Water scarcity is a relevant issue in a scenario of climate change not only for animal production but also for many regular human activities. Thus, an adequate estimation of its demand is of high importance due to the high requirement for animal protein worldwide forecasted for the next decades. This study highlights that pen surface temperature can be used to predict daily water intake and tympanic temperature in feedlot steers. Commercial feedlot producers can estimate their water requirements under scenarios of heat stress with a simple tool, contributing to keeping sustainable production systems.Adequate estimation of water demand in cattle production feed yards is of high importance, especially due to reduced water availability because of changes in rain precipitation patterns and amounts. The pen surface in feed yards receives and reflects solar radiation, affecting along with other factors the microclimate to which cattle are exposed. This study aimed to describe the relationship between the pen surface temperature with the daily water intake and the tympanic temperature of finishing steers. Climate variables, including solar radiation, air temperature, relative humidity, and wind speed plus pen surface temperature and soil temperature at 10.2 cm depth were recorded. Data were collected from a weather station located in the feedlot in Concord NE, whereas daily water intake was estimated from a set of experimental pens sharing a waterer in two adjacent pens. Simple and polynomial linear regressions were assessed from data collected in different experiments conducted from 2003 to 2006. Two models to predict daily water intake were developed for finishing steers using the pen surface temperature as the predictor variable. The first one included data for the period May-October (overall model) and the second one for the summer period (June-August). The best fit for the overall model was a quadratic fit (r(2) = 0.86), whereas the best-fit model for the summer model was the cubic (r(2) = 0.72). Subsequently, both models were validated with data from an independent experiment conducted in the summer of 2007 in the same facilities. Both models tended to slightly overestimate daily water intake when they were validated (14.6% and 12.6%, respectively). For tympanic temperature, the best-fit model was linear, explaining 76% of the observed variability. When the dataset was split into night-time and daytime the best-fit model for the night period was a quadratic one and a linear one for the daytime, both improving the explanation of the variability observed. In conclusion, pen surface temperature can be used to predict both daily water intake and tympanic temperature in feedlot steers without access to shade.