Translator Disclaimer
1 June 2017 Yellow Rail (Coturnicops noveboracensis) Occupancy in the Context of Fire in Mississippi and Alabama, USA
Author Affiliations +
Abstract

The Yellow Rail (Coturnicops noveboracensis) is a migratory bird with many aspects of its ecology poorly understood. The objective of this study was to examine effects of fire, vegetation structure, and landscape variables on site occupancy and detection probabilities for Yellow Rails overwintering in coastal pine savannas of Mississippi and Alabama. Between December and April, 2012–2013, dragline surveys for Yellow Rail were conducted at three conservation areas: two in Jackson County, Mississippi, and one in Mobile County, Alabama, USA. Site occupancy for Yellow Rail was 0.81 ± 0.06 (SD) with detection probabilities of 0.79 ± 0.09 (SD). Yellow Rail occupancy related negatively with time since fire, indicating fire provides conditions attractive to Yellow Rail overwintering throughout the study area. Yellow Rail use of wetland and fire-maintained habitats within coastal Mississippi during winter, coupled with continued loss of open grasslands and inadequate management of fire-dependent pine savanna habitats throughout the southeastern USA highlights the continued need to prioritize the conservation and effective management of herbaceous-dominated ecosystems.

Longleaf pine (Pinus palustris) ecosystems, pine savannas in particular, were once among the most expansive ecosystems in North America (Landers et al. 1995). It is estimated that prior to European settlement these ecosystems covered over 30 million ha in the southeastern United States (Frost 1993). These fire-dependent ecosystems frequently experienced low to moderate intensity fires produced by lightning and Native American activities (Brockway and Lewis 1997). Disruption of historical fire regimes, coupled with anthropogenic activities including habitat fragmentation and degradation, has been deemed responsible for the rapid decline of longleaf ecosystems (Brockway and Lewis 1997). Currently, longleaf pine ecosystems are among the most ecologically imperiled of all forested ecosystems in North America (Noss 1989; Simberloff 1993), with an area currently estimated to be less than 1.2 million ha (Brockway and Lewis 1997; Outcalt 2000; Frost 2006).

Disturbance, most notably fire, helped shape both historic and present-day vegetation structure of longleaf pine ecosystems (Platt 1999). Temporal and spatial heterogeneity in pyrogenic activity directly influences avian communities that inhabit fire-dependent ecosystems (Wiens 1974). Studies indicate prescribed fire is a beneficial management tool for avian species associated with longleaf pine systems (Tucker et al. 2003; Bechtoldt and Stouffer 2005). Subsequently, present day fire management programs aim to reintroduce fire into the landscape, resulting in dynamic habitat mosaics (Parr and Brockett 1999). However, information is lacking relating the effects of time since fire on occupancy for many avian species associated with longleaf pine ecosystems.

The Yellow Rail (Coturnicops noveboracensis), the second smallest of the North American rails, is a highly secretive marsh bird whose breeding range includes Canada and the northern Great Lakes region. The species winters along the Atlantic and Gulf Coastal Plains region from Texas to North Carolina, USA, primarily selecting high marsh habitat with dense, low undergrowth (Lima 1993; Butler et al. 2014; Leston and Bookhout 2015). As such, periodic disturbance including fire, herbivory, and flooding across breeding and wintering grounds appears critical in maintaining Yellow Rail habitat suitability (Burkman 1993; Mizell 1998; Austin and Buhl 2013). These disturbances recycle nutrients, increase ground cover biodiversity, remove litter, and top-kill woody vegetation, limiting encroachment (Platt 1999). In pine savanna habitats, fire suppresses woody encroachment and promotes herbaceous emergence (Lewis and Harshbarger 1976; Wilson et al. 1995). Burkman (1993) found that on the breeding grounds, Yellow Rail were more likely to be found in burned habitats having lower percentage of shrubs and higher percentage of sedges (Carex lasiocarpa) than control plots, suggesting suitability of habitat for Yellow Rail diminishes with the encroachment of woody plant species. Similarly, two previous studies conducted on Yellow Rail in Michigan noted a negative relationship between time since fire and Yellow Rail presence (Burkman 1993; Austin and Buhl 2013). Austin and Buhl (2013) found fire was the most important factor explaining the presence of Yellow Rail on their breeding grounds.

The goal of this study was to estimate occupancy rates for Yellow Rail as related to habitat features and management regimes in pine savannas along the northern Gulf Coast of Mississippi and Alabama, USA. We hypothesized that time since fire influenced Yellow Rail occupancy through its effects on habitat structure and composition. Given the discussion from above, we predicted Yellow Rail occupancy should: 1) decrease with time since fire; 2) be negatively related to woody vegetative cover and woody height; and 3) be positively related to herbaceous vegetation density.

