Improved understanding of the responses of stem radial growth to climates is necessary for modeling and predicting the response of forest ecosystems to future climate change. We used dendrochronological methods to study climate effects on the radial growth of a subalpine deciduous conifer, Larix potaninii. Tree-ring residual chronologies were developed for five sites at the upper distributional limits in the Central Hengduan Mountains, Southwestern China. Redundancy analysis and response function were used to compare inter-annual variability in growth sensitivity among the chronologies and to identity key climatic factors controlling tree radial growth. The results showed that both precipitation and temperature influenced tree growth, and response patterns were consistent for five chronologies. During the current year’s early growing season (T_mean in May and T_max in June), temperature positively affected the radial growth of L. potaninii, while September T_min and October precipitation in the previous year and May and June precipitation in the current year all had negative impacts on its radial growth. L. potaninii growth appeared to be mainly limited by photothermal conditions in May and June. In the context of increasing CO₂ concentrations accompanied with warmer temperatures, future climate change would likely stimulate the radial growth of L. potaninii in Central Hengduan Mountain.

Responses of tree radial growth to climate are usually species-specific. Northwestern Yunnan has become a hotspot for the study of dendrochronology due to its sensitivity to climate change and the relative integrity of vegetation preservation. In this paper, we take three dominant conifers—Pinus armandii, Pinus yunnanensis and Picea likiangensis—as the research objects and analyze their tree-ring width chronologies in order to reveal the main climate factors affecting tree growth in northwestern Yunnan and to evaluate species-specific variation in climate response. The results showed that the radial growth of the three tree species was co-regulated by temperature and precipitation but that the growth response patterns were varied. Specifically: (1) The radial growth of the three species of conifers was significantly and negatively correlated with the July average maximum temperature (T_max) and the October Palmer Drought Severity Index (PDSI) in the current year. (2) Current May precipitation significantly promoted P. armandii growth and inhibited P. likiangensis growth, and a wet July was beneficial for both P. yunnanensis and P. likiangensis growth, while the radial growth of P. yunnanensis and P. armandii showed a significant and positive correlation with the August T_max in the current year. (3) The sliding analysis supported the results of the response function by showing stable relationships with climate factors which significantly affected tree growth. Results from redundancy analysis (RDA) and response function analysis were basically consistent, demonstrating that these two methods could complement each other in the understanding of relationships between tree radial growth and climatic factors. This study elucidates the climate–growth relationship of the main tree species in the study area and provides theoretical guidance and scientific evidence for regional forest management.