Using stable isotope ratios to evaluate dietary breadth in Oryzomys palustris sanibeli

Material Information

Using stable isotope ratios to evaluate dietary breadth in Oryzomys palustris sanibeli
Seelig, Alexandra Lynn
Publication Date:


The Sanibel Island rice rat (SIRR; Oryzomys palustris sanibeli) is a subspecies marsh rice rat endemic to Sanibel Island, currently listed as a Species of Special Concern in the state of Florida. Rice rats inhabit Sanibel's interior freshwater marshes, buttonwood shrublands [Conocarpus erectus], and exterior mangrove swamps. The population of SIRR is estimated to be well under 100 individuals and little is known about their ecological niche. We examined stable isotope ratios (δ13C and δ15N) of SIRR guard hair and samples of potential diet items to determine the influence of seasonality (summer or winter) and habitat on trophic level and niche breadth. We used Levene's test for homogeneity of variance to determine if SIRR dietary niche breadth varied seasonally. Analysis showed that dietary niche breadth varied between summer and winter samples for δ15N (p=0.018) but not δ13C (p=0.068), showing that SIRR diet was more diverse during summer than winter periods. We found that SIRR isotopic ratios, using separate models for δ15N and δ13C, overlapped with the animal prey base for δ15N (f-value = 6.598, p=0.014) but not δ13C (f-value = 3.682, p=0.062), and showed no overlap with a plant prey base (δ13C f-value = 15.394, p=0.0003; δ15N f-value = 9.241, p=0.004). Seasonality did not explain variation in isotopic δ15N (t-value = -0.328, p = 0.745) or δ13C (t-value = 1.573, p = 0.127). These findings show that SIRR significantly prefers animal prey items over plant diet items when available. As such, we can conclude a food web ranking above that of primary consumers. ( en )
General Note:
Awarded Bachelor of Science, summa cum laude, on May 8, 2018. Major: Biology. Emphasis/Concentration: Natural Science
General Note:
College or School: College of Agricultural and Life Sciences
General Note:
Advisor: Robert McCleery. Advisor Department or School: Wildlife Ecology & Conservation

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Alexandra Lynn Seelig. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.

UFDC Membership

UF Undergraduate Honors Theses


This item is only available as the following downloads:

Full Text


1 Correspondence concerning this manuscript should be addressed to Alexandra Seelig 1 Department of Wildlife Ecology & Conservation, University of Florida 110 Newins Ziegler 2 Hall, PO Box 110430, G ainesville, FL 32611 0430. 3 Contact phone: 269 405 3025 4 Contact email: 5 6 Running heading: Stable isotopes reveal seasonal trophic niche variation 7 8 Using stable isot ope ratios to evaluate dietary breadth in Oryzomys palustris sanibeli 9 10 Alexandra L. Seelig*, Wesley W. Boone IV and Robert A. McClee ry 11 12 D epartment of Wildlife Ecology and Conservation, University of Florida, Gainesville, FL 32611, 13 USA ( ALS, WWB, RAM ) 14 15 Correspondent: 16 17 The Sanibel Island rice rat (SIRR; Oryzomys palustris sanibeli ) is a subspecies marsh rice rat 18 endemic to Sanibel Island, currently listed as a Species of Special Concern in the state of Florida. 19 buttonwood shrublands [ Conocarpus 20 erectus ] and exterior mangrove swamps. The population of SIRR is estim ated to be well under 21 100 individuals and little is known about their ecological niche. We examined stable isotope 22 ratios ( 13 C and 15 N) of SIRR guard hair and samples of potential diet items to determine the 23


2 influence of seasonality (summer or winter) an d habitat on trophic level and niche breadth. We 24 determine if SIRR dietary niche breadth 25 varied seasonally. Analysis showed that dietary niche breadth varied between summer and winter 26 samples 15 N (p=0. 018) but not 13 C ( p=0.068 ) showing that SIRR diet was more diverse 27 during summer than winter periods. We found that SIRR isotopic ratios using separate models 28 15 13 C, overlap ped with the animal prey base 15 N (f value = 6.598, p=0.0 14 ) but 29 not 13 C (f value = 3.682, p=0.06 2 ) and showed no overlap with a plant prey base ( 13 C f value 30 = 15.394, p=0.0003 ; 15 N f value = 9.241, p= 0.004 ). Seasonality did not explain variation in 31 15 N ( t value = 0.328, 13 C ( t value = 1.573, p = 0.127). These findings 32 show that SIRR significantly prefers animal prey items over plant diet items when available. As 33 such, we can conclude a fo od web ranking above that of primary consumers. 34 35 Key words: Florida, diet selection, Oryzo mys palustris sanibeli, Sanibel Island, stable isotopes 36 37


