Grantee Research Project Results
2000 Progress Report: Environmental Factors That Influence Amphibian Community Structure and Health as Indicators of Ecosystems
EPA Grant Number: R825867Title: Environmental Factors That Influence Amphibian Community Structure and Health as Indicators of Ecosystems
Investigators: Beasley, Val , Richards, Carl , Johnson, Lucinda , Piwoni, Marvin , Schoff, Pat , Cole, Rebecca
Current Investigators: Beasley, Val , Richards, Carl , Schotthoefer, Anna , Lieske, Camilla , Johnson, Catherine , Murphy, Joe , Johnson, Lucinda , Piwoni, Marvin , Schoff, Pat , Cole, Rebecca
Institution: University of Illinois Urbana-Champaign , Illinois Waste Management and Research Center , United States Geological Survey , University of Minnesota - Duluth
EPA Project Officer: Packard, Benjamin H
Project Period: June 1, 1998 through May 31, 2001 (Extended to September 30, 2002)
Project Period Covered by this Report: June 1, 1999 through May 31, 2000
Project Amount: $1,299,991
RFA: Ecosystem Indicators (1997) RFA Text | Recipients Lists
Research Category: Aquatic Ecosystems , Ecological Indicators/Assessment/Restoration
Objective:
The overall goal of this research is to assess the relative influence of landscape patterns, biotic interactions, water quality, and contaminants on the health and community structure of amphibians. The objectives are to evaluate: (1) the relative influence of wetland- and watershed-scale factors (i.e., landscape and ecological data) on amphibian community structure and health; and (2) whether amphibian community structure and health are indicative of ecological integrity.
Progress Summary:
Spatial and Ecological Assessments. We assessed landscape
fragmentation patterns, connectivity of habitat patches, and other landscape
metrics for our 64 Tier 1 (T1) wetlands in Illinois, Wisconsin, and Minnesota.
Summaries of transportation, geology, and wetland coverage data were used with
site-specific biological and physical data to develop species-specific habitat
models (Johnson, et al., in press), and to assess relationships between
amphibian communities and environmental factors at various scales. Collection
for spatial and ecological data for the 36 Tier 2-Tier 3 (T2-3) wetlands in
Minnesota was completed, and land cover data were analyzed to develop local
landscape metrics. Aerial photos for 1 km buffers around each site were used to
describe fine-scale land cover, types and locations of structures, and other
forms of human disturbance. Soil samples collected in 1999 were analyzed for
phosphorus, organic matter, and texture. Specimens for semiquantitative
macroinvertebrate characterization at T2-3 sites during 1999 were processed in
the laboratory, and specimens were identified to genus where possible. In 2000,
macroinvertebrate surveys were again performed at these sites and individuals
were identified to the lowest possible taxon in the field. Data on amphibian
larvae and egg masses encountered during dipnetting surveys also were recorded.
Amphibian Calling Surveys and Malformation Assessments. Calling
surveys were completed for the 36 T2-3 sites during April, May-June, and July
2000. Because of year-to-year variability of amphibian occurrences in wetlands,
calling surveys also were conducted at 64 T1 sites. Over the 3 years, we
identified 11 anuran species in our 80 study sites (T1 and T2-3 sites). The
number of species apparently breeding (based on calling surveys) in each wetland
ranged from 1 to 9, with most wetlands supporting 4 to 6 species. Chorus frogs
(Pseudacris triseriata) were most common, occurring in 70 sites at least
1 year of the study, while pickerel frogs (Rana palustris) were heard
only one time. Based on calling surveys, species turnover was high at most
sites. However, the spring peeper (Pseudacris crucifer) was the most
consistent species across the 3 years, and was heard calling during all 3 years
at 19 sites.
