Enhancing predictive performance for spectroscopic studies in wildlife science through a multi-model approach: A case study for species classification of live amphibians.

Near infrared spectroscopy coupled with predictive modeling is a growing field of study for addressing questions in wildlife science aimed at improving management strategies and conservation outcomes for managed and threatened fauna. To date, the majority of spectroscopic studies in wildlife and fisheries applied chemometrics and predictive modeling with a single-algorithm approach. By contrast, multi-model approaches are used routinely for analyzing spectroscopic datasets across many major industries (e.g., medicine, agriculture) to maximize predictive outcomes for real-world applications. In this study, we conducted a benchmark modeling exercise to compare the performance of several machine learning algorithms in a multi-class problem utilizing a multivariate spectroscopic dataset obtained from live animals. Spectra obtained from live individuals representing eleven amphibian species were classified according to taxonomic designation. Seven modeling techniques were applied to generate prediction models, which varied significantly (p < 0.05) with regard to mean classification accuracy (e.g., support vector machine: 95.8 ± 0.8% vs. K-nearest neighbors: 89.3 ± 1.0%). Through the use of a multi-algorithm approach, candidate algorithms can be identified and applied to more effectively model complex spectroscopic data collected for wildlife sciences. Other key considerations in the predictive modeling workflow that serve to optimize spectroscopic model performance (e.g., variable selection and cross-validation procedures) are also discussed.

The impact of time and environmental factors on the mitochondrial vesicle and subsequent motility of amphibian sperm.

For amphibian species that display external fertilization in an aquatic environment, hypoosmotic shock to sperm cells can quickly result in damage to cellular structure and function. This study sought to determine how fertilization media osmolality, temperature, and time impact the stability of the mitochondrial vesicle's association with the sperm head and thus motility and quality of forward progression. The presence of the mitochondrial vesicle and its relationship with sperm motility and quality of forward progression were analyzed in sperm samples from the Fowler's toad (Anaxyrus fowleri) (n = 10) when held for six hours under two temperatures and four osmolalities. Results indicated that the presence of the mitochondrial vesicle is needed for sperm motility over time (p < 0.001), where higher osmolalities (p < 0.001) and lower temperatures (p < 0.001) correlated with maintaining the vesicle attachment to the spermatozoa. The higher osmolality of spermic urine was the most important factor for maintaining higher quality of forward progressive motility (p < 0.01) of spermatozoa. Sperm samples held at 4 °C and 40 mOsm/kg had the longest half-life for motility (540 min) and quality of forward progression (276 min), whereas sperm held at 22 °C and 2.5 mOsm/kg had the shortest half-life for motility (7 min) and quality of forward progression (18 min). Special attention should be given to the osmolality and temperature of fertilization solutions, or breeding tank water, when developing cold storage protocols for toad sperm or reproducing animals to ensure the retention of the mitochondrial vesicle for maximum fertilization capability.

Lemur paparazzi: Arboreal camera trapping and occupancy modeling as conservation tools for monitoring threatened lemur species.

Primate species face growing risks of extinction throughout the world. To better protect their populations, effective monitoring techniques are needed. The goal of this study was to evaluate the use of arboreal camera traps and occupancy modeling as conservation tools for threatened lemur species. This project aimed to (1) estimate the occupancy and detection probabilities of lemur species, (2) investigate factors potentially affecting lemur habitat use, and (3) determine whether ground or arboreal cameras are better for surveying lemur assemblages. We conducted camera trapping research in five forest fragments (total trap nights = 1770; 900 arboreal trap nights (134 photo events); 870 ground trap nights (2 photo events)) and reforestation areas (total trap nights = 608; 1 photo event) in Kianjavato, Madagascar from May to September 2019. We used arboreal trap data from fragments to estimate occupancy for five species: the red-fronted brown lemur (Eulemur rufifrons; ψ = 0.54 ± SD 0.03), Jolly's mouse lemur (Microcebus jollyae; ψ = 0.14 ± 0.17), the greater dwarf lemur (Cheirogaleus major; ψ = 0.42 ± 0.30), the red-bellied lemur (Eulemur rubriventer; ψ = 0.24 ± 0.03), and the black-and-white ruffed lemur (Varecia variegata; ψ = 0.24 ± 0.08). Tree diameter, elevation, distance to village, and canopy connectivity were important predictors of occupancy, while camera height, canopy connectivity, fragment ID, and fragment size predicted detection. Arboreal cameras recorded significantly higher species richness compared with ground cameras. We suggest expanded application of arboreal camera traps in future research, but we recommend longer trapping periods to better sample rarer species. Overall, arboreal camera trapping combined with occupancy modeling can be a highly efficient and useful approach for monitoring and predicting the occurrence of elusive lemur species and has the potential to be effective for other arboreal primates and canopy taxa across the globe.