Methods

Study Area

The study areas comprised more than 9,150 ha of fire maintained pine savanna habitats in three conservation areas along the Gulf Coast of the USA currently managed for the restoration or maintenance of wet pine savanna habitat: two in Mississippi (Mississippi Sandhill Crane National Wildlife Refuge (NWR) and Jackson County Mitigation Bank), and the Grand Bay Savanna Forever Wild complex in Alabama (Fig. 1). Located in the Gulf Coastal Plain ecoregion, these study areas are characterized by low topography and infertile, acidic, saturated clay soils, with a temperate climate including hot summers (mean = 27 °C; precipitation = 32 cm) and mild, wet winters (mean = 9 °C; precipitation = 38 cm) (National Climatic Data Center 2015). Dominant pine savanna herbaceous cover includes wiregrass (Aristida stricta), muhly grass (Muhlenbergia spp.), toothache grass (Ctenium aromaticum), bluestem grasses (Andropogon spp.) and panic grasses (Panicum spp.) (Peet and Allard 1993). Dominant woody vegetation consists of St. John's-wort (Hypericum brachyphyllum), gallberry (Ilex glabra), wax myrtle (Morella cerifera), sweetbay magnolia (Magnolia virginiana), yaupon (Ilex vomitoria) and smilax (Smilax domingensis) (Peet and Allard 1993).

Mississippi Sandhill Crane NWR is managed with prescribed fire applied at 3-year return intervals where 1/3 of management units (40 total) are burned annually (Wilder 2008). Jackson County Mitigation Bank is just north of the Grand Bay National Estuarine Research Reserve/NWR (Fig. 1). This area extends over 150 ha and is managed periodically by prescribed burning (3–5 year fire return interval). Grand Bay Savanna Forever Wild complex consists of four contiguous tracts, totaling more than 2,000 ha (Fig. 1) and is managed with frequent prescribed burns (2–3 year fire return interval).

Survey Methods

Within the study area, we conducted surveys for Yellow Rails between December and April, 2012–2013. We defined a study site as a singular management unit within one of the three conservation areas and survey plot as the area surveyed within each study site, where each study site is associated with one survey plot. We selected 13 study sites based on fire history and feasibility to survey selected habitat, with 11 sites located within Mississippi Sandhill Crane NWR, one site in Jackson County Mitigation Bank and one site in the Grand Bay Savanna Forever Wild complex (Table 1). The size of each survey plot varied (1–5 ha) conditional upon logistical constraints, such as tree-line boundaries (Table 1). Primary habitat type of all survey plots was pine savanna.

Figure 1.

Study areas in Mississippi and Alabama, USA, 2012–2013.

f01_95.jpg

We conducted surveys with a four-person field crew using a 15-m weighted dragline (Mizell 1998; Grace et al. 2005; Butler et al. 2010) with attached noisemakers (500-mL plastic bottles containing rocks and attached to the dragline at 0.5-m intervals) to create noise and disturbance throughout the vegetation, causing birds to flush from ground cover. Each member of the field crew used a Kelly's K-Light (5-LED 21 V Std Light) or a Boss Light (24v7). Two members of the crew pulled each end of the dragline at a walking speed (˜1.6 kmph), while the other two members were spread equidistant behind the dragline. Surveys began 30 min after sunset and stopped after 3 hr or upon coverage of the entire survey plot (Mizell 1998; Grace et al. 2005; Butler et al. 2010).

During each transect pass, the survey crew walked in a straight line navigating to a fixed position established by a lantern hanging from a Shepherd's hook placed at each transect origin. At the end of each transect pass, the crew measured the distance (m) between each end of the bowed dragline. The crew then shifted perpendicular to the transect, moving one dragline length (15 m) to the right or left. After shifting over, the crew placed the lantern at the end of the dragline for navigating each pass. To determine the total area covered for each survey, we used a composite of dragline widths and the linear distances covered by each transect. Transect lengths were collected using the Tracks function on a handheld GPS unit (Garmin GPSMap76CSX). The crew then proceeded parallel to the transect using the lantern to navigate. For each survey plot, we standardized starting locations and the direction of line shift.

The survey crew directly noted Yellow Rail detections, using field identification characteristics (i.e., white secondary wing patches) and flight behavior (i.e., rapid, shallow wing-beats in a low arcing path) particular to this species. We recorded the initial flush location of detected Yellow Rails using a handheld GPS unit and placed a 1-m PVC pole vertically in the ground to facilitate subsequent collection of vegetation metrics during daylight hours. We surveyed each survey plot three times throughout the field season (December to April), with surveys at the same site all 14 or more days apart.

