Printer Friendly

Understory Vegetation Structure and Soil Characteristics of Geomys pinetis (Southeastern Pocket Gopher) Habitat in Southwestern Georgia.


Geomys pinetis (southeastern pocket gopher) is a fossorial rodent historically associated with Pinus palustris (longleaf pine) communities characteristic of the Coastal Plain physiographic province in southeastern Alabama, southern Georgia, and northern and central Florida (Golley, 1962; Pembleton and Williams, 1978; Wilkins, 1987). Within longleaf pine communities, southeastern pocket gophers not only contribute to the integrity of the historic faunal community but also enhance habitat quality through creation of refugia for amphibian (Funderburg and Lee, 1968; Blihovde, 2006), reptile (Mount, 1963; Funderburg and Lee, 1968) and several obligate arthropod commensals (Pembleton and Williams, 1978; Skelley and Kovarik, 2001) and through soil turnover (Simkin and Michener, 2005). Longleaf pine communities have been highly impacted by conversion to other land uses and fragmentation, altering the amount and distribution of southeastern pocket gopher habitat. Patches of suitable habitat apparently sustain isolated populations, but the species is absent from a large portion of its historical range (Southern Wildlife Consultants, 2008). It is therefore important to understand characteristics of suitable habitat so areas meeting these criteria can be maintained or restored in future conservation efforts.

For many wildlife species, suitable habitat is better characterized by habitat structure than plant species composition (Garden et al., 2007; Stostad and Menendez, 2014). The historic association of the southeastern pocket gopher with the longleaf pine community is well-established (Golley, 1962; Pembleton and Williams, 1978). Studies have documented southeastern pocket gopher presence in additional habitats such as agricultural fields and utility right-of-ways (Avise and Laerm, 1982; Southern Wildlife Consultants, 2008); however, future pocket gopher conservation likely will be focused on areas with a predominately longleaf pine land cover. Older thinned P. elliottii (slash pine) and P. taeda (loblolly pine) stands with open canopies and sufficient understory growth also may provide suitable habitat because southeastern pocket gophers likely select for understory vegetation structure typical of open pine communities rather than presence of longleaf pine (Southern Wildlife Consultants, 2008). Ford (1980) found no associations between southeastern pocket gopher presence and specific plant species but found southeastern pocket gophers avoided areas that lacked grasses and/or were covered by root systems that impeded burrowing. Similarly, G. bursarius ozarkensis (Ozark pocket gophers) select areas based on availability of abundant forage rather than presence or absence of specific plant species (Connior et al., 2010). Therefore, within areas managed to maintain a predominately pine land cover, understory vegetation structure is a potentially important determinate of suitable southeastern pocket gopher habitat.

The fossorial lifestyle of the southeastern pocket gopher suggests soil characteristics also may be important in defining suitable habitat. During burrow construction, aboveground openings are plugged with soil (Hickman and Brown, 1973a). Their ability to breathe is dependent on diffusion of gasses through the soil, limiting pocket gopher occurrence to highly porous soils with low water-holding capacity (McNab, 1966). The association of southeastern pocket gophers with areas of xeric sandy soils is well supported in the literature (McNab, 1966; Wilkins, 1985; Wilkins, 1987; Simkin and Michener, 2005; Southern Wildlife Consultants, 2008). However, specific soil characteristics, such as soil texture, or the chemical parameters of pH, nitrogen, and carbon content, have not been described at the spatial scale of the individual burrow, except in the context of comparing gopher mound soil to the surrounding matrix (Simkin et al, 2004).

It is difficult, if not impossible, to conserve a species when characteristics determining suitable habitat are not understood. Within areas managed to maintain a predominately pine land cover, southeastern pocket gopher habitat may be further defined by understory vegetation structure, soil characteristics, or a combination of the two (McNab, 1966; Ford, 1980; Wilkins, 1985; Wilkins, 1987; Simkin and Michener, 2005; Southern Wildlife Consultants, 2008). Therefore, we quantified understory vegetation structure in terms of ground cover categories, and soil texture, pH, nitrogen, and carbon at multiple depths, in areas with gopher activity and those lacking activity within a landscape dominated by an open pine over story. We used a modeling approach to evaluate understory vegetation structure, soil characteristics, and a combination of vegetation and soil characteristics, as predictors of gopher presence.


