Creek, S = Shell Beach, C = Chaffinch Island, G = Guilford Marina, and E = East River. Patches are designated as Sa = Spartina alterniflora, sSa = short Spartina alterniflora, Sp = Spartina patens, Ds = Distichlis spicata, Sp/Ds = Distichlis spicata and Spartina patens mix, Sp/sS = Spartina patens and short Spartina alterniflora mix, and Pa = Phragmites australis. Arrows designate the general distribution of vegetation patch types. All areas abutting tidal creeks were comprised of S. alterniflora and bare patches, whereas the high-marsh areas are mosaics of short S. alterniflora, S. patens, and D. spicata. “Back” and “Front” designates within-study area sub-sites. 260 Northeastern Naturalist Vol. 20, No. 2 sea level, respectively. Salinities at all sites ranged ≈19–28 psu. Uca pugnax is the most prevalent fiddler crab in the study region, and our study focused on this species. Over the sampling period, prior to that, and afterwards we observed no Uca pugilator Bosc (Sand Fiddler Crab) and only a few (less than 10) Uca minax LeConte (Red Jointed Fiddler Crab) at the study sites. The study sites were chosen to represent a spectrum of varying salt marsh landscape structures and conditions. The vegetation patch structure of the marsh systems was analyzed to provide context for assessing the spatial ecology of Uca pugnax. High-resolution (0.5-ft pixel) aerial photographs of the sites obtained in 2004 were entered into a geographic information system (GIS; ESRI ArcGIS 9.2) and initial patch types were identified based on differences in vegetative cover in the aerial photographs and preliminary field observations. Primary vegetative cover types were determined through extensive field observations using global positioning system (GPS) both prior to and during sampling crab populations. Patches identified in the GIS interpretations were located in the field, and the dominant grass types were determined based on percent cover in multiple 0.25-m2 quadrats. Patches at the study sites were categorized into nine types: bare (unvegetated), tall Spartina alterniflora, short S. alterniflora, S. patens, Distichlis spicata, D. spicata and S. patens mix, S. patens and short S. alterniflora mix, Phragmites australis (Cav.) Trin. ex Steud. (Commom Reed), and salt pannes. Bare patches are muds devoid of live, aboveground vegetation typically found in low-marsh areas. Field sampling and analysis methods Sampling was conducted during the summer and fall of 2009. Data were collected over two days each month, and each of the main field sites was surveyed at least three times during the study period (except PB, which was sampled only twice). Surveys were conducted during the day at low tides, generally over several hours around slack low tide. Weather conditions on sampling days were generally similar: sunny or slightly overcast and no precipitation. Sampling was done during both spring and neap tidal periods. Because we focused on burrow counts and not actual counts of fiddler crabs, we do not feel that differences in tidal stage and daily weather conditions would bias our counts. Sampling was conducted several times in each main marsh system studied over the study period in order to incorporate any longer-term changes in burrows that may be occurring. A total of 5–8 transects were sampled in each of the front and back areas of the main study salt marshes in different portions of each area. The secondary sites were sampled only once, and data from these were consolidated due to the low number of sampling points per patch types. For each survey, transects were laid out from low-marsh areas along tidal creeks, mosquito ditches, or the marsh front dominated by tall Spartina alterniflora and bare muds into high-marsh areas where various types of vegetation patches where found, to the upland edge (if present depending on the orientation of the transect). This layout was chosen in order to sample patch types (determined by their dominant vegetation) along general elevation gradients. The lengths of transects varied depending on location. Within each patch type, one to three GPS points were taken depending on 2013 Y.C. Luk and R.N. Zajac 261 the size of the patch. At each GPS point, three 0.25-m2 quadrats were deployed haphazardly around the GPS point but within the same vegetation patch type. A total of 730 quadrats were sampled over the course of the study. In each quadrat, the number of burrow openings ≤2 cm in diameter was counted. We did not count larger-sized burrows in order to exclude other crab species such as Sesarma reticulatum Say (Purple Marsh Crab) which also inhabits these marshes, although none have been observed on the BF site, and to not count burrows or burrow-like features that may have been created and/or enlarged by erosion and were potentially not occupied by U. pugnax. The dominant vegetation types in the quadrat were recorded, as well as the presence or absence of live U. pugnax (males and/ or females) in the vicinity of the GPS point in the observer’s field of view in the specific patch being sampled. Our data on live crabs is presence/absence data, not counts, and given our sampling methods, multiple times of sampling, and other observations (see Results), we feel the presence/absence data provide an unbiased indication of the spatial distribution of live crabs on the marsh. Uca pugnax burrows are generally 1–2 cm in diameter at the opening (Crane 1975), and we used 2 cm as our cutoff size for burrow counts. The visual