The purpose of this study is to (1) demonstrate the viability of detecting terrestrial caves at thermal-infrared wavelengths, (2) improve our understanding of terrestrial cave thermal behavior, (3) identify times of day when cave openings have the maximum thermal contrast with the surrounding surface regolith, and (4) further our understanding of how to detect caves on Earth, the Moon and Mars. We monitored the thermal behavior of two caves in the Atacama Desert of northern Chile. Through this work, we identified times when temperature contrasts between entrance and surface were greatest, thus enabling us to suggest optimal overflight times. The largest thermal contrast for both caves occurred during mid-day. One cave demonstrated thermal behavior at the entrance suggestive of cold-trapping, while the second cave demonstrated temperature shifts suggestive of airflow. We also collected thermograms without knowing optimal detection times; these images suggest both caves may also be detectable during off-peak times. We suggest cave detection using thermal remote sensing on Earth and other planetary objects will be limited by (1) capturing imagery in the appropriate thermal wavelength, (2) the size of cave entrance vs. the sensor's spatial resolution, (3) the viewing angle of the platform in relation to the slope trajectory of the cave entrance, (4) the strength of the thermal signal associated with the cave entrance, and (5) the time of day and season of thermal image capture. Through this and other studies, we will begin to identify the range of conditions under which caves are detectable in the thermal infrared and thus improve our detection capabilities of these features on Earth, the Moon and Mars.