Habitat and Site Assessment

We measured plant community structure the day following each survey, at two locations per observed bird: 1) each location where we detected a Yellow Rail; and 2) a paired control point. We determined control point locations using a random numbers table, selecting a number between 1–360 for compass bearing and a number between 20–100, as a distance (in m) from the detection point. We then established circular plots (10-m radius) centered on each Yellow Rail or control point. Using a 2-m pole placed at the circular plot centroid, we measured point-intercept density by determining the lowest decimeter where the pole was greater than 50% obscured and recorded maximum height of woody and herbaceous vegetation within 25 cm of the pole (Robel et al. 1970). We measured vegetation height at 2-m intervals along four transects, one in each of the four cardinal directions radiating from the centroid (four measurements/transect), and recorded point-intercept density at the end of each 10-m transect, as observed toward the center of the plot at a height of 1 m. Additionally, we recorded diameter at breast height (DBH) of all trees greater than or equal to 7.5 cm DBH within the plot and visually estimated percent coverage of dominant woody (tree and shrub species) and herbaceous vegetation to the nearest 1%, totaling 100% across the plot. We also measured habitat characteristics at study plots where we detected no Yellow Rail (hereafter, “non-detection location”) throughout the field season by selecting five locations within each study plot using a random numbers table. To select each of the five points, we randomly selected one transect from among those traversed during the final survey, and then randomly selected a distance along that transect. We repeated this process each time we selected a sampling point.

Table 1.

Characteristics associated with Yellow Rail study sites within study areas in Mississippi and Alabama, 2012– 2013. Time since fire is averaged among three site visits.

t01_95.gif

For analytical purposes, we summarized vegetation characteristics by averaging all vegetation metrics collected at all random points to describe the vegetation at the study plot level associated with a given study site. At sites where we detected no Yellow Rail, we summarized site characteristics by averaging all vegetation metrics collected at all random points associated with a given site. We included site-averaged metrics in subsequent occupancy models.

For each study site, we calculated patch size (ha) and survey area (ha) using ArcGIS (Environmental Systems Research Institute 2011). We used 2013 high resolution orthoimagery (U.S. Geological Survey 2013) with 1-m resolution to calculate the area of open grassland in each study site using the area measure tool in ArcGIS (Environmental Systems Research Institute 2011). We calculated survey area using a composite of dragline widths and the linear distance covered by each transect for each survey event. We eliminated any areas of transect overlap from the calculations. We averaged survey areas among site visits for use in modeling siteoccupancy by Yellow Rail.

Statistical Methods

We employed Markov chain Monte Carlo (MCMC) methods within a Bayesian framework using statistical package JAGS (Plummer 2013) and statistical package R2jags (Su and Yajima 2012) in statistical program R (R Development Core Team 2013) to model Yellow Rail detection probability and site occupancy (Kéry and Schaub 2012). Bayesian models followed the same maximum likelihood approach as described by MacKenzie et al. (2002). Site-occupancy models use replicated surveys within a survey plot to resolve the uncertainty of a recorded absence, which can occur when a species is absent or present but not detected (MacKenzie et al. 2006). Modeling Yellow Rail occupancy in the Bayesian framework allows us to provide direct probability statements about specific models while incorporating uncertainty into each model (Ellison 2004; Martin et al. 2005). This method extends logistic regression models with Bernoulli distributions for state (Ψ) and detection (ρ), accounting for error in detection (MacKenzie et al. 2002). Occupancy models provide estimates for Ψ, the estimated proportion of study sites occupied by the species, and ρ, the probability of detecting the species given it is present during a survey event. Assumptions of occupancy models include a closed population, absence of false presence, and that selection and scale of sampling areas are ecologically significant to the study species (MacKenzie et al. 2006).

We constructed occupancy models using a two-step selection process. First, we modeled ρ while Ψ was held constant (MacKenzie et al. 2002), only including covariates that could influence ρ. We considered those factors potentially influencing the detection probability of Yellow Rail using a dragline method to be DENSITY and WOOD. These covariates were therefore included as sample-specific, or detection-model covariates. Site covariates, those considered to affect Ψ, included herbaceous and woody height, herbaceous and woody percent cover, point-intercept density, patch size, survey area, stand DBH and time since fire (FIRE) (Table 2). Prior to running models, we scaled all continuous site covariates (mean = 0 and to unit SD) and examined correlations between all predictor variables for multicollinearity (r > 0.5) (Table 3) (Gehring et al. 2015).