quantification of fiddler crab burrows is a noninvasive and consistent approach to estimate the relative number of fiddler crabs in a given area (Bertness and Miller 1984, Jordao and Oliveira 2003). Krebs and Valiela (1978) found a significant correlation between the number of U. pugnax burrows and the number of U. pugnax in sample plots. However, the number of burrows in a quadrat may not equal the number of individuals in that area. An individual crab may be responsible for excavating numerous burrows, while many of them are left unoccupied at any given time; each burrow may also have more than one opening (Crane 1975). Despite the potential over-estimation of Uca population size based on burrow counts (Macia et. al. 2001, Macintosh 1988), this method provides information regarding the relative abundance of the species and can be used to compare populations of U. pugnax across different spatial scales on salt marshes. Because our focus was on the spatial distribution of Uca pugnax relative to salt marsh landscape patch structure, data from all sample dates were combined for each study site. Differences in burrow abundance among and within several spatial scales were analyzed using nested analysis of variance (ANOVA). Variance partitioning was performed to determine the contribution of each spatial scale of the nested, spatial hierarchy to the overall variability in burrow abundance. In these analyses, the region was comprised of three marshes (Banca, Pleasant Point, and Chaffinch) within which were nested the sub-sites (e.g., BF, BB). Vegetation patches were nested within each of the sub-sites, and sets of quadrats were nested within patches. Based on the results of the nested ANOVA and the variance component analysis, one-way ANOVAs were performed to test differences among patch types. If significant differences were found, Tukey-Kramer post hoc tests were used to assess differences among specific patch types. The presence/absence of live U. pugnax in different vegetation patch types was examined by calculating the percentage of sampling points with live crabs present for each patch type. A chi-square test was performed to determine the association between the presence/ 262 Northeastern Naturalist Vol. 20, No. 2 absence of live U. pugnax and the burrow counts in the patches. All analyses were conducted using either SPSS or NCSS statistical software. Results Salt marsh patch structure As is typical of southern New England salt marshes, low-marsh areas at the study sites were comprised of a mix Spartina alterniflora and bare patches, and were ≈4–6 m wide in most areas along the wider tidal creeks (Fig. 1). Along tidal creeks, and sometimes mosquito ditches, the majority of bare patches consisted of relatively smooth unvegetated muds, but along low-marsh areas directly on Long Island Sound, as at the BF study site, and along some portions of tidal creeks, these bare areas were highly eroded muds and peats with much higher surface roughness, often with spatially complex labyrinths of holes (≈2–15 cm in diameter) permeating the peat (for images of patch structure, see Supplemental File 1, available online at https://www.eaglehill.us/NENAonline/ suppl-files/n20-2-1156-Zajac-s1, and, for BioOne subscribers, at http://dx.doi. org/10.1656/N1156.s1). Low-marsh areas along mosquito ditches are limited to a narrow, ≈1–2-m-wide, dense band of tall S. alterniflora. At the BF site, the low marsh is relatively broad with S. alterniflora extending 20–30 m upland from the highly eroded marsh face, likely as a result of the greater spatial extent of tidal inundation. In several areas of BB and PF, small clumps and individual stems of S. alterniflora were found at the low-marsh/high-marsh transition and encroaching into the high marsh, indicating localized expansion of S. alterniflora into higher marsh areas, as has been found in other salt marshes (Donnelley and Bertness 2001; Supplemental File 1, available online at https://www.eaglehill.us/ NENAonline/suppl-files/n20-2-1156-Zajac-s1, and, for BioOne subscribers, at http://dx.doi.org/10.1656/N1156.s1). At no site was Phragmites australis present along low-marsh areas and creeks, as has been found in some brackish, disturbed portions of northeast salt marsh systems (e.g., Weis and Weis 2003, Windham and Lathrop 1999). Overall, there were relatively narrow low-marsh zones in each study area, except BF, with distinct low-marsh patch characteristics, and no indication that these areas were moving up-slope in any significant way at these sites. At BF, the broader low marsh is the result of a lower slope and related elevations across this section of marsh that fronts LIS. The high-marsh areas of the primary study marshes were complex patch mosaics of short Spartina alterniflora, S. patens, Distichlis spicata, pools and bare areas (Fig. 1; Supplemental File 1, available online at https://www.eaglehill.us/ NENAonline/suppl-files/n20-2-1156-Zajac-s1, and, for BioOne subscribers, at http://dx.doi.org/10.1656/N1156.s1). The BB, PF, and PB sites are dominated by extensive patches of short S. alterniflora, with some small patches of S. patens (generally less than 400 m2). The PF site had several larger patches of D. spicata along the upland border mixed with some S. patens. The BF site had only a very narrow high marsh comprised of small patches of short S. alterniflora and a few patches of S. patens. The CB high-marsh areas were comprised primarily of large patches of Distichlis spicata