Intermittent flooding lowers the impact of elevated atmospheric CO₂ on CH₄ emissions from rice paddies

Highlights

• The first study to compare the effect of elevated CO₂ on GHG emissions under CF and IF.

• Elevated CO₂ stimulated CH₄ emissions under CF, but had no effect under IF.

• Elevated CO₂ did not affect N₂O emissions under CF and IF.

• Elevated CO₂ increased the abundance of methanogens under CF only.

• Current estimates of CO₂ effects on CH₄ emissions from rice paddies may be too high.

Abstract

Atmospheric CO₂ concentrations and water management practices both affect greenhouse gas (GHG) emissions from rice paddies, but interactive effects between these two factors are still unknown. Here, we show the first study to compare the impact of elevated atmospheric CO₂ (eCO₂) on GHG emissions under continuously flooded irrigation (CF) and under intermittently flooded (IF) conditions. Elevated CO₂ stimulated CH₄ emissions under CF by 50% in a field experiment and by 46% in a pot experiment, but it had no effect under IF in both experiments. Elevated CO₂ had no effect on N₂O emissions in either the field or pot experiment. Rice root biomass, aboveground biomass and grain yield increased with eCO₂, but were not affected by water management. Elevated CO₂ only stimulated the abundance of methanogens under CF, suggesting that increased soil O₂ availability with IF limited methanogenic activity under eCO₂. Our findings suggest that estimates of CH₄ emissions from global rice agriculture with eCO₂ need to account for recent changes in water management.

Introduction

By burning fossil fuels, logging forests, and changing land use in other ways, humans are causing a buildup of CO₂ in the atmosphere (IPCC, 2021). Elevated CO₂ concentrations (eCO₂) can influence soil abiotic and biotic conditions, e.g., soil carbon (C) availability, nitrogen (N) availability and microbial activity (van Groenigen et al., 2014, Pei et al., 2020, Wu et al., 2021). In turn, these changes can affect methane (CH₄) emissions and nitrous oxide (N₂O) emissions (Inubushi et al., 2003, van Groenigen et al., 2011, van Groenigen et al., 2013, Liu et al., 2018). As a major source of human caused GHG emissions, rice paddies account for ~9% anthropogenic CH₄ (IPCC, 2021) and ~11% cropland N₂O emissions (US-EPA, 2006).

Methanogenic archaea produce CH₄ under anaerobic conditions (Conrad, 2007, Hou et al., 2022), and their activity is affected by soil C and O₂ availability. Exudates of roots are a major substrate for methanogenesis (Conrad, 2007). Because eCO₂ generally increases rice root biomass, it often stimulates CH₄ emissions as well (van Groenigen et al., 2013). Agronomic practices in rice fields strongly affect CH₄ emissions as well (Liao et al., 2021). For example, water saving practices such as intermittent flooding (IF) strongly reduce CH₄ emissions relative to continuously flooded (CF) irrigation practices by increasing O₂ availability (Meijide et al., 2017). The last few decades have witnessed a shift in water management of rice fields from CF to IF practices (Yan et al., 2009, Lampayan et al., 2015). However, whether the response of CH₄ emissions to eCO₂ depend on water management is still unclear. This is because studies assessing the effect of eCO₂ on CH₄ emissions have so far been conducted almost exclusively under CF conditions, and possible interactive effects between eCO₂ and water management have not been studied directly.

Microbial nitrification and denitrification are the main processes to produce soil N₂O (Bouwman, 1998), and both these processes are influenced by soil C and N availability (Jiang et al., 2016, Wang et al., 2021). Soil C availability is often increased by eCO₂, thereby facilitating N₂O production (van Groenigen et al., 2011). A recent global meta-analysis suggests that eCO₂ stimulates soil N₂O emissions by 4.6% (Liu et al., 2018). However, only 10% of the observations in this analysis were from rice paddies, and results from primary studies are inconsistent (Wang et al., 2018a, Sun et al., 2018). On one hand, eCO₂ can decrease N₂O production by increasing N uptake of plant and reducing the amount of soil N available for denitrification (Sun et al., 2018). On the other hand, eCO₂ can increase soil C availability, which may result in a higher N₂O production (Wang et al., 2018a). As with CH₄, it is unclear whether the response of N₂O emissions to eCO₂ depends on water management.

Elevated CO₂ typically increases soil C inputs, thereby providing substrate for methane producing microbes (van Groenigen et al., 2013). However, IF can reduce the effect of soil C availability on CH₄ emissions from paddies (Jiang et al., 2019a). Moreover, compared to CF, IF often increases soil O₂ availability which can mitigate CH₄ emissions and stimulate N₂O emissions from paddies (Feng et al., 2013, Jiang et al., 2019a). Thus, we hypothesized that 1) IF lowers the response of GHG emissions to eCO₂, and 2) the effects of CO₂ concentrations and water management on CH₄ emissions were mediated through methanogens and methanotrophs in the rhizosphere. To test these hypotheses, we conducted a field experiment to assess the effect of eCO₂ on GHG emissions and rice yield under CF and IF. We used a free air carbon dioxide enrichment system (FACE), allowing us to study the effect of eCO₂ under realistic environmental conditions. To identify the soil processes underlying treatment effects, we also conducted a pot experiment in walk-in chambers using the same treatment combinations. To the best of our knowledge, this is the first study to show the interactive effects of CO₂ concentrations and water practices on GHG emissions from paddies.