3 I NTRODUCTION 38 T he Sanibel Island Rice Rat (SIRR, Oryzomys palustris sanibeli ) is a subspecies of marsh rice 39 rat [ Oryzomys palustris ] endemic to Sanibel Island, Florida and listed by the State of Flor ida as a 40 Species of Special Concern (Hipes et al. 2000) The marsh rice rat ( Oryzomys palustris ) is a 41 semi aquatic, medium sized rat distributed throughout the Eastern United States, from 42 Pennsylvania to Florida, and west to Texas ( Svihla 1931; Hamilton 19 46; Esher et al. 1978; 43 Wolfe 1982). However, SIRR have evolved independently from other rice rat subspecies due to 44 geographic isolation, increasing the threat of extinction to this population (Indorf and Gaines 45 2013). Though it morphologically resemble s other Oryzomys subspecies, O. p. sanibeli is 46 characterized by an amber brown pelage, an average body length of 263mm, tail length of 47 125mm, and 33mm hind foot (Hamilton 1955 ; Indorf and Gaines 2013 ). 48 While the life history of SIRR is not yet well document ed, much is known about the ecology 49 of other O. palustris populations O. palustris has been suggested to be largely carnivorous 50 (Sharp 1967), while other sources have recorded a diet high in seeds and succulent plant parts 51 (Svihla 1931 ; Lowery 1974; Hamil ton and Whitaker 1979 ). Negus et al. (1961) recorded a diet 52 ranging from 85% vegetation and seeds to 75% arthropods, with significant seasonal variation. 53 Insects, fiddler crabs, and snails are other common food items, as well as fish, clams, and the 54 carcas ses of small birds or mammals ( Lowery 1974 ; Hamilton and Whitaker 1979). Little is 55 known in comparison about plant diet items, but consumption of Spartina alterniflora Spartina 56 glabra, Salicornia europea, Tripsacum sp., and Elymus sp. is presumed (Hamilton 1946). 57 These differences may be dependent on what foods are available Presumably, flooded 58 periods offer increased carnivorous food sources because aquatic macroinvertebrates would be in 59 higher abundance, while dry periods may result in shifts towards plant food sources. This is of 60


4 particular interest on Sanibel Island whic h experiences distinct rainy, flooded (summer) and dry, 61 non flooded (winter) seasons. 62 Due to these conflicting reports and the lack of specific knowledge on the SIRR we 63 investigated the dietary breadth and diet item selection We measured stable isotope r 13 C 64 15 N), which are commonly used for evaluating trophic relationships and diet base, to 65 compare samples of SIRR guard hair samples in summer and winter with potential diet items. 66 Specifically, we tested four predictions: 1) SIRR would feed on a wider variety of diet items in 67 summer months compared to winter months due to greater availability of macroinvertebrates 2) 68 SIRR feeds primarily on animal prey, in accordance with the findings of Sharp (1967) for 69 Oryzomys palustris 3) SIRR found in fr eshwater grasslands would have isotopic signature s more 70 closely related to grass samples than those found in mangrove swamps and 4) SIRR would 71 consume a greater amount of animal prey in mangrove wetlands than elsewhere due to the 72 limited diversity of plan t species present 73 74 M ATERIALS AND M ETHODS 75 Study area Sanibel Island is a small (<4,900 ha) barrier island in southwest Florida (City 76 of Sanibel 2013). Conservation lands account for ~ 50% of the island, with remaining lands used 77 for residential development (City of Sanibel 2013). Remnant sand ridges ~1 2 m above mean sea 78 level trap ped 79 variable water depths (Boggess 1974). A series of low ridges and swales occur r e d within the 80 creating a diversity of freshwater plant communities (City of Sanibel 2013). 81 Although historic accounts document nearly 82 ( Hammond 1970), woody species have become dominant on many ridges. The freshwater 83