Anuran malformation surveys were conducted in August 2000, following the same protocol used in 1999. Metamorphs collected from each wetland were sent to the University of Illinois and the National Wildlife Health Center (NWHC) for pathology and parasitology studies. Additional metamorphs captured at each site were measured, examined for malformations in the field, and released (target n = 100 total). Based on initial, external examinations, we identified 17 metamorphs with malformations in 2000, comprising 1.47 percent of the total number of frogs captured. Over the 3-year period of this study, 3,188 frogs (primarily leopard frogs, Rana pipiens) were caught and examined for malformations. Field examinations indicated an overall malformation rate of 1.19 percent. This number may change because laboratory analyses may indicate that some external abnormalities are caused by disease or predation, and other individuals that appear to be normal may have internal abnormalities. Malformation rates were not consistent within a particular site across years, and no site had consistently high malformation rates.
Skeletal Analysis. Frogs (254) collected from Tier 2-3 sites in 1999 were cleared and stained to reveal bone and cartilage, and six showed skeletal malformations, including hind leg, foot, pelvis, and/or lower jaw defects. Of 229 R. pipiens examined, 4 exhibited malformations, including missing metatarsals and phalanges, spinal torsion and coccyx malformation, and lower jaw malformation. Other Ranid species, collected much less frequently, added disproportionately to the total malformation percentage. Only 8 Rana clamitans and 16 R. septentrionalis were collected, but one of the R. clamitans had a slightly reduced hind foot, and one R. septentrionalis had malformed vertebrae and was missing a phalanx. The total malformation rate was 2.4 percent. Variations in structure of the coccyx and the articulation between the sacral vertebrae and coccyx were common (13 of 254 specimens); however, these were judged to be variations of normal bone position and growth and did not produce torsion or mal-position of the pelvis or legs.
Pathology Studies 1999 Frogs. Preliminary histologic evaluation of the 1999 tissues from Tier 2-3 sites has been completed. Data currently are being checked and diagnoses confirmed. The most commonly identified histologic finding was hepatitis, which was identified in at least one frog (and usually nearly all frogs) at sites from which at least five frogs were evaluated. The intensity of the hepatitis differed across sites. It was most often characterized by mononuclear cell infiltrates consistent with a chronic infection. The cause of the hepatitis probably differs across sites. Intracytoplasmic inclusion bodies consistent with viral infection were identified at 18 sites (across all watersheds), and hepatitis also was associated with parasitism and micro-granulomas. Most gonadal nodules noted at necropsy last season consisted of immature gonadal tissue. However, 15 individuals from 8 sites had intersex gonads: primarily testes with ova present (ovotestes). The incidence of intersex in frogs from all sites was 2 percent. At sites with at least one intersex individual, the incidence of intersex metamorphs was 3 percent to 16 percent.
Pathology Studies 2000 Frogs. In 2000, methods for gross pathology studies and collections of tissues for histopathology were essentially the same as in 1999. Frogs from 24 sites, and tadpoles from 2 of these sites, were maintained at a cool temperature and transported to the University of Illinois until examination. Sites with high and low capture numbers as well as no capture were represented in all the watersheds. At 10 of the 36 sites, the capture target of frogs for pathology studies was met in 2000. There was a severe drought in much of the study area in 2000; 6 of the 36 study sites became dry; and no frogs were captured at 12 sites. Again in 2000, the vast majority of frogs captured were the target species R. pipiens. Overall, 44 percent of the frogs captured were female, 50 percent were male, and 6 percent were undetermined (most had immature gonads; and sex determination awaits histologic assessment). This is similar to the sex distribution in 1999 (female 49, male 51). A comparison of R. pipiens weights for each site from 1999 and 2000 revealed that the frogs weighed an average of 45 percent less in 2000; while differences in weights of frogs from individual sites in 2000 frogs ranged from 68 percent less to 8 percent more than 1999 frogs. The average snout to vent lengths at sites also decreased this season, and R. pipiens in 2000 were 12 percent shorter than at the same sites in 1999. The reductions in length and mass are probably related to accelerated metamorphosis to accommodate the drought.
Malformations or deformations were identified in 11 frogs from 5 sites. Of these, four had apparent malformations, including one with a third hind limb, one with hemimely of the right hind limb, and two with shortened right hind limbs. Six had lesions that could be either malformation or deformation, including two frogs with ectrodactyly of the right hind phalanges, and four with anophthalmia. Another frog had partial loss of a hind limb with external fibrosis consistent with traumatic loss. Except for one R. septentrionalis, frogs with malformations or deformations were R. pipiens.