mixed with smaller patches of S. patens and short 2013 Y.C. Luk and R.N. Zajac 263 S. alterniflora. Some patches had mixtures of these species. At the Banca and Pleasant Point sites, the seaward/creekward edges of many of the larger S. patens patches were characterized by small hummocks of S. patens with areas of bare mud among the hummocks which then transitioned to more typical, densely vegetated S. patens with a flatter topography and a thicker continuous root mat. The smaller S. patens patches, especially in the BF site, were totally hummocked in this fashion (Supplement File 1, available online at https://www.eaglehill.us/ NENAonline/suppl-files/n20-2-1156-Zajac-s1, and, for BioOne subscribers, at http://dx.doi.org/10.1656/N1156.s1). The upland transition of the main study marshes was dominated by patches of Phragmites australis of varying width, or other upland plants where there were sharp increases in elevation, as along the eastern side of Pleasant Point where the marsh abuts a headland and the northern border of BB, which abuts a raised railroad bed. Spatial distribution of Uca pugnax burrows Differences in Uca pugnax burrow density were not consistent among the spatial scales within the central Connecticut salt marsh landscapes examined (Table 1). There were no significant differences in burrow density at the largest spatial scales, neither among marsh systems nor among the sub-sites within a marsh system. In contrast, there were highly significant differences in burrow density among patches within sub-sites and among different locations within specific patch types (Table 1). A variance component analysis indicated that the greatest of variation in burrow density is accounted for at the patch and withinpatch level on the salt marshes studied (Table 2), with variation among patch types within sub-sites accounting for ≈58% of the variation and variation within patches explaining ≈29% of the variation. Crab-burrow densities were significantly different among patches within each field sub-site (one-way ANOVA: P < 0.001 for each sub-site). In all the marshes Table 1. Nested ANOVA results testing differences in Uca pugnax burrow abundance at different spatial scales. Spatial scale Type III SS DF MS F P-value Marsh study sites 154.292 2 77.15 0.036 0.965 Sub-sites within marsh study sites 4778.339 2 2389.17 1.417 0.256 Vegetation patch types within sub-sites 72,409.970 32 2262.81 9.586 0.0001 Locations within patches 55,029.834 216 254.77 6.507 0.0001 Table 2. Variance component analysis of Uca pugnax burrow abundance at different spatial scales. The minimum norm quadratic unbiased estimation method (weight = 1 for random effects and residual) was used. Estimate of Percent variation Spatial scale variation explained Marsh study sub-sites 1.34 0.6% Vegetation patch types within sub-sites in a field site 132.47 54.5% Locations within patches 70.11 28.8% Error 39.32 16.2% 264 Northeastern Naturalist Vol. 20, No. 2 studied, the highest number of burrows was almost always found in low-marsh habitats comprised of bare muds and Spartina alterniflora patches along tidal creeks and mosquito ditches (Fig. 2). An exception was in low-marsh areas on portions of marshes fronting LIS exhibiting extensive erosion, as in BF and the portion of PF fronting LIS. There was a significant difference (one-way ANOVA: P < 0.001) in burrow abundances among these different types of low-marsh habitats (Fig. 3), with burrow counts significantly higher along tidal creek low-marsh areas than along mosquito ditches and marsh fronts; marsh front areas had significantly lower burrow counts than the other two low-marsh areas (Tukey-Kramer multiple-comparison tests: P < 0.05). Figure 2. Uca pugnax burrow densities in different salt marsh patches on the Banca Marsh system (top), Pleasant Point Marsh system (middle), and Chaffinch Marsh and other marsh systems. Patch designations as in Fig. 1. B = Bare, unvegetated muds in low marsh. SE = standard error. No burrows were found in Phragmites australis patches on Pleasant Point Front, so this patch type was omitted from the figure. Results of posthoc Tukey-Kramer tests are given as either uppercase (for front portions of marsh systems and other marshes) or lowercase (for back portions of marsh systems) letters. Patch types with different letter designations had significantly different mean burrow abundances. 2013 Y.C. Luk and R.N. Zajac 265 In high-marsh areas of the study sites, the numbers of burrows were, on the most part, lower than in bare mud and S. alterniflora patches in low-marsh areas. Few burrows were found in short S. alterniflora patches that comprised significant portions of the salt marsh landscape at all sites. The numbers of burrows in other patch types on the high marsh varied among the marshes, and relatively high burrow counts were found in certain patches on specific marshes. These included the Distichlis spicata/S. patens patches on BF and PB, as well as S. patens and D. spicata patches on PB (Fig. 2). High abundances were also found in Phragmites australis patches along the upland border on BF and similar upland-transition Phragmites patches on several of the other smaller marshes surveyed (Shell Beach and Hoadley Creek; Fig. 2). There were no burrows found in upland-transition Phragmites australis patches on PF. Several patch types on the high-marsh portions of the study sites were characterized by distinct variations in topography wherein portions of the patches, generally along