Objective Ecological stress memory (ESM) is an important mechanism by which trees adapt to recurrent drought. However, against the backdrop of increasingly frequent drought events driven by global climate change, the characteristics and drivers of ESM formation in Picea schrenkiana remain unclear.

Method This study investigated Picea schrenkiana at different elevations in the western Tianshan Mountains of Xinjiang. Drought events were identified using the Palmer Drought Severity Index (PDSI) during the tree growing season. The existence of ecological stress memory (ESM) was assessed based on the difference in resistance (ΔRt) and the Wilcoxon paired signed-rank test, and correlation analysis and generalized estimating equations (GEE) were used to investigate the driving factors of ESM.

Result (1) The formation of ESM in Picea schrenkiana showed a clear dependence on drought severity. The ΔRt was highest (0.121) under moderate drought (−3 ≤ PDSI < −2), whereas its median value was lowest (−0.422) under severe drought (−4 ≤ PDSI < −3), indicating a significant negative legacy effect. As drought severity increased from mild to moderate or from moderate to severe, the resistance difference (ΔRt) increased, indicating the formation of ESM in Picea schrenkiana. (2) The environmental gradient represented by elevation is an important spatial factor influencing ESM formation, with the ESM occurrence rate decreasing significantly with increasing elevation (P < 0.001). The ESM occurrence rates at low-, middle-, and high-elevation sites were 100%, 66.7%, and 33.3%, respectively. (3) Individual traits such as diameter at breast height (DBH) and growth rate had relatively weak effects on whether snow spruce developed ESM, whereas the within-tree effect of age was positively associated with ESM, indicating that ESM tended to increase continuously within the same tree as age and cumulative drought experience increased.

Conclusion The formation of ESM in Picea schrenkiana is jointly regulated by drought intensity and the environmental gradient represented by elevation, with individual traits exerting a certain moderating effect. These findings enrich the theory of growth stress memory in trees and provide an important scientific basis for climate-adaptive management of Picea schrenkiana forests in the Tianshan Mountains.