At least one lesion was identified in 63 percent of the 391 R. pipiens that underwent full necropsy. Of frogs with at least one lesion, 25 percent were in poor body condition. The most common lesions included inflammation (45 percent), parasitism (36 percent), and trauma (3 percent). Inflammation was the most common lesion identified for the kidney and liver. Parasites were commonly identified in the body cavity, gastrointestinal tract, musculoskeletal system, and skin. Enlargement was the primary abnormality noted in the spleen. Gonadal nodules were sometimes noted in the reproductive system. As in 1999, little homogeneity of lesion distribution was noted within watersheds.
Parasitology Studies 1999 Frogs. Metamorphs (Gosner stages 43-46) collected for parasitology studies in 1999 from 26 of the 36 Tier 2-3 sites have been thoroughly examined for parasites. Muscle tissues were viewed at the time of necropsy, and closer inspection for parasites took place after specimens were cleared and stained as previously described (Hanken and Wassersug, 1981); Sessions, et al., 1990). Blood films also were collected from each frog; and 80 percent have been scanned for parasites. In total, 21 parasite taxa were found across all sites in 1999. Nine were represented by larval trematodes (echinostomes, R. septentrionalis, Alaria sp., Fibricola sp., Apharynogstrigea pipientis, Clinostomum sp., an unidentified strigeid, and two unidentified); four were adult trematodes that mature in frogs (Haematoloechus sp., Megalodiscus temperatus, Cephalogonimus sp., and gorgoderid); and five were nematodes (Oswaldocruzia pipiens, Falcaustra sp., Rhabdias ranae, Cosmocercoides dukae, and Spiroxyx contortus). A larval cestode, leech, and mites also were found. Echinostomes in the kidneys were the most commonly encountered parasites, with prevalences from 40 to 100 percent and mean intensities from 8 to 321 metacercariae/frog. R. septentrionalis was found at 4 sites at intensities up to 88 metacercariae per frog and prevalences between 6 and 100 percent. However, the occurrence of this parasite was not associated with any malformations.
Parasitology in 2000 Frogs. Metamorphs and a small number of tadpoles were collected for parasitology studies from 17 of the 36 Tier 2-3 sites. All frogs collected were R. pipiens except for seven R. clamitans. Data were gathered on malformations, body weight, snout-to-vent length, snout-to-tail length, liver, kidney, and fat body weights, as well as weight of carcass after removal of visceral organs, and examination of all tissues for parasites. Blood films were collected. Parasites were preserved and will be identified to the lowest possible taxonomic group.
Malformations or deformations were observed in 3.2 percent of the frogs examined in 2000, and these were found at 6 of the 17 sites from which frogs were collected. Of these, one was apparently due to trauma; one was missing a limb, one had premature termination of the forelimb, and one had phocomelia (abnormal foot attached to limb of disorganized, unrecognizable bones) of a hind limb. Two frogs from two different sites had a bony protrusion from the skull behind the eye. One frog was bilaterally asymmetrical, with the right side of the body smaller than the left, and was missing the lens of the right eye. In addition, one frog had an asymmetrical pelvis. R. septentrionalis was observed again in 2000, and again there was no association with malformations. It occurred only at two sites with prevalences of 8.7 percent and 25 percent, and a mean intensity of two metacercariae/frog.
Parasitology in 2000 Snails. Snails were collected from Tier 2-3 sites in May and August 2000. Water containing the snails was examined after 24-36 hours for cercariae (released larval trematodes). A key was used to identify cercariae (Schell, 1985). For more common snail species, 30 percent of the individuals not shedding cercariae were killed and examined for trematode infection. Other snails were fixed in 80 percent ethanol. Nine species of freshwater gastropods, comprising three families: Planorbidae, Physidae, and Lymnaeidae, were found across all sites. In addition, representatives of the families Sphaeriidae (freshwater bivalves) and Succinidae (amphibious gastropods) were detected. The most widely distributed snail species were Helisoma trivolvis, Physa gyrina, Stagnicola exilis, and Gyralus parvus. Infections with echinostomes were the most frequently encountered in snails of H. trivolvis, P. gyrina, and S. exilis. At some sites, nearly all of the H. trivolvis examined were infected with an echinostome. Other commonly observed larval trematode infections were of the Brevifurcate-apharyngate, Strigea and Armatate types (Schell, 1984).