the creekward edges, were comprised of hummocks of the grasses giving way to flatter, continuous mats (Fig. 4). These areas included patches of S. patens, D. spicata, and mixed S. patens/D. Spicata patches. When collecting burrow abundance data, these topographic differences were noted for each quadrat sampled in these patch types. Analysis of burrow count data among hummock and mat areas (across all marshes) indicated that burrow abundances were significantly higher in the hummock areas in all cases (Fig. 5), and that the densities of burrows were often similar to levels found in bare mud and S. alterniflora patches. Uca presence /absence in patch types In order to obtain a fuller assessment of the spatial distribution of Uca pugnax across the salt marsh landscapes, the percent occurrence of live crabs by Figure 3. Uca pugnax burrow counts in different types of low-marsh areas in the study areas. SE = standard error. Patch types with different letter designations had significantly different mean burrow abundances based on post-hoc Tukey-Kramer tests. 266 Northeastern Naturalist Vol. 20, No. 2 patch types was examined at each site. Live crabs on the surface of the marsh were found at 40% or more of the sample points in each patch types across all marshes (Fig. 6). As might be expected, the highest percentage was found in bare patches along creek banks where most burrows were found. However, crabs were present at similar levels in high-marsh patches of short Spartina alterniflora, S. patens, and Distichlis spicata and mixes of the latter two as in S. alterniflora in the low marsh. A very high percentage was also found in the upland-transition Phragmites australis patches that were sampled. Overall, there was a significant Figure 4. Photographs showing hummocks of Spartina patens on one of the central Connecticut study sites. Hummocks are ≈10–20 cm wide, ≈3–10 cm high (measured from bare sediment to sediment level at top of hummock) and ≈5–20 cm apart. Upper photo: Arrows point to fiddler crab burrows (≈2 cm in diameter). Lower photo: H indicates hummocks, B indicates bare sediments. See also Fig. ES-3, and ES-7 in Supplemental File 1, available online at https://www.eaglehill.us/ NENAonline/suppl-files/ n20-2-1156-Zajac-s1, and, for BioOne subscribers, at http://dx.doi.org/10.1656/ N1156.s1). 2013 Y.C. Luk and R.N. Zajac 267 Figure 5. Differences in Uca pugnax burrow densities among mat and hummock areas in different salt marsh patches. Sp = Spartina patens; Ds = Distichlis spicata; Sp/Ds = mixture of S. patens and D. spicata. Asterisks indicate significant difference in burrow density among hummock and mat areas; *P < 0.05; **P < 0.01; ***P < 0.001. SE = standard error. Figure 6. Frequency of live Uca pugnax on different patch types in salt marsh systems along the central Connecticut coast of Long Island Sound. Patch designations as in Fig. 1. B = bare, unvegetated muds in low marsh. Table 3. Relationships between the presence/absence of Uca pugnax burrow and the presence/ absence of live U. pugnax in the primary marsh study systems and results of associated chisquare tests. Presence of live crabs Absence of live crabs Banca Marsh (χ2 = 4.425, P = 0.035) Presence of burrows 41.3% (n = 31) 29.3% (n = 22) Absence of burrows 9.3% (n = 7) 20.0% (n = 15) Chaffinch Marsh (χ2 = 0.973, P = 0.324) Presence of burrows 30.0% (n = 12) 25.0% (n = 10) Absence of burrows 17.5% (n = 7) 27.5% (n = 11) Pleasant Point Marsh (χ2 = 7.793, P = 0.005) Presence of burrows 39.6% (n = 42) 25.5% (n = 27) Absence of burrows 11.3% (n = 12) 23.6% (n = 25) 268 Northeastern Naturalist Vol. 20, No. 2 association between the presence of live Uca and presence of burrows at Banca and Pleasant Point salt marshes but not at Chaffinch (Table 3). There were some differences among marsh sub-sites, specifically that a higher percentage of live crabs were found on high-marsh patches on BB, PB, and CB than on PF and BF. Discussion The desnity of Uca pugnax burrows was not significantly different among salt marsh systems nor among sub-sites within marshes across the region of the LIS coast studied (Table 1). The general salt marsh landscape structure is similar among the study sites, as are other gross characteristics including tidal amplitude, water temperature, salinity, and some geomorphological and hydrologic characteristics. Uca pugnax populations appear to be responding to landscape patch structure and regional environmental conditions similarly across these larger spatial scales (on the order of kms to 100s of m). In contrast, burrow density was significantly different among patch types within sub-sites, and among different locations within specific patches. These spatial scales (10s of m to less than 10 m) accounted for the greatest proportion of spatial variability in U. pugnax burrow abundance (Table 2). This finding suggests that variation in burrow densities reflect meso- to local-scale differences in the mix and characteristics (e.g., elevation, root mat density, sediment type, type and degree of vegetative cover) of patch types within specific portions of the marshes, and differences in physical dynamics associated with tides and hydrology. The higher densities of Uca pugnax burrows found within low-marsh patches relative to most high-marsh areas is not surprising and consistent with previous reports (e.g., Bertness and Miller 1984, McCaffrey 1977, Ringold 1979, Teal 1958). Bertness and Miller (1984) found that U. pugnax prefer to burrow in