Study Site.--Our study was conducted from August 2012 through December 2013 at the Joseph W. Jones Ecological Research Center at Ichauway in Baker County, Georgia, U.S.A (31[degrees]14'17.2"N, 84[degrees]28'52.3"W). Ichauway is situated within the Dougherty Plain physiographic district, which is characterized by marine and fluvial deposited parent materials that have developed into Entisols and Ultisols over highly fractured Ocala limestone with a flat to rolling karst topography (Beck and Arden, 1983; Hayes et al, 1983; Couch et al. 1996). Ichawaynochaway creek runs north to south through the center of the property; the property is bordered by the Flint River to the southeast. Ichauway covers 117 km" of predominately longleaf pine forest. Other land cover types found at Ichauway are slash and loblolly pine stands, mixed pine hardwoods, riparian hardwood forests, Quercus virginianus (live oak) depressions, isolated depressional wetlands, creek swamps, agricultural fields, shrub-scrub uplands, and areas impacted by human development (Goebel et al, 1997). Aristida beryichiana (wiregrass) is the dominate understory species on approximately 1/3 of the property. The site is managed with prescribed fire to maintain vegetation composition and structure. Stands are burned at least every other year, primarily during March and April (Atkinson et al, 1996).

Sample Site Selection.--We selected 50 locations exhibiting fresh pocket gopher mounding activity (hereafter, active sites). Mounds were detected opportunistically from roads while trapping gophers for a concurrent telemetry study. Gopher presence was confirmed at each active site by successful capture of a gopher, or the gopher filling traps with soil. We also identified 50 locations with no evidence of pocket gopher activity (hereafter, inactive sites). To identify inactive sites, we first used ArcGIS version 10.x (Esri, Redlands, California) to overlay a grid of random points generated by Geospatial Modeling Environment ( onto the study site with greater than 250 m between random points and greater than 250 m between active sites and random points. We selected 250 m as the minimum distance based on the farthest known dispersal distance by a southeastern pocket gopher (Hickman and Brown, 1973b). We then surveyed random points in random order by walking 50 m transects in each cardinal and sub-cardinal direction until 50 locations with no apparent gopher activity were identified.

Understory Vegetation Structure and Soil Sampling.--At each active and inactive site, we randomly selected five 1-[m.sup.2] subplots within an 18 m radius based on the mean home range of gophers in a concurrent telemetry study (Warren, 2014). The radius was centered on the centroid of activity (mounds or telemetry locations) for active sites and on the plotted point for inactive sites. Understory vegetation structure was quantified by visibly estimating percent cover of pine straw, hardwood leaf litter, woody vegetation, forbs and vines, wiregrass, and other grass species in each quadrant of the subplot and averaging the quadrants to represent the subplot. We recognize pocket gophers can occur in agricultural fields, wildlife openings, and roadsides on the study area. However, our objective was to examine factors related to pocket gopher presence in an open pine-dominated landscape. We assumed over story conditions were suitable for pocket gophers across the majority of the site, therefore, only included understory variables in models. We used the Clip tool in ArcMap 9.3.1 to determine the percent of land cover types represented within circular buffers with an 18 m radius around active and inactive sites.

At each active and inactive site, we used a 7 cm diameter bucket auger to collect soil samples at depths of 0-10, 15-25, 40-50, 65-75, and 90-100 cm. We used the qualitative field texture method to estimate soil texture at each depth (Thien, 1979). We used the estimated soil textures at each depth to create a texture profile for each site and selected representatives of each unique texture profile for particle size analysis at a commercial testing laboratory (Waters Agricultural Testing Lab, Camilla, Georgia) using the hydrometer method (Gee and Bauder, 1986). We assigned the quantitative results for each representative profile to the remaining samples from the sites that shared the same profile based on field characterization. Percent nitrogen and carbon of each soil sample was determined using a Flash 2000 carbon nitrogen analyzer (CE Elantech, Lakewood, New Jersey) at the University of Georgia Forest Soils Laboratory (Athens, Georgia). We determined soil pH for each sample using a 5 g of soil to 10 ml deionized water ratio and immersing a pH electrode in the mixture (McLean, 1982).

Data Analysis.--Prior to constructing statistical models, we used Pearson's product moment correlation to examine variables for collinearity ([absolute value of r] > 0.80). Correlation analyses revealed multicollinearity among the five depths for each of the soil variables (percent sand, percent silt, percent clay, pH, nitrogen, and carbon). Therefore, we represented each variable in the models with mean values from the five depths. Percent sand was negatively collinear to percent clay and was excluded from the models because the correlation between percent sand and percent silt was higher than correlation between percent clay and percent silt.

Using the 11 remaining variables (Table 1), we constructed 15 models testing hypotheses, based on previous literature, regarding the relative importance of understory vegetation structure, soil characteristics, and a combination of vegetation and soil characteristics on predicting southeastern pocket gopher presence or absence. These models included a global model of the 11 variables and a null model. Each variable was present in an equal number of models and had equal probability of occurring in the best model to determine the predictive value of each measured variable (Conner and Godbois, 2003). We performed logistic regression to test the models with gopher presence or absence as the binary response variable. We determined the best overall model for predicting southeastern pocket gopher presence by selecting the model that returned the lowest Akaike's Information Criterion (AIC) score adjusted for small sample size (AICc) and used less than two [DELTA][AIC.sub.c] as the cutoff for a set of competing models (Akaike, 1973; Burnham and Anderson, 2002). We used model averaging to calculate parameter estimates, 95% confidence intervals, and variable weights ([summation][w.sub.i]) for each variable. We considered variables with confidence intervals excluding zero to be informative predictors.


