Assessment of the methane mitigation potentials of alternative water regimes in rice fields using a process‐based biogeochemistry model

Abstract

Rice production is a substantial source of atmospheric CH 4 , which is second only to CO 2 as a contributor to global warming. Since CH 4 is produced in anaerobic soil environments, water management is expected to be a practical measure to mitigate CH 4 emissions. In this study, we used a process-based biogeochemistry model (DNDC-Rice) to assess the CH 4 mitigation potentials of alternative water regimes (AWR) for rice fields at a regional scale. Before regional application, we tested DNDC-Rice using site-scale data from three rice fields in Japan with different water regimes. The observed CH 4 emissions were reduced by drainage of the fields, but were enhanced by organic amendments. DNDC-Rice gave acceptable predictions of variation in daily CH 4 fluxes and seasonal CH 4 emissions due to changes in the water regime. For regional application, we constructed a GIS database at a 1 Â 1 km mesh scale that contained data on rice field area, soil properties, daily weather, and farming management of each cell in the mesh, covering 3.2% of the rice fields in Japan's Hokkaido region. We ran DNDC-Rice to simulate CH 4 emissions under five simulated water regimes: the conventional water regime and four AWR scenarios with gradually increasing drainage. We found that AWR can reduce CH 4 emission by up to 41% compared with the emission under conventional water regime. Including the changes in CO 2 and nitrous oxide emissions, potential mitigation of greenhouse gas (GHG) was 2.6 Mg CO 2 Eq. ha À1 yr À1 . If this estimate is expanded to Japan's total rice fields, expected GHG mitigation is 4.3 Tg CO 2 Eq. yr À1 , which accounts for 0.32% of total GHG emissions from Japan. For a reliable national-scale assessment, however, databases on soil, weather, and farming management must be constructed at a national scale, as these factors are widely variable between regions in Japan.

Validation of the DNDC-Rice model by using CH4and N2O flux data from rice cultivated in pots under alternate wetting and drying irrigation management

“…Li et al (2005) ) were well correlated (n ¼ 9, r 2 ¼ 0.93, p < 0.01), but the estimation errors ranged from À220% to þ28.6%. Fumoto et al (2010) tested the DNDC-Rice model by using data from a paddy field in Japan; they found simulated seasonal N 2 O emissions of 0.08 to 0.65 kg N ha…”

“…Fumoto et al (2008) validated the DNDC-Rice model by using CH 4 data collected from 12 sites in Japan and China; the gross simulated CH 4 emissions during the cropping period were 53% lower to 162% higher than the observed values. Fumoto et al (2010) also reported validation results by using data collected from three rice fields in Hokkaido, Japan, to assess the CH 4 mitigation potentials of alternative water regimes for rice fields; the simulated values were 50% lower to 169% higher than the observed values. Smakgahn et al (2009) validated the DNDC-Rice model by using CH 4 emission data from nine paddy fields in Thailand under a continuous flooding treatment; the simulated values were positively correlated with the observed values, although the simulation results included some uncertainty caused by the incorporation of rice straw in the field, as well as uncertainty in the reducible iron (Fe) content and rice root biomass.…”

“…Simulation of N 2 O fluxes by using the DNDC model or the DNDC-Rice model has been reported by Li et al (2005), Babu et al (2006), and Fumoto et al (2010). Li et al (2005) reported that the DNDC model was capable of estimating the seasonal magnitudes of N 2 O fluxes from paddy sites, although discrepancies existed for about 20% of their tested cases.…”

“…Babu et al (2006) also validated the model, by using data from Delhi and Ludhiana, India, and reported that the total observed and simulated seasonal N 2 O emissions were strongly correlated, although the daily fluxes revealed discrepancies and the simulated seasonal N 2 O emissions were 248% lower to 28.6% higher than the observed values. Fumoto et al (2010) validated the DNDC-Rice model by using data from Japan; they reported that the simulated N 2 O emissions were 66% lower to 265% higher than the observed values, and they hypothesized that modifying the model's prediction of the nitrogen release rate from nitrogen fertilizers would decrease the discrepancy.…”

“…The model was developed to simulate these emissions from upland fields (Li et al 1992(Li et al , 1994, and it was revised by Li et al (2004) to simulate the anaerobic biogeochemistry of paddy fields. Fumoto et al (2008) subsequently revised the model to improve its ability to estimate CH 4 emission from paddy fields under continuously flooding management, and the revised model (''DNDC-Rice''; Fumoto et al 2010), was validated by using CH 4 data from paddy fields in Thailand (Smakgahn et al 2009) and Japan and China (Fumoto et al 2008(Fumoto et al , 2010. Fumoto et al (2008) validated the DNDC-Rice model by using CH 4 data collected from 12 sites in Japan and China; the gross simulated CH 4 emissions during the cropping period were 53% lower to 162% higher than the observed values.…”