Using a high-resolution (40*40 km) chemical transport model coupled with the Regional Emission inventory in Asia (REAS), we simulated surface ozone concentrations ([O 3]) and evaluated O 3-induced wheat production loss in China and India for the years 2000 and 2020 using dose-response functions based on AOT40 (accumulated [O 3] above 40 ppb) and POD Y (phytotoxic O 3 dose, accumulated stomatal flux of O 3 above a threshold of Y nmol m -2 s -1). Two O 3 dose metrics (90 days AOT40 and POD 6) were derived from European experiments, and the other two (75 days AOT40 and POD 12) were adapted from Asian studies. Relative yield loss (RYL) of wheat in 2000 was estimated to be 6.4-14.9% for China and 8.2-22.3% for India. POD 6 predicted greater RYL, especially for the warm regions of India, whereas the 90 days AOT40 gave the lowest estimates. For the future projection, all the O 3 dose metrics gave comparable estimates of an increase in RYL from 2000 to 2020 in the range 8.1-9.4% and 5.4-7.7% for China and India, respectively. The lower projected increase in RYL for India may be due to conservative estimation of the emission increase in 2020. Sensitivity tests of the model showed that the POD Y -based estimates of RYL are highly sensitive to perturbations in the meteorological inputs, but that the estimated increase in RYL from 2000 to 2020 is much more robust. The projected increase in wheat production loss in China and India in the near future is substantially larger than the uncertainties in the estimation and indicates an urgent need for curbing the rapid increase in surface [O 3] in these regions.
