The University of Maine DigitalCommons@UMaine Senator George J. Mitchell Center for Sustainability Publications Solutions 11-2014 Modeling nutrient transport and transformation by pool-breeding amphibians in forested landscapes using a 21 year dataset Krista A. Capps University of Maine Keith Berven University of Maine Scott D. Tiegs Oakland University Follow this and additional works at:https://digitalcommons.library.umaine.edu/ mitchellcenter_pubs Repository Citation Capps, Krista A.; Berven, Keith; and Tiegs, Scott D., "Modeling nutrient transport and transformation by pool-breeding amphibians in forested landscapes using a 21 year dataset" (2014).Publications. 29. https://digitalcommons.library.umaine.edu/mitchellcenter_pubs/29 This Article is brought to you for free and open access by DigitalCommons@UMaine. It has been accepted for inclusion in Publications by an authorized administrator of DigitalCommons@UMaine. For more information, please [email protected]. Submitted to Freshwater Biology Modeling nutrient transport and transformation by pool-breeding amphibians in forested landscapes using a 21 year dataset Krista A Capps 1, Keith A Berven 2, and Scott D Tiegs 2 1 Sustainability Solutions Initiative, University of Maine, Orono, ME, U.S.A 2 Department of Biological Sciences, Oakland University, Rochester MI, U.S.A. 1 Summary Introduction Methods Results Discussion Acknowledgements References Figures 2 SUMMARY 1. Migrations of animals can transfer energy and nutrients through and among terrestrial and aquatic habitats. Pool-breeding amphibians, such as the wood frog (Lithobates sylvaticus), make annual breeding migrations to ephemeral wetlands in forest habitats in the eastern and midwestern United States and Canada. 2. To model the influence of wood frogs on nutrient transport and transformation through time, we coupled long-term population monitoring data (1985–2005) from a wood frog population with estimates of the elemental composition of wood frog egg masses and emerging juveniles. 3. Over the 21-year study period, 8.8 kg carbon (C), 2.0 kg nitrogen (N) and 0.20 kg phosphorus (P) were transported from the terrestrial to the aquatic habitat and approximately 21 kg C, 5.5 kg N and 1.2 kg P were exported to the surrounding terrestrial habitat by wood frogs. 4. During the study period, the average net flux of C, N and P was from aquatic to terrestrial habitats, but the magnitude and direction of the net flux was element dependent. Thus, the net flux of C, N and P did not always flow in the same direction. 5. Predicting long-term trends in nutrient and energy flux by organisms with biphasic life cycles should rely on long-term population data to account for temporal variability. This is especially true for organisms that are sensitive to long- term shifts in temperature and precipitation patterns, such as amphibians that breed in ephemeral pools. 3 Keywords: animal migrations, nutrient dynamics, resource subsidies, vernal pool, wood frog 4 INTRODUCTION Flows of organic matter across habitat and ecosystem boundaries, or spatial subsidies, have the potential to alter ecosystem processes as they can represent an ecologically important flux of energy and nutrients across landscapes (Polis, Anderson & Holt, 1997). These subsidies can influence abundance and interactions of species in recipient habitats (Nakano & Murakami, 2001; Baxter, Fausch & Saunders, 2005; Rubbo, Cole & Kiesecker, 2006). Consumers with biphasic life cycles can provide spatial subsidies by transporting and transforming energy and nutrients across system boundaries (Regester, Whiles & Lips, 2008). Investigations of cross-habitat flows have often described the movement of energy and nutrients from terrestrial to aquatic habitats as aquatic systems are predicted to receive higher amounts of materials from terrestrial environments than vice versa (Lindeman, 1942; Bartels et al., 2012). However, the quality of these subsidies can differ, since the stoichiometric composition of spatial subsidies is highly variable across ecosystems and types of subsidies (Earl & Semlitsch, 2012). Materials moving from terrestrial to aquatic habitats are commonly dominated by detritus (Bartels et al., 2012). Leaf litter and other forms of detritus can have high carbon (C) to nitrogen (N) or phosphorus (P) ratios (C:N:P) and are low-quality subsidies. Conversely, subsidies moving from aquatic to terrestrial environments are predominantly created by the movement of living organisms (Bartels et al., 2012) and typically have much lower C:N:P ratios, forming higher-quality subsidies. For instance, in lotic 5 systems, migrating salmon can generate large fluxes of elements across habitat boundaries and represent important components of food webs in recipient ecosystems (Tiegs et al., 2009, 2011). Examinations of the flux of energy and nutrients across freshwater–terrestrial boundaries have primarily been conducted in lotic systems (Bartels et al., 2012), yet spatial subsidies can be important fluxes of elements into and out of other kinds of freshwater habitats (Regester, Lips & Whiles, 2006; Earl & Semlitsch, 2012; Reinhardt et al., 2013). Seasonal wetlands occur globally and often support high densities of organisms adapted to breeding in ephemeral habitats (Leibowitz, 2003). Examples of seasonal wetlands include playas, prairie potholes, Carolina bays and vernal pools – the subject of this study. Vernal pools occur in a diversity of landscape settings from isolated upland