Recent research suggests that canopy photosynthesis drives soil respiration on short timescales (hours to days) in some temperate forests, presumably by affecting root respiration. However, measurements of fine root respiration (R_r) have lacked the sampling intensity and temporal precision to detect diel patterns, if present. We used a novel automated sampling system with temperature controlled cuvettes to make >7,000 measurements of R_r in a loblolly pine (Pinus taeda) forest exposed to elevated carbon dioxide and nitrogen fertilization. We detected diel patterns of R_r that track soil temperature as well as daily net ecosystem exchange (NEE) as measured by an on-site eddy-covariance tower. The direct response of R_r to temperature explained most of the diel variation in R_r on days with low NEE, but not on days with high NEE. The magnitude of the diel variation in R_r was correlated with the current and previous days’ NEE (r^₂=0.83, p<0.001 and r^₂=0.57, p<0.01, respectively), with no significant correlation with NEE from preceding days (p>0.1). Glucose additions to excised roots in solution consistently stimulated R_r measured in oxygen electrodes, suggesting that R_r was substrate limited and that the observed diel patterns in R_r could be explained by sugar supply from recent canopy photosynthesis. These results suggest a surprisingly close linkage between canopy and soil carbon processes, with implications for predicting R_r from measurements of NEE or photosynthesis. The combination of elevated [CO₂] and N fertilization reduced basal R_r by 61% (p=0.07) indicating that R_r may be decreased by global change.