Due to the inverse correlation of herbaceous cover and woody cover among sites, we selected woody cover to remain in the global model as previous studies indicated a negative relationship between Yellow Rail occurrence and woody cover (Burkman 1993; Martin 2012; Austin and Buhl 2013). Subsequently, we designated woody cover as a proxy for collected woody features and selected density to remain in the model to represent habitat structure. By including woody cover and density as habitat features in modeling occupancy, we were able to assess Yellow Rail occupancy as it relates to both vegetative structure and composition (Ribic et al. 2009). As patch size was highly correlated with survey area, we selected patch size to remain in occupancy modeling to evaluate the strength toward area sensitivity for Yellow Rail in pine savanna habitats (Robbins et al. 1989). Area sensitivity occurs when the size of a fragmented area influences a species' occupancy or abundance (Robbins et al. 1989).

Table 2.

Metrics used in generating site-occupancy estimates of Yellow Rail in pine savanna habitats of Mississippi and Alabama, USA, 2012–2013.

t02_95.gif

We constructed a singular global model that included those variables we considered biologically significant to explain Yellow Rail occupancy. Rather than applying criteria to select among competing models, we based inference on output from this single model (Carillo-Rubio et al. 2014). This global model included the covariates DENSITY and WOOD on detection probability and DBH, DENSITY, FIRE, woody cover, and patch size as covariates on occupancy. We assigned all model parameters flat priors (Uniform (0, 0.01)). For the candidate model, we sampled from three MCMC chains for 50,000 iterations, discarding the first 5,000 iterations as “burn-in”, and a thinning factor of 10 to minimize autocorrelation among iterations, producing 13,500 estimates for each model parameter. We evaluated chain convergence using R-hat <1.1 (Gelman and Hill 2007), where R-hat values <1.1 indicate adequate convergence to the principal distribution. For the global model, we considered parameters in which the 90% credible intervals did not overlap zero to affect Yellow Rail detection and occupancy. We also generated Bayesian P values (Gelman et al. 2014) to determine goodness of fit and to assess the proportion of posterior distribution values that were either > 0 or < 0 if parameter 90% credible intervals overlapped zero.

Results

Throughout our study, we conducted a total of 53 surveys covering 485 ha (Table 1). During these surveys, we detected a total of 85 Yellow Rails across all sites and surveys. We detected multiple (two to six) rails per survey at sites G-05, G-11SE, G-11SW, G-13 and G-14 of the Mississippi Sandhill Crane NWR, at the Jackson County Mitigation Bank and at the Grand Bay Savanna Forever Wild complex. Individual birds were detected during surveys at G-05, G-11SW, G-13 and G-14 of the Mississippi Sandhill Crane NWR, at the Jackson County Mitigation Bank and at the Grand Bay Savanna Forever Wild complex, while no individuals were detected during some surveys at sites G-07, G-11SE, G-11SW, G-12, O-07, O-10 and O-13 of the Mississippi Sandhill Crane NWR.

Table 3.

Correlation matrix for habitat characteristics collected in pine savanna habitats of Mississippi and Alabama, USA, 2012–2013. See Table 2 for parameter definitions.

t03_95.gif

We detected Yellow Rail at least once at nine of our study sites (69%), for a total of 22 detections in 2012–2013. Mean occupancy probability among all sites, as derived from global model estimates, was 0.81 ± 0.06 (± SD) with a detection probability of 0.79 ± 0.09 (± SD). Posterior summaries and derived parameters from the global model indicated increased likelihood of Yellow Rail occupancy proximate to when supporting habitat had been last burned (Table 4), where the probability of Yellow Rail occupancy significantly decreased with time post burn (Table 4; Fig. 2). None of the other covariates considered were found to affect Yellow Rail occupancy or their detection.

Table 4.

Mean, standard deviation (SD), and 90% credible intervals (CRI) of landscape and habitat metrics included in the global model for Yellow Rail occupancy in pine savanna habitats of Mississippi and Alabama, USA, 2012–2013. Two asterisks (**) indicates parameter where the 95%- CRI does not overlap zero. Single asterisk (*) indicates support of a moderate effect on occupancy where the 90% CRI does not overlap zero. fi01_95.gif values for all parameter estimates were ≤ 1.02. Mean and variance of all predictor variables were standardized to improve model fit.

t04_95.gif

Discussion

This study reinforces the importance of implementing prescribed fire in managing pine savanna habitats for Yellow Rail in pine savanna habitats of the Gulf Coast of the USA. Similar to Burkman (1993), we found Yellow Rail occurrence was greatest at less than 2 years post burn, while we detected individuals in few sites at greater than 2 years post burn. Austin and Buhl (2013) reported similar results where, at a larger scale, time since fire was the most important factor explaining Yellow Rail presence during the breeding season. In addition, shorter fire return intervals in pine savanna ecosystems have been shown to influence the occurrence of other winter grassland bird species of conservation concern such as Henslow's Sparrow (Ammodramus henslowii) (Tucker et al. 2003; Bechtoldt and Stouffer 2005).

