Greenhouse gas emissions, irrigation water use, and arsenic concentrations; a common thread in rice water management

Authors: ANDERS, MERLE - University Of Arkansas; LINQUIST, BRUCE - University Of California; ADVIENTO-BORBE, M.A. - University Of California; McClung, Anna; VAN KESSEL, C - University Of California; Chaney, Rufus; HENRY, C.G. - University Of Arkansas; KERR, S. - University Of Arkansas; HENDRIX, D. - University Of Arkansas

Submitted to: Rice Technical Working Group Meeting Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 12/2/2012
Publication Date: 12/15/2014
Citation: Anders, M.M., Linquist, B.A., Adviento-Borbe, M.A., McClung, A.M., van Kessel, C., Henry, C., Chaney, R. L., Kerr, S., Hendrix, D. 2014. Greenhouse gas emissions, irrigation water use, and arsenic concentrations; a common thread in rice water management. Proc. 35th Rice Tech. Work. Group Meet., New Orleans, LA, p.115-116. Feb. 18-21, 2014. CDROM.

Interpretive Summary: Rice is commonly grown under field flooded conditions as a means to control weeds and optimize yield potential. However, research has shown that the oxygen-depleted soil conditions that result from flooded fields result in increased methane (greenhouse gas) emissions and increased availability of naturally occurring soil arsenic for plant uptake in rice fields. Because competition for water resources is increasing with greater urban growth, it is important to minimize irrigation use in rice fields while maintaining high productivity. A multi-year study was conducted to determine the impact of water saving practices on rice yields, grain arsenic, and greenhouse gas (GHG) emissions. The study consisted of six water treatments that ranged in soil wetness throughout the season. These included the standard flood system, an alternate wetting and drying system, and a system where the rice was grown in elevated beds and irrigated via furrows. Results demonstrated that with decreasing water application there was a decrease in yield, grain arsenic, and GHG emissions. However, management methods were identified that reduced water use, grain arsenic, and GHG emissions but were not significantly different in yield as compared to the flooded system. In total, these results illustrate a strong relationship between water use efficiency, GHG emissions, and grain arsenic content and suggest that adopting water saving irrigation approaches can address environmental, production, and grain quality concerns.

Technical Abstract: Rice has historically been grown as a flooded crop in the United States. As competition for water resources has grown, there is interest in reducing water use in rice production so as to maintain a viable and sustainable rice industry into the future. An irrigation study was established in 2011 at the University of Arkansas, Rice Research and Extension Center to evaluate the effect of reduced water use on rice production, water efficiency, greenhouse gas emissions, and grain arsenic content. The study consisted of six water treatments, from wettest to driest: 1) flood, 2) AWD/40-Flood, 3) AWD/60, 4) AWD/40, 5) RR/60, and 6) RR/40. Alternate wetting and drying (AWD) treatments were flooded to a 10-cm depth at the V4-V5 growth stage and held at the 10-cm water depth for 10 days after which they were allowed to dry to the targeted percent of soil water holding capacity ( /60 or /40). The AWD/40-flood treatment was managed as AWD until the plants reached the R0-R1 growth stage then flooded until time of harvest when they were drained. Row-watered (RR) treatments were planted in a bed-furrow system with water applied in the furrow whenever the designated soil water holding capacity was reached. Three hybrid varieties (CLXL745, XL723, XL753) were used over the four years. Four replications were used with each variety planted into 4.24 x 30.5-m plots. Plots were planted following a rice-soybean (R-S) rotation all three years with an additional continuous rice rotation treatment added in 2013. Average irrigation water use across all treatments and years was 6,396-m3 ha-1 while grain yields averaged 9,362 kg ha-1 with an irrigation water efficiency 712 m3 water Mg-1 grain. For the rice-soybean rotation, grain yields averaged over three years were 10,703 kg ha-1 for the standard flood treatment followed by 10,543, 10,119, 9,828, 7,362, and 7,116 kg ha-1 for the AWD/40-Flood, AWD/60, AWD/40, RR/60, and RR/40 treatments, respectively. There were no significant differences in grain yields between the flood and AWD treatments. Irrigation water applied to each treatment ranged from 8,087 m3 for the flooded treatment to 6,031, 5,574, 4,527, 8,675, and 6,636 for the AWD/40- Flood, AWD/60, AWD/40, RR/60, and RR/40 treatments, respectively. Both of the RR treatments used nearly the same amount of irrigation water as the flood treatment but had significantly lower grain yields. Reductions in water use from the flooded treatment were 25, 31, and 44% for the AWD/40-flood, AWD/60, and AWD/40 treatments respectively. There was a corresponding increase in water efficiency measured as m3 irrigation water Mg-1 rice for the AWD/40-flood, AWD/60, and AWD/40 over the flooded treatment of 31, 36, and 62%, respectively. Greenhouse gas measurements were collected on CLXL745 in the flood, AWD/40-flood, AWD60, and AWD/40 treatments during the 2012 and 2013 seasons. Total global warming potential (GWP) (methane plus nitrous oxide) averaged 2,878 kg CO2 eq ha-1 season-1 for the flooded R-S rotation and 4,804 kg CO2 eq ha-1 season in the continuous rice rotation. These values were reduced in the R-S rotation to 1,678, 295, and 435 kg CO2 ec ha-1 season-1 for the AWD/40-flood, AWD/60, and AWD/40 treatments, respectively. For the continuous rice rotation, these values were 2,397, 485, and 611. In all AWD treatments, nitrous oxide emissions were greater than in the flood with these values more than offset by reductions in methane emissions. Methane emissions were greater and nitrous oxide emissions lower in the continuous rice rotation compared to the rice-soybean rotation. Whole-grain arsenic levels were measured from all 2011 and 2012 treatments. Arsenic content, over all treatments was similar for XL753 and CLXL745 while it was significantly lower for XL723. Arsenic levels decreased from 0.24 mg kg-1 in the flood treatment to 0.02 mg kg-1 in the RR/40 treatment. Th