Patuxent Wildlife Research Center

NAAMP III Archive - calling surveys
Home | Archive by Alphabetical Order | Archive by Category

Anurans as Indicators of Wetland Condition in the Prairie Pothole Region of North Dakota

Dorothy G. Bowers¹, David E. Andersen^1,2, and Ned H. Euliss, Jr.³

¹Department of Fisheries and Wildlife
University of Minnesota, St. Paul, MN 55108

²U.S. Geological Survey-Biological Resources Division
Minnesota Cooperative Fish and Wildlife Research Unit, Department of Fisheries and Wildlife
University of Minnesota, St. Paul, MN 55108
email:dea@fw.umn.edu

³U.S. Geological Survey-Biological Resources Division
Northern Prairie Science Center
Jamestown, ND 58401
[ Abstract ]

Introduction
In 1995, we initiated a two-year pilot Environmental Monitoring and Assessment Program (EMAP) study in the North Dakota Prairie Pothole Region (PPR) to address the following goals:

1) describe anuran relative abundance, species composition, and species distribution in selected study sites within the PPR of North Dakota;

2) determine a suitable road survey method(s) for estimating anuran relative abundance and describing anuran distribution; and

3) determine if relationships exist between land-use patterns and anuran distribution and relative abundance.

Standardized survey methods for anurans in the PPR do not exist, and applicability of survey methodology from dissimilar landscapes (e.g., eastern and northwestern U.S.) was not clear. Within the constraints of the EMAP, we modified existing anuran survey methodology for application in North Dakota, and present a description of those methods and a summary of survey results from 1995 and 1996.

Study Area:
The North Dakota PPR was divided into three ecoregions extending roughly north to south based on physiography (Figure 1. Moving from east to west across the state, these include the Red River Valley, the Drift Plain, and the Missouri Coteau. Nine species of anurans inhabit the North Dakota PPR and occur in at least one of the selected study sites. These species include the plains spadefoot toad (Spea bombifrons), American toad (Bufo americanus), Great Plains toad (B. cognatus), Canadian toad (B. hemiophrys), Woodhouse's toad (B. woodhousii), western chorus frog (Pseudacris triseriata), northern leopard frog (Rana pipiens), wood frog (R. sylvatica), and gray treefrog (Hyla versicolor/chrysoscelis).

Methods:
The EMAP sampling framework consists of a triangular grid randomly placed over the conterminous United States, with 40-km² hexagons surrounding each grid point. We selected a minimum of six hexagons/ecoregion from this framework as study sites for anuran surveys (Fig.1).

We visited each hexagon three times during the anuarn calling season. At each hexagon, four types of surveys were conducted: road calling surveys, wetland calling surveys, night driving surveys, and automated recording surveys (Fig. 2).

Anuran Calling Surveys
Procedures for wetland and road calling surveys were identical, the methods varying only by site selection. Within each hexagon, we randomly chose six wetland survey points along rights-of-way within 50 m of seasonal and semi permanent wetlands. Three of these survey points were located adjacent to wetlands surrounded by cropland and three adjacent to wetlands surrounded by grassland. Grasslands included uplands composed of native grasslands or deciduous woodlands, Conservation Reserve Program lands, hayland, and pastures and croplands included upland crops such as wheat, barley, field corn, soy beans, sunflowers, and sugar beets. In addition, ten survey points were established at 0.8 km (0.5 mile) intervals along an 8 km (5 mile) survey route within each hexagon. The survey route was a randomly located segment of a pre established Breeding Bird Survey route.

We performed ten-minute calling surveys at each survey point between dusk and 0300 the next morning. We identified calling anurans to species, and recorded both estimated number of individuals and call index value (CIV). CIVs were adopted from Wisconsin Department of Natural Resources survey procedures and ranged from one to three (Scott and Woodward, 1994). We also noted weather conditions and land-use patterns over an estimated 0.4 km (0.25 mile) radius in each of four quadrants (NW, NE, SE, SW) surrounding wetland and road survey points

Night Driving Surveys
Night driving surveys included recording sightings or aural detections of anurans in the hexagons at times other than during calling surveys. We noted collection time, location, land-use pattern, age class, sex, and species. Additionally, in 1996 45 hexagons were visited in August or September and night driving surveys were conducted on the 8 km (5 mile) road calling survey route. We recorded species, number, location along the survey route, and size class for each anuran observation. In addition, we recorded the duration of the survey and weather conditions at the start and end of each survey.

Automated Recording Systems
We assembled programmable analog recorders using a modification of Peterson and Dorcas' (1994) automated recorder design. We collected recorder data with paired wetland recordings (PWR) and method variance recordings (MVR). Paired wetland recordings were sampled at thirteen pairs of seasonal wetlands in 1995 and fourteen pairs in 1996. Each pair was of similar size, with one wetland located in cropland and one located in grassland. Method variance recordings were sampled at points midway between road calling survey points and wetland calling survey points at sixteen sites in 1995 and 21 sites in 1996. In 1995, we performed automated recording surveys only during the last visit to hexagons. We evaluated species presence/absence from both the PWR and MVR recordings.