Natural longleaf pine and longleaf pine-hardwood mixed forests composed 76.4% of buffers surrounding active sites but only 25.8% of buffers surrounding inactive sites (Table 2). Active sites generally were characterized by higher percent ground cover of wiregrass and other grasses and lower percent ground cover of hardwood leaf litter. Active sites had higher percent sand and lower percent clay than inactive sites, and soil texture at active sites was loamy sand and at inactive sites was sandy loam. Percent silt was similar between active and inactive sites. Active sites had average pH within the range of 5.1-5.5, while inactive sites had an average range from 4.5-5.0. Soil nitrogen and carbon content generally was less for active sites than inactive sites (Table 3). Soil profiles at active sites were sand at 0-10 cm, then loamy sand from 15-100 cm, whereas inactive sites were loamy sand at 0-25 cm, sandy loam at 40-50 cm, and sandy clay loam from 65-100 cm (Table 4).

The best soil characteristics model, which was also the best overall model, combined all five soil variables and had an Akaike weight (w,) of 0.892. The best vegetation structure model had no Akaike weight (Table 5). Model averaged 95% confidence intervals for percent nitrogen, percent clay, pH, and percent silt excluded zero, indicating that these were informative predictors (Table 6). Nitrogen, pH, percent clay, carbon, and percent silt had the highest variable weights ([summation][w.sub.i]).


Our results suggest soil characteristics are more important than understory vegetation structure for predicting southeastern pocket gopher presence on our study area. Percent clay, percent silt, percent nitrogen, and pH were informative predictors of southeastern pocket gopher presence. Percent silt, percent nitrogen, and pH covaried (r [greater than or equal to] 0.24) with percent clay. Clay particles have a much larger surface area to volume ratio than sand particles, therefore a higher cation exchange capacity, causing clayey soils derived from the same parent material to generally have a lower pH than sandy soils (Manahan, 2001). In addition, soils with higher clay content typically have higher plant productivity than sandy soils, creating more organic matter in the soil that contributes to higher nitrogen content (Oik, 2008). As such it is difficult to identify a single soil parameter that influences gopher use, but there is a clear preference for sandier substrates, with southeastern pocket gopher presence in this study being higher in loamy sands compared to sandy loams. This suggest soils that promote burrow integrity are important; if the soil is too loose the burrow will collapse but if too compact it may create conditions unfavorable for digging.

The association between southeastern pocket gophers and sandy soils is well-documented (McNab, 1966; Wilkins, 1985; Wilkins, 1987; Simkin and Michener, 2005; Southern Wildlife Consultants, 2008), but no previous study has quantified preferred texture. Based on our results, a 7.4% decrease in sand and a 7.5% increase in clay, the difference between a loamy sand and a sandy loam, differentiated unsuitable southeastern pocket gopher habitat from suitable habitat (Table 3). Texture profiles of soil collected at 0-10, 15-25, 40-50, 65-75, and 90-100 cm indicated southeastern pocket gopher presence was higher in areas with sand-loamy sand throughout the profile relative to areas with increases in clay content at deeper horizons (Table 4). Increasing clay with depth is characteristic of Ultisols. However, the sandier profiles in this study should not be interpreted as only representing soil series in the Entisol order as series in the Ultisol order regionally (e.g., Troup series) can have clay layers below the 1 m depth sampled in this study.

Southeastern pocket gophers may select loamy sands based on metabolic needs. The southeastern pocket gopher has a low basal metabolism, a convergent trait shared among fossorial rodents allowing them to tolerate the hypoxic environments of their burrows and tunnels. However, southeastern pocket gophers dissipate heat by increasing circulation in their tails which limits reduction in metabolic rate and mandates that substantial gas exchange through the soil is needed to meet metabolic needs (McNab, 1966). Therefore, southeastern pocket gopher presence may be related to the greater porosity of sand relative to silt or clay. Gas exchange through the soil increases with increasing percent sand, allowing pocket gophers greater flexibility with regards to activity level and burrow depth.

Pocket gophers also may select for loamy sands throughout the upper meter to decrease energy expended when burrowing. According to Vleck (1981), foraging in Thomomys bottae (Botta's pocket gopher) is a balance between acquiring energy from food and expending energy while expanding tunnels in search of food. Therefore, pocket gophers may minimize energy expenditure by selecting sandier soils. The relative energetic cost of digging in clayey versus sandy soils was quantified in burrowing Geolycosa spp. (wolf spiders) that expended 5.6 J of energy burrowing in clayey subsoils but only 1.9 J in sandy/sandy loam subsoils (Suter et al, 2011). This difference in energy expenditure likely explains why areas with pocket gopher burrow systems decrease with increasing soil clay content (Romanach et al., 2005; Warren, 2014).