5 hydrology on Sanibel Island has been greatly altered by the construction of ponds and drainage 84 85 while largely intact, have also exper ienced degradation due to mosquito control de watering 86 projects. Historically, rice rats have been known to occur within these freshwater wetlands 87 (Humphrey et al. 1986), while recent research has uncovered their use of mangrove forests on 88 Sanibel Island ( Boone, Unpublished data ). 89 Sanibel Island lies within a tropical climate with summers being significantly wetter than 90 winters, with fall to late spring only containing 15% of annual rainfall (Kushlan 1987; Duever et 91 al. 1994) Summers are characterized by frequent thunderstorms and occasional tropical cyclones 92 (Duever et al. 1994) Occasionally, winter frontal systems result in higher than average winter 93 rainfall (Duever et al. 1994) However, human alteration of the islands may significantly alter 94 hydroper occurring wetlands (City of 95 Sanibel 2013). It is currently unknown how these hydroperiod variations impact SIRR 96 distribution (Abuzeineh et al. 2007; van der Merwe 2016). 97 For our research, we categ orized SIRR habitats on the island into three distinct communities: 98 buttonwood shrublands, mangrove swamps, and inland grass marshes. The freshwater wetlands 99 make up much of the inland habitat for SIRR and consist of both ridges with short hydroperiods 100 and swales which remain dry for much of the year (City of Sanibel 2013). Swales are 101 characterized by water dependent plant species and ridges contain more mesic species (Boone, 102 Unpublished data). Common freshwater wetland species in the region include cordgra sses 103 [ Spartina sp. ] sawgrasses [ Cladium sp.] and leather ferns [ Acrostichum aureum ] (Boone, 104 Unpublished data). The buttonwood shrublands consist of wetland ridges that have transitioned 105 from grassy vegetation to woody species, especially buttonwood [ Conocarpus erectus ], in the 106


6 last eighty years (Hammond 1970). These buttonwood ridges are seasonally flooded, leading to a 107 transitional hydrologic profile (FNAI 2015) Coastal mangrove swamps are abundant on the 108 island, but SIRR has not been observed in th ese areas in the past (F lorida Fish and Wildlife 109 Commission 2013). These swamps are characterized by red [ Rhizophora mangle ], black 110 [ Avicennia germinans ], and white mangroves [ Avicennia marina ] (Boone, Unpublished data). 111 We can refer to these communities a s freshwater, transitional, and saltwater, respectively. We 112 classified collection sites into vegetative communities using vegetative data from the Florida 113 Natural Areas Inventory (F lorida Natural Areas Inventory 2015) in ArcGIS 114 Data collection To underst and the relationship between SIRR diet selection, seasonality, 115 and plant community we took hair samples from SIRR found in interior freshwater marshes, 116 buttonwood shrublands, and exterior mangrove swamps in summer and winter We conducted 117 s mall mammal trap ping on 54 grids located on conservation lands on Sanibel Island, Florida. We 118 placed 54 grids total, 18 grids in each of 3 communities; fresh water marshes, buttonwood 119 shrub lands, and mangrove swamps. Each grid consisted of 25 Sherman box traps (8 cm 9 cm 120 23 cm; H.B. Sherman Traps, Tallahassee, Florida, USA) in a 5x5 ar r angement with 15m between 121 traps. We baited traps with birdseed and ran them for four consecutive nights at a time. We 122 collected approximately 1 cm 2 of dorsal guard hair to allow adequate material for stable isotope 123 analysis ( Darimont & Reimchen 2002; van der Merwe and Hellgren 2016). To do this we 124 scruffed each animal upon trapping and cut dorsal guard hair with scissors before releasing the 125 animal. We placed the hair in sterile plastic v ials and froze until processing. We collected hair 126 samples during 2 trapping seasons (June August 2017 and December 2017 February 2018). All 127 samples were collected from conservation lands 128 on Sanibel Island, Florida as part of an ongoing study. 129


7 We chose hair for our stable isotope analysis to reflect a seasonal dietary turnover (Schwertl 130 et al. 2003 ; Sponheimer et al. 2003 b ; Cerling et al. 2006). Though SIRR hair molting frequencies 131 are cur rently unrecorded, we assumed a similar molting cycle to golden mice ( Ochrotomys 132 nuttali ) and deer mice ( Peromyscus spp.). We only caught adult rats, allowing us to operate on 133 the assumption of biannual adult molts in the spring and fall (Linzey and Linzey 1967 ; Miller et 134 al. 2008). By these assumptions and because we sampled in summer and winter hair samples can 135 be used to estimate diet over the prior 2 4 months since the most recent spring/winter molt 136 We collected potential food items from each area i ncluding vegetation, epifauna, fish, and 137 macroinvertebrates Due to increased availability in both macroinvertebrate and plant diet items, 138 we chose to collect diet items in the summer season. We clip ped vegetation and use d mesh dip 139 nets to sample for macro invertebrates, fish, and insects in each of 3 sampling communities 140 because these items are known components of SIRR diets (van der Merwe and Hellgren 2016 ). 141 Invertebrates included insects ( mostly dragonflies, damselflies, aquatic beetles and all within the 142 orders Hemiptera, Coleoptera and Odonata ) gastropods ( from families Littorinidae, Ellobiidae, 143 Thiaridae and Mytilidae ) and crustaceans ( mangrove crabs, crayfish, and fiddler crabs from 144 families Sesarmidae, Cambaridae, and Ocypodidae, respectively ). Vertebrates included fish 145 (Gambusia and Poecilia) and anoles ( Anolis sagrei ). Vegetation samples include d mangrove 146 shoots [ Avicennia germinans; A. marina; Rhizophora mangle ] vines or propagules, buttonwood 147 [ Conocarpus erectus ] sedges [family Cyperaceae] cordgrasses [ Spartina sp.] purslane [family 148 Portulacaceae] sawgrass [ Cladium sp.] and flowering plants including Tillandsia recurvata, 149 Eustoma exaltatum Bacopa monnieri and Parthenocissus quinquefolia 150 We developed sampling processing protocol in acc ordance with van der Merwe and Hellgren 151 (2016). We froze s amples within 6 hours of collection except for hair samples which we stored at 152


