Circadian rhythms regulate the environmental responses of net CO2 exchange in bean and cotton canopies

Studies on the dependence of the rates of ecosystem gas exchange on environmental parameters often rely on the up-scaling of leaf-level response curves ('bottom-up' approach), and/or the down-scaling of ecosystem fluxes ('top-down' approach), where one takes advantage of the natural diurnal covariation between the parameter of interest and photosynthesis rates. Partly independent from environmental variation, molecular circadian clocks drive similar to 24 h oscillations in leaf-level photosynthesis, stomatal conductance and other physiological processes in plants under controlled laboratory conditions. If present and of sufficient magnitude at ecosystem scales, circadian regulation could lead to different results when using the bottom-up approach (where circadian regulation exerts a negligible influence over fluxes because the environment is modified rapidly) relative to the top-down approach (where circadian regulation could affect fluxes as it requires the passage of a few hours). Here we dissected the drivers of diurnal net CO2 exchange in canopies of an annual herb (bean) and of a perennial shrub (cotton) through a set of experimental manipulations to test for the importance of circadian regulation of net canopy CO2 exchange, relative to that of temperature and vapor pressure deficit, and to understand whether circadian regulation could affect the derivation of environmental flux dependencies. Contrary to conventional wisdom, we observed how circadian regulation exerted controls over net CO2 exchange that were of similar magnitude to the controls exerted by direct physiological responses to temperature and vapor pressure deficit. Diurnal patterns of net CO2 exchange could only be explained by considering effects of environmental responses combined with circadian effects. Consequently, we observed significantly different results when inferring the dependence of photosynthesis over temperature and vapor pressure deficit when using the top-down and the bottom up approaches. (C) 2017 Elsevier B.V. All rights reserved.