Spatiotemporal Variability and Environmental Controls of Temperature Sensitivity of Ecosystem Respiration across the Tibetan Plateau

Abstract

Warming-induced carbon loss via ecosystem respiration (R_e) is probably intensifying in the alpine grassland ecosystem of the Tibetan Plateau owing to more accelerated warming and the higher temperature sensitivity of R_e (Q₁₀). However-little is known about the patterns and controlling factors of Q₁₀ on the plateau, impeding the comprehension of the intensity of terrestrial carbon–climate feedbacks for these sensitive and vulnerable ecosystems. Here, we synthesized and analyzed multiyear observations from 14 sites to systematically compare the spatiotemporal variations of Q₁₀ values in diverse climate zones and ecosystems, and further explore the relationships between Q₁₀ and environmental factors. Moreover-structural equation modeling was utilized to identify the direct and indirect factors predicting Q₁₀ values during the annual-growing, and non-growing seasons. The results indicated that the estimated Q₁₀ values were strongly dependent on temperature- generally, with the average Q₁₀ during different time periods increasing with air temperature and soil temperature at different measurement depths (5 cm, 10 cm, 20 cm). The Q₁₀ values differentiated among ecosystems and climatic zones, with warming-induced Q₁₀ declines being stronger in colder regions than elsewhere based on spatial patterns. NDVI was the most cardinal factor in predicting annual Q₁₀ values, significantly and positively correlated with Q₁₀. Soil temperature (T_s) was identified as the other powerful predictor for Q₁₀, and the negative Q₁₀–T_s relationship demonstrates a larger terrestrial carbon loss potentiality in colder than in warmer regions in response to global warming. Note that the interpretations of the effect of soil moisture on Q₁₀ were complicated, reflected in a significant positive relationship between Q₁₀ and soil moisture during the growing season and a strong quadratic correlation between the two during the annual and non-growing season. These findings are conducive to improving our understanding of alpine grassland ecosystem carbon–climate feedbacks under warming climates.