The effects of forest fires on above- and below-ground processes may vary widely depending on their intensity. Fire intensity is therefore an important parameter to consider when monitoring and interpreting fire effects. Because fire intensity is, in practice, a difficult variable to measure, researchers often rely on indicators of fire intensity, such as maximum temperature or time x temperature series. A variety of techniques for estimating fire intensity and/or fire temperatures have been developed, for example, fuel consumption, pyrometers (heat-sensitive paints), caloriometers (water-loss from open containers), thermocouples, and infrared cameras. These techniques range widely in both their cost and accuracy, and therefore, have different applications for research and monitoring. We evaluated the usefulness of four of these methods in 9 experimental fires in a longleaf pine upland in southern Alabama. At 140 1 m2 plots, we measured pre- and post-burn fuel conditions, maximum fire temperature (ceramic tile and metal pyrometers), energy release (caloriometers), and fire temperature over time (Type-K thermocouples). Results showed that although metal pyrometers consistently underestimated maximum temperature, they were better indicators of fire intensity than ceramic tile pyrometers, caloriometers, or fuel consumption indicators. Results also revealed that rather simple fire behavior parameters, such as flame length, were sufficient to show significant differences among different fires. Semivariance analysis indicated that fire temperatures were spatially dependent over scales less than 15 m, suggesting that the spatial arrangement of the sampling strategy needs to be considered in order to adequately compare fires of different intensities.