The CO₂ production from agriculture accounts for 23% of anthropogenic greenhouse gas emissions (Smith et al., 2007). If intercropping with legumes or cereals is introduced (Hu et al., 2016; Cui et al., 2019), this emission can be substantially reduced (Gan et al., 2011), because such systems have a high potential to sequester soil organic carbon (SOC) by reducing soil respiration (Chai et al., 2014; Hu et al., 2015). Namely, pea - maize or wheat - maize intercropping reduces soil respiration from maize strips during the growing season (Li et al., 2001; Qin et al., 2013); pea - oat intercropping reduces CO₂ emission during the period with higher precipitation; barley - pea intercropping also results in 10% higher soil C sequestration than barley monoculture (Chapagain and Riseman, 2014).

Agricultural ecosystems increase SOC sequestration up to 4% if intercropping with legumes or cereals is introduced compared to crop monoculture (Cong et al., 2015). This is associated with: i) regulation of crop growth by intercropping, and thus, reduction of root exudates and following CO₂ efflux (Dyer et al., 2012; Qin et al., 2013); ii) changes in the composition of microbial community structure and decrease in biomass and functional diversity under one of the species (Zhou et al., 2019b) or stimulation of soil microbial biomass growth under intercropping (Latati et al., 2017), and iii) the regulation of soil CO₂ efflux by plant species composition, which may be suppressed by reduce of net primary production because of water availability shortage (Zhou et al., 2019a). Therefore, various intercropping systems (e.g., rapeseed (pea) - maize, wheat-soybean (maize)) can have the potential to reduce the soil CO₂ efflux from farmland (Li et al., 2001; Chai et al., 2014).

Soil hydrothermal factors play an important role in CO₂ efflux from farmland (Hursh et al., 2017). Temperature is the most important factor affecting soil respiration, and there is a positive relationship between them (Hursh et al., 2017). Plant community composition affects soil respiration's temperature sensitivity. The temperature sensitivity of soil respiration is affected by plant community composition (Mauritz and Lipson, 2021). Higher temperatures are generally expected to enhance soil C losses due to increased soil decomposition (Crowther et al., 2016; Bond-Lamberty et al., 2018). The sensitivity of respiration to temperature changes with the soil water content, substrate availability, and species composition (Geist and Lambin, 2004). The combined factors of soil temperature and moisture would better predict soil respiration (Feng et al., 2018). Soil moisture also strongly affects the changes in SOM (Jassal et al., 2008; Wang et al., 2014), respiration (Bouma et al., 1997), and microbial activity (Hallett and Young, 1999; Drenovsky et al., 2004), and the synergistic relationship between soil respiration and moisture can greatly increase or decrease the decomposition rate of SOM, depending on the direction of moisture change (Sierra et al., 2015). However, it is not yet clear how hydrothermal factors will be changed in the intercropping of the legume with brassica and what the response of CO₂ emission will be.

Milk vetch (Astragalus sinicus L.) intercropping with rapeseed (Brassica napus L.) can enhance farmland productivity (Zhou et al., 2018), change the microbial community structure and decrease microbial biomass and functional activity in the rapeseed rhizosphere (Zhou et al., 2019b). To verify the potential of intercropping to reduce the CO₂ emission, the soil respiration from milk vetch - rapeseed system was monitored together with the temperature and moisture for the entire development of crops and compared to the monoculture. Considering that intercropping can improve water use efficiency (Ren et al., 2017), and can decrease soil temperature (Gong et al., 2019) because of the large soil surface cover, it was hypothesized that soil CO₂ efflux would be lower compared to the monocultures. The objective of the experiment was to explore the effects of intercropping on soil respiration and to determine the relationship between soil respiration and hydrothermal factors.

