Marks National Wildlife Refuge Research Articles

Disentangling direct and indirect effects of extreme events on coastal wetland communities.

One of the primary ways in which climate change will impact coastal freshwater wetlands is through changes in the frequency, intensity, timing and distribution of extreme weather events. Disentangling the direct and indirect mechanisms of population- and community-level responses to extreme events is vital to predicting how species composition of coastal wetlands will change under future conditions. We extended static structural equation modelling approaches to incorporate system dynamics in a multi-year multispecies occupancy model to quantify the effects of extreme weather events on a coastal freshwater wetland system. We used data from an 8-year study (2009-2016) on St. Marks National Wildlife Refuge in Florida, USA, to quantify species-specific and community-level changes in amphibian and fish occupancy associated with two flooding events in 2012 and 2013. We examine how physical changes to the landscape, including potential changes in salinity and increased wetland connectivity, may have contributed to or exacerbated the effects of these extreme weather events on the biota of isolated coastal wetlands. We provide evidence that the primary effects of flooding on the amphibian community were through indirect mechanisms via changes in the composition of the sympatric fish community that may have had lethal (i.e. through direct predation) or non-lethal (i.e. through direct or indirect competitive interactions) effects. In addition, we have shown that amphibian species differed in their sensitivity to direct flooding effects and indirect changes in the fish community and wetland-specific conductance, which led to variable responses across the community. These effects led to the overall decline in amphibian species richness from 2009 to 2016, suggesting that wetland-breeding amphibian communities on St. Marks National Wildlife Refuge may not be resilient to predicted changes in coastal disturbance regimes because of climate change. Understanding both direct and indirect effects, as well as species interactions, is important for predicting the effects of a changing climate on individual species, communities and ecosystems.

Seeking shelter from the storm: Conservation and management of imperiled species in a changing climate.

Climate change is anticipated to exacerbate the extinction risk of species whose persistence is already compromised by habitat loss, invasive species, disease, or other stressors. In coastal areas of the southeastern United States (USA), many imperiled vertebrates are vulnerable to hurricanes, which climate models predict to become more severe in the 21st century. Despite this escalating threat, explicit adaptation strategies that address hurricane threats, in particular, and climate change more generally, are largely underrepresented in recovery planning and implementation. We provide a basis for stronger emphasis on strategic planning for imperiled species facing the increasing threat of catastrophic hurricanes. Our reasoning comes from observations of short‐term environmental and biological impacts of Hurricane Michael, which impacted the Gulf Coast of the southeastern USA in October 2018. During this storm, St. Marks National Wildlife Refuge, located along the northern Gulf of Mexico's coast in the panhandle region of Florida, received storm surge that was 3.0–3.6 m (NAVD88) above sea level. Storm surge pushed sea water into some ephemeral freshwater ponds used for breeding by the federally threatened frosted flatwoods salamander (Ambystoma cingulatum). After the storm, specific conductance across all ponds measured varied from 80 to 23,100 µS/cm, compared to 75 to 445 µS/cm in spring 2018. For 17 overwashed wetlands that were measured in both spring and fall 2018, posthurricane conductance observations were, on average, more than 90 times higher than in the previous spring, setting the stage for varying population responses across this coastal landscape. Importantly, we found live individual flatwoods salamanders at both overwashed and non‐overwashed sites, although we cannot yet assess the demographic consequences of this storm. We outline actions that could be incorporated into climate adaptation strategies and recovery planning for imperiled species, like A. cingulatum, that are associated with freshwater coastal wetlands in hurricane‐prone regions.

Life history plasticity does not confer resilience to environmental change in the mole salamander (Ambystoma talpoideum).