muddy sediments where there was an intermediate root density of S. alterniflora and in bare patches alongside structural support, such as provided by mussels and underground structures of S. alterniflora. We found the highest density of burrows almost always in bare patches of low-marsh muds, and at densities higher than in low-marsh patches of S. alterniflora. Structural elements such as mussels or roots were present in some of these patches, but these were not always evident in others. Nomann and Pennings (1998) suggested that plant structures do not necessarily support burrow walls, and that the association of burrows with vegetation may be a predator-avoidance response. In contrast to the high burrow abundance found along low-marsh creek banks and mosquito ditches, significantly lower burrow abundances were found where low-marsh areas are being actively eroded such as at BF and portions of PF (Fig. 3). At BF, the erosion rate of the fronting low marsh has been estimated to be ≈1.17 m yr-1 between 1994 and 2009 (R.N. Zajac, unpubl. data). In these low-marsh areas, the marsh surface is comprised of a mix of steep banks, shell deposits, sediments riddled with holes of varying sizes, and a lack of fine sediments (Supplemental File 1, available online at https://www.eaglehill.us/NENAonline/suppl-files/n20-2-1156-Zajac-s1, and, for BioOne subscribers, at http://dx.doi.org/10.1656/N1156.s1). These conditions apparently discourage burrowing by Uca, and indeed very few crabs were ever seen moving about on the sediment surface in these areas. 2013 Y.C. Luk and R.N. Zajac 269 The high-marsh areas of the study sites were characterized by heterogeneous mosaics of different vegetation patches, panes, and pools (Fig. 1). The lowest burrow densities were found in short S. alterniflora high-marsh patches (Fig. 2). Burrow abundances were also relatively low in the other types of high-marsh patches sampled, but not in all cases. In many areas, moderate to high burrow densities were found in portions of S. patens and Distichlis spicata patches, particularly at PB, where grasses were limited to hummocks with bare sediments between them (Fig. 4; see also Smith et al. 2012). These hummock areas are not the same as salt marsh panes, which are relatively large unvegetated patches. Some S. patens and D. spicata patches, especially at BF, were entirely comprised of hummocks with no continuous mat. These hummock areas had burrow densities that were equivalent to, and in some cases exceeded, densities found in low-marsh S. alterniflora patches. Experiments by Bertness and Miller (1984) indicated that U. pugnax can inhabit high-marsh areas if substrate they can burrow into is available. This study indicates that such vegetated hummock areas provide suitable substrates for U. pugnax to burrow into, and that they are occupying these types of high-marsh areas in numbers not previously reported. For example, Bertness and Miller (1984) found ≈18 burrows m-2 in the S. patens/D. spicata zone they sampled, and McCaffrey (1977) found ≈2 burrows m-2 in a nearby Connecticut salt marsh. In our study, we found ≈60 burrows m-2 in the high-marsh hummock areas. Root density appeared sparser in the areas among the hummocks, which may be forming due to increased inundation and peat collapse and erosion of the S. patens and D. spicata root mats (e.g., DeLaune et al. 1994, Smith et al. 2012, Warren and Niering 1993), but may also involve herbivory by Sesarma (Smith et al. 2012). The grasses on the hummocks may provide cover from predators. We also found high densities (≈52 m-2) of crab burrows in some patches of Phragmites australis along the upland transition, notably at BF and several of the secondary marshes sampled. Although not sampled, we observed numerous burrows in portions of upland-transition P. australis patches on BB. Although live Uca pugnax appeared to be concentrated in the low marsh, especially along tidal creeks and mosquito ditches where Spartina alterniflora and bare patches are the dominant patch types, our results indicate that significant numbers of U. pugnax are also active in the high-marsh areas of the salt marshes studied. We frequently observed high abundances of U. pugnax at the edges of pools on the high marsh surrounded by short S. alterniflora and Distichlis spicata, in the absence of burrows. When approached, most U. pugnax scrambled into adjacent vegetation patches for cover, and some individuals burrowed temporarily into the soft substrate in the pools. Although U. pugnax may not be creating many burrows in the extensive short S. alterniflora patches on the study marshes, the crabs are actively accessing these high-marsh habitats and feeding there, based on our observations of extensive pellets they create when feeding on the sediment surface (Grimes et al. 1989). Uca pugnax populations along the central Connecticut coast of LIS exhibit spatial patterns of population abundance that are generally similar at large, amongmarsh system scales, but are quite variable relative to within-marsh, meso-scale and local patch composition and conditions. As found in other studies, most 270 Northeastern Naturalist Vol. 20, No. 2 individuals occupy low-marsh habitats. However, our results indicate that high abundances are also found in high-marsh patch types and upland-transition habitats at some sites, and that some low-marsh habitats have become uninhabitable by U. pugnax. These patterns suggest that U. pugnax populations on southern New