Figure 2.

Prediction of the covariates time since fire as relevant to site-occupancy (solid line [with uncertainty 90% credible intervals, dashed line]) of Yellow Rail in pine savanna habitats of Mississippi and Alabama, USA, 2012–2013.

f02_95.jpg

Information on the effects of habitat fragmentation and patch size on wintering bird species, including Yellow Rail, is imperative for the management of grassland habitats (Herkert et al. 1996). Of interest, we did not find support for patch size as a predictor of Yellow Rail occupancy during the nonbreeding season. In an evaluation of Yellow Rail breeding habitat in Michigan, USA, Austin and Buhl (2013) found landscape-level composition (measured within 300 m of survey points) was not strongly tied to Yellow Rail occupancy. Rather, Yellow Rail presence was influenced by local conditions (within 100 m of survey points), where the primary predictor of Yellow Rail presence was fire history. Within the historical wintering range of Yellow Rail in the southeastern USA, fires occurred in intervals of 1–3 years in pine savanna habitats (Frost 1995). Gonzalez-Benecke et al. (2015) found throughout the southeastern USA that increased fire frequency (2–4 year return interval) significantly reduced woody ground cover, but had little effect on herbaceous ground cover. They also found reduced fire frequency (> 3-year return interval) significantly increased basal area of trees throughout the study area (Gonzalez-Benecke et al. 2015). Thus, a more thorough exploration of the relationship between fire return intervals and Yellow Rail occupancy, both on wintering and breeding grounds, is justified.

Survey protocols for Yellow Rail on their breeding grounds incorporate the vocalization of the species for detection (Bazin and Baldwin 2007; Austin and Buhl 2013). However, anecdotal observations of wintering Yellow Rails suggest they rarely vocalize during the winter in pine savanna habitats (M. S. Woodrey, pers. commun.), observations consistent with a general pattern that temperate birds infrequently vocalize during the nonbreeding season (Fletcher et al. 2000). Thus, surveys targeting winter grassland bird species, such as the Yellow Rail, will require a specific method not involving the use of callbroadcast techniques.

In pine savanna habitats, fire and hydrology are two major factors controlling the distribution of vegetation and habitat structure (Noss 2013). Although we observed most study sites saturated with water at the soil level, rarely was there measureable surface water. On their wintering grounds in coastal Texas, Mizell (1998) found standing water depth was a strong determinant of Yellow Rail habitat use (mean = 1.3 cm). However, working in similar coastal prairie areas in eastern Texas, Grace et al. (2005) did not find a strong relationship between water depth and Yellow Rail habitat use. Yellow Rails are most often associated with a variety of emergent wetland habitats (Leston and Bookhout 2015), including pine savanna systems. Due to our inability to adequately assess hydrological conditions associated with Yellow Rail habitat use in pine savanna habitats, we suggest further research into the relationship between soil moisture and winter Yellow Rail distribution and abundance. A couple of potentially fruitful approaches could be the use of fine scale delineation of plant species according to their wetland indicator status or assessing soil moisture directly by collecting soil samples and drying them in a laboratory setting.

Fire is a key ecological process central to the restoration and management of longleaf pine systems (Platt 1999; Noss 2013). Based on evidence provided here coupled with historical fire intervals, we recommend managers target 2-year intervals to maintain habitat features attractive to wintering Yellow Rail in pine savanna habitats. Targeting a 2-year return interval would likely result in a series of burns implemented on a 3- to 4-year interval, given the vagaries of weather, the narrow environmental conditions necessary for burn permits, and required financial and personnel resources. Further, although a 2-year fire interval could provide habitats directly supporting Yellow Rail, a mosaic of habitat succession should be established to support broader longleaf pine management objectives. At the local level, managers can also reduce habitat fragmentation by combining management units and, where possible, removing unnecessary fire breaks. While fire breaks allow for the compartmentalization and control of prescribed fire, they also promote the fragmentation of habitat, disrupt hydrology and enable the spread of invasive species (Noss 2013).