Plasticity in life history strategies can be advantageous for species that occupy spatially or temporally variable environments. We examined how phenotypic plasticity influences responses of the mole salamander, Ambystoma talpoideum, to disturbance events at the St. Marks National Wildlife Refuge (SMNWR), FL, USA from 2009 to 2014. We observed periods of extensive drought early in the study, in contrast to high rainfall and expansive flooding events in later years. Flooding facilitated colonization of predatory fishes to isolated wetlands across the refuge. We employed multistate occupancy models to determine how this natural experiment influenced the occurrence of aquatic larvae and paedomorphic adults and what implications this may have for the population. We found that, in terms of occurrence, responses to environmental variation differed between larvae and paedomorphs, but plasticity (i.e. the ability to metamorphose rather than remain in aquatic environment) was not sufficient to buffer populations from declining as a result of environmental perturbations. Drought and fish presence negatively influenced occurrence dynamics of larval and paedomorphic mole salamanders and, consequently, contributed to observed short-term declines of this species. Overall occurrence of larval salamanders decreased from 0.611 in 2009 to 0.075 in 2014 and paedomorph occurrence decreased from 0.311 in 2009 to 0.121 in 2014. Although variation in selection pressures has likely maintained this polyphenism previously, our results suggest that continued changes in environmental variability and the persistence of fish in isolated wetlands could lead to a loss of paedomorphosis in the SMNWR population and, ultimately, impact regional persistence in the future.

Correlation of water quality indicators for coastal marshes

Encompassing 68,000 acres spread out between Wakulla, Jefferson, and Taylor counties along the Gulf Coast of northwest Florida, St. Marks National Wildlife Refuge was established in 1931 to provide wintering habitat for migratory birds. This study evaluated the water quality of coastal marshes at St. Marks National Wildlife Refuge in terms of coliform and chlorophyll a, along with dissolved oxygen, temperature, pH, conductivity, and total suspended solid. Coliform and chlorophyll a were correlated to water physical and chemical parameters based on statistical analysis. Coliform showed direct correlations with turbidity. However, chlorophyll a displayed no correlation with turbidity. Instead, chlorophyll a had a correlation with pH. With the increase of pH, chlorophyll a exponentially decreased.

Assessment of observed and perceived changes in ecosystems over time, with special reference to the Sylt-Rømø Bight, German Wadden Sea

Examples of state changes in three aquatic ecosystems (the Neuse River estuary, NC, USA, the Kromme River estuary (St. Francis Bay, South Africa), the St. Marks National Wildlife Refuge, FL, USA) over time have been quantitatively assessed using ecological network analysis (ENA). A suite of ENA indices were compared among the multiple states of these ecosystems to illustrate the usefulness of ENA for describing differences among system states. Quantitative network models were constructed for the Sylt-Rømø Bight ecosystem and for the mussel bed subsystem of the Bight, depicting standing stocks of the living and non-living components and flow of carbon [a surrogate for energy] between them. These models consist of 59 compartments, and were assessed by means of ENA protocols. The impact of invasive species, such as the Pacific oyster, Crassostrea gigas, on existing mussel beds [of Mytilus edulis], and on predators [mainly birds] dependent on mussels for energy were assessed using AUTOMOD, a routine that predict the impact of species on each other within the ecosystem model. Predictive modelling clearly shows variability in bird biomass due to change in abundance of their important prey species at lower trophic levels. The simulations illustrated that the numbers of eider duck and oystercatcher in the Bight could be reduced to 50% of their current abundance within about 10 years at a 20%–40% reduction in prey biomass. Ecosystem properties calculated using ENA for the initial and alternate state of the Bight and the mussel beds after a simulated reduction of 40% in prey biomass show a decline in virtually all metrics (Total System Throughput, Development Capacity, Ascendency, Redundancy), system trophic efficiency, cycling, and system ratios (e.g. Average Mutual Information, Flow Diversity, Food Web Connectance).

Phenology and Egg Production of the Cactus Moth (Lepidoptera: Pyralidae): Comparison of Field Census Data and Life Stage Development in the Field