England salt marshes are expanding into high-marsh patches, and in some cases upland transitions, in significant numbers as these portions of salt marsh landscapes become increasingly accessible and inhabitable due to changes in environmental conditions. Alternatively, the observed distributions may be typical for northeast US salt marsh systems but not previously observed due to the lack of study of different marsh landscapes in this region. Based on the patch structure and other salt marsh characteristics at our study sites, we suggest that it is more likely that the U. pugnax distributions we found are responses to salt marsh change. The marsh patch structure at the study sites does not conform to the typical banding pattern commonly associated with New England salt marshes. Although most low-marsh areas are comprised of Spartina alterniflora and bare areas (apart from the low-marsh areas undergoing significant erosion), the high marsh areas are complex mosaics of relatively large patches of short S. alterniflora and mixes of smaller patches of S. patens and Distichlis spicata. Expansion of short S. alterniflora patches and reduction in the patch sizes of S. patens was suggested by Warren and Niering (1993) to be the result of rising sea level and increased tidal inundation on a Connecticut salt marsh east of our study sites. At most of our sites, we found S. alterniflora expanding into high-marsh areas beyond the low/high-marsh transition area (see Supplemental File 1, available online at https://www.eaglehill.us/NENAonline/suppl-files/n20-2-1156-Zajac-s1, and, for BioOne subscribers, at http://dx.doi.org/10.1656/N1156.s1) which has been associated with responses to sea-level rise (e.g., Donnelly and Bertness 2001), as well as low-marsh areas at several sites with characteristics that are consistent with salt marsh erosion/loss linked to sea-level rise that have been observed in the region (Hartig et al. 2002, Tiner et al. 2006). On an overall basis, local anthropogenic modifications, increased rate of sea-level rise, and changing ecological dynamics are likely affecting the structure and dynamics of salt marshes in Connecticut and southern New England. Due to increased frequency and duration of tidal inundation over a greater extent of high-marsh areas and subsequent effects on salinity, nutrient levels, and substrate redox potential, the typical vegetative banding of southern New England salt marshes is being modified. Extensive herbivory by the crab Sesarma as found on some Cape Cod salt marshes (Holdredge et al. 2009) can also increase patch heterogeneity, creating bare areas that may affect other marsh characteristics and dynamics, although the extent to which this occurs in LIS salt marshes is not known at this time. As such, the distribution and abundance of salt marsh fauna, such as U. pugnax, which are adapted to specific marsh environments, may be in a state of flux as they respond to changing conditions across salt marsh landscapes. The relatively high density of U. pugnax burrows in some high-marsh patches, particularly in S. patens patches that exhibit a hummock structure, and in upland-transition areas dominated by Phragmites autralis, indicate that Uca is occupying high-marsh and uplandtransition patches that are becoming more favorable habitats. The presence of 2013 Y.C. Luk and R.N. Zajac 271 live U. pugnax in all high-marsh and some upland-transition areas also supports that these crabs may be undergoing habitat expansion as salt marshes change. In some cases, the spatial patterns found appear to constitute a shift in their distribution, as found at BF where the low marsh is so eroded that it is not inhabitable by the crabs and, with increased tidal inundation, the crabs are now occupying high-marsh and upland-transition patches. Crab burrowing can affect vegetation and marsh erosion (Bortolus and Iribarne 1999, Hughes et al. 2009, May 2002, Wilson et al. 2012). It is interesting to speculate that if U. pugnax populations increase across a greater extent of southern New England salt marsh landscapes, particularly into high-marsh patches as our study suggests, whether their burrowing may exacerbate local erosion and waterlogging which may affect vegetation and create localized foci of marsh change. Spatially expanding populations may also increase potential negative effects on other critical aspects of marsh dynamics such as plant recruitment. Smith and Tyrrell (2012) found that U. pugnax burrowing and foraging activity disturbed salt marsh soils and this had negative impacts on the establishment of halophyte seedlings. Alternatively, increased burrowing may enhance the productivity of marsh grasses by increasing drainage, aeration, and decomposition of organic matter, although this effect may not be as significant in high-marsh areas (Bertness 1985). Salt marshes are dynamic ecosystems where a multitude of physical, geochemical, and biological factors, as well as human activities, are constantly reshaping the marsh landscape (e.g., Gedan et al. 2009, Holdredge et al. 2009, Orson 1999, Orson et al. 1985, Pennings and Bertness 2001). As sea-level rise and other factors impacts these systems, many physical changes are occurring, and it is essential to understand how Uca pugnax, an integral member of these ecosystems, is responding. The presence of live U. pugnax and high burrow densities in high-marsh patches suggests a correlation with physical conditions related to salt marsh change. There