Are trapping data still suited for home range estimation? An analysis with various estimators, asymptotic models and data ordering procedures
Bibliographic record
Abstract
Abstract Understanding the size of animals’ home ranges is vital for studies in ecology and conservation. Trapping datasets are an important source of information when targeting the biodiversity of an area, inconspicuous species, or high numbers of individuals in contrast to more expensive telemetry-based methods such as radio- or GPS-collaring. Currently, studies relying on trapping lack an evaluation of the performance of existing home range estimation procedures comparable to those developed for telemetry. Using animal movement simulations, we evaluate three variables reflecting the trade-offs faced by ecologists when designing a trapping study, 1) the number of observations obtained per individual, 2) the trap density and 3) the proportion of the home range area falling inside of the trapping grid. We compare the performance of five estimators on these conditions, four commonly used (AKDE, KDE, MCP, LoCoH) and a possible alternative for situations with low trap density or high number of observations (bicubic interpolation). We further test suggested benefits of using asymptotic models (Michaelis-Menten and monomolecular) to assess the total home range area when information obtained per individual is scarce, as this situation might be common in trapping datasets. In addition, we propose sorting the observations based on the distance between locations to improve the performance of asymptotic models’ estimates. Using the results of the different procedures we constructed a generalized additive model (GAM) that allows predicting the bias in home range size under the different scenarios investigated. Our results show that the proportion of the area covered by the trapping grid and the number of observations were the most important factors predicting the accuracy and reliability of the estimates. The use of asymptotic models helped obtaining an accurate estimation at lower sample sizes and this effect was further improved by distance-ordering. The autocorrelation informed KDE was the estimator performing best under most conditions evaluated. Nevertheless, bicubic interpolation can be an alternative under common trapping conditions with low density of traps and low area covered. We provide the current results to the constructed GAM as a prospective tool for ecologists planning a new study or with already collected datasets that aim at assessing the potential biases in their estimates. Reliable and accurate home range estimates using trapping data can optimize monetary costs of home range studies, potentially enlarging the span of species, researchers and questions studied in ecology and conservation.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.139 | 0.320 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.004 |
| Bibliometrics | 0.003 | 0.003 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".