This paper develops and validates a method of using time-at-temperature (TAT) histograms from satellite transmitter tags to describe the dive activity patterns and approximate depth distributions of five deep-diving toothed whale species in the northern Bahamas. TAT histograms represent a bandwidth-conserving method of recovering a long-term proxy record of dive activity. However, using temperature to interpret TAT on a scale of approximate depths required the complex estimation of TAT histogram bin boundary depths in a dynamic oceanographic region. Here we evaluated the relative performance of four interpolation methods and a global reanalysis data assimilation model in estimating climatological isotherm depth surfaces within our study area. TAT-derived approximate time-at-depth (TAD) distributions aligned closely with directly observed TAD distributions from a smaller sample of depth-recording satellite tags deployed on separate individuals of each species. TAT-derived approximate depth distributions were also consistent with various published accounts for this suite of species. Estimating dive ranges and time budgets are important components of (1) understanding habitat overlap between species, (2) evaluating the potential role of these predators in meso- and bathypelagic ecosystems, and (3) assessing vulnerability and exposure to anthropogenic impacts.