The air supply mode of air conditioners creates a dynamic indoor thermal environment, which can significantly affect the thermal comfort needs of elderly individuals. However, the traditional predicted mean vote (PMV) index is designed for steady-state conditions and fails to consider the impact of spatial and temporal fluctuations in thermal environments. This study investigates the thermal needs of the elderly in dynamic air-conditioned environments and proposes evaluation metrics to quantify the effects of these fluctuations. Seven elderly participants were recruited for a climate chamber experiment. Thermal environment parameters were monitored at 24 points (across three heights and eight directions) surrounding the subjects under four typical air supply modes of room air-conditioners in cooling conditions. The temporal and spatial characteristics of the thermal environment were quantified using two proposed indicators: uniformity (defined as the difference in environmental parameters between the chest and the back of the subjects) and fluctuation (defined as the average variability of environmental parameters across the eight directions). Results showed significant uniformity and fluctuation in the thermal environment surrounding the subjects. Thermal perception evaluation showed that reduced environmental fluctuation and improved uniformity were associated with more neutral thermal and draught sensations, thereby enhancing thermal comfort. Correlation analysis further confirmed a significant relationship between thermal perception and the proposed uniformity and fluctuation indicators. These findings suggest that the uniformity and fluctuation indexes can be used as effective evaluation metrics for assessing the heat demand and thermal comfort of elderly individuals in dynamic indoor air-conditioned environments.