Fishing for Anglers in a Sea of Data: Using Mobility Data to Identify and Track Marine Recreational Fishing

Analysing recreational anglers in marine waters presents several challenges. First, while there are well‐established approaches to estimate fishing effort, they are only as good as the coverage of the underlying data, which invariably has blind spots that can bias estimates. Additionally, tracking anglers' movements over time and their fishing locations in the open water is nearly impossible. This paper demonstrates the potential value of passively collected mobility data, GPS coordinates with timestamps from smartphones, for the analysis of marine recreational fishing. Using a classification process that includes supervised machine learning algorithms, we identify over 16,000 recreational fishing trips in the Gulf of Mexico from 2019 to 2022. We then categorise the identified trips into two groups: ‘station trips’, which are from places where creel surveys are conducted, and ‘non‐station trips’, which originate from locations not covered by creel data. We find that about 75% of all trips in the Gulf are station trips, though there is substantial variation across states. We examine differences in the on‐water behaviour of the two groups and study the spatial and temporal nature of these fishing trips. For validation of our proposed method, we compare the spatial and temporal variation of our identified trips based on mobility data with administrative data from Texas, Alabama, and Mississippi collected by government agencies. Correlations range from 0.63 to 0.94, providing strong evidence that our algorithm yields valid measures of recreational fishing effort, which can inform marine resource managers where such data are not available.