have been relatively few studies of how resident fauna are responding to apparent sea-level rise-related changes that are occurring on many salt marsh systems. This study revealed a spatially complex pattern of the burrow distribution of the fiddler crab U. pugnax across southern New England salt marsh landscapes not reported in previous studies, and hopefully will act as a framework for future studies of how fauna are responding to salt marsh change. Acknowledgments We thank D. Steven Brown and Stephen Klepner for assistance in the field and Brittney Gibbons for help in the lab. This work was supported in part by a University Research Scholar award to R.N. Zajac from the University of New Haven. Comments and suggestions by two anonymous reviewers and Melisa Wong greatly improved the manuscript. Literature Cited Adam, P. 2002. Saltmarshes in a time of change. Environmental Conservation 29:39–61. Aspey, W.P. 1978. Fiddler crab behavioral ecology: Burrow density in Uca pugnax (Smith) and Uca pugilator (Bosc) (Decapoda Brachyura). Crustaceana:235–244. 272 Northeastern Naturalist Vol. 20, No. 2 Baily, B., and A.W. Pearson. 2007. Change detection mapping and analysis of salt marsh areas of central southern England from Hurst Castle Spit to Pagham Harbour. Journal of Coastal Research 23:1549–1564. Barnwell, F., and C.L.I.I. Thurman. 1984. Taxonomy and biogeography of the fiddler crabs (Ocypodidae: genus Uca) of the Atlantic and Gulf coasts of eastern North America. Zoological Journal of the Linnean Society 81:23–87. Bergey, L.L., and J.S. Weis. 2008. Aspects of population ecology in two populations of fiddler crabs, Uca pugnax. Marine Biology 154:435–442. Bertness, M.D. 1985. Fiddler crab regulation of Spartina alterniflora production on a New England salt marsh. Ecology:1042–1055. Bertness, M.D. 1991. Interspecific interactions among high marsh perennials in a New England salt marsh. Ecology 72:125–137. Bertness, M.D. 1992. The ecology of a New England salt marsh. American Scientist 80:260–268. Bertness, M.D., and A. Ellison. 1987. Determinants of pattern in a New England salt marsh plant community. Ecological Monographs 57:129–147. Bertness, M.D., and T. Miller. 1984. The distribution and dynamics of Uca pugnax (Smith) burrows in a New England salt marsh. Journal of Experimental Marine Biology and Ecology 83:211–237. Bortolus, A., and O. Iribarne. 1999. Effects of the SW Atlantic burrowing crab Chasmagnathus granulata on a Spartina salt marsh. Marine Ecology Progress Series 178:79–88. Cammen, L.M., E.D. Seneca, and L.M. Stroud. 1984. Long-term variation of fiddler crab populations in North Carolina salt marshes. Estuaries and Coasts 7:171–175. Castillo, J., C. Luque, E. Castellanos, and M. Figueroa. 2000. Causes and consequences of salt-marsh erosion in an Atlantic estuary in SW Spain. Journal of Coastal Conservation 6:89–96. Crane, J. 1975. Fiddler Crabs of the World: Ocypodidae: Genus Uca. Princeton University Press, Princeton, NJ. 736 pp. DeLaune, R., J. Nyman, and W. Patrick, Jr. 1994. Peat collapse, ponding, and wetland loss in a rapidly submerging coastal marsh. Journal of Coastal Research:1021–1030. Donnelly, J., and M. Bertness. 2001. Rapid shoreward encroachment of salt marsh cordgrass in response to accelerated sea-level rise. Proceedings of the National Academy of Sciences 98:14,218. Gallagher, J.L., G.F. Somers, D.M. Grant, and D.M. Seliskar. 1988. Persistent differences in two forms of Spartina alterniflora: A common garden experiment. Ecology 69:1005–1008. Gedan, K., B.R. Silliman, and M.D. Bertness. 2009. Centuries of human-driven change in salt marsh ecosystems. Annual Review of Marine Science 1:117–141 Gedan, K.B., A.H. Altieri, and M.D. Bertness. 2011. Uncertain future of New England salt marshes. Marine Ecology Progress Series 434:229–237. Greenberg, R., J.E. Maldonado, S.A.M. Droege, and M.V. McDonald. 2006. Tidal Marshes: A Global Perspective on the Evolution and Conservation of their Terrestrial Vertebrates. BioScience 56:675–685. Grimes, H.H., M.T. Huish, J. Kerby, and D. Moran. 1989. Species profiles: Life histories and environmental requirements of coastal fishes and invertebrates (Mid-Atlantic). Atlantic Marsh Fiddler. US Fish and Wildlife Service, Biological Report 82(11.114). Lafayette, LA. 18 pp. Available online at http://www.nwrc.usgs.gov/wdb/pub/species_ profiles/82_11-114.pdf. 2013 Y.C. Luk and R.N. Zajac 273 Hartig, E.K., V. Gornitz, A. Kolker, F. Mushacke, and D. Fallon. 2002. Anthropogenic and climate-change impacts on salt marshes of Jamaica Bay, New York City. Wetlands 22:71–89. Holdredge, C., M. Bertness, and A. Altieri. 2009. Role of crab herbivory in die-off of New England salt marshes. Conservation Biology 23:672–679. Hughes, Z.J., D.M. FitzGerald, C.A. Wilson, S.C. Pennings, K. Wieski, and A. Mahadevan. 2009. Rapid headward erosion of marsh creeks in response to relative sea-level rise. Geophysical Research Letters 36:L03602. Jaramillo, E., and K. Lunecke. 1988. The role of sediments in the distribution of Uca pugilator(Bosc) and Uca pugnax (Smith) (Crustacea, Brachyura) in a salt marsh of Cape Cod. Meeresforschung 32:46–52. Jordao, J. and R.F. Oliveira. 2003. Comparison of non-invasive methods for quantifying population density of the fiddler crab Uca tangeri. Journal of the Marine Biological Association of the UK 83:981–982 Katz, L. 1980. Effects of burrowing by the fiddler crab Uca pugnax (Smith). Estuarine and Coastal Marine Science 11:233–237. Kirwan, M.L., A.B. Murray, J.P. Donnelly, and D.R. Corbett. 