Genotypic Variation in the Response of Rice (Oryza sativa L.) to Increased Atmospheric Carbon Dioxide and its Physiological Basis

Increasing atmospheric carbon dioxide concentration (C_a) is one aspect of global change that will have a significant impact on the productivity of agricultural crops. Crop yields have been shown to increase with increasing C_a. The magnitude of yield response to increased C_a could vary depending on genotypic and environmental factors. The objective of this study was to determine the genotypic variation of yield response and its physiological basis in rice (Oryza sativa L.) through an initial varietal screening using 16 genotypes. The genotypes were grown under two concentrations of C_a, i.e. 370 ± 28 (ambient) and 570 ± 42 (elevated) μmol mol^?1, in open top chambers under lowland field conditions at the Rice Research and Development Institute in Sri Lanka (7°50′N, 80°50′E) from May to August 2001 (yala season) and from November 2001 to March 2002 (maha). C_a within chambers was maintained around target concentrations by a computer‐based real‐time data acquisition and control system. There was significant variation between genotypes in the response of yield to elevated C_a, with absolute increases up to 530 g m^?2 in yala and 347 g m^?2 in maha. In relative terms, percentage yield increases from ambient to elevated C_a ranged from 4 % to 175 % in yala and from 3 % to 64 % in maha. Genotypic variation in yield showed significant positive correlations with light‐saturated net photosynthetic rate of the flag leaf during the grain‐filling stage. This indicated that increased assimilate supply and its genotypic variation contributed to the observed genotypic variation in yield response to elevated C_a. Furthermore, the capacity to develop a larger reproductive sink through increased panicle number per m² and increased number of grains per panicle contributed to greater yield at elevated C_a. There was a significant genotype × season interaction with genotypes responding differentially to increased C_a in the two seasons. This was mainly due to inter‐seasonal variation in incident radiation during the grain‐filling stage. Our results demonstrate the significant genotypic variation that exists within the rice germplasm, in the response to increased C_a of yield and its correlated physiological parameters. A subset of genotypes from screening trials such as the present study can be used for more in‐depth analysis of the influence of elevated C_a on processes responsible for yield determination in rice and for molecular studies to elucidate the genetic basis of the response to increased C_a. This could pave the way for breeding genotypes which are more productive in a future high CO₂ environment, provided that genotypes with greater flexibility in their physiology are selected to counter the genotype × environmental interaction.