Although our results suggest southeastern pocket gophers select areas with lower nitrogen and carbon content and higher pH, presence of pocket gophers may be responsible for the observed differences between active sites and inactive sites. Burrowing by fossorial mammals can have extensive effects on the physical, chemical, and vegetation properties of their environments, and several fossorial mammals, including pocket gophers (Reichman and Seabloom, 2002), have even been labeled as ecosystem engineers (Zhang et al, 2003; Hagenah et al., 2013). Studies involving T. talpoides (northern pocket gophers) in Colorado (Litaor et al., 1996) and southeastern pocket gophers in Georgia (Simkin et al, 2004) similarly have reported lower levels of nitrogen and carbon related to pocket gopher activity. Both studies examined the soil of gopher mounds as compared to the surrounding matrix. Differences in nitrogen and carbon were not explained by differences in soil texture but rather as a result of pocket gophers bringing deeper, less nutrient dense soils to the surface. An increase in pH also has been observed relative to activity of other fossorial mammals, such as Parotomys brantsii (Brant's whistling rat; Desmet and Cowling, 1999) and Taxidea taxus (American badger; Eldridge and Whitford, 2009). Higher pH associated with badger activity is a result of increased soil aeration, increased microbial activity, and mixing of litter caused by digging (Eldridge and Whitford, 2009). It is possible that a history of southeastern pocket gopher activity at active sites could affect soil nitrogen, carbon, and pH in similar ways. Therefore, it may be inappropriate to use pH and/or soil nitrogen alone to evaluate potential southeastern pocket gopher habitat.

Understory vegetation structure was not important for predicting southeastern pocket gopher presence on our study site. However, our results should not be interpreted to suggest habitat suitability for the southeastern pocket gopher is determined by soil attributes alone. We observed differences in some ground cover categories between active and inactive sites (Table 3), but the predictive abilities of these variables were outweighed by soil variables. Further, our study was conducted on a site that is actively managed (Atkinson, 1996) for longleaf pine, and pocket gopher activity is observed over most of the site. Although buffers around inactive sites were composed of more varied land cover than those around active sites, two of 15 potential land cover types comprised 25% of buffers around inactive sites (Table 2). Therefore, the variation in understory vegetation structure throughout our study site may not have been sufficient to accurately demonstrate the importance of vegetation structure in southeastern pocket gopher habitat selection. Understory vegetation structure variables may be more important in predicting habitat selection at more heterogeneous sites, or at larger scales.

Recognition of the vital importance of longleaf pine communities as floral biodiversity hotspots (Peet and Allard, 1993) and critical habitat for rare fauna such as Gopherus polyphemus (gopher tortoise; U. S. Fish and Wildlife Service, 1990) and Picoides borealis (Red-cockaded Woodpecker; U. S. Fish and Wildlife Service, 2003) has resulted in the promotion of longleaf pine habitat restoration (Van Lear et al., 2005; Aschenbach et al., 2010). As areas of longleaf pine are restored in the Southeast, suitable habitat for southeastern pocket gopher colonization likewise will expand. The highly fragmented nature of newly available habitats, however, will likely prevent natural colonization of southeastern pocket gopher populations into restored habitats, making translocations necessary. Our study suggests suitability of potential southeastern pocket gopher habitat is dependent on soil characteristics through the upper 1 m and these characteristics should be considered when developing translocation protocols. Soil texture is the most useful for evaluating areas for potential southeastern pocket gopher relocation because pH, nitrogen, and carbon levels found in areas of southeastern pocket gopher activity likely covary with soil texture (Manahan, 2001; Oik, 2008), or result from pocket gopher activity (Litaor et al, 1996; Desmet and Cowling, 1999; Simkin et al, 2004; Whitford, 2009). Due to the ability of longleaf pine to proliferate on a variety of soils, from the xeric loamy sands southeastern pocket gophers prefer (McNab, 1966; Wilkins, 1985; Wilkins, 1987; Simkin and Michener, 2005; Southern Wildlife Consultants, 2008) to more clayey soils, such as those of the Georgia Red Hills (Outcalt, 2000), translocation of the southeastern pocket gopher will not be appropriate at all longleaf pine restoration sites within the southeastern pocket gopher's geographic range. Our results will help focus southeastern pocket gopher translocations into restored areas with higher potential for success.

Acknowledgments.--We thank Jean Brock for her assistance with ArcMAP, Man Cobb for helping prepare soil samples and Brandon Crouch for analyzing soil samples. We also thank the Joseph W. Jones Ecological Research Center at Ichauway, the Warnell School of Forestry and Natural Resources, and the University of Georgia Graduate School for funding and support.


Akaike, H. 1973. Information theory as an extension of the maximum likelihood principle, p. 267-281. In: Petrov B.N. and F. Csaki (eds). Sec. Inter. Symp. on Info. Theory. Akademiai Kaido, Budapest, Hungary. 451 pp.

