The rotational temperature obtained from the rotational population distribution in the bands of the hydroxyl airglow has been shown to be a suitable proxy for the temperature at a height of 87 km [She and Lowe, 1998>. In this paper we examine in detail simultaneous observations on November 2--3, 1997, at Fort Collins, Colorado (41 ¿N, 105 ¿W), with both a sodium temperature lidar and the Coupling, Energetics, and Dynamics of Atmospheric Regions (CEDAR) OH mesospheric temperature mapper during which significant differences between the hydroxyl and lidar temperatures occur. The large differences are associated with a major temperature enhancement in the region of the peak of the hydroxyl emission. We model the effect on the shape of the emission rate profile of the hydroxyl airglow caused by the large temperature enhancement observed on this night by the lidar. As a result of the temperature sensitivity of the processes that give rise to the airglow, the profile shows major distortions from its normal shape. These distortions in turn lead to hydroxyl rotational temperatures that differ significantly from the 87-km lidar observations. The mean rotational temperature deduced in this way agrees well with the observed values. Such deviations in the temperature are expected to be rare, occurring only when a large temperature enhancement occurs near the peak of the airglow emission profile. ¿ 2001 American Geophysical Union