ACKNOWLEDGMENTS

We thank everyone at the Mississippi Sandhill Crane National Wildlife Refuge, Grand Bay National Estuarine Research Reserve/National Wildlife Refuge and Alabama Department of Natural Resources and Conservation for their support throughout this project especially, Angela Dedrickson and John Trent. This study was not possible without the countless volunteers, interns and field technicians that assisted with this project, especially Jared Feura, Eamon Harrity, Matt Boone and Jake Walker. Several anonymous reviewers provided comments valuable to strengthening this manuscript. Financial support was provided by the Mississippi Ornithological Society and the U.S. Fish and Wildlife Service National Wildlife Refuge System's Southeast Inventory & Monitoring Network. In addition, this material is based upon work that is supported by the National Institute of Food and Agriculture, U.S. Department of Agriculture, Hatch Project funds awarded to the Mississippi Agricultural and Forestry Experiment Station and NOAA Award # NA12NOS4200106 to the Mississippi Department of Marine Resources' Grand Bay National Estuarine Research Reserve. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the authors and do not necessarily reflect the view of the U.S. Government.

Literature Cited

1.

Austin, J. E. and D. A. Buhl. 2013. Relating Yellow Rail (Coturnicops noveboracensis) occupancy to habitat and landscape features in the context of fire. Waterbirds 36: 199–213. Google Scholar

2.

Bazin, R. and F. B. Baldwin. 2007. Canadian Wildlife Service standardized protocol for the survey of Yellow Rails (Coturnicops noveboracensis) in prairie and northern region. Unpublished report, Environment Canada, Winnipeg, Manitoba. Google Scholar

3.

Bechtoldt, C. L. and P. C. Stouffer. 2005. Home-range size, response to fire, and habitat preferences of wintering Henslow's Sparrows. Wilson Bulletin 117: 211–225. Google Scholar

4.

Brockway, D. G. and C. E. Lewis. 1997. Long-term effects of dormant-season prescribed fire on plant community diversity, structure and productivity in a longleaf pine wiregrass ecosystem. Forest Ecology and Management 96: 167–183. Google Scholar

5.

Burkman, M. A. 1993. The use of fire to manage breeding habitat for Yellow Rails. M.S. Thesis, Northern Michigan University, Marquette. Google Scholar

6.

Butler, C. J., J. K. Wilson, C. R. Brower and S. R. Frazee. 2014. Age ratios, sex ratios, and a population estimate of Yellow Rails at San Bernard National Wildlife Refuge, Texas. Southeastern Naturalist 59: 319– 324. Google Scholar

7.

Butler, C. J., L. H. Pham, J. N. Stinedurf, C. L. Roy, E. L. Judd, N. J. Burgess and G. M. Caddell. 2010. Yellow Rails wintering in Oklahoma. Wilson Journal of Ornithology 122: 385–387. Google Scholar

8.

Carillo-Rubio, E., M. Kéry, S. J. Morreale, P.J. Sullivan, B. Gardner, E. G. Cooch and J. P. Lassoie. 2014. Use of multispecies occupancy models to evaluate the response of bird communities to forest degradation associated with logging. Conservation Biology 28: 1034–1044. Google Scholar

9.

Ellison, A. M. 2004. Bayesian inference in ecology. Ecology Letters 7: 509–520. Google Scholar

10.

Environmental Systems Research Institute (ESRI). 2011. ArcGIS v. 10. ESRI, Redlands, California. Google Scholar

11.

Fletcher, R. J., Jr. , J. A. Dhundale and T. F. Dean. 2000. Estimating non-breeding season bird abundance in prairies: a comparison of two survey techniques. Journal of Field Ornithology 71: 321–329. Google Scholar

12.

Frost, C. C. 1993. Four centuries of changing landscape patterns in the longleaf pine ecosystem. Pages 17– 43 in The Longleaf Pine Ecosystem: Ecology, Restoration, and Management ( S. M. Hermann, Ed.) Proceedings of the 18th Tall Timbers Fire Ecology Conference, Tall Timbers Research Station, Tallahassee, Florida. Google Scholar

13.

Frost, C. C. 1995. Presettlement fire regimes in southeastern marshes, peatlands, and swamps. Pages 39– 60 in Fire in Wetlands: A Management Perspective ( S. I. Cerulean and R. T. Engstrom, Eds.). Proceedings of the 19th Tall Timbers Fire Ecology Conference, Tall Timbers Research Station, Tallahassee, Florida. Google Scholar

14.

Frost, C. C. 2006. History and future of the longleaf pine ecosystem. Pages 9–48 in The Longleaf Pine Ecosystem: Ecology, Silviculture, and Restoration ( S. Jose, E.J. Jokela and D. L. Miller, Eds.). Springer, New York, New York. Google Scholar

15.