Aschenbach, T. A., B. H. Foster, and D. W. Imm. 2010. The initial phase of a longleaf pine-wiregrass savannah restoration: species establishment and community response. Restor. Ecol., 18:762-771.

Atkinson J. B., J. Brock, and R. Smith. 1996. Operational longleaf pine management at Ichauway. p. 43-45. In: J. S. Rush (ed.). Longleaf Pine: A Regional Perspective of Challenges and Opportunities,

Mobile, Alabama, 17-19 September 1996. Proc. of the First Longleaf Alliance Conf. Rep. No. 1. The Longleaf Alliance, Solon Dixon Forestry Center, Andalusia, Alabama.

Avise, J. C. and J. Laerm. 1982. Gophers of the southeastern United States. Flor. Nat., 55:7-10.

Beck, B. F. and D. D. Arden. 1983. Hydrology and geomorphology of the Dougherty Plain. Southwest Georgia. Southeastern Geological Survey, Tallahassee Florida. 59 pp.

Blihovdf, W. B. 2006. Terrestrial movements and upland habitat use of gopher frogs in central Florida. Southeast. Nat., 5:265-276.

Burnham K. P. and D. E. Anderson. 2002. Model selection and inference: a practical information-theoretic approach, 2nd ed. Springer-Verlag, New York, New York. 488 pp.

Conner, L. M. and I. A. Godbois. 2003. Habitat associated with daytime refugia of fox squirrels in a longleaf pine forest. Am. Mull. Nat.. 150:123-129.

Connior, M. B., A. A. Kershen, R. E. Medlin Jr. D. A. Eerod, D. B. Sasse, and T. S. Risen. 2010. Distribution and habitat attributes of at? endemic subspecies of pocket gopher. Am. Midi. Nal., 164:217-229.

Colch, G. A., E. H. Hopkins, and P. S. Hardy. 1996. Influences of environmental settings on aquatic ecosystems in the Apalachicola-Chattahoochee-Flint River basin. U. S. Geological Survey, Atlanta, Georgia. 58 pp.

Desmet, P. G. and R. M. Cowling. 1999. Patch creation by fossorial rodents: a key process in the revegetation of phytotoxic and soils. J. Arid Environ., 43:35-45.

Eldridge, D.J. and W. G. Whitford. 2009. Badger (Taxidea taxus) disturbances increase soil heterogeneity in a degraded shrub-steppe ecosystem. J. Arid Environ., 73:66-73.

Ford, C. R. 1980. The status and preservation of the colonial pocket gopher. MS Thesis, University of Georgia, Athens.

Funderburg, J. B. and D. S. Lee. 1968. The amphibian and reptile fauna of pocket gopher (Geomys) mounds in central Florida. J. Help., 1:99-100

Garden, J. G., C. A. McAlpine, H. P. Possingham, and D. N. Jones. 2007. Habitat structure is more important than vegetation composition for local-level management of native terrestrial reptile and small mammal species living in urban remnants: a case study from Brisbane, Australia. Austral Erol.. 32:669-685.

Gee, G. W. and J. W. Balder. 1986. Particle-size analysis, p. 383-411. In: A. Klute (ed). Methods of soil analysis, part 1. Physical and mineralogical methods. Agronomy Mono. No. 9, second edition. ASA-SSSA, Madison, Wisconsin. 1173 pp.

Goebel, P. C., B. J. Palik, L. K. Ki reman, and L. West. 1997. Ecosystem types and classification of the J. W. Jones Ecological Research Center at Ichauway. Internal Site Description, J. W. Jones Ecological Research Center, Ichauway, Newton, Georgia. 99 pp.

Golley, F. B. 1962. Mammals of Georgia: a study of their distribution and functional role in the ecosystem. University of Georgia Press, Athens, Georgia. 468 pp.

Hagenah, N., N. C. Bennett, and A. Kitchener. 2013. Mole rats as ecosystem engineers within a biodiversity hotspot, the Cape Frynbos. J. Zool, 289:19-26.

Hayes, L. R., M. L. Maslia, and W. C. Meeks. 1983. Hydrology and model evaluation of the principal artisan aquifer, Dougherty Plain, Southwest Georgia. U. S. Geological Survey, Atlanta, Georgia. 91 pp.

Hickman, G. C. and L. N. Brown. 1973a. Mound-building behavior of the southeastern pocket gopher (Geomys pinetis). J. Mammal., 54:786-790.

Hickman, G. C. and L. N. Brown. 1973b. Pattern and rale of mound production in the southeastern pocket gopher (Geomys pinetis). J. Mammal., 54:971--975.

Litaor, M. L, R. Mancinelu, and J. C. Halfpenny. 1996. The influence of pocket gophers on the status of nutrients in Alpine soils. Geoderma, 70:37-48.