8 ambient temperature We stored m acroinvertebrate samples in ethanol for identification before 153 being processed. We soake d h air in an acetone bath for 2 hours and we rinsed all samples for 10 154 minutes in deionized water before drying them for 72 hours at 60C. Followin g oven drying, we 155 also dried and ground some samples in a cryogenic freezer mill. We homogenized and ground 156 t he remainder of the samples with a mortar and pestle. We stored s ubsamples (~2.0mg for 157 vegetation; ~0.35 0.45 mg for hair and invertebrates) in glass vials while waiting for isotopic 158 analysis. tional Animal Care and 159 Use Committee (IACUC; Protocol 201709811 ). We then analyzed stable isotope ratios of hair 160 161 typical trophic level of SIRR in each community. 162 Stable isotope analysis To understand relationships between ri ce rats, diet items and 163 season, we applied 15 13 C values to our samples. Values of 15 N from animal tissue 164 with 3 4% 165 enrichment with values increasing as trophic level inc reases ( DeNiro and Epstein 1981 ; 166 Minagawa and Wada 1984 ; Peterson and Fry 1987 ). The 15 N values also serve to elucidate 167 complex food web interactions that otherwise are difficult to quantify (Kling et al. 1992). 168 13 C levels remain stable between trophic positions but vary among primary 169 producers, and by extension, organisms feeding on primary producers ( Rounick and Winterbourn 170 1986 ; Peterson and Fry 1987 ; France and Peters 1997 ). In terrestrial systems, 13 C can also be 171 used to discriminat e between vegetation characterized by different photosynthetic pathways (C 3 172 vs. C 4 vs. CAM) and therefore, is useful for further characterizing the nature of plants in a diet 173 Becau se C3 174 plants discriminate against 13 C more than C4 plants, a higher ratio of 13 C to 1 2 C (denoted as 13 C ) 175


9 in hair samples would indicate a diet higher in C4 plants like grasses and sedges (Sponheimer et 176 al. 2003 a ). However, we cannot use these values alone to determine trophic level without 177 appropriate isotopic baselines for comparison purposes (Post 2002). 178 Samples were analyzed at the University of Florida Light Stable Isotope Mass Spect r ometry 179 Lab using a Thermo Electron DeltaV Advantage isotope ratio mass spectrometer coupled with a 180 ConFlo II interface linked to a Carlo Erba NA 1500 CNHS Elemental Analyzer. Samples were 181 loaded into tin capsules and placed in a 50 position automated Zero Blank sample carousel on a 182 Carlo Erba NA1500 CNS elemental analyz er. After combustion in a quartz column at 1020 o C in 183 an oxygen rich atmosphere, the sample gas was transported in a He carrier stream and passed 184 through a hot reduction column (650 o C) consisting of elemental copper to remove oxygen. The 185 effluent stream t hen passed through a chemical (magnesium perchlorate) trap to remove water 186 followed by a 0.7 meter GC column at 120 o C to separate N 2 from CO 2 The sample gas next 187 passed into a ConFlo II preparation system and into the inlet of a Thermo Electron Delta V 188 A dvantage isotope ratio mass spectrometer running in continuous flow mode where the sample 189 gas was measured relative to laboratory reference N 2 and CO 2 gases. All carbon isotopic 190 results are expressed in standard delta notation relative to VPDB. All nitro gen isotopic results 191 are expressed in standard delta notation relative to AIR (Curtis, personal communication). 192 Data analysis. We investigated a metric of trophic niche breadth by quantifying the 193 variance of stable isotopic values among individual s (Bearh op et al. 2004). We compared stable 194 isotopic niche breadth 15 13 C between SIRR summer and winter samples and between 195 samples collected in each of the 3 vegetative communities, 196 of variance (Flaherty and Ben Davi d 2010). 197 15 13 C isotopic ratios overlapped sampled plant and animal prey items. W e 198