Gehring, E., G. B. Pezzatti, P. Krebs, S. Mazzoleni and M. Conedera. 2015. On the applicability of the pipe model theory on the chestnut tree (Castanea sativa Mill.). Trees 29: 321–332. Google Scholar

16.

Gelman, A. and J. Hill. 2007. Data analysis using regression and multilevel/hierarchical models. Cambridge University Press, New York, New York. Google Scholar

17.

Gelman, A., J. B. Carlin, H. S. Stern and D. B. Rubin. 2014. Bayesian data analysis, 3rd ed. CRC Press, Boca Raton, Florida. Google Scholar

18.

Gonzalez-Benecke, C. A., L. J. Samuelson, T. A. Stokes, W. P. Cropper, Jr. , T. A. Martin and K. H. Johnsen. 2015. Understory plant biomass dynamics of prescribed burned Pinus palustris stands. Forest Ecology and Management 344: 84–94. Google Scholar

19.

Grace, J. B., L. K. Allain, H. Q. Baldwin, A. G. Billock, W. R. Eddleman, A. M. Given and R. Moss. 2005. Effects of prescribed fire in the coastal prairies of Texas. USGS Open-File Report 2005–1287, U.S. Department of the Interior, Geological Survey, Reston, Virginia. Google Scholar

20.

Herkert, J. R., D. W. Sample and R. E. Warner. 1996. Management of grassland landscapes for the conservation of migratory birds. Pages 89–116 in Managing Midwest Landscapes for the Conservation of Neotropical Migratory Birds ( F. R. Thompson, III , Ed.). General Technical Report NC-187, U.S. Forest Service, North Central Forest Experiment Station, St. Paul, Minnesota. Google Scholar

21.

Kéry, M. and M. Schaub. 2012. Bayesian population analysis using WinBUGS: a hierarchical perspective. Academic Press, Waltham, Massachusetts. Google Scholar

22.

Landers, J. L., D. H. Van Lear and W. D. Boyer. 1995. The longleaf pine forests of the Southeast: requiem or renaissance? Forestry 9: 39–44. Google Scholar

23.

Leston, L. and T. A. Bookhout. 2015. Yellow Rail (Coturnicops noveboracensis). No. 139 in The Birds of North America Online ( A. Poole, Ed.). Cornell Lab of Ornithology, Ithaca, New York,  https://birdsna.org/Species-Account/bna/species/yelrai, accessed 18 May 2015. Google Scholar

24.

Lewis, C. E. and T. J. Harshbarger. 1976. Shrub and herbaceous vegetation after 20 years of prescribed burning in the South Carolina coastal plain. Journal of Range Management 29: 13–18. Google Scholar

25.

Lima, S. L. 1993. Ecological and evolutionary perspectives on escape from predatory attack: a survey of North American birds. Wilson Bulletin 105: 1–47. Google Scholar

26.

MacKenzie, D. I., J. D. Nichols, G. B. Lachman, S. Droege, J. A. Royle and C. A. Langtimm. 2002. Estimating site-occupancy when detection probabilities are less than one. Ecology 83: 2248–2255. Google Scholar

27.

MacKenzie, D. I., J. D. Nichols, J. A. Royle, K. H. Pollock, L. L. Bailey and J. E. Hines. 2006. Occupancy estimation and modeling: inferring patterns and dynamics of species occurrence. Elsevier, Burlington, Massachusetts. Google Scholar

28.

Martin, K. A. 2012. Habitat suitability of the Yellow Rail in south-central Manitoba. M.S. Thesis, University of Manitoba, Winnipeg. Google Scholar

29.

Martin, T. G., P. M. Kuhnert, K. Mengersen and H. P. Possingham. 2005. The power of expert opinion in ecological models using Bayesian methods: impact of grazing on birds. Ecological Applications 15: 266– 280. Google Scholar

30.

Mizell, K. L. 1998. Effects of fire and grazing on Yellow Rail habitat in a Texas coastal marsh. Ph.D. Dissertation, Texas A&M University, College Station. Google Scholar

31.

National Climatic Data Center (NCDC). 2015. NCDC climatic data online.  http://www.ncdc.noaa.gov/, accessed 11 March 2015. Google Scholar

32.

Noss, R. F. 1989. Longleaf pine and wiregrass: keystone components of an endangered ecosystem. Natural Areas Journal 9: 211–213. Google Scholar

33.

Noss, R. F. 2013. Forgotten grasslands of the south: natural history and conservation. Island Press, Washington, D.C. Google Scholar

34.

Outcalt, K. W. 2000. The longleaf pine ecosystem of the South. Native Plants Journal 1: 42–53. Google Scholar

35.