Manahan, S. E. 2001. Fundamentals of environmental chemistry, 2nd ed. CRC Press LLC, Boca Raton, Florida. 743 pp.

McLean, E. O. 1982. Soil pH and lime requirement, p. 199-223. In: A. L. Page, R. H. Miller, and D. R. Keeney, (eds). Methods of soil analysis, part 2. Chemical and microbial properties. Agronomy Mono. No. 9, 2nd ed. ASA-SSSA, Madison, Wisconsin. 1159 pp.

McNab, B. K. 1966. The metabolism of fossorial rodents: a study of convergence. Erol., 47:712-733.

Mount, R. H. 1963. The natural history of the red-tailed skink, Eumeces egregius Baird. Am. Midl. Nat., 70:356-385.

Olk, D. C. 2008. Organic forms of nitrogen, p. 57-100 In: SchepersJ.S. and W.R. Raun (eds) Nitrogen in agricultural systems. Agronomy Mono. No. 49. ASA-CSSA-SSSA, Madison, Wisconsin. 984 pp.

Outcalt, K. W. 2000. The longleaf pine ecosystem of the South. Native Plants Journal 1:1.

Peet, R. K. and D.J. Allard. 1993. Longleaf pine vegetation of the Southern Atlantic and Eastern Gulf Coast Regions: a preliminary classification, p. 45-81 In: Herman S.M. (ed) Proceedings of the 18th Tall Timbers Fire Ecology Conference, the longleaf pine ecosystem: ecology, restoration and management. Tall Timbers Research, Inc., Tallahassee, Florida. 418 pp.

Pembleton, E. F. and S. L. Williams. 1978. Geomys pinetis. Mammal. Sp., 86:1-3.

Reichman, O. J. and E. W. Seabloom. 2002. The role of pocket gophers as subterranean ecosystem engineers. Trends Ecol. and Evol., 17:44-49.

Romanach, S. S., E. W. Seabloom, O. J. Reichman, W. E. Rogers, and G. N. Cameron. 2005. Effects of species, sex, age, and habitat on geometry of pocket gopher foraging tunnels. J. Mammal, 86:750-756.

Simkin, S. M. and W. K. Michener. 2005. Faunal soil disturbance regime of a longleaf pine ecosystem. Southeast. Nat., 4:133-152.

Simkin, S. M., W. K. Michener, and W. R. Wyatt. 2004. Mound microclimate, nutrients and seedling survival. Am. Midl. Nat., 152:12-24.

Skelley, P. E. and P. W. Kovarik. 2001. Insect surveys in the southeast: investigating a relictual entomofauna. El. Entomol., 84:552-555.

Southern Wildlife Consultants. 2008. Survey of the current distribution of the southeastern pocket gopher (Geomys pinetis) in Georgia. Final report to Georgia Department of Natural Resources: 1-16.

Stostad, H. N. and R. Menendf.z. 2014. Woodland structure, rather than tree identity, determines the breeding habitat of Willow Warblers Phylloscopus trochilus in the northwest of England. Bird Study, 61:246-254.

Suter R. B., G. E. STRvrroN, and P. R. Miller. 2011. Mechanics and energetic of excavation by burrowing wolf spiders, Geolycosa spp. J. Insect Sci., 11:22.

Thien, S.J. 1979. A flow diagram for teaching texture feel analysis. J. Agron. Edu. 8:54-55.

U.S. Fish and Wildlife Service. 1990. Gopher tortoise (Gophmis polyphemus) recovery plan. U.S. Fish and Wildlife Service, Jackson, Mississippi. 28 pp.

U.S. Fish and Wildlife Service. 2003. Recovery plan for the red-cockaded woodpecker (Picoides borealis): second revision. U.S. Fish and Wildlife Service, Atlanta, Georgia. 296 pp.

Van Lear, D. H., W. D. Carroll, P. R. Kapeluck, and R. Johnson. 2005. History and restoration of the longleaf pine-grassland ecosystem: implications for species al risk. For. Ecol. Manage., 211:150-165.

Vleck, D. 1981. Burrow structure and foraging costs in the fossorial rodent, Thomomys bottae. Oecologia. 44:391-396.

Warren, A. E. 2014. Ecology of the southeastern pocket gopher (Geomys pinetis) in southwestern Georgia. M.S. Thesis, University of Georgia, Athens.

Wilkins, K. T. 1985. Variation in the southeastern pocket gopher, Geomys pinetis, along the St. Johns River in Florida. Am. Midi. Nat., 114:125-134.

Wilkins, K. T. 1987. Zoographic analysis of variation in recent Geomys Pinetis (Geomvidae) in Florida. Florida Slate Museum. Bull. Biol. Sciences, 30:1-28.