Results and Discussion
Surveys were conducted between 5 May and 12 July 1995 and 18 April and 23 June 1996. Our results at this stage are based primarily on 1995 data and are preliminary; they are presented primarily as descriptive summaries.

Range Expansions/Contractions
Ranges of Spea bombifrons, Rana pipiens, Bufo hemiophrys, and B. americanus varied from those previously described by Wheeler and Wheeler (1966) and Hoberg and Gause (1992). We located S. bombifrons in hexagons outside the species' previously recorded range. This extension of S. bombifrons' range could be a result of either range expansion or incomplete past delineation of breeding range; it is most likely the result of incomplete records for this species. The ranges of R. pipiens, B. hemiophrys, and B. americanus were less extensive than previously described. Failure to detect these species in hexagons within their previously described ranges suggests either inadequate sampling techniques or decreased ranges. Data from 1996 surveys should clarify this issue.

Calling Activity
Our results from calling surveys indicate that Bufo americanus, B. cognatus, B. woodhousii, Hyla versicolor/chrysoscelis, Pseudacris triseriata, Rana pipiens, and R. sylvatica called between 2130 and 0059, with some species beginning earlier and some continuing later. Species such as B. cognatus, B. woodhousii, and H. versicolor/chrysoscelis that initiate calling late in the season began nightly calling at later hours than species such as B. americanus and P. triseriata, which initiate calling early in the season. In the early breeding season when night temperatures fell below 8 °C, P. triseriata, R. pipiens, and R. sylvatica were also actively calling during the day. Calling season varied among Pseudacris triseriata, Rana pipiens, Rana sylvatica, B. americanus, B. cognatus, B. woodhousii, and Hyla versicolor/chrysoscelis ((Figure 3). Pseudacris triseriata, R. pipiens, and R. sylvatica were all heard calling prior to the first survey in 1995. Rana sylvatica was not heard after 10 May in 1995. In 1996, R. sylvatica was heard between 23 April and 2 May, and then again on 24 May. Pseudacris triseriata was heard throughout the survey period in 1995 and 1996. However, peak calling for P. triseriata was completed by mid June in 1995. Rana pipiens was only recorded on six nights in 1995. The calling season of B. americanus lasted approximately two weeks, from 10 May to 23 May, in 1995 and only five days, from 25 to 30 May, in 1996. Bufo cognatus and B. woodhousii calling seasons extended for three weeks from 18 May through 8 July in 1995, but in 1996, B. cognatus was heard from 24 May to 16 June and B. woodhousii was heard over a 2 week period from 3-15 June. We first detected H.versicolor/chrysoscelis in the third week of June and recorded calls through the end of the survey period in 1995, and in 1996 we detected H. versicolor/chrysocelis from 30 May to 19 June. In 1995, associations between temperature and humidity and mean estimated number of calling individuals/survey point varied among species. Overall estimated number of calling anurans/survey point was highest between 5 and 25 °C, but anurans called when temperatures ranged from <5 to >25 °C. Bufo americanus, Pseudacris triseriata, Rana pipiens, and R. sylvatica called most actively at temperatures <10 °C, B. cognatus and B. woodhousii at temperatures ranging from 5-25 °C, and Hyla versicolor/chrysoscelis at temperatures between 15 and 25 °C. Bufo cognatus, B. woodhousii, and P. triseriata called actively across the widest range of temperatures. Calling anurans were detected over a humidity range of <45 to 95%. Mean estimated number/survey point of B. americanus, B. woodhousii, and R. pipiens was greatest at humidity less than 65%. I and P. triseriata were detected across the full range of humidity, H. versicolor/chrysoscelis from 66-85% humidity, and R. sylvatica from 56-90% humidity.

Influence of Weather on Calling Activity High wind seems to affect calling activity and call detection more than low temperatures (5-8 °C) and rainfall. We observed several full choruses of Pseudacris triseriata and Bufo americanus on calm nights when temperatures were below 8 °C and choruses of B. cognatus when temperatures were below 15 °C. Thus, early season surveys can be conducted at temperatures at or below 5 °C if wind and rainfall do not interfere with audition. It is important to complete late season surveys prior to the onset of summer weather conditions. During periods of warm and dry weather, we recorded few calling anurans. Our preliminary analysis indicates that anurans in the North Dakota PPR call most actively when temperatures are below 25 °C and humidity ranges from <45-85%. During the third visit in 1995, even when humidity fell in these ranges, we recorded few anurans. In the North Dakota PPR, we recommend that calling surveys be completed prior to June 30, although breeding seasons may extend later depending on annual weather conditions.

Survey Techniques
Our primary survey emphasis was placed on wetland and road calling surveys. We detected eight of nine anuran species present in the PPR of North Dakota from calling surveys. Only Bufo hemiophrys was not recorded using these techniques. We observed all nine species during night driving surveys; however, the frequency of observations during night driving surveys was less than that of calling surveys. Data gathered in 1995 indicate that a combination of calling surveys and night driving surveys provides a more complete description of species composition and distribution within hexagons than either technique used alone. Calling surveys are unreliable for detecting explosive breeding species but provide indications of relative abundance, whereas night driving surveys are successful in detecting all species but provide data insufficient to evaluate relative abundance.

