Photosynthetic limitation by CO₂ diffusion in drought stressed orange leaves on three rootstocks

Abstract

Photosynthetic limitations under moderate water deficit were evaluated in ‘Valência’ orange trees grafted on three different rootstocks, in pots. Net CO₂ assimilation rate (A _N), stomatal conductance (g _s), and photosystem II (PS II) operating efficiency ( ${F}\ifmmode{'}\else$'$\fi_{{\text{q}}} /{F}\ifmmode{'}\else$'$\fi_{{\text{m}}}$) in response to changing intercellular CO₂ partial pressure (C _i) were analyzed under controlled conditions. Drought decreased A _N and g _s, whereas ${F}\ifmmode{'}\else$'$\fi_{{\text{q}}} /{F}\ifmmode{'}\else$'$\fi_{{\text{m}}}$ remained unchanged. This resulted in a higher ratio between electron transport rate (ETR) and gross CO₂ assimilation rate (A _G). Since the comparison of A _N–C _i gas exchange curves can lead to incorrect conclusions, a normalization of C _i values ($C_{{{\text{i}}_{{\text{f}}} }}$) of stressed leaves was applied. Then, the relationship established for irrigated trees between the ETR/A _G ratio and C _i was used to estimate the $C_{{{\text{i}}_{{\text{f}}} }}$ from ETR/A _G ratios measured under water stress. The response of A _N to $C_{{{\text{i}}_{{\text{f}}} }}$ suggests that the CO₂ diffusional restriction is the main factor that limits photosynthesis in orange leaves under moderate water deficit.

Appendix

Estimation of g _m and conversion of A _N–C _i curves to A _N–C _c curves

The effect of CO₂ on g _m was considered using the variable J method (Harley et al. 1992) and A _N, R _D and ETR data as follows:

$$ g_{{\text{m}}} = A_{\rm N} /{\left( {C_{{\text{i}}} - {\left( {\Upgamma ^{*} {\left( {{\text{ETR}} + 8{\left( {A_{{\text{N}}} + R_{{\text{D}}} } \right)}} \right)}} \right)}/{\left( {ETR - 4{\left( {A_{{\text{N}}} + R_{{\text{D}}} } \right)}} \right)}} \right)} $$

where the chloroplastic CO₂ compensation point (Γ*) was estimated and corrected for the leaf temperature (31°C) according to Bernacchi et al. (2001). From these g _m values, A _N–C _i curves were transformed to A _N–C _c curves after calculating C _c as:

$$ C_{{\text{c}}} = C_{{\text{i}}} - (A_{{\text{N}}} /g_{{\text{m}}} ) $$

However, the g _m estimates are very sensitive to small errors in the assumed parameters (ETR, Γ* and R _D) and are depended on C _i. Therefore, the sensitivity to errors in the g _m and C _c estimates was analyzed by the dC _c/dA _N test proposed by Harley et al. (1992). Thus, only data that satisfy the criterion of 10 < dC _c/dA _N < 50 are considered reliable and outside this range they are judged unacceptable.

The dependence of g _m on C _i and the corresponding A _N–C _c curve obtained by this method are shown in Fig. 5. Unfortunately, none of the g _m and C _c values were considered reliable, once dC _c/dA _N was always less than 10. Nevertheless, even thus the absolute values might be wrong, the A _N–C _c curve indicated that very little CO₂ was diffusing up to the chloroplast. All data were conformed to a single curve, suggesting that the carboxylation efficiencies were the same under both watering treatments. As expected, the g _m values and the corresponding C _c values were lower in drought stressed leaves than in irrigated leaves.