Zhang, Y., Z. Zhang, and J. Liu. 2003. Burrowing rodents as ecosystem engineers; the ecology and management of plateau zokors Myospalax fontanimi in alpine meadow ecosystems on the Tibetan Plateau. Mammal Rev., 33:284-294.

Submitted 23 March 2017

Accepted 30 June 2017


Warnell School of Forestry and Natural Resources, University of Georgia, ?SO E Green Street, Athens 30602



Joseph W. Jones Ecological Research Center at Ichauway, 3988Jones Center Drive, Newton, Georgia 39870

(1) Current Address: Florida Fish and Wildlife Conservation Commission, 3911 Hwy 2321, Panama City, Florida 32409

(2) Corresponding Author:
Table 1.--Descriptions of understory vegetation structure and soil
characteristic variables measured within an 18 m radius of 50 sites
with southeastern pocket gopher activity and 50 sites without
activity within the Jones Ecological Research Center in Baker
County, Georgia, 2012-2013

Variable                Description

Vegetation structure
  Pine Straw            Percent of 5 1-[m.sup.2] plots
                        covered by pine straw.
  Leaf Litter           Percent of 5 1-[m.sup.2] plots
                        covered by hardwood leaf litter.
  Woody Vegetation      Percent of 5 1-[m.sup.2] plots
                        covered by woody vegetation.
  Forbs/Vines           Percent of 5 1-[m.sup.2] plots
                        covered by forbs and vines.
  Wiregrass             Percent of 5 1-[m.sup.2] plots
                        covered bv wiregrass.
  Other Grass           Percent of 5 1-[m.sup.2] plots covered
                        by grasses other than wiregrass.

Soil characteristics

  Silt                  Mean percent sill of soil collected at
                        10, 25, 50, 75, and 100 cm.
  Clay                  Mean percent clav of soil collected at
                        10, 25, 50, 75, and 100 cm.
  Nitrogen              Mean percent nitrogen of soil collected
                        at 10, 25, 50, 75, and 100 cm.
  Carbon                Mean percent carbon of soil collected
                        at 10, 25, 50, 75, and 100 cm.
  pH                    Mean pH of soil collected at 10, 25,
                        50, 75, and 100 cm.

Table 2.--Percent of landcover types in 18 m radius plots
surrounding 50 sites with southeastern pocket gopher activity
(Active Sites) and 50 sites without activity (Inactive Sites)
within the Jones Ecological Research Center in Baker County,
Georgia, 2012-2013

Landcover                                   Active     Inactive
                                           sites (%)   sites (%)

Longleaf Forest                              60.5        24.3
Longleaf/Hardwood Forest                     15.9         1.5
Hardwood/Other Pine Forest                    7.1        17.6
Shrub/Scrub                                   3.7         4.1
Urban/Built-up                                3.0         1.0
Evergreen Coniferous Plantation               2.5        11.3
Other Pine Forest                             2.4         5.3
Other Pine/Hardwood Forest                    2.3         9.3
Agricultural                                  1.8         5.9
Evergreen and Deciduous Hardwood Forest       0.5        10.1
Non-forested Wetland                          0.1         5.2
Wildlife Food Plot                            0.1         0.2
Cypress/Tupelo Forest                         0.0         1.5
Deciduous Hardwood Forest                     0.0         2.0
Open Water                                    0.0         0.7

Table 3.--Means and standard errors (se) for variables measured
within an 18 m radius of 50 sites with southeastern pocket gopher
activity (Active Sites) and 50 sites without activity (Inactive
Sites) within the Jones Ecological Research Center in Baker County,
Georgia, 2012-2013. All units are percent except pH

Variable                Active sites    Inactive sites

                        Mean     SE     Mean     SE

Vegetation structure
  Pine Straw             9.8     1.1    11.4     2.2
  Leaf Litter            6.3     1.2    18.0     3.1
  Woody                  8.2     0.9     8.9     1.3
  Forbs/Vines           21.7     1.5    22.9     2.0
  Wiregrass              8.8     2.4     2.1     0.9
  Other Grass           28.6     2.6    22.8     3.1

Soil characteristics

  Sand                  85.7     0.8    78.3     2.2
  Clay                   7.1     0.6    14.6     2.0
  Silt                   7.2     0.3     7.1     0.3
  pH                    5.40    0.07    5.0]    0.07
  Nitrogen              0.035   0.001   0.052   0.004
Carbon                  0.385   0.019   0.497   0.041

Table 4.--Means and standard errors (sf.) for percent sand, clay,
and silt, and corresponding textures measured at 50 sites with
southeastern pocket gopher activity (Active) and 50 sites without
activity (Inactive) within the Jones Ecological Research Center
in Baker County, Georgia, 2012-2013

Depth (cm)          Active

                    Mean ([+ or -] SE)