Road Calling Surveys
Conducting road calling surveys at 0.8 km (0.5 mile) intervals and recording data for each quadrant worked well in North Dakota where roads follow a north-south, east-west grid system. Roads form convenient boundaries for changes in land use; typically, each quadrant is composed of either cropland or grassland. When this is not the case, there is usually a clear, linear division between land-use categories. Assessing land-use pattern is relatively simple in this landscape. However, in many other regions of the United States, roads do not follow grid systems, and more attention is required to classify land use. A similar survey technique used by the Ontario Task Force on Declining Amphibian Populations may be better suited for these areas. This system indicates habitat types and both distance from survey point and location of calling species (Gartshore et al., 1995). However, with this system and in any area with structures blocking the field of view, detailed habitat data will be difficult to acquire without entering private land.

Automated Recording Surveys
While recording surveys are useful for sampling periodically throughout the night and detecting species that call irregularly, the amount of information that can be obtained from them is limited. Wind and heavy rainfall decrease recording quality and reduce detection distances. CIV and estimated number of calling anurans is difficult to determine when more than five individuals are calling at one time. Recordings are best used for presence/absence data only.

Overall species richness within hexagons as measured by calling and night driving surveys ranged from one to six. Species richness in hexagons ranged from three to six in the Red River Valley ecoregion, from one to five in the Drift Plain ecoregion, and from zero to three in the Missouri Coteau ecoregion. These data indicate that the most agriculturally impacted ecoregion sampled, the Red River Valley, supports the highest species richness of anurans. This may reflect physiographic boundaries or anuran associations with soil type or climatic conditions. It is also possible that agricultural practices may increase habitat suitability for anurans. Analyses of relative abundance data are incomplete, and may reveal a different trend.

Potential indicator species
Ideally, biological indicators should be sufficiently sensitive to provide an early warning of change, be distributed over a broad geographic area, and provide continuous assessment over a wide range of stress (Noss, 1990). Based on the latter two criteria, Pseudacris triseriata is the best potential anuran indicator in the North Dakota PPR because it is the most ubiquitous species and its calling activity extends over the longest interval. However, P. triseriata distribution does not appear to correlate with land-use pattern. Our preliminary analysis of 1995 relative abundance data suggests that three of the seven species detected from road calling surveys may associate with land-use pattern in the PPR; Bufo americanus and Hyla versicolor/chrysoscelis appear to associate with grassland areas, while B. cognatus seems to associate with croplands. Relative abundance of Pseudacris triseriata, B. woodhousii, Rana pipiens, and R. sylvatica appear unrelated to upland landscape pattern. The two species that may associate with grasslands, B. americanus and H. versicolor/chrysoscelis, both have restricted ranges in the North Dakota PPR and short calling seasons, and thus would not be suitable as regional indicators. However, B. cognatus is widely distributed and has a relatively long calling season, and thus may prove to be a suitable indicator of agricultural landscapes.

Acknowledgments
This research was cosponsored by the U.S. Environmental Protection Agency through Interagency Agreement Reference Number DW14935541-01 and the U.S. National Biological Service (formerly the U.S. Fish and Wildlife Service) Reference Number 14-48-0009-92-1929. All work followed protocols approved by the University of Minnesota Animal Care and Use Committee.

Literature Cited
Gartshore, M. E., M. J. Oldham, R. van der Ham, F. W. Scheuler, C. A. Bishop, and G. C. Barrett. 1995. Amphibian road call counts participants manual. Ontario Task Force on Declining Amphibian Populations, Burlington, Ontario. 27 pp.

Hoberg, T. and C. Gause. 1992. Reptiles and amphibians of North Dakota. North Dakota Outdoors 55:7-18.

Noss, R. F. 1990. Indicators for monitoring biodiversity: a hierarchical approach. Conserv. Biol. 4:355-364.

Peterson, C. R. and M. E. Dorcas. 1994. Automated data acquisition. Pp.47-57. In Heyer, W. R., M. A. Donnelly, R. W. McDiarmid, L. C. Hayek, and M. S. Foster (eds.). Measuring and monitoring biological diversity. Standard methods for amphibians. Smithsonian Institution, Washington, D.C.

Scott, N. J., Jr. and B. D. Woodward. 1994. Surveys at breeding sites. Pp. 118-125. In Heyer, W. R., M. A. Donnelly, R. W. McDiarmid, L. C. Hayek, and M. S. Foster (eds.). Measuring and monitoring biological diversity. Standard methods for amphibians. Smithsonian Institution, Washington, D.C.

Wheeler, G. C. and J. Wheeler. 1966. The amphibians and reptiles of North Dakota. The University of North Dakota Press, Grand Forks. 104 pp.

U.S. Department of the Interior
U.S. Geological Survey
Patuxent Wildlife Research Center
Laurel, MD, USA 20708-4038
http://www.pwrc.usgs.gov/naamp3/naamp3.html
Contact: Sam Droege, email: Sam_Droege@usgs.gov
Last Modified: June 2002