Timber harvesting can have strong effects on terrestrial salamanders, which are critical components of forest ecosystems and indicators of environmental change. Effects of harvest methods such as clearcutting have been studied in the short term, but few studies have examined salamander trends throughout the decade following timber harvest. The effects of other harvest methods, such as patch cuts and shelterwoods, also remain unclear. We examined the effects of clearcuts, patch cuts, and shelterwood harvests on salamander relative abundance one year before and up to eleven years after harvest in clearcuts, shelterwoods, and patch cuts at the Hardwood Ecosystem Experiment (HEE). A total of 41,858 salamanders representing ten species were captured under artificial coverboards over this period with eastern red-back (Plethodon cinereus) and zigzag (Plethodon dorsalis) salamanders dominating. No significant declines occurred in the first three years following harvest, but salamander captures declined in the 4–6 and 7–11 year periods in clearcuts and patch cuts, and no signs of post-harvest recovery were found. Rapid vegetation recovery and high volumes of coarse woody debris (CWD) following harvest likely protected the microclimate used by salamanders; when CWD levels dropped 4–6 years after harvest, a corresponding decline in salamander captures occurred. These years also corresponded to a regional drought that likely increased temperature and decreased moisture in harvest openings, further impacting terrestrial salamanders. Neither the first nor second shelterwood harvests produced declines in salamanders, suggesting that canopy retention could prevent declines due to changing microclimate or drought. Our results highlight the need to consider compound effects of disturbances such as drought and canopy loss and the importance of long-term salamander monitoring following timber harvest to capture lagged effects and eventual recovery.