Data from: Using Large-Scale Community Science Data and Computer Vision to Evaluate Thermoregulation as an Adaptive Driver of Physiological Color Change in Anolis carolinensis

This dataset contains all research-grade iNaturalist observations of Anolis carolinensis from 20 counties ending at the collection date of December 12, 2023. Included is a full list of downloaded observations, segments that passed filtering, and a final list of observations where color and temperature was successfully determined. The full list consists of each observation's ID, downloadable image link, and associated metadata, including but not limited to the citizen scientist contributor, datetime, coordinates, and county. The segments are derived from these images using the BigCritterColor package developed by Jacob Idec, which utilizes the Segment Anything Model to create a segment of the anole in the image. Then, AffinityPropogation is used to cluster these segments and remove undesirable clusters to filter out incorrect segments. The segments included in this dataset are those remaining after filtering and human verification. The final list includes the IDs and metadata for all observations that not only were successfully segmented, but also obtained a color designation (green or brown) and an estimated temperature. This data is what was used when modeling color interactions.