Background: The objective of this study was to examine associations of daily averages and daily variations in ambient relative humidity (RH), temperature, and PM2.5 on the obstructive sleep apnea (OSA) severity. Methods: A case-control study was conducted to retrospectively recruit 8628 subjects in a sleep center between January 2015 and December 2021, including 1307 control (apnea-hypopnea index (AHI) < 5 events/h), 3661 mild-to-moderate OSA (AHI of 5–30 events/h), and 3597 severe OSA subjects (AHI > 30 events/h). A logistic regression was used to examine the odds ratio (OR) of outcome variables (daily mean or difference in RH, temperature, and PM2.5 for 1, 7, and 30 days) with OSA severity (by the groups). Two-factor logistic regression models were conducted to examine the OR of RH with the daily mean or difference in temperature or PM2.5 with OSA severity. An exposure-response relationship analysis was conducted to examine the outcome variables with OSA severity in all, cold and warm seasons. Results: We observed associations of mean PM2.5 and RH with respective increases of 0.04–0.08 and 0.01–0.03 events/h for the AHI in OSA patients. An increase in the daily difference of 1 % RH increased the AHI by 0.02–0.03 events/h in OSA patients. A daily PM2.5 decrease of 1 μg/m3 reduced the AHI by 0.03 events/h, whereas a daily decrease in the RH of 1 % reduced the AHI by 0.03–0.04 events/h. The two-factor model confirmed the most robust associations of ambient RH with AHI in OSA patients. The exposure-response relationship in temperature and RH showed obviously seasonal patterns with OSA severity. Conclusion: Short-term ambient variations in RH and PM2.5 were associated with changes in the AHI in OSA patients, especially RH in cold season. Reducing exposure to high ambient RH and PM2.5 levels may have protective effects on the AHI in OSA patients.