Sensitivity Analysis of a CO2 Assimilation - Stomatal Conductance Approach to Monitor Water Stress.

During crop growth water is needed for the maintenance of turgor and cooling of leaves. When water supply becomes inadequate, stomata will close and leaf stomatal conductance, g_s, can be used as a water stress indicator. As the outward diffusion of evaporated water in stomata and the uptake of CO₂, used in photosynthesis, have one and the same pathway, the CO₂ net assimilation (A_n) can be used to parameterize g_s. This is called the A_n-g_s approach. An important advantage of this approach is that it describes synergetic interactions between the most important stimuli, including light, temperature and CO₂. It is important to understand the sensitivity of the A_n-g_s approach to deviations in the real meteorological data caused by random or systematic observation errors. A sensitivity analysis, where the uncertainty in the output is apportioned to the uncertainty in the system input parameters, is an element of verification in which the response of the model to the input is the expected one. On the other hand, verifying that the model does not exhibit unexpectedly strong dependencies upon non-influential parameters is a valuable element of quality assurance. Systematic sensitivity studies for the A_n-g_s approach are not available.

In this study we present results of modeling with an A_n-g_s approach and perform a sensitivity analysis for several meteorological input data. For calibration and testing, we used data of two greenhouse experiments with bean plants (Phaseolus vulgaris L.) in pots submitted to two different water regimes (full and deficit irrigation). We discuss the sensitivity of the evaluated parameters.