Global climate change, rice productivity and methane emissions: comparisonof simulated and experimental results

Irrigated rice production is a major food source for a large portion of the world's population, and a major anthropogenic source of the greenhouse gas methane (CH4). Potential impacts of global climate change [elevated carbon dioxide (CO2) and/or elevated temperature] on rice can be predicted with s imulation models, but experiments are necessary to determine how well these models mimic the responses of the field crop. This paper compares grain yi eld, biomass, and methane emissions from experiments at the international R ice Research Institute (IRRI) at Los Banos, the Philippines, with potential responses based on simulations using the ORYZA1 process model and the clim ate data from those experiments. Yield and biomass were compared for the 19 95 and 1996 dry seasons (DS) and the 1994 wet season (WS). Emissions of CH4 from rice fields were evaluated for the 1995 WS and 1996 DS, Simulated and experimental responses (adjusted for effects of the open-top chambers on p lant growth) differed with climate change scenario, response parameter, and season Under current climate conditions (ambient CO2 and ambient temperatu re), simulated grain yield was 14% lower than the adjusted experimental gra in yield in the 1996 DS, but was 17 and 37% higher than experimental grain yield in the 1995 DS and 1994 WS, respectively. With current climate, simul ations underestimated experimental aboveground, belowground, and total biom ass. The simulated CH4 emissions were the same as the experimental emission s, assuming CH4 emissions were 2.9% of the simulated total biomass carbon. With elevated CO7 and ambient temperature, simulations predicted greater in creases (compared with current climate) in grain yield, aboveground biomass , and total biomass, but generally smaller increases in belowground biomass and CH4 emissions than the significant (at p < 0.05) increases that were f ound experimentally. With ambient CO2 and elevated temperature, both simula tions and experiments generally showed either no change or a decrease in gr ain yield and biomass, but none of the responses in the experiments wen sta tistically significant. Simulated ambient CO2 and elevated temperature resu lted in a smaller decrease in CH4 emissions than the significant decrease f ound in the experiments. For both elevated CO2 and elevated temperature, si mulated grain yield increased in all three seasons, whereas there were no s ignificant effects on experimental grain yield. The simulations predicted s maller increases in belowground biomass and CH4 emissions with elevated CO2 and elevated temperature than the significant increases in the experiments . To better correspond to experimental results, this study suggested that c urrent simulation models could be improved in terms of effects of temperatu re on grain yield and use of belowground biomass to estimate CH4 emissions. (C) 1999 Elsevier Science B.V. All rights reserved.