Modeling Recreational Visitation at Bureau of Land Management Sites

Citation

Hanson, D., Wood, S. A., Rappaport, S., Wilkins, E. J., & Schuster, R. M. (2026). Modeling recreational visitation at Bureau of Land Management sites. Scientific Reports, 16(1), 18848. doi.org/10.1038/s41598-026-43154-y

Link to file

Public land managers need to know how many people are visiting places like national parks and BLM land. Right now, they mostly count visitors using automatic counters or other on-the-ground equipment. That works, but it’s a hassle—especially for remote spots or places with lots of entry points.

Researchers looked at whether other kinds of data could help estimate visitation more easily. They studied 70 Bureau of Land Management sites across the U.S., using 1,328 months’ worth of actual visitor counts. Then they built computer models that combined three types of digital data—cell phone location pings, geotagged social media posts, and community science observations (like iNaturalist)—along with 15 basic characteristics of each site (like size, location, and facilities).

Models that included site characteristics did better than models that only used digital data. When researchers tested the models on sites they hadn’t seen before, accuracy varied—meaning the models work best when they’ve been trained on similar types of sites. Relying on digital data alone isn’t enough, but combining it with other information is promising.

This approach offers a scalable way to estimate visitation where traditional counting is tough or impractical. It could help land managers make better decisions without needing to install and maintain counters everywhere.

Abstract

Estimates of recreational visitation are essential for public land management. Visitation is typically estimated using devices such as automated counters that require logistics and effort, particularly at remote locations or those with several access points. In this study, we investigate the utility of alternative data sources and statistical models for estimating visitation to public lands in the United States, through an analysis of data from 70 Bureau of Land Management sites. We compile 1328 site-months of visitor count data collected on-site, which are used to train and evaluate three random forest models incorporating combinations of 15 site-level characteristics and three sources of digital mobility data—mobile device locations, geolocated social media, and community science observations. Models including site characteristics perform better than models relying on mobility data alone. Cross-validation using held-out sites reveal varying prediction accuracy, suggesting that model generalizability depends on the inclusion of characteristically similar sites in the training data. These results underscore the limitations of relying solely on mobility data for visitation estimation and highlight the benefits of combining diverse data sources. Our approach provides a scalable, data-driven framework for estimating visitation where traditional monitoring is challenging or infeasible, supporting broader applications in recreation management.