Amphibian Diversity in Central Vietnam and the Distribution Pattern and Natural History of Granular Spiny Frogs Quasipaa verrucospinosa (Anura: Dicroglossidae)

Abstract

Tropical forests have rich biodiversity including many amphibian species However amphibian declines are a global phenomenon that is occurring at an alarming speed Habitat fragmentation or loss global climate change over-exploitation and disease are the main causes that have led to these declines Thus measuring biodiversity and identifying the rarity and commonness of species play essential roles in both community ecology and conservation However the biology and ecology of most native amphibians in central Vietnam are largely unknown and information on amphibian species composition diversity and distribution is lacking This includes the granular spiny frog Quasipaa verrucospinosa (Bourret 1937) which is already classified as “Near Threatened” due to loss of forest and stream habitats environmental degradation and human exploitation This dissertation investigates species composition and the number of frogs of each amphibian species in primary and secondary forests Based on these data I estimated species diversity (i e species richness heterogeneity and evenness) as well as rarefaction curves and determined the lognormal distribution of amphibian species I also estimated site occupancy rates and the distribution of granular spiny frogs and described certain aspects of the population biology and life history of this species These include the dietary composition spatiotemporal and size-dependent variation seasonal patterns of reproductive biology acoustic features and variation of male calls and patterns of tadpole metamorphosis The results of this study indicate that the overall similarity of species composition of amphibians between the two regions (0 86) within a landscape scale was considerably higher than that between primary and secondary forests (0 03) I caught 13 529 individuals of 101 species in 33 genera eight families two orders (Anura and Gymnophiona) The amphibian community in secondary forests contained a number of species characteristic of primary rain forest habitat (28 7%) The species richness and relative abundance were significantly higher in primary forests than in secondary forests (P <0 001) The expected number of amphibian species in primary forests was significantly richer than in secondary forests The expected total number of species theoretically available for observation in the entire community (including the rare species not yet collected) indicated that the “universe” comprises approximately 106 species with the number of species predicted to fall below the veil line about 4 48 species Based on the detection/non-detection data for each site over multiple visits for granular spiny frogs from the best model among all candidate models I estimated a site occupancy rate of 0 632 that was higher than the na?ve occupancy estimate of 0 403 and a 57% increase over the rate of sites at which frogs were actually observed The site variable of primary forest was an important determinant of site occupancy whereas occupancy was not associated with the variable of secondary forest The species detection model p(temperature humidity rainfall) included 90 9% of the total weight providing clear evidence that environmental conditions were important sample covariates in modeling detection probabilities Dietary analysis of granular spiny frogs identified 2645 prey items from 27 orders and nine classes Prey of granular spiny frogs comprised mainly invertebrates but also included fishes frogs and conspecific sub-adults The major prey items as determined by frequency of occurrence item count and percent volume were spiders beetles crabs hymenopterans grasshoppers crickets and cicadas Insects alone accounted for an importance value of 59 8% The mean monthly prey volume consumed was positively and negatively correlated to temperature and rainfall respectively The volume of prey consumed was positively correlated with snout-vent length and mouth width of granular spiny frogs supporting the gape limitation hypothesis Male frogs reached sexual maturity at 98 36 mm snout-vent length (SVL) and an age of nine months Females reached sexual maturity at 109 45 mm SVL and an age of 11 months Histological analyses revealed that sperm were not present throughout the year indicating discontinuous reproductive activity for adult males Reproductive females were observed only during months of the main rainy season (September to December) indicating that females reproduce seasonally The gonads of females began to increase in volume in July and maximal egg production occurred during the main rainy season Clutch size was correlated with female SVL and varied between seasons and months Multiple regression analyses indicated that precipitation but not temperature was associated positively with testicular and ovarian volume Analysis of acoustic parameters revealed that granular spiny frogs emitted two call types (advertisement and aggressive calls) Advertisement calls had a mean dominant frequency of 836 Hz and aggressive calls had a mean dominant frequency of 1020 Hz Call duration number of pulses pulse rate and pulse interval were highly variable within individuals (dynamic properties) whereas the dominant frequency was the most stereotyped property (a static property) Pulse length exhibited an intermediate level of variation within individuals Male SVL negatively influenced dominant frequency supporting a body-size constraint for this call characteristic The average number of tadpoles per clutch was 518 individuals the average survival rate at the complete stage of metamorphosis was 80% and the total time to metamorphosis was 55 8 days Multiple regression results for possible effects of water temperature dissolved oxygen and pH values on survival rates and the total time of metamorphosis of tadpoles were significant among localities The process of metamorphosis is completed in stage 46 at which juvenile frogs have a mean body mass of 2 3 g and a mean SVL of 25 8 mm Using a two-way multivariate analysis of variance revealed that body sizes of tadpoles varied significantly among years sites and by site-year interaction

Date of Award	2015 Jul 22
Original language	English
Supervisor	Ya-Fu Lee (Supervisor)