10 assumed that a smaller breadth of variance was the result of specialization on a limited number 199 of food sources, and larger variance resulted from a more generalist diet. For all analyses, we 200 used hair to ; Weiser and 201 Powell 2011 ; Osterback et al. 2015). We also investigated whether vegetative community 202 15 13 C between samples 15 N and 203 13 C) using a Tukey Kramer Pairwise Multiple Comparison Test to account for unequal sample 204 sizes between groups (Sokal and Rohlf 1995). We also used gener alized linear models 205 15 13 C ( van der Merwe and Hellgren 2016 ), in program R (R version 3.4.2, 206 www.r project. org, accessed 4 March 2017) to determine if seasonality explained variation in 207 15 13 C sample values. 208 209 R ESULTS 210 We collected 15 SIRR hair samples in summer and 15 in winter Of these, we collected 211 23 5 and 2 hair samples from fresh water marshes, buttonwood shrub lands, and mangrove 212 swamps respectively. We collected 28 plant and 19 animal specimens for isot opic analysis. We 213 averaged isotopic values when multiple samples of a single species were available so that only 1 214 value was used in the statistical analyses. This yielded 25 plant and 15 animal specimens 215 included in statistical analyses. 216 Rice rat f ur isotopic ratios were found to overlap with that of an animal prey base for 217 13 C ( f value = 3.682, p= 0.062 ) but differed significantly for 15 N ( f value = 6.59 8, p= 0.0 14 ) 218 Fur isotopic ratios differed significantly from a plant food base 13 C f value = 15.394, 219 p=0.0003 ; 15 N f value = 9.241, p= 0.004 ; Fig. 1 ) Hair sample 15 N varied significantly 220 between summer (mean = 5.37 ) and winter ( mean = 5.51 p=0.018), while 13 C did not ( summer 221


11 mean = 23.06, winter mean = 23.86, p= 0.068 ; Fig. 2 ). Ve getative communit y modeled using a 222 Tukey Kramer Pairwise Multiple Comparison Test was significantly correlated with 13 C 223 isotopic ratios for pairwise comparisons because the 95% confidence interval (CI) was positive 224 and did not pass through zero for freshwater marsh buttonwood shrubland (mean = 1.604 CI = 225 0. 3 2.90 9 ) and buttonwood shrubland mangrove swamp (mean = 4.375 CI = 2.16 4 6.587 ) but 226 not for mangrove swamp freshwater marsh whose CI contained zero (mean = 2.771 CI = 227 4.720 0.822 ; Fig. 3 ). T h e 15 N isotopic ratio pairwise 228 comparisons between any vegetative communities because all comparisons crossed zero 229 (freshwater marsh buttonwood shrubland mean = 0.566, 95 CI = 1.96 3 0.83 1 ; buttonwood 230 shrubland mangrove sw amp mean = 1.526, CI = 3.89 5 0.84 2 ; mangrove swamp freshwater 231 marsh mean = 0.96, CI = 1.12 7 3.04 7 ) 15 N ( t 232 value = 0.328, 13 C ( t value = 1.573, p = 0.127). 233 234 D ISCUSSION 235 As predicted, SIRR isotopic values indicated a diet higher in animal prey than vegetation, 236 regardless of season or community. These findings were in agreement with those in Sharp (1967) 237 that rice rats are primarily carnivorous when animal prey items are available However animal 238 prey items found in mangrove areas (crabs, snails, etc) had high 13 C values (> 18), in line with 239 those of grassland vegetation, that were not detected in the hair of rodents from mangrove areas 240 indicating a potential avoidance of animal food source s in mangrove areas Therefore, it appears 241 that most reliance on animal prey occur s not in mangrove 242 wetlands as we had predicted This emphasis on freshwater animal prey is worrisome for a 243