The high CO₂ efflux from Intercrop at the early growth stages are explained by low competition for the soil resources between species (Mushagalusa et al., 2008), and intensive plant growth. Intercropping with RS may decrease the impact of MV on soil respiration at the later stage ( Figure 3 ) i) due to inhibition of MV growth (Zhou et al., 2018), thus, decreasing the release of root exudates from MV into the soil (Liu et al., 2013) and ii) due to suppress microbial biomass and activity in the RS rhizosphere in the presence of MV (Zhou et al., 2019b), and thus, reducing rhizosphere respiration (Blaise et al., 2021; Yang et al., 2021). The RS and Intercrop had similar CO₂ efflux rates after the seedling stage, whereas maximum values were found from MV ( Figure 3 ). This is explained by higher root exudation under N₂ fixing plant species (Zúñiga-Feest et al., 2018) and thus, higher rhizosphere respiration (Becker and Holz, 2021). This showed the positive effect of intercropping on the SOM accumulation because the CO₂ efflux was reduced compared to N-fixing plant monoculture.

Seasonal and interannual variations of CO₂ emission are related to the soil temperature and moisture (Suseela et al., 2012) because these parameters directly regulate microbial biomass and activity (Zhou et al., 2019b). Sensitivity of soil respiration to temperature was lower in Intercrop than in MV and was independent of temperature changes ( Figure 6 ). Intercrop also decreased soil moisture, especially at the seedling and stem elongation stages, compared to the monoculture ( Figure 6 ). This explains why in the Intercrop the growth of MV was suppressed (Zhou et al., 2018). High CO₂ emission rate under MV observed at the flowering and podding stages can be explained more by the high soil moisture at the initial plant growth stages, then by the temperature. In contrast, the decreased CO₂ efflux rate under Intercrop at later stages can be directly affected by variations in moisture or temperature, and probably other factors, such as changes in the composition of microbial communities (Zhou et al., 2019b), a decrease of rhizosphere C flux (Suseela and Dukes, 2013) and plant species-species interactions (Dijkstra et al., 2010).

The similar trends in CO₂ efflux and temperature suggested that soil temperature was still the most important factor affecting soil respiration ( Supplementary Figure 1 ). Soil respiration was positively correlated with soil temperature, and a negative correlation of soil respiration to soil moisture was also found, especially in Intercrop ( Figure 5 ). It illustrated that the effect of soil moisture on soil respiration is more important in intercropping systems than in monoculture. Soil respiration responses to increases in temperature are constrained by soil moisture (Conant et al., 2004). If the soil moisture is often lower than the soil water holding capacity, the soil respiration cannot be high enough to reach the limiting point due to reduced oxygen diffusion into the soil and inhibited substrate decomposition (Tang et al., 2006; Hu et al., 2017). Furthermore, as microbial respiration is linearly related to soil water content and log-linearly related to water potential (Cook and Orchard, 2008), the decreased soil moisture will directly lead to the decrease of soil microbial biomass and functional activity (Zhou et al., 2019b). Thus, the variation in soil moisture can be the real reason for inhibited soil respiration by milk vetch intercropping with rapeseed.

Soil CO₂ efflux from Intercrop was 1.4 times higher than from mean of MV and RS at seedling and stem elongation stages, however, soil CO₂ efflux from MV was 1.6 times higher than Intercrop after that. Cultivation of legume in monoculture, although there is a positive contribution to soil N balance, can promote SOM losses compared to Brassica. In contrast, intercropping of Legume with Brassica is a beneficial agricultural practice to reduce the rate of CO₂ efflux, which is related to the flowering and podding stages of plant growth. The sensitivity of soil respiration to temperature decreased in Intercrop, in which the variation of soil moisture was the primary factor to inhibit soil respiration. Therefore, milk vetch-rapeseed intercropping could be a potential approach to produce low CO₂ emissions from farmland, however soil moisture should be adequately regulated so that agricultural intercropping systems can be well adaptable in the face of frequent global droughts.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by QZ, AG and JC. The first draft of the manuscript was written by QZ and all authors commented on previous versions of the manuscript. All authors contributed to the article and approved the submitted version.