Recovering Hidden Wildlife--Vehicle Collisions from Existing Crash Data: A Cross-State Framework for Detection, Prevalence Estimation, and Under-Reporting Audits
Abstract
Wildlife--vehicle collisions (WVCs) impose a large toll on U.S. highways, yet state crash databases often fail to record animal involvement, leaving agencies uncertain of the true missing count. This study evaluates whether existing crash records can identify individual WVCs and estimate overall unrecorded rates. We analyzed 4.9 million crash records from Colorado, Utah, Iowa, and Illinois (2007-2026). Each state's schema underwent a label-leakage screen to exclude explicit animal fields before harmonization into a shared vocabulary. Calibrated gradient-boosted classifiers were trained per state, interpreted using Shapley additive explanations, and evaluated for prevalence estimation via Adjusted Classify-and-Count and Maximum-Likelihood techniques across in-domain, cross-state, and synthetic reporting-gap scenarios. Detection performance depends heavily on record completeness rather than collision rarity. At ~4% prevalence, feature-rich Utah achieved an average precision of 0.88 versus 0.64 for feature-poor Colorado, and outperformed Iowa despite Iowa's 3x higher WVC rate. Within-state richness curves confirm feature depth drives accuracy. While classifiers ranked records accurately across states, probability calibration did not transfer directly. However, in-domain prevalence estimation recovered true collision rates within 0.001. This work shifts WVC analysis from spatial clustering to record-level detection and missing-data estimation using legacy datasets. Transportation agencies can retrospectively flag unrecorded WVCs, quantify total collision burden to support mitigation funding, and identify high-value reporting fields prior to revising crash-report manuals.
Related
- The Two Cultures of Prevalence Mapping: Small Area Estimation and Model-Based Geostatistics
- Comment on: "The Two Cultures of Prevalence Mapping: Small Area Estimation and Model-Based Geostatistics"
- A comparative analysis of methodological practices used in capture–recapture population size estimation: opioid use disorder prevalence in New York State, 2020
- Kalman Filtering and Smoothing for Improving Precision in Horvitz--Thompson Estimation of Infectious Disease Prevalence
- Quickest Change-point Detection in Generalized Linear Models with Variational Estimation
BibTeX
@misc{bello2026recovering,
title = {{Recovering Hidden Wildlife--Vehicle Collisions from Existing Crash Data: A Cross-State Framework for Detection, Prevalence Estimation, and Under-Reporting Audits}},
author = {Rasheed Bello and Judith Mwakalonge and Vaighese Vaidyan and Vijay Bendigeri and Anthony Dontoh and Gurcan Comert and Miriam Mkufya},
howpublished = {arXiv},
year = {2026},
month = nov,
doi = {10.48550/arxiv.2611.03658},
eprint = {2611.03658},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2611.03658},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Recovering Hidden Wildlife--Vehicle Collisions from Existing Crash Data: A Cross-State Framework for Detection, Prevalence Estimation, and Under-Reporting Audits
AU - Rasheed Bello
AU - Judith Mwakalonge
AU - Vaighese Vaidyan
AU - Vijay Bendigeri
AU - Anthony Dontoh
AU - Gurcan Comert
AU - Miriam Mkufya
PY - 2026
DA - 2026/11/01
PB - arXiv
DO - 10.48550/arxiv.2611.03658
UR - https://arxiv.org/abs/2611.03658
AB - Wildlife--vehicle collisions (WVCs) impose a large toll on U.S. highways, yet state crash databases often fail to record animal involvement, leaving agencies uncertain of the true missing count. This study evaluates whether existing crash records can identify individual WVCs and estimate overall unrecorded rates. We analyzed 4.9 million crash records from Colorado, Utah, Iowa, and Illinois (2007-2026). Each state's schema underwent a label-leakage screen to exclude explicit animal fields before harmonization into a shared vocabulary. Calibrated gradient-boosted classifiers were trained per state, interpreted using Shapley additive explanations, and evaluated for prevalence estimation via Adjusted Classify-and-Count and Maximum-Likelihood techniques across in-domain, cross-state, and synthetic reporting-gap scenarios. Detection performance depends heavily on record completeness rather than collision rarity. At ~4% prevalence, feature-rich Utah achieved an average precision of 0.88 versus 0.64 for feature-poor Colorado, and outperformed Iowa despite Iowa's 3x higher WVC rate. Within-state richness curves confirm feature depth drives accuracy. While classifiers ranked records accurately across states, probability calibration did not transfer directly. However, in-domain prevalence estimation recovered true collision rates within 0.001. This work shifts WVC analysis from spatial clustering to record-level detection and missing-data estimation using legacy datasets. Transportation agencies can retrospectively flag unrecorded WVCs, quantify total collision burden to support mitigation funding, and identify high-value reporting fields prior to revising crash-report manuals.
ER -