12 barrier island frequently impacte d by hurricanes and that is vulnerable to rising sea levels (Titus 244 1990) 245 We also found that SIRR diet breadth variance does indeed vary seasonally as we 246 predicted V ariation of winter isotopic ratios (both 13 C and 1 5 N) in hair samples was less than 247 that of summer samples This wide summer scatter confirms that SIRR diet is much broader and 248 more diverse in summer months than in winter months. T his may indicate seasonal variation in 249 their reliance on macroinvertebrates (high 1 5 N values) which are pres umably more abundant 250 during wet summer months and a greater abundance and diversity of plant food sources in 251 summer (low 1 5 N values) (Neckels et al. 1990) If macroinvertebrate populations decrease in 252 drier months, SIRR would be forced to consume more plants especially grasses and sedges 253 leading to a more consistent diet profile across all sites (Neckels et al. 1990) 254 13 C composition varied significantly depending on the 255 community in which they were caught (Fig. 3 ) Because C4 plants like grasses and sedges favor 256 the heavier 13 C compared to C3 plants, C4 plants will exhibit a higher 13 C ratio in the hair 257 samples (Sponheimer et al. 2003 a ). The lower values of 13 C came from hair samples co llected 258 in the mangrove swamps, so we can conclude that that population shows the least signature of a 259 grass based diet. SIRR caught in grassy freshwater marshes show a comparatively stronger grass 260 diet signature as the 13 C are higher than those in mangrove swamps However, SIRR trapped in 261 the buttonwood shrubland (transitional) areas 13 C values and therefore had a 262 larger signature of a grass diet than those caught in the grasslands. This may be because 263 gra sslands are flooded more frequently, and therefore likely support a greater abundance and 264 diversity of macroinvertebrate and fish prey T herefore SIRR occupying grasslands eat more 265 animals and less grass than those in the buttonwood shrubland. Furthermore this may indicate 266


13 that SIRR selectively choose C4 plants ( grasses over other vegetation types ) in the shrublands 267 that may be more prevalent. 268 Additionally, the lack of overlap between SIRR hair isotopic ratios in different vegetative 269 communities implies that movement between these areas may be limited. This finding is 270 noteworthy because it indicates that multiple populations of SIRR may occur separately on 271 Sanibel Island and maintain foraging area fidelity increasing the likel i hood of local extirpation. 272 It is not known if movements are restricted as a result of anthropogenic alteration of the island 273 (i.e. isolation by roads, development, canals, etc). 274 Our findings suggest that SIRR are capable of exploiting seasonally available food 275 sources, with dietary breadth maximized during summer months when much of the island is 276 inundated with fresh water. Animal prey is likely an important dietary component during all 277 seasons in accord ance with Sharp (1967) Further research regarding the effects of water level 278 modification on aquatic invertebrate diversity and abundance are needed to ensure adequate prey 279 levels are maintained. Finally, the apparent lack of movement of rice rats between various habitat 280 types, at least over time spans of 3 4 months, requires further investigation to determine if 281 dispersal barriers have been induced by anthropogenic change. 282 283 A CKNOWLEDGMENTS 284 We than k ndergraduate Honors 285 Program U.S. Fish and W ildlife S ervice, and Florida Fish and Wildlife Conservation 286 Commission for 287 Sanibel Captiva Conservation Foundation, the City of Sanibe l and The Sanctuary Golf Club for 288 their collaborative efforts and property access. We thank the Ding Darling Wildlife Society for 289


14 providing housing throughout our research. Individually we would like to thank Audrey 290 Albrecht, Jennifer Bernatis, Joelle Ca rbonell, Nate Caswell, Mark Clark, Jeremy Conrad, Jason 291 Curtis, Sarah Lathrop Chris Lechowicz, Holly Milbrandt, Kyle 292 Sweet, Paul Tritaik, and Toni Westland, 293 294 LITERATURE CITED 295 296 Abuzeineh, A. A., R. D. Owen, N. E. McIntyre, C. W. Dick, R. E. Strauss, and T. Holsomback. 297 2007. Response of marsh rice rat (Oryzomys palustris ) to inundation of habitat. 298 Southwestern Naturalist 52:75 78. 299 300 Bearhop, S., Adams, C. E., Waldron, S., Fuller, R. A. and Macleod, H. 2004 Determining 301 trophic niche width: a novel approach using stable isotope analysis. Journal of Animal 302 Ecology 73: 1007 1012. 303 304 Brown, L. N. 1997 Mammals of Florida. Windward Publishing Inc., Miami. 305 306 Boggess, D. H. 1974. The shallow fresh water system of Sani bel Island, Lee County, Florida, 307 with emphasis on the sources and effects of saline water. Florida Bureau of Geology 308 Report of Investigation 69. Tallahassee, Florida. 309 310


15 Cerling, T. E., G. Wittemyer, H. B. Rasmussen, F. Vollrath, C. E. Cerling, T. J. Robinso n, and I. 311 Douglas Hamilton. 2006. Stable isotopes in elephant hair document migration patterns 312 and diet changes. Proceedings of the National Academy of Sciences USA 103:371 373. 313 314 City of Sanibel. 2013. Sanibel plan : the comprehensive land use plan of the City of Sanibel, 315 Florida. Sanibel, Florida. 316 317 Darimont, C. T., and T. E. Reimchen. 2002. Intra hair stable isotope analysis implies seasonal 318 shift to salmon in gray wolf diet. Canadian Journal of Zoology 80:1638 1642. 319 320 DeNiro, M. J., and S. Epstein. 1978. I nfluence of diet on the distribution of carbon isotopes in 321 animals. Geochimica et Cosmochimica Acta 42:495 506. 322 323 DeNiro, M. J., and S. Epstein. 1981. Influence of diet on the distribution of nitrogen isotopes in 324 animals. Geochimica et Cosmochimica Acta 45 :341 351. 325 326 Duever, M. J., Meeder, J. F., Meeder, L. C., & McCollom, J. M. 1994. The climate of south 327 Florida and its role in shaping the Everglades ecosystem. Everglades: The ecosystem and 328 its restoration, 225 248. 329 330 Ellstrand, N.C., and D.R. Elam. 1993. Po pulation genetic consequences of small population size: 331 Implications for plant conservation. Annual Review of Ecology and Systematics 24:217 332 242. 333


16 334 Esher, R.J. J. L. Wolfe, and J. N. Layne. 1978. Swimming Behavior of Rice Rats ( Oryzomys 335 palustris ) and Cott on Rats ( Sigmodon hispidus ), Journal of Mammalogy 59(3):551 8. 336 337 Flaherty, E. A., and M. Ben David. 2010. Overlap and partitioning of the ecological and isotopic 338 niches. Oikos 119:1409 1416. 339 340 Florida Fish and Wildlife Conservation Commission. 2013 A species action plan for the Sanibel 341 Island rice rat. Tallahassee, Florida. 342 343 Florida Natural Areas Inventory. 2015. Cooperative land cover map. Tallahassee, Florida. 344 345 France, R. L., and R. H. Peters. 1997. Ecosystem differences in the trophic enrichment of 13 C in 346 aquatic food webs. Canadian Journal of Fisheries and Aquatic Sciences 54 :1255 1258. 347 348 G o ldman E. A. 1918. The rice rats of North America. North American Fauna 43:1 100. 349 350 Hamilton, W. 1946. Habits of the Swamp Rice Rat, Oryzomys Palustris Palustri s (Harlan ). The 351 American Midland Naturalist 36 (3) : 730 736. 352 353 Hamilton, W. J. 1955. Two new rice rats (genus Oryzomys ) from Florida. Proceedings of the 354 Biological Society Washington 68:83 86. 355 356


17 Hamilton Jr. W. J. and J. O W hitaker J r 1979. Mammals of the eastern United States. Second 357 Ed. Cornell University Press. 358 359 Hammond, E. A. 1970. Sanibel Island and its vicinity, 1833, a Document, The Florida Historical 360 Quarterly 48:392 411. 361 362 Hipes, D., Jackson, D. R., NeSmith, K., Printiss, D., & Brandt, A. 2000. Field guide to the rare 363 animals of Florida. Florida Natural Areas Inventory Tallahassee, Florida 364 365 Humphrey, S. R., R. W. Repenning, and H. W. Setzer. 1986. Status survey of five Florida 366 mammals. University of Florida Cooperative Fish and Wildlife Resear ch Unit, Technical 367 Report No. 22, Gainesville, Florida. 368 369 Indorf, J.L. & Gaines, M.S. 2013 Genetic divergence of insular marsh rice rats in subtropical 370 Florida. Journal of Mammalogy 94(4) : 897 910. 371 372 opes and planktonic trophic structure 373 in arctic lakes. Ecology 73:561 566. 374 375 Kushlan, J. A. 1987. External threats and internal management: the hydrologic regulation of the 376 Everglades, Florida, USA. Environmental Management 11:109 119. 377 378


18 Linzey, D. W., and A. V. Linzey. 1967. Maturational and seasonal molts in the golden mouse, 379 Ochrotomys nuttalli. Journal of Mammalogy 48:236 241. 380 381 Lowery, G. H. 1974. Mammals of Louisiana and its adjacent waters. Louisiana State Univ. Press, 382 Baton Rouge 383 384 Miller, J. F., J. S. Millar, and F. J. Longstaffe. 2008. Carbon and nitrogen isotope tissue diet 385 discrimination and turnover rates in deer mice, Peromyscus maniculatus Canadian 386 Journal of Zoology 86: 685 691 387 388 Minagawa, M., and E. Wada 1984. Stepwise en richment of 15 N along food chains: further 389 evidence and the relation between 15 N and animal age. Geochimica et Cosmochimica 390 Acta 48 :1135 1140. 391 392 Myers, D. 2010. Oxygen and hydrogen stable isotope ratios in Mississippi River floodplain 393 invertebrates: implica tions for dispersal and food web analysis. Thesis. Southern Illinois 394 University, Carbondale, Illinois, USA. 395 396 N eckles H. A., M urkin H. R., & C ooper J. A. 1990. Influences of seasonal flooding on 397 macroinvertebrate abundance in wetland habitats. Freshwater B iology, 23(2) : 311 322. 398 399