Natural phenology and development of the cactus moth, Cactoblastis cactorum (Berg) (Lepidoptera: Pyralidae), was studied under field conditions in St. Marks National Wildlife Refuge, St. Marks, FL, from July 2006 to September 2007. Pads of cactus (Opuntia stricta Haw. [Cactaceae]) were visually surveyed weekly for presence of moth immature stages. Adult male C. cactorum populations were surveyed using a pheromone lure and wing style sticky traps. The field census data identified 3 generations per year which generally occurred in August to September, October to April, and May to July. Numbers of eggsticks peaked in midAugust, midOctober and midApril. High numbers of early larval immatures (1st to 3rd instars) were recorded in October 2006, May 2007 and September 2007. High numbers of late-stage larvae (4th to 5th instar) were recorded in September to October 2006, December to January 2006, June to July 2007. Peak numbers of Cactoblastis male adults occurred approximately in midOctober 2006, April to May 2007, and July to August 2007. Pupae were cryptic and difficult to sample. Development in field cages was studied by introducing cohorts of moth eggs on potted cactus plants (Opuntia ficus-indica (L.) Mill.) into the field at approximately the same times they were found to occur naturally. During the course of each generation, the potted plants were returned to the laboratory for determination of moth lifestage. Body weights and lengths also were recorded. In the field cages, 3 generations were observed and these coincided with the populations that were observed in the open field. However, the generations were more clearly defined because eggs were introduced as cohorts and most insects were recovered. Measurements of larval head capsules and body lengths show that development in the winter generation is slower than that of the summer and fall generations, although peak measurements are approximately identical in all 3 generations. Female pupae weighed 225.25 (± 4.8 SE) mg which was significantly heavier than males weighing 138.01 (± 5.5) mg. Pupal weights differed according to generation, but there was no effect on adult weight. Female adults weighed 130.3 (± 11.3) mg, compared with males, which weighed 60.81 (± 2.4) mg. Adult sex ratio was approximately 1:1. Male adults appeared to have emerged earlier than females in the fall generation.

Changes in a Northwestern Florida Gulf Coast Herpetofaunal Community Over a 28-y Period

ABSTRACT Population declines of amphibians and reptiles throughout the world have led to the initiation of projects to monitor their status and trends. Historical collections give an indication of which species occurred in an area at one time, although the ambiguity surrounding locations and environmental conditions associated with collection decreases the value of this information source. Resampling using the same general protocols can give valuable insights to changes in community structure. However, this is only feasible when sampling methodology and exact site locations are known. From 2002–2005 we resampled 12 sites in St. Marks National Wildlife Refuge in Florida's panhandle, an area in which intensive herpetological surveys were conducted in 1977–1979. We documented a general decrease in species richness among the diversely managed sites, changes in dominant species and diversity and an increasing trend toward homogeneity of the herpetofaunal community among habitats. Changes were attributed to fou...

Modeling soil salinity distribution along topographic gradients in tidal salt marshes in Atlantic and Gulf coastal regions

Soil salinity plays a very important role in determining the distribution of vegetation, plant productivity, and biogeochemical processes in coastal marsh ecosystems. Salinity gradients and salinity–vegetation associations in salt marshes have often been observed but rarely explained. A quantitative and systematic study on the soil salinity distribution in salt marshes is not only important to the understanding of coastal marsh ecosystems but also to the development of a potentially useful ecological and environmental indicator. In this research, we developed a salt marsh soil salinity model based on an existing salt and water balance model with modifications to several key features to examine the impacts of tidal forcing, climate, soil, vegetation, and topography on soil salinity distributions of the Atlantic and Gulf coastal marshes. This model was calibrated and validated using field observations from the St. Marks National Wildlife Refuge (NWR) of northwestern Florida, USA. The results showed that the model had good agreement ( r 2 = 0.84, n = 15, P < 0.001) with field observations. We found that the mean higher high water (MHHW) level determines the location of the salinity maximum in a coastal salt marsh. Simulations indicate that tidal irregularity primarily controls the width of the salinity maximum band. Evapotranspiration, temperature, hydraulic conductivity, and incoming tidal salinity significantly affect the salinity maximum band, which may lead to the formation of salt barrens/flats when reaching a threshold level.

Determining the trophic guilds of fishes and macroinvertebrates in a seagrass food web

We established trophic guilds of macroinvertebrate and fish taxa using correspondence analysis and a hierarchical clustering strategy for a seagrass food web in winter in the northeastern Gulf of Mexico. To create the diet matrix, we characterized the trophic linkages of macroinvertebrate and fish taxa present inHalodule wrightii seagrass habitat areas within the St. Marks National Wildlife Refuge (Florida) using binary data, combining dietary links obtained from relevant literature for macroinvertebrates with stomach analysis of common fishes collected during January and February of 1994. Heirarchical average-linkage cluster analysis of the 73 taxa of fishes and macroinvertebrates in the diet matrix yielded 14 clusters with diet similarity ≥ 0.60. We then used correspondence analysis with three factors to jointly plot the coordinates of the consumers (identified by cluster membership) and of the 33 food sources. Correspondence analysis served as a visualization tool for assigning each taxon to one of eight trophic guilds: herbivores, detritivores, suspension feeders, omnivores, molluscivores, meiobenthos consumers, macrobenthos consumers and piscivores. These trophic groups, corss-classified with major taxonomic groups, were further used to develop consumer compartments in a network analysis model of carbon flow in this seagrass ecosystem. The method presented here should greatly improve the development of future network models of food webs by providing an objective procedure for aggregating trophic groups.

