| Abstract
| - The measurement of the response of net photosynthesis to leaf internal CO2 (i.e. A-Ci curves) is widely used for ecophysiological studies. Most studies did not consider CO2 exchange between the chamber and the surrounding air, especially at the two extremes of A-Ci curves, where large CO2 gradients are created, leading to erroneous estimations of A and Ci. A quantitative analysis of CO2 leakage in the chamber of a portable open gas exchange system (Li-6400, LI-COR Inc., NE, USA) was performed. In an empty chamber, the measured CO2 leakage was similar to that calculated using the manufacturer's equations. However, in the presence of a photosynthetically inactive leaf, the magnitude of leakage was substantially decreased, although still significant. These results, together with the analysis of the effects of chamber size, tightness, flow rate, and gasket material, suggest that the leakage is larger at the interface between the gaskets than through the gaskets. This differential leakage rate affects the parameterization by photosynthesis models. The magnitude of these errors was assessed in tobacco plants. The results showed that leakage results in a 10% overestimation of the leaf maximum capacity for carboxylation (Vc,max) and a 40% overestimation of day respiration (Rl). Using the manufacturer's equations resulted in larger, non-realistic corrections of the true values. The photosynthetic response to CO2 concentrations at the chloroplast (i.e. A-Cc curves) was significantly less affected by leakage than A-Ci curves. Therefore, photosynthetic parameterization can be improved by: (i) correcting A and Ci values for chamber leakage estimated using a photosynthetically inactive leaf; and (ii) using A-Cc instead of A-Ci curves.
|
