Does growth irradiance affect temperature dependence and thermal acclimation of leaf respiration? Insights from a...

Understanding the response of leaf respiration (R) to changes in irradiance and temperature is a prerequisite for predicting the impacts of climate change on plant function and future atmospheric CO2 concentrations. Little is known, however, about the interactive effects of irradiance and temperature on leaf R. We investigated whether growth irradiance affects the temperature response of leaf R in darkness (Rdark) and in light (Rlight) in seedlings of a broad-leaved evergreen species, Quercus ilex. Two hypotheses concerning Rdark were tested: (1) the Q10 (i.e. the proportional increase in R per 10 degrees C rise in temperature) of leaf Rdark is lower in shaded plants than in high-light-grown plants, and (2) shade-grown plants exhibit a lower degree of thermal acclimation of Rdark than plants exposed to higher growth irradiance. We also assessed whether light inhibition of Rlight differs between leaves exposed to contrasting temperatures and growth irradiances, and whether the degree of thermal acclimation of Rlight is dependent on growth irradiance. We showed that while growth irradiance did impact on photosynthesis, it had no effect on the Q10 of leaf Rdark. Growth irradiance had little impact on thermal acclimation when fully expanded, pre-existing leaves were exposed to contrasting temperatures for several weeks. When Rlight was measured at a common irradiance, Rlight/Rdark ratios were higher in shaded plants due to homeostasis of Rlight between growth irradiance treatments and to the lower Rdark in shaded leaves. We also showed that Rlight does not acclimate to the same degree as Rdark, and that Rlight/Rdark decreases with increasing measuring and growth temperatures, irrespective of the growth irradiance. Collectively, we raised the possibility that predictive carbon cycle models can assume that growth irradiance and photosynthesis do not affect the temperature sensitivity of leaf Rdark of long-lived evergreen leaves, thus simplifying incorporation of leaf R into such models.