Water Quality Assessments. The deployment scheme for water quality instrumentation was greatly expanded for 2000 to compensate for problems in 1999 due to defective instrumentation. This allowed us to deploy 38 instruments in summer 2000 to monitor dissolved oxygen (DO) in mg/L, temperature (?C), and pH at hourly intervals. There was a severe drought in summer 2000 in the study area and a modified deployment method (each probe suspended from a float) enabled data acquisition during larval development (from May 24 to June 24). Graphs of DO, temperature, and pH were produced for each pond. Means for DO, temperature and pH for each pond, the pond's association with agriculture (farm pond or not farm pond), rankings for pond size as small or large, and rankings for two dominant aquatic macrophytes, coontail (Ceratophyllum demersum) and duckweed (Lemna sp.), as absent, or low, medium or high amounts were evaluated for a relationship to leopard frog absence or presence (catch or no catch) using stepwise logistic regression. The analysis found that only DO (p<0.01) and coontail (p<0.05) were significantly associated with numbers of frogs captured. In a further test of interactive terms using a Poisson distribution, DO and coontail were significantly (p<0.001) associated with frog capture. Interpolated contour plots were produced depicting the interactions between DO and temperature as well as DO and pH for all ponds in the analysis. The optimal mean DO value was approximately 8-14 mg/L, the optimal temperature value was around 12?C, and although the logistic regression analysis did not identify pH as a significant term in the model, it suggested that a pH around 9.0 was very well tolerated. Approximately 33 percent of the ponds analyzed had a mean pH near 8.0. The interpolation analysis suggests that an abundance of aquatic plants may have contributed to greater frog catch. In contrast to coontail, duckweed (also abundant at many of the ponds) was not significantly associated with frog catch at the ponds. Unlike coontail, dense growths of duckweed reduce sunlight to plants in the water column and DO often declines.
Analytical Toxicology. For 2000, a refined analyte list was developed based on findings last summer, information on chemical use, ecotoxicology data, and environmental fate information gathered by project staff. Changes to the sampling approach included sediment sampling only during the June collection period of the 2000 season, and deletion of some organic compounds from the analyte list during the July period because they were so unlikely to be detected. Sediment collection procedures were developed to insure that the first 8-11cm of pond sediment were recovered. Fast delivery of samples to the laboratory was implemented to address concerns with nitrite stability. Water and sediment samples were obtained at each pond over a 2-week period beginning in June and water samples over a 1-week period beginning in July. Several ponds dried precluding water sampling. Only selected glyphosate and diquat/paraquat samples were taken in the July collection period. Also, for 2000, we increased the number of field spikes. Many of the analytes targeted for this study were absent from or present only at very low levels in study ponds in 1999. Field spiking of selected samples verified that collection, handling, and analytical procedures were preserving these analytes in the samples.
For anion analysis, samples were filtered in the field immediately, and filtrates in vials were immediately iced down, then shipped on ice overnight at 2-day intervals to our laboratory. The samples were analyzed for sulfate, chloride, phosphate, nitrate, and nitrite by ion chromatography the morning they were received. All anion analyses are completed. There were differences in common anion composition of the ponds in the St. Cloud area as compared to ponds near Minneapolis. The highest nitrate levels were observed at two ponds, with one having 24 and 13 mg/L in June and July, and the other having 16 mg/L in July. The elevated concentrations likely were associated with rainfall events.
Methods for metals analysis were similar to those used for 1999 samples but were more quantitative. Pond waters were digested and analyzed beginning upon delivery of samples in June and have been completed. Metals analysis on 2000 sediment samples is complete, and data are in review. Work on tissue samples is continuing. Aquatic metal data are being evaluated. Zinc concentrations were elevated in several ponds. Certain of these ponds also had comparatively high levels of nickel and cobalt.