Parr, C. L. and B. H. Brockett. 1999. Patch-mosaic burning: a new paradigm for savanna fire management in protected areas? Koedoe 42: 117–130. Google Scholar

36.

Peet, R. K. and D. J. Allard. 1993. Longleaf pine vegetation of the southern Atlantic and eastern Gulf Coast regions: a preliminary classification. Pages 45–81 in The Longleaf Pine Ecosystem: Ecology, Restoration, and Management ( S. M. Hermann, Ed.). Proceedings of the 18th Tall Timbers Fire Ecology Conference, Tall Timbers Research Station, Tallahassee, Florida. Google Scholar

37.

Platt, W. J. 1999. Southeastern pine savannas. Pages 23– 51 in Savannas, Barrens, and Rock Outcrop Plant Communities of North America ( R. C Anderson, J. S. Fralish and J. M. Baskin, Eds.). Cambridge University Press, Cambridge, U.K. Google Scholar

38.

Plummer, M. 2013. JAGS: just another Gibbs sampler v. 3.4.0.  http://mcmc-jags.sourceforge.net/, accessed 2 October 2014. Google Scholar

39.

R Development Core Team. 2013. R: a language and environment for statistical computing v. 3.0.2. R Foundation for Statistical Computing, Vienna, Austria.  http://www.R-project.org/, accessed 20 August 2014. Google Scholar

40.

Ribic, C. A., R. R. Koford, J. R. Herkert, D. H. Johnson, N. D. Niemuth, D. E. Naugle, K. K. Bakker, D. W. Sample and R. B. Renfrew. 2009. Area sensitivity in North American grassland birds: patterns and processes. Auk 126: 233–244. Google Scholar

41.

Robbins, C. S., D. K. Dawson and B. A. Dowell. 1989. Habitat area requirements of breeding forest birds of the middle Atlantic states. Wildlife Monographs 103: 3–35. Google Scholar

42.

Robel, R. J., J. N. Briggs, A. D. Dayton and L. C. Hulbert. 1970. Relationships between visual obstruction measurements and weight of grassland vegetation. Journal of Range Management 23: 295–297. Google Scholar

43.

Simberloff, D. 1993. Species-area and fragmentation effects on old-growth forests: prospects for longleaf pine communities. Pages 1–13 in The Longleaf Pine Ecosystem: Ecology, Restoration, and Management ( S. M. Hermann, Ed.). Proceedings of the 18th Tall Timbers Fire Ecology Conference, Tall Timbers Research Station, Tallahassee, Florida. Google Scholar

44.

Su, Y. S. and M. Yajima. 2012. R2jags: a package for running jags from R. R package v. 0.03–08. R Foundation for Statistical Computing, Vienna, Austria.  http://CRAN.R-project.org/package=R2jags/, accessed 2 October 2014. Google Scholar

45.

Tucker, J. W., Jr. , G. E. Hill and N. R. Holler. 2003. Longleaf pine restoration: implications for landscapelevel effects on bird communities in the Lower Gulf Coastal Plain. Southern Journal of Applied Forestry 27: 107–121. Google Scholar

46.

U.S. Geological Survey. 2013. EarthExplorer. U.S. Department of the Interior, Geological Survey, Earth Resources Observation and Science Center, Sioux Falls, South Dakota.  http://earthexplorer.usgs.gov/, accessed 1 August 2012. Google Scholar

47.

Wiens, J. A. 1974. Habitat heterogeneity and avian community structure in North American grassland birds. American Midland Naturalist 91: 195–213. Google Scholar

48.

Wilder, T. 2008. Fire management plan for the Mississippi Sandhill Crane National Wildlife Refuge and the Grand Bay National Wildlife Refuge. U.S. Department of the Interior, Fish and Wildlife Service, Gautier, Mississippi. Google Scholar

49.

Wilson, C. W., R. E. Masters and G. A. Buckenhofer. 1995. Breeding bird response to pine-grassland community restoration for Red-Cockaded Woodpeckers. Journal of Wildlife Management 59: 56–67. Google Scholar
Kelly M. Morris, Mark S. Woodrey, Scott G. Hereford, Eric C. Soehren, Tara J. Conkling, and Scott A. Rush "Yellow Rail (Coturnicops noveboracensis) Occupancy in the Context of Fire in Mississippi and Alabama, USA," Waterbirds 40(2), 95-104, (1 June 2017). https://doi.org/10.1675/063.040.0202
Received: 18 July 2015; Accepted: 1 January 2017; Published: 1 June 2017
JOURNAL ARTICLE
10 PAGES


SHARE
ARTICLE IMPACT
Back to Top