        Sand        Clay         Silt       Texture

0-10    88.1(0.7)   4.1(0.4)     7.8(0.4)   Sand
15-25   86.9(0.7)   4.6(0.4)     8.5(0.4)   Loamy Sand
40-50   86.9(1.1)   6.6(0.9)     6.5(0.4)   Loamy Sand
65-75   83.2(1.2)   9.7(1.2)     7.1(0.4)   Loamy Sand
90-100  83.5(1.5)   10.5(1.3)    6.0(0.3)   Loamy Sand

Depth (cm)                       Inactive

                    Mean ([+ or -] SE)

        Sand        Clay         Silt       Texture

0-10    82.2(1.9)   8.4(1.4)     9.4(0.6)   Loamy Sand
15-25   83.7(2.1)   9.0(1.9)     7.3(0.4)   Loamy Sand
40-50   78.3(2.7)   15.2(2.7)    6.5(0.3)   Sandy Loam
65-75   73.0(2.6)   20.4(2.6)    6.6(0.4)   Sandy Clay Loam
90-100  74.2(2.3)   20.0(2.2)    5.8(0.4)   Sandy Clay Loam

Table 5.--Variables, number of variables in the model (K), Akaike's
Information Criterion adjusted for small sample size (AICc),
difference in AICc value between the model and the model with the
lowest AICc value ([DELTA]AICc), and Akaike weight ([w.sub.i]) for
models used to predict southeastern pocket gopher presence within
the Jones Ecological Research Center in Raker County, Georgia,
2012-2013. Models are categorized as testing the competing
hypotheses that vegetation structure (Vegetation) or soil
characteristics (Soil) better predict gopher activity

Model                    K    AICc      [DELTA][DELTA]ICc

Silt+Clay+pH+Nitrogen+   5    92.807    0.000
 Carbon (a,c)
Global                   11   97.024    4.217
Nitrogen                 1    110.516   17.708
pH                       1    123.080   30.272
Clay                     1    124.852   32.044
LL (b)                   1    128.122   35.314
WG                       1    132.577   39.769
Carbon                   1    132.819   40.011
PS+LL+WO+FV+WG+OG        6    133.403   40.596
OG                       1    137.724   44.916
Null                     0    137.816   45.009
WO                       1    139.444   46.636
PS                       1    139.535   46.727
FV                       1    139.560   46.752
Silt                     1    139.837   47.029

Model                    [w.sub.i]   Category

Silt+Clay+pH+Nitrogen+   0.892       Soil
 Carbon (a,c)
Global                   0.108       Combination
Nitrogen                 0.000       Soil
pH                       0.000       Soil
Clay                     0.000       Soil
LL (b)                   0.000       Vegetation
WG                       0.000       Vegetation
Carbon                   0.000       Soil
PS+LL+WO+FV+WG+OG        0.000       Vegetation
OG                       0.000       Vegetation
Null                     0.000
WO                       0.000       Vegetation
PS                       0.000       Vegetation
FV                       0.000       Vegetation
Silt                     0.000       Soil

(a) Best soil model

(b) Best vegetation model

(c) Best overall predictive model

(d) PS = Pine Straw, LL = Leaf Litter, WO = Woody, FV =
Forb/Vine, WG = Wiregrass, OG = Other Grass

Table 6.--Variable, parameter estimate, lower 95% confidence
interval (CI), upper 95% CI, and variable weight
([summation][w.sub.i]) for variables in models used to predict
southeastern pocket gopher presence within the Jones Ecological
Research (".enter in Baker County, Georgia, 2012-2013

Variable       Estimate    Lower 95% CI   Upper 95% CI   [summation]

Nitrogen (a)   -142.4593      -234.703        -50.216        1.0000
PH (a)           1.7143          0.430          2.999        0.9999
Clay (a)        -0.1216         -0.213         -0.030        0.9999
Carbon          -0.3600         -5.403          4.683        0.9999
Silt (a)         0.5964          0.115          1.077        0.9999
LL              -0.0062         -0.019          0.007        0.1083
WG               0.0003         -0.008          0.008        0.1083
OG               0.0011         -0.005          0.007        0.1083
WO               0.0022         -0.008          0.012        0.1083
PS               0.0003         -0.006          0.006        0.1083
FV              -0.0038         -0.013          0.006        0.1083

(a) Variables with confidence intervals that exclude zero, indicating
that they are informative predictor

(b) PS = Pine Straw, LL= Leaf Litter, WO = Woody, FV = Forb/Vine,
WG = Wiregrass, OG = Other Grass
COPYRIGHT 2017 University of Notre Dame, Department of Biological Sciences
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2017 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Warren, Ashley E.; Castleberry, Steven B.; Markewitz, Daniel; Conner, L. Mike
Publication:The American Midland Naturalist
Date:Oct 1, 2017
Previous Article:Winter Breeding and Summer Nonbreeding Home Ranges of Bald Eagles from Louisiana.
Next Article:Responses of Eastern Gray Squirrels (Sciurus carolinensis) to Predator Calls and Their Modulation by Coat Color.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters