Abiotic pond characteristics potentially influencing breeding of Houston toads (Bufo houstonensis).
TEXAS J. SCI. 55(4):315-322
The federally endangered Houston toad (Bufo houstonensis) is found only in Texas, with the largest population occurring in Bastrop County. They require pine and/or oak woodlands underlain by pockets of deep sandy soils, with permanent or ephemeral pools of water available for breeding. Despite extirpation from several counties within their historical range and documented decreases in the most well studied population in Bastrop country, relatively little field research has been published on the reproductive ecology of this toad. One of the current methods attempted to stabilize the populations is to create artificial breeding ponds in the hope of their subsequent colonization and breeding use by Houston toads. Unfortunately little published data is available documenting the nature of ponds used by Houston toads. This study establishes base-line data of physical characteristics of ponds used by the Houston toad for breeding in contrast to adjacent ponds that are currently unused by the toad for reproduction.
The 4826 acre Griffith League Ranch (GLR) lies in the small "Lost Pines" area of Bastrop County. The ranch contains 18 permanent and ephemeral ponds scattered throughout each of 7 soil types. These soil types include: Patilo series (3054 acres), Demona Series (297 acres), Silstid Series (1042 acres), Axtell Series (190 acres), Tabor Series (121 acres), Sayers Series (80 acres) and Jedd Series (42 acres) (Baker 1979). The geology of the GLR consists of the Carrizo Sand formation which divides the Wilcox group to the west and the Reklaw formation to the east. With the exception of two ponds (Pond 1 and Pond 4), all the ponds examined lie in predominantly deep sandy surface soils regardless of the underlying geology due to surface overrunning by adjacent sands. Ponds 1 and 4 are over predominately clay surface soils of the Reklaw formation.
The majority of the GLR is native forest with several large pastures dividing the ranch from the southwest to the northeast. The forests are composed of loblolly pine (Pinus taeda), post oak (Quercus stellata) and eastern red cedar (Juniperus virginiana). The majority of the ponds on the site are either adjacent to (within 100 m) or within forested patches with the exception of one pond. One pond lies more than 100 m from the forest, separated by pasture. This last pond (#17) has a small patch (less than 0.25 hectares) of trees and shrubs on its northern shore. Thus, with one exception the ponds are relatively homogeneous in their surrounding upland vegetational composition.
Determining the use of a given pond by the Houston toad is a long-term endeavor. Based on four breeding season surveys it appears that all ponds on the GLR could be easily reached by Houston toad males during the breeding season. Initially Houston toads were located by audio surveys of male choruses during the breeding season. This was enhanced by verifying the presence of females, eggs or juveniles at each pond each year. Often chorusing of Houston toads is taken as evidence of breeding. This has not been true during the four years of study of the 18 ponds on the Griffith League Ranch. While chorusing has been detected in all but three ponds (Ponds 1, 4 and 17) females, egg strands and/or tadpoles have only been found in eight of the remaining 15 ponds ([approximately equal to]50%). A pond with chorusing was considered to be a positive pond and a pond which has shown evidence of successful reproduction to be a breeding pond. However, despite the presence of eggs at several ponds each year, successful emergence of juveniles is not common even in those ponds that have had eggs deposited in them. Usually only one of the breeding ponds in any given year has shown successful emergence. Thus, from 2000 to 2003 only four "emergence" ponds have been detected on the GLR.
MATERIALS AND METHODS
Of the eighteen ponds found on the ranch, fourteen were chosen as providing generally consistent slopes along each of the four cardinal points for the comparisons in this study. While the study of biotic factors, including competition and predation, are important in order to understand the dynamics regulating community structure, the study of abiotic factors must also be included as possible controlling variables (Dunson & Travis 1991). Two abiotic components, embankment slope and pH, were examined as possible variables influencing successful B. houstonensis reproduction.
Littoral slope has been found to be a major factor in determining species richness, fish sizes and densities (Randall et al. 1996), and the biomass of submerged macrophyte communities (Duarte & Kalff 1986). Banks & Beebee (1987) found the endangered Natterjack toad (Bufo calamita) sought ponds with particular physical and chemical properties that included steepness of the pond bank and pH. These researchers concluded that B. calamita discriminated in its choice of breeding sites and sought out unshaded pools with shallow margins and avoided shady or steep-sided ponds and ponds with acidic waters.
From four compass bearings (N, S, E, W) the slope of each of the 14 ponds surveyed at GLR was measured from the pond's edge to a depth of 60.0 cm. The slope transects were measured from June to September 2001, when the ponds were at their mean pool height. The mean pool height was determined using the edges of the ponds indicated by the spring 2000 and 2001 levels.
Bufo tadpoles tend to form aggregations around the edges of ponds during the day and migrate into deeper water at night (Beiswenger 1977; Beebee 1985), thus 60.0 cm was chosen as the maximum slope depth because it was within the areas known to be used by other Bufo tadpoles during their diel migrations (Griffiths et al. 1988; Bardsley & Beebee 1998). The slope measurements were then categorized in terms of degrees as North slope, South slope, East slope and West slope. The pH of each pond was measured with a Sper Scientific pH meter 30 cm from the pond's edge and 2 cm below the surface at morning in the breeding season from January to April 2002. All ponds were monitored extensively for a total of four breeding seasons to document chorusing and discriminate between positive ponds and breeding ponds.
The littoral slope and pH at two cm depth for the 14 ponds measured at Griffith League Ranch are summarized in Table 1. Chorus and reproductive use of those ponds across four breeding seasons are also summarized in Table 1. Comparisons of pond slopes between ponds used for breeding by B. houstonensis and ponds not used by B. houstonensis indicated that except for the Southern slopes of ponds (Mann-Whitney U test, P = 0.01-0.02, Z = -2.24), there was no significant difference in slopes between breeding and nonbreeding ponds. The pH range for ponds used by B. houstonensis for breeding ranged from 5.33 to 6.70. All seven nonbreeding ponds that were surveyed and are not currently used by the toad for breeding fell within this same pH range. The pH was found to vary to a small degree across the breeding season and to an even lesser degree between night and daytime measurements.
Pool height varied dramatically in late 2002 and 2003 from earlier years. Each of the measurements taken at the 2000-2002 mean pool heights were submerged deeper than 60 cm depth during the 2003 breeding season. With significant rains in the summer of 2002 and again during late winter of that year all of the ponds rose to maximum pool heights. It is apparent that the bank slopes at the mean pool height become irrelevant at maximum pool heights for each pond. Hence, while slope results are relevant in this study for three of the four study years, the final year demonstrated that longer cycle fluctuations in pool height are important to take into consideration. In all but one case (8 of 9 or 88%), egg strings from Houston toads were deposited on banks with a slope of 10[degrees] or less (Table 1). In at least one pond (#2) the variation in pool height created a shallow slope bank at full pool that had been very steep (25[degrees]) at mean pool height.
The spatial distribution of Bufo tadpoles varies within ponds and the larvae are known to undergo a diel cycle of distribution and activity (Beiswenger 1977; Beebee 1985). Bufo americanus tadpoles spend the night in the deeper and warmer areas of the pond and begin migrating to the shallow areas when light intensity increases and warms the water (Beiswenger 1977). Bufo houstonensis observed in this study appear to undergo a similar migration in which they form aggregations that feed along the shore during the day but appear to disperse into the deeper portions of the pond at night. These tadpoles were observed feeding upon pine pollen that has accumulated along the shores as well as material attached to vegetation, on the surface of water, and along the shores (Hillis et al. 1984). Bardsley & Beebee (1998) identified four microhabitats utilized by B. calamita, which ranged from a shallow vegetated depth of two cm to a deeper vegetated depth of 40 cm.
For the 14 ponds examined here, the general trend detected supports higher Houston toad egg string occurrence for those banks with a slope less than 10[degrees] (Table 1). It is also apparent that multiple breeding seasons are required to evaluate the suitability of a pond as a potential breeding site. Often the data across years provides a contrary result to that predicted from a single year. For example, Pond 16 had successful Houston toad reproduction in 2000 but no juveniles emerged. Subsequent chorusing did not occur at that location in the following three seasons. Likewise, Pond 2 appears to have steeper slopes than the other B. houstonensis breeding ponds (Table 1). Yet it has demonstrated both reproduction and emergence with greater consistency than any other location on the GLR. On a closer examination, three of the four cardinal point measurements for this pond had very gradual slopes ([approximately equal to] 10[degrees]), while one measurement was taken at a very steep embankment ([approximately equal to] 25[degrees]). Within the pond, no Houston toad larvae or emergents were located at the steeply sloped area. However, even the steep bank recorded on the West slope varied dependent upon pool height of the pond. In another example, Pond 17 is an ephemeral pond, which was dry or nearly dry by the time breeding season for B. houstonensis began in 2000 and 2001. However, the pond held water continuously during the breeding seasons of 2002 and 2003.
The gross characterization of pond slopes at four cardinal points was not an adequate descriptive measure for which area within a pond was most likely to be used by Houston toads when breeding did occur. Despite the significant difference of South slopes among the positive and negative ponds, this was not predictive of where within those ponds egg strings might be deposited. This is important, as cohorts of Houston toad tadpoles remain within two meters of the site of egg strand deposition as they mature and emerge from the ponds. So while slopes do seem to show an effect on egg string placement within a pond, the slope differences among ponds are not enough to discriminate between breeding and non-breeding ponds.
The pH of ponds fluctuates on a seasonal and to a smaller degree, on a daily basis due to biotic and abiotic factors including thermal stratification, aquatic vegetative photosynthesis, and direct input from rain and nutrients carried into the ponds from the watershed. The total range of pH values determined during this study was nevertheless narrow. Bufo houstonensis tadpoles were found in ponds across this range on the GLR, therefore, pH does not seem to be a factor preventing the Houston toad from successfully breeding in any of the ponds. Further work will examine trends in slope for these ponds over time, characterize the remaining ponds on the Ranch and seek information regarding other factors which may influence Houston toad reproduction. These will include vegetation characteristics in the ponds, relative levels of dissolved oxygen, pond fish composition and densities, invertebrate predation, and interspecific anuran competition.
This study would indicate that ponds with a more gradual slope are indicative of aquatic habitat preferred by Houston toads for breeding. While it's not self evident at this point why slopes less than 10[degrees] appear to be important for successful B. houstonensis reproduction, further investigation may provide some understanding of this abiotic factor. In any case, this result corresponds to the conclusions reached by researchers regarding gentle slopes preferred by a different endangered toad species (Banks & Beebee 1987).
One aspect of the study that was not expected was the rarity of breeding compared with chorusing and similarly infrequent successful emergence when breeding had occurred. The variability of successful reproduction by year and by pond was also an unexpected result. It is apparent that this aspect of the reproductive biology of the Houston toad is a result that needs to be considered in future management and monitoring programs. Male chorusing is not an acceptable indicator of reproduction. It is speculated here that drought cycles effect breeding pond "cycling" as a consequence of pool levels and correlated effects on adult toad survivability in the upland habitat surrounding each pond. If borne out this would mean that breeding pond surveys would need to include drought and wet periods to accurately judge the distribution of breeding sites on a landscape scale. Assessment of successful reproduction rather than chorusing may provide a better measure of the status of Houston toad populations, while emergence studies would be even more descriptive.
It is apparent that fluctuations in pool height are important in considering effective slope for a pond. Artificial ponds constructed to provide breeding habitat for the Houston toad should take into account such fluctuations and plan bank slopes accordingly. As this study represents only 14 ponds, it is important to emphasize the necessity of including more known breeding and nonbreeding ponds within and outside of Bastrop County to evaluate the conclusions proposed herein. Likewise, defining the aforementioned biotic factors influencing reproductive success for the Houston toad are a logical extension of this study.
Table 1. Abiotic characteristics for 14 ponds on the Griffith League Ranch, Bastrop County, Texas correlated to data for Houston toad (Bufo houstonensis) chorusing and breeding across four years. Six of these ponds supported successful reproduction of the endangered Houston toad and all ponds are no farther than 1000m from one or more known Houston toad breeding ponds. Pond Degree Slope N S E W 1 8.53 14.03 9.09 19.29 2 9.09 * 8.53 * 14.5 25.17 4 5.71 9.09 6.84 7.97 5A 9.09 6.84 7.41 * 6.84 * 6 7.05 * 6.33 3.60 5.37 * 7 4.00 3.43 6.28 * 5.71 9 7.41 5.71 5.71 16.17 10 9.65 6.27 7.97 9.09 11 2.86 7.41 6.28 * 9.65 12 11.86 18.78 28.37 6.84 14 6.28 9.09 4.57 6.28 15 17.22 8.53 14.57 9.09 16 21.80 22.30 13.50 * 12.95 17 3.74 3.74 3.81 3.81 P <.05 .01-.02 <.05 <.05 Pond pH Houston Toad Chorusing at 2cm depth 2000 2001 2002 2003 1 6.67 no no no no 2 5.77 yes yes * yes * yes * 4 6.57 no no no no 5A 6.70 no yes * yes * yes * 6 5.33 yes yes * yes * yes 7 5.63 yes * yes yes yes 9 6.31 no yes yes yes 10 6.11 no yes yes yes 11 6.38 yes yes yes yes * 12 6.74 yes no yes yes 14 5.79 no yes yes no 15 5.98 no no no no 16 6.45 yes * yes no no 17 5.63 no no no no P * Indicates location of Houston toad egg strings or emergence.
We would like to thank the CAC-BSA for their continued support for Houston toad research and collaboration on projects taking place on the GLR. T. Swannack, K. Jones and P. Koepp assisted with field work, B. Lanier and A. Marquardt with laboratory procedures. We are grateful to the BSA volunteers, Texas State University students and Bastrop County citizens that assisted with this project. This research was directly supported by Federal and State Section 6 funding through the USFWS Region 2, TPWD and the CAC-BSA. We thank the personnel in Austin office of each agency for their help. This work was completed under permits TEO39544-0 and SPR-0102-191.
MRJF at: firstname.lastname@example.org
Baker, F. E. 1979. Soil survey of Bastrop County, Texas. U.S. Department of Agriculture, Soil Conservation Service. Washington D.C.
Banks, B. & T. J. C. Beebee. 1987. Factors influencing site choice by the pioneering amphibian Bufo calamita. Holarctic Ecology, 10:14-21.
Bardsley, L. & T. J. C. Beebee. 1998. Interspecific competition between Bufo larvae under conditions of community transition. Ecology, 79:1751-1759.
Beebee, T. J. C. 1985. Natterjack (Bufo calamita) tadpole behavior in captivity. British Herpetological Society, 13:15-18.
Beiswenger, R. E. 1977. Diel patterns of aggregation behavior in tadpoles of Bufo americanus, in relation to light and temperature. Ecology, 58:98-108.
Duarte, C. M. & J. Kalff. 1986. Littoral slope as a predictor of the maximum biomass of submerged macrophyte communities. Limnology and Oceanography, 31:1072-1080.
Dunson, W. & J. Travis. 1991. The role of abiotic factors in community organization. The American Naturalist, 138:1067-1091.
Griffiths, R. A., J. M. Getliff & V. J. Mylotte. 1988. Diel patterns of activity and vertical migration in tadpoles of the common toad, Bufo Bufo. Herpetological Journal, 1:223-226.
Hillis, D. M., A. M. Hillis & R. F. Martin. 1984. Reproductive ecology and hybridization of the endangered Houston toad (Bufo houstonensis). Journal of Herpetology, 18:56-72.
Randall, D. J., C. K. Minns, V. W. Cairns & J. E. Moore. 1996. The relationship between an index of fish production and submerged macrophytes and other habitat features at three littoral areas in the Great Lakes. Canadian Journal of Fisheries and Aquatic Sciences, 53:35-44.
Michael R. J. Forstner and Theo L. Ahlbrandt
Department of Biology
Texas State University-San Marcos
San Marcos, Texas, 78666