Monitoring the recovery of Juncus roemerianus marsh burns with the normalized difference vegetation index and Landsat Thematic Mapper data

Nine atmospherically corrected Landsat Thematic Mapper images were usedto generate mean normalized difference vegetation indices (NDVI) at 11burn sites throughout a coastal Juncus roemerianus marsh in St. MarksNational Wildlife Refuge, Florida. Time-since-burn, the time lapse from thedate of burn to the date of image collection, was related to variation inmean NDVI over time. Regression analysis showed that NDVI increasedfor about 300 to 400 days immediately after the burn, overshooting thetypical mean NDVI of a nonburned marsh. For about another 500 to 600days NDVI decreased until reaching a nearly constant NDVI of about0.40. During the phase of increasing NDVI the ability to predicttime-since-burn was within about ±60 days. Within the decreasingphase this dropped to about ±88 days.Examination of each burn site revealed some nonburn related influences onNDVI (e.g., seasonality). Normalization of burn NDVI by site-specificnonburn control NDVI eliminated most influences. However, differentialresponses at the site-specific level remained related to either storm impactsor secondary burning. At these sites, collateral data helped clarify theabnormal changes in NDVI. Accounting for these abnormalities,site-specific burn recovery trends could be broadly standardized into fourgeneral phases: Phase 1 – preburn, Phase 2 – initial recovery (increasingNDVI), Phase 3 – late recovery (decreasing NDVI), and Phase 4 – finalcoalescence (unchanging NDVI). Phase 2 tended to last about 300 to 500days, Phase 3 an additional 500 to 600 days, and finally reaching Phase 4,900 to 1,000 days after burn.

Organizing and understanding a winter’s seagrass foodweb network through effective trophic levels

Trophic structure of ecosystems is a unifying concept in ecology; however, the quantification of trophic level of individual components has not received the attention one might expect. Ecosystem network analysis provides a format to make several assessments of trophic structure of communities, including the effective trophic level (i.e. non-integer) of these components. We applied network analysis to a Halodule wrightii community in Goose Creek Bay, St. Marks National Wildlife Refuge, Florida, USA, during January and February 1994 where we sampled a wide variety of taxa. Unlike most applications of network analysis, the field sampling design was specific for network construction. From these data and literature values, we constructed and analyzed one of the most complex, highly articulated and site specific foodweb networks to be done. Care was taken to structure the network to reflect best the field data and ecology of populations within the requirements of analysis software. This involved establishing internally consistent rules of data manipulation and compartment aggregation. Special attention was paid to the microbial components of the food web. Consumer compartments comprised effective trophic levels from 2.0 (herbivore/detritivore) to 4.32 (where a level 4.0 represents ‘secondary carnivory’), and these values were used to organize data interpretation. The effective trophic levels of consumers tended to aggregate near integer values, but the spread from integer values increased with increasing level. Detritus and benthic microalgae acted as important sources of food in the extended diets of many consumers. ‘Bottom-up’ control appeared important through mixed trophic impact analysis, and the extent of positive impacts decreased with increasing trophic level. ‘Top-down’ control was limited to a few consumers with relatively large production or biomass relative to their trophic position. Overall, ordering results from various network analysis algorithms by effective trophic level proved useful in highlighting the potential influence of different taxa to trophodynamics. Although the calculation of effective trophic level has been available for some time, its application to the evaluation of other analyses has previously not received due consideration.

Assessment of Spatial and Temporal Variability in Ecosystem Attributes of the St Marks National Wildlife Refuge, Apalachee Bay, Florida

Six carbon-flow networks of a seagrass ecosystem were constructed from comprehensive databases collected at three different sites during January and February 1994, respectively. The flow models, which consisted of 51 compartments each, were analysed by means of network analysis and the resultant outputs compared on spatial and temporal scales. Results on the biomass, species diversity, input, rates of exchange and export of material showed distinct differences between the sites in a month, as well as between months at the same site. System-level attributes, such as total system throughput, ascendancy, the structure and magnitude of recycling, flow diversity, etc., derived from network analysis, also showed noticeable differences from site to site and between months. There was an increase in ecosystem activity and in the magnitude of its global properties between months which are ascribed to a rise in temperature between months, increased rates in respiration and growth of resident species populations, as well as to the immigration of large numbers of fish and birds.

The influence of methane oxidation on the stable isotopic composition of methane emitted from Florida swamp forests

This study reports the first measurements of the δ 13C of CH 4 emitted from seasonally flooded swamp forests in the southeastern United States. The seasonally averaged δ 13C of CH 4 emitted from a north Florida swamp forest located in the St. Marks National Wildlife Refuge was −52.7 ± 6.11%. (error is ± one standard deviation throughout, n = 28), a value 13C-enriched, relative to typical wetland emissions. In an Everglades cypress dome, the average δ 13C of emitted CH 4 was −52.5 ± 6.7%.( n = 3). Consistent with attenuation of CH 4 emission by CH 4 oxidation in these environments, CH 4 emitted via diffusion from the St. Marks swamp forest was enriched in 13C by 6.4 ± 5.8%. ( n = 28) and D by 57 ± 36%. ( n = 6) relative to sedimentary CH 4. Methane emitted from the cypress dome had also been altered by oxidation, as it was enriched in 13C by 12.1 ± 4.3%. relative to sedimentary CH 4. Emission experiments, performed in situ with inhibitors of aerobic CH 4 oxidizing bacteria, were used to calculate the fractionation factors (α) for stable carbon and hydrogen isotopes of CH 4 undergoing transport and oxidation. Values ranged from 1.003 to 1.021 and 1.050 to 1.129, respectively. The best estimates for carbon and hydrogen α values were 1.020 and 1.068, respectively. The δ values of produced (sedimentary) CH 4 were relatively constant in the St. Marks subtropical swamp forest. Additionally, because the transport of CH 4 to the atmosphere was dominated by molecular diffusion, variations in the magnitude of CH 4 oxidation appeared to be the primary factor controlling the δ values of emitted CH 4. This contrasts with systems dominated by bubble ebullition, where variations in CH 4 production mechanisms have been hypothesized to be the primary factor controlling the δ values of emitted CH 4.

Comments on a Site in the St. Marks National Wildlife Refuge, Wakulla County, Florida

In a recent issue of American Antiquity John M. Goggin has described certain materials from a site in the St. Marks National Wildlife Refuge, Wakulla County, Florida. The present note is a continuation of Goggin's remarks, based on certain materials not available to him at the time of his writing, and on a more recent surface collection of sherds from the site.Goggin mentions a copper plate ornamented by incision, having human and animal figures depicted on the surface, which he had not seen. When this object was originally found in 1938 by Bill Kary of Tallahassee, it was examined by J. Clarence Simpson of the Florida Geological Survey. Simpson made a cast of the object, as well as a detailed drawing. The drawing reproduced in this note (Fig. 41, b) was made from Simpson's sketch and a copy of the cast in the possession of Dr. Mark F. Boyd of Tallahassee. The writer is grateful to both of these gentlemen for the use of their material.

Manifestations of a South Florida Cult in Northwestern Florida

A number of unusual ornaments of gold and copper were found several years ago in a site on Apalachee Bay, Wakulla County, Florida (Fig. 74). This site lies in the St. Marks National Wildlife Refuge a few hundred feet east of its lookout tower, several hundred yards south of the headquarters, and a mile northwest of the St. Marks lighthouse. The site is a burial ground, showing no signs of habitation other than some sherds and flakes of chert scattered on the surface. The burials are found close under the sandy surface on a small natural ridge which lies along the shore where high land merges into extensive marshes. Nearby are two other sites, both shallow refuse areas. One lies about 150 yards to the west, near the gate of the present lookout tower. The other is under the buildings of the present refuge headquarters.