The ratio of leaf internal (ci ) to ambient (ca ) partial pressure of CO2 , defined here as χ, is an index of adjustments in both leaf stomatal conductance and photosynthetic rate to environmental conditions. Measurements and proxies of this ratio can be used to constrain vegetation model uncertainties for predicting terrestrial carbon uptake and water use. We test a theory based on the least-cost optimality hypothesis for modelling historical changes in χ over the 1951-2014 period, across different tree species and environmental conditions, as reconstructed from stable carbon isotopic measurements across a global network of 103 absolutely dated tree-ring chronologies. The theory predicts optimal χ as a function of air temperature, vapour pressure deficit, ca and atmospheric pressure. The theoretical model predicts 39% of the variance in χ values across sites and years, but underestimates the intersite variability in the reconstructed χ trends, resulting in only 8% of the variance in χ trends across years explained by the model. Overall, our results support theoretical predictions that variations in χ are tightly regulated by the four environmental drivers. They also suggest that explicitly accounting for the effects of plant-available soil water and other site-specific characteristics might improve the predictions.

中文翻译：

---

光合作用的气孔限制的历史变化：最优性原理的经验支持。

CO2的叶片内部（ci）与环境（ca）分压之比（在此定义为χ）是叶片气孔导度和光合速率对环境条件的调节指标。该比率的测量值和近似值可用于约束植被模型的不确定性，以预测陆地碳的吸收和用水。我们测试了一种基于最小成本最优假设的理论，以对1951-2014年间χ在不同树种和环境条件下的历史变化进行建模，这是根据103个绝对过时的树的全球网络中稳定的碳同位素测量值重建而来的。环形年表。该理论预测最佳χ是空气温度，蒸气压赤字，ca和大气压的函数。该理论模型预测了跨站点和跨年的χ值的方差的39％，但低估了重建的χ趋势的站点间的变异性，导致该模型解释的跨年的χ趋势的方差只有8％。总体而言，我们的结果支持理论预测，即χ的变化受四个环境驱动因素的严格控制。他们还建议，明确考虑植物可利用的土壤水和其他特定地点特征的影响可能会改善预测。