Search for: All records

Ecologists interested in monitoring the effects caused by climate change are increasingly turning to passive acoustic monitoring, the practice of placing autonomous audio recording units in ecosystems to monitor species richness and occupancy via species calls. However, identifying species calls in large datasets by hand is an expensive task, leading to a reliance on machine learning models. Due to a lack of annotated datasets of soundscape recordings, these models are often trained on large databases of community created focal recordings. A challenge of training on such data is that clips are given a "weak label," a single label that represents the whole clip. This includes segments that only have background noise but are labeled as calls in the training data, reducing model performance. Heuristic methods exist to convert clip-level labels to "strong" call-specific labels, where the label tightly bounds the temporal length of the call and better identifies bird vocalizations. Our work improves on the current weakly to strongly labeled method used on the training data for BirdNET, the current most popular model for audio species classification. We utilize an existing RNN-CNN hybrid, resulting in a precision improvement of 12% (going to 90% precision) against our new strongly hand-labeled dataset of Peruvian bird species.Jacob Ayers (Engineers for Exploration at UCSD); Sean Perry (University of California San Diego); Samantha Prestrelski (UC San Diego); Tianqi Zhang (Engineers for Exploration); Ludwig von Schoenfeldt (University of California San Diego); Mugen Blue (UC Merced); Gabriel Steinberg (Demining Research Community); Mathias Tobler (San Diego Zoo Wildlife Alliance); Ian Ingram (San Diego Zoo Wildlife Alliance); Curt Schurgers (UC San Diego); Ryan Kastner (University of California San Diego) 

Tree growth and longevity trade-offs fundamentally shape the terrestrial carbon balance. Yet, we lack a unified understanding of how such trade-offs vary across the world’s forests. By mapping life history traits for a wide range of species across the Americas, we reveal considerable variation in life expectancies from 10 centimeters in diameter (ranging from 1.3 to 3195 years) and show that the pace of life for trees can be accurately classified into four demographic functional types. We found emergent patterns in the strength of trade-offs between growth and longevity across a temperature gradient. Furthermore, we show that the diversity of life history traits varies predictably across forest biomes, giving rise to a positive relationship between trait diversity and productivity. Our pan-latitudinal assessment provides new insights into the demographic mechanisms that govern the carbon turnover rate across forest biomes. 

Abstract Accurate estimates of survival are crucial for many management decisions in translocation programs. Maximizing detection probabilities and reducing sampling biases for released animals can aid in estimates of survival. One important source of sampling bias is an animal’s behavior. For example, individuals that are consistently more exploratory or active may be more likely to be detected visually. Behavioral traits can be related to survival after reintroduction, and because many pre‐release treatments aim to manipulate animal behavior, it is critical to tease apart relationships between behavior and detection probability. Here, we assessed the repeatability (intra‐individual consistency and inter‐individual variation) of behavioral traits for an endangered amphibian, the mountain yellow‐legged frog (Rana muscosa). Because new technological tools offer one potential solution for reducing sampling biases while increasing detection, we also tested whether a long‐range passive integrated transponder (PIT) tag reader could enhance surveys for these individuals after translocation into the wild. After confirming thatex situbredR. muscosaexhibit repeatable behavioral traits (repeatability = 0.25–0.41) and releasing these frogs (N = 196) into the wild, we conducted post‐release surveys visually and with the long‐range PIT tag reader. Integrating the long‐range reader into surveys improved detection probability four‐fold in comparison to visual surveys alone (~0.09 to ~0.36). Moreover, mark–recapture modeling revealed that tag reader detection probability was not biased toward detecting individuals of specific behavioral types, while visual detection was significantly related to behavioral traits. These results will enable a more accurate understanding of individual differences in post‐release success in translocations. This may be particularly important for amphibian species, which can be difficult to detect and are expected to increasingly be involved in human‐managed breeding and translocation programs due to their vulnerable conservation status. 

Abstract Camera trapping has revolutionized wildlife ecology and conservation by providing automated data acquisition, leading to the accumulation of massive amounts of camera trap data worldwide. Although management and processing of camera trap‐derived Big Data are becoming increasingly solvable with the help of scalable cyber‐infrastructures, harmonization and exchange of the data remain limited, hindering its full potential. There is currently no widely accepted standard for exchanging camera trap data. The only existing proposal, “Camera Trap Metadata Standard” (CTMS), has several technical shortcomings and limited adoption. We present a new data exchange format, the Camera Trap Data Package (Camtrap DP), designed to allow users to easily exchange, harmonize and archive camera trap data at local to global scales. Camtrap DP structures camera trap data in a simple yet flexible data model consisting of three tables (Deployments, Media and Observations) that supports a wide range of camera deployment designs, classification techniques (e.g., human and AI, media‐based and event‐based) and analytical use cases, from compiling species occurrence data through distribution, occupancy and activity modeling to density estimation. The format further achieves interoperability by building upon existing standards, Frictionless Data Package in particular, which is supported by a suite of open software tools to read and validate data. Camtrap DP is the consensus of a long, in‐depth, consultation and outreach process with standard and software developers, the main existing camera trap data management platforms, major players in the field of camera trapping and the Global Biodiversity Information Facility (GBIF). Under the umbrella of the Biodiversity Information Standards (TDWG), Camtrap DP has been developed openly, collaboratively and with version control from the start. We encourage camera trapping users and developers to join the discussion and contribute to the further development and adoption of this standard.