Methods for gas chromatographic mass spectrometry (GCMS) analyses were similar to those described for 1999, except, to improve recoveries, extractions were changed to liquid/liquid methods instead of extraction disks. Analysis of pond water extracts for base neutral compounds by GC/MS is completed and data are being review. Sediment samples have been dried, ground, and homogenized, and extraction methods have been evaluated and defined. Thus far, the only notable organic contaminants observed in the limited number of samples assayed were tentatively identified as atrazine and atrazine breakdown products. Analysis for organic contaminants is continuing.
For 2000, the analytical methods used for glyphosate, carbamate pesticides, diquat, and paraquat were unchanged from those employed for the 1999 samples. Ten n-methylcarbamoyloxime and n-methylcarbamate pesticides were monitored. Glyphosate and diquat and paraquat were not found in 1999 samples. Except for a few selected samples, these analytes were dropped for the July 2000 collection period samples. No glyphosate was detected in any of the field samples. In nearly all of the samples, paraquat and diquat were below method detection limits. All analyses of pond water samples for these analytes have been completed for both 1999 and 2000 specimens.
Quality assurance (QA) results to date generally demonstrate the usefulness of the field spiking procedures. In particular, for nitrite, the field spiking results validated the sample handling procedures employed in the study. Recoveries for field-spiked samples and blanks for nitrite were near 100 percent. Complete QA results will be included in the final report.
References:
Hanken J, Wassersug RJ. The visible skeleton. Functional Photography 1981;16:22-26.
McGarigal K, Marks BJ. FRAGSTATS: spatial pattern analysis program for quantifying landscape structure. General Technical Report PNW-GTR-351. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station, 1995, 122 pp.
Schell SC. Handbook of Trematodes of North America North of Mexico. University Press of Idaho, Moscow, ID, 1985, pp. 10-17.
Sessions SK, Ruth S. Explanation of naturally occurring supernumerary limbs in amphibians. Journal of Experimental Zoology 1990;254:38-47.
Future Activities:
Finer scale landscape metrics within concentric buffers (within 100, 200, 500, and 1000 m of each site) will be developed. Landscape metrics, soil assay data, quantitative vegetation and macroinvertebrate surveys, and findings of water quality assessments will be analyzed with more general ecological data collected annually at each site (e.g., surrounding land use, wetland type, hydrologic regime, general water quality, and other aquatic fauna). Also, ongoing efforts will focus on completing histologic assessment of 1999 frogs, and histological preparation and examinations of 2000 frogs. When completed, data will be analyzed and compared with spatial, ecological, water quality, parasitologic, and contaminant findings. Our focus during the next year in the parasitology area will be on completing the data assemblage and describing relationships between parasite intensity and taxa richness with ecological conditions. Analysis of extractable organics from pond waters is continuing, as is sediment analysis for metals. Organic analyses of sediments and tissues and metals analysis of tissues has begun, with completion expected in late spring/early summer 2001. In the coming months, we will complete our analyses and will identify indicators of wetland ecosystem integrity based on amphibian communities and amphibian health parameters.
Journal Articles:
No journal articles submitted with this report: View all 47 publications for this projectSupplemental Keywords:
water, watersheds, sediments, exposure, ecological effects, animal, sex, chemicals, toxics, heavy metals, organics, pathogens, ecosystem, indicators, aquatic, habitat, integrated assessment, ecology, epidemiology, pathology, remote sensing, analytical, LANDSAT, satellite, modeling, Midwest., RFA, Scientific Discipline, Ecosystem Protection/Environmental Exposure & Risk, Ecology, Ecosystem/Assessment/Indicators, exploratory research environmental biology, Ecological Effects - Environmental Exposure & Risk, Microbiology, Ecological Risk Assessment, Biology, Ecological Indicators, ecological exposure, pesticide exposure, aquatic biota , landscape indicator, watersheds, amphibians, ecosystem integrity, parasitic infection, multiple spatial scales, biotic integrity, contaminant impact, water qualityRelevant Websites:
http://www.nrri.umn.edu/cwe/hostedpages/amphweb.html
Progress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.