2011. Rapid wetland expansion during European settlement and its implication for marsh survival under modern sediment delivery rates. Geology 39:507–510. Krebs, C.T., and I. Valiela. 1978. Effect of experimentally applied chlorinated hydrocarbons on the biomass of the fiddler crab Uca pugnax (Smith). Estuarine and Coastal Marine Science 6:375–386. Macia, A., I. Quincardete, and J. Paula. 2001. A comparison of alternative methods for estimating population density of the fiddler crab Uca annulipes at Saco Mangrove, Inhaca Island (Mozambique). Hydrobiologia 449:213–219. Macintosh, D. 1988. The ecology and physiology of decapods of mangrove swamps. Symposium of the Zoological Society of London 59:315–341. May, M.K. 2002. Pattern and process of headward erosion in salt marsh tidal creeks. M.Sc. Thesis. East Carolina University, Greenville, NC. 132 pp. McCaffrey, R.J. 1977. Record of the accumulation of sediment and trace metals in a Connecticut, USA, salt marsh. Ph.D. Dissertation. Yale University, New Haven, CT. 156 pp. Miller, K.G., and D. Maurer. 1973. Distribution of the fiddler crabs, Uca pugnax and Uca minax, in relation to salinity in Delaware rivers. Chesapeake Science 14:219–221. Miller, W., S. Neubauer, and I. Anderson. 2001. Effects of sea-level-induced disturbances on high salt marsh metabolism. Estuaries and Coasts 24:357–367. Miller, W.R., and F.E. Egler. 1950. Vegetation of the Wequetequock-Pawcatuck tidal marshes, Connecticut. Ecological Monographs 20:143–172. Montague, C. 1980. A natural history of temperate western Atlantic fiddler crabs (genus Uca) with reference to their impact on the salt marsh (Uca pugnax, Uca pugilator, Uca minax). Contributions in Marine Science 23:25–55. Nabout, J.C., L.M. Bini, and J.A.F. Diniz-Filho. 2010. Global literature of fiddler crabs, genus Uca (Decapoda, Ocypodidae): Trends and future directions. Iheringia Série Zoologia 100:463–468. Niering, W., and R. Warren. 1980. Vegetation patterns and processes in New England salt marshes. BioScience 30:301–307. Nomann, B.E., and S.C. Pennings. 1998. Fiddler crab-vegetation interactions in hypersaline habitats. Journal of Experimental Marine Biology and Ecology 225:53–68. Orson, R.A. 1999. A paleoecological assessment of Phragmites australis in New England tidal marshes: Changes in plant community structure during the last few millennia. Biological Invasions 1:149–158. 274 Northeastern Naturalist Vol. 20, No. 2 Orson, R., W. Panageotou, and S.P. Leatherman. 1985. Response of tidal salt marshes of the US Atlantic and Gulf coasts to rising sea levels. Journal of Coastal Research 1:29–37. Pennings, S.C., and M.D. Bertness. 2001. Salt marsh communities. Pp. 289–316, In M.D. Bertness, S.D. Gaines, and M.E. Hay (Eds.). Marine Community Ecology. Sinauer Associates, Sunderland, MA. 550 pp. Ringold, P. 1979. Burrowing, root-mat density, and the distribution of fiddler crabs in the eastern United States. Journal of Experimental Marine Biology and Ecology 36:11–21. Rozas, L.P., and D.J. Reed. 1993. Nekton use of marsh-surface habitats in Louisiana (USA) deltaic salt marshes undergoing submergence. Marine Ecology-Progress Series 96:147–147. Sanford, E., S. Holzman, R. Haney, D. Rand, and M. Bertness. 2006. Larval tolerance, gene flow, and the northern geographic range limit of fiddler crabs. Ecology 87:2882–2894. Smith, S.M. 2009. Multi-decadal changes in salt marshes of Cape Cod, MA: Photographic analyses of vegetation loss, species shifts, and geomorphic change. Northeastern Naturalist 16:183–208. Smith, S.M. and M.C. Tyrrell. 2012. Effects of mud fiddler crabs (Uca pugnax) on the recruitment of halophyte seedlings in salt marsh dieback areas of Cape Cod (Massachusetts, USA). Ecological Research 27:233–237. Smith, S.M., K.C. Medeiros, and M.C. Tyrrell. 2012. Hydrology, herbivory, and the decline of Spartina patens (Aiton) Muhl. in outer Cape Cod Salt Marshes (Massachusetts, USA). Journal of Coastal Research 28:602–612. Teal, J.M. 1958. Distribution of fiddler crabs in Georgia salt marshes. Ecology 39:186–193. Teal, J.M. 1962. Energy flow in the salt marsh ecosystem of Georgia. Ecology:614–624. Tiner, R., I. Huber, T. Nuerminger, and E. Marshall. 2006. Salt marsh trends in selected estuaries of southwestern Connecticut. US Fish and Wildlife Service, Hadley, MA. 20 pp. Vernberg, F., and W. Vernberg. 1975. Adaptations to Extreme Environments. Physiological Ecology of Estuarine Organisms. University of South Carolina Press, Columbia, SC. 397 pp. Warren, R.S., and W.A. Niering. 1993. Vegetation change on a northeast tidal marsh: Interaction of sea-level rise and marsh accretion. Ecology 74:96–103. Weis, J.S., and P. Weis. 2003. Is the invasion of the Common Reed, Phragmites australis, into tidal marshes of the eastern US an ecological disaster? Marine Pollution Bulletin 46:816–820. Wilson, C., Z. Hughes, and D. FitzGerald. 2012. The effects of crab bioturbation on Mid-Atlantic saltmarsh tidal creek extension: Geotechnical and geochemical changes. Estuarine, Coastal and Shelf Science 106:33–44. Windham, L. and R.G. Lathrop. 1999. Effects of Phragmites australis (Common Reed) invasion on aboveground biomass and soil properties in brackish tidal marsh of the Mullica River, New Jersey. Estuaries and Coasts 22:927–935. Wolf, P.L., S.F. Shanholtzer, and R.J. Reimold. 1975. Population estimates for Uca pugnax (Smith, 1870) on the Duplin estuary marsh, Georgia, USA (Decapoda Brachyura, Ocypodidae). Crustaceana 29:79–91.