19 Negus, N. C., Gould, E., & Chipman, R. K. 1961. Ecology of the rice rat, Oryzomys palustris 400 (Harlan), on Breton Island, Gulf of Mexico, with a critique of the social stress 401 theory. Tulane Studies in Zoology 8 (4) : 95 123 402 403 Nifong, J .C., Layman, C.A., & Silliman, B.R. 2015 Size, sex and individual level behaviour 404 drive intrapopulation variation in cross ecosystem foraging of a top predator. Journal of 405 Animal Ecology 84 : 35 48. 406 407 O'Leary, M. H., S. Madhavan, and P. Paneth. 1992. Physical and chemical basis of carbon 408 isotope fractionation in plant s. Plant, Cell and Environment 15 :1099 1104. 409 410 Osterback, A. K., D. M. Frechette, S. A. Hayes, S. A. Schaffer, and J. W. Moore. 2015. Long 411 term shifts in anthropogenic subsidi es to gulls and implications for imperiled fish. 412 Biological Conservation 191:606 613. 413 414 Peterson, B. J., and B. Fry. 1987. Stable isotopes in ecosystem studies. Annual Review of 415 Ecology and Systematics 18 :293 320. 416 417 Post, D. M. 2002. Using stable isotopes to estimate trophic position: models, methods, and 418 assumptions. Ecology 83:703 718. 419 420 Rounick, J. S., and M. J. Winterbourn. 1986. Stable carbon isotopes and carbon flow in 421 ecosystems. BioScience 36 :171 177. 422


20 423 Schwertl, M., K. Auerswald, and H. Schnyder. 2003. Reconstruction of the isotopic history of 424 animal diets by hair segmental analysis. Rapid Communications in Mass Spectrometry 425 17:1312 1318. 426 427 Sharp Jr H. F. 1967. Food ecology of the rice rat, Oryzomys palustris (Harlan), in a Georgia salt 428 marsh. Journal of Mammalogy 48 (4) : 557 563. 429 430 Sokal, R., and F. Rohlf. 1995. Biometry. 3 rd edition. Freedman and Company, New York. 431 432 433 Lower Keys. Journal of Mammalogy 59:787 792. 434 435 Spo nheimer, M., Lee Thorp, J.A., DeRuiter, D.J., Smith, J.M., Van Der Merwe, N.J., Reed, K., 436 Grant, C.C., Ayliffe, L.K., Robinson, T.F., Heidelberger, C. and Marcus, W. 2003. Diets 437 of southern African Bovidae: stable isotope evidence. Journal of Mammalogy 84( 2) : 471 438 479. 439 440 Sponheimer, M., T. Robinson, L. Ayliffe, B. Roeder, J. Hammer, B. Passey, A. West, T. Cerling, 441 D. Dearing, and J. Ehleringer. 2003. Nitrogen isotopes in mammalian herbivores: hair 442 nal of Osteoarchaeology 443 13:80 87. 444 445


21 Steenweg, R. J., R. A. Ronconi, and M. L. Leonard. 2011. Seasonal and age dependent dietary 446 partitioning between great black backed and herring gulls. The Condor 113:795 805. 447 448 Svihla, A. 1931. Life History of the Texas Rice Rat (Oryzomys palustris texensis). Journal of 449 Mammalogy 12 (3) : 238 242. 450 451 Titus, J. G. 1990. Greenhouse effect, sea level rise, and barrier islands: Case study of Long 452 Beach Island, New Jersey. 453 454 Van der Merwe, J., and E. C. Hellgren. 2016. Spatial variation in trophic ecology of small 455 mammals in wetlands: support for hydrological drivers. Ecosphere 7:e01567. 456 457 Weiser, E. L., and A. N. Powell. 2011. Evaluating gull diets: a comparison of conventional 458 methods and stable isotope analysis. Journal of Field Ornithology 82:297 310 459 460 Wolfe, J. L. 1982. Oryzomys palustris. Mammalian Species 176:1 5. 461 462


22 463 Fig. 1. Isotopic ratios of a ll samples (animal, hair, and plant) from both seasons. 464 465 Fig. 2. Hair sample isotopic ratios sorted by the season in which they were collected. 466


23 467 Fig. 3. Isotopic ratios of hair samples sorted by the vegetative community in which they were 468 captured 469