depressions to larger wetland complexes (Calhoun & Demaynadier, 2008). Because of their ephemeral nature, vernal pools are free of fish and provide critical breeding habitat for organisms sensitive to predation by fishes and other predators that require permanently inundated habitats. Pool-associated species include invertebrates such as fairy shrimp (Eubranchipus spp.) and amphibians such as mole salamanders (Ambystoma spp.) and wood frogs (Lithobates sylvaticus) (Semlitsch, 2000). Animals using both aquatic and terrestrial habitats as part of their life cycle transport elements across habitat boundaries (Nakano & Murakami, 2001; Regester et al., 2008; Kraus & Vonesh, 2012). An important example is organisms with biphasic life 6 cycles that use aquatic habitats for larval development and emerge as juveniles or adults into terrestrial environments (Regester et al., 2008; Hoekman et al., 2011; Kraus & Vonesh, 2012). Pool-breeding amphibians depend upon aquatic habitats for reproduction and require adjacent upland habitat for dispersal, foraging and hibernation (Semlitsch, 2002; Faccio, 2003). Movement of amphibians between aquatic and terrestrial habitats in the forest– wetland matrix may generate a substantial flow of nutrients and energy between breeding pools and adjacent terrestrial ecosystems (Regester & Whiles, 2006; Earl et al., 2011; Earl & Semlitsch, 2012). However, the flow of energy via dispersing amphibians has been quantified only for a few species (e.g. Burton & Likens, 1975; Regester et al., 2006), and generalised patterns in the variability, magnitude and direction of the flux of amphibian-derived elements are poorly understood. Moreover, most investigations have been limited to one- to three-year study periods (e.g. Seale, 1980; Regester et al., 2006; Reinhardt et al., 2013). Often, pool-breeding amphibians are characterised by boom-and-bust population cycles that are driven by variability in the numbers of breeding adults, the extent of successful larval development and the recruitment of emergent juveniles (Berven, 1990, 1995; Whiteman & Wissinger, 2005). However, long-term studies are needed to understand variability in the quality and quantity of spatial subsidies produced by populations of organisms with biphasic life cycles. The purpose of this study was to model the influence of pool-breeding wood frogs on 7 nutrient transport and transformation through time. We coupled long-term population monitoring data (1985–2005) of a single wood frog population with data on the body elemental composition of wood frog egg masses and emergent juveniles to estimate the magnitude of elemental flux between terrestrial and aquatic habitats. We predicted that wood frogs would be a substantial, yet highly variable, flux of C, N and P between terrestrial and aquatic habitats. We hypothesised this effect would vary among elements and depend upon life-stage stoichiometry. METHODS Study site and field methods We collected population data from a single population of wood frogs in University of ʹ ʹ Michigan’s Saginaw Forest Preserve (42°40N, 83°13W) between 1985 and 2005. The forest preserve is ~32 hectares and is surrounded by a matrix of woodland, pasture and commercial and residential developments. Wood frog breeding in this forest 2 occurs predominantly in an ephemeral pond (~2912 m area; 1.5 m maximum depth) that was the subject of our study (Berven & Boltz, 2001; Berven, 2009). All adults entering the pond and all juveniles leaving the pond were collected in a pit-fall array created using aluminum window screening (height: 75 cm) and wooden stakes. The fence completely encircled the pond and was buried to a depth of 20 cm. Pitfall traps were left open and monitored daily during the breeding season (1 March–30 April) and during the period of juvenile emergence (1 June–28 July); however, the 8 traps were closed during the rest of the year to permit organisms to freely enter and exit the pool (Berven, 2009). We measured the wet masses each year of a sub-sample of the breeding adults (~140 individuals of each sex) in the laboratory and emergent juveniles (~100) in the field using an electronic balance. To estimate dry mass from wet mass, a subset of eggs and juveniles collected in 2012 were oven-dried to a constant mass, cooled in a desiccator and weighed to the nearest 1 mg. We used the wet mass/dry mass relationship to convert wet mass to dry mass for each sample year. Total egg mass number was estimated as the total number of females returning to the pool each year, as all females entering the pool were gravid (Berven, 1981). We determined clutch size by pairing females (average of 40 females each year; range 13–79) with a male in a pan of water and allowing them to deposit their eggs. After egg deposition, we counted the number of eggs in each clutch. We then used the relationship between female wet mass and egg number to determine the average clutch size for the average-size female breeding each year. Wet mass of the egg masses was estimated using the relationship between the total egg mass (difference between mass of females before and after egg deposition) and 2 wet mass of fully expanded egg masses (y = 16.14x + 17.63; r = 0.79). We estimated the dry mass of egg masses using the relationship between wet and dry mass (y = 2 0.0174x + 0.393; r = 0.92). Individual adult remineralisation rates were estimated using five male frogs collected in 2013 from the Michigan population using methods modified from Vanni et al. 9