Changes in GHG Emissions Based on Irrigation Water Quality in Short-Term Incubated Agricultural Soil of the North China Plain

Abstract

A worsening water shortage is threatening the sustainable development of agriculture in the North China Plain (NCP). How to make effective use of inferior water resources and alleviate the impact of insufficient water resources on agricultural environments is one of the urgent problems in agricultural production. Although agriculture plays an important role in greenhouse gas (GHG) emissions, the effects of irrigation water quality on such emissions in the NCP are not clear. In this study, we used a short-term incubation experiment to test the effects of the irrigation water quality (underground water (UW), saline water (SW), and reclaimed water (RW)) and frequency (high (H) and low (L)) on regulating the soil GHG emissions of the NCP. The results indicated that RW treatment increased the CO₂ and N₂O emissions by 15.00% and 20.81%, respectively, and reduced the CH₄ uptake by 12.50% compared with the UW treatment. In addition, SW treatment decreased the CO₂ and N₂O emissions and CH₄ uptake by 35.18%, 40.27%, and 20.09% against UW treatment, respectively. The high-frequency water added to the soil significantly increased the GHG emissions for all water qualities applied. Compared with UW, the global warming potential was significantly increased by RW_H and RW_L with 26.48% and 14.5% and decreased by SW_H and SW_L with 32.13% and 43.9%, respectively. Compared with the increase brought by reclaimed water, changing irrigation water sources from conventional groundwater to saline water (4 g L⁻¹) will moderately reduce GHG emissions under the worsening water shortage conditions occurring in the NCP.

1. Introduction

The continued emissions of greenhouse gases (GHG) will cause further warming and long-lasting changes in all components of the climate system, increasing the likelihood of severe, pervasive, and irreversible impacts for humans and ecosystems [1]. Agriculture has been reported to release significant amounts of GHG, particularly CH₄ and N₂O [2,3]. Therefore, investigating the sources of GHG emissions from agricultural soil is important for estimating their impact on climate change and creating strategies for their mitigation.

It is generally known that the agriculture industry is the largest consumer of freshwater, with 70% of all freshwater withdrawals occurring through irrigation [4]. Irrigation, one of the most important practices in agricultural management, is an important measure for increasing grain production. Owing to extensive use of irrigation, China can feed 21% of the world’s population with only 6% of the world’s freshwater resources and 9% of its arable land [5].

We had reported that drip irrigation has the potential to mitigate the soil CO₂, N₂O, and CH₄ emissions from the winter wheat field. In drip irrigation, a soil-wetting pattern is constricted and keeps the larger portion of the soil profile and soil surface dry compared to flood irrigation [6]. Irrigation can strongly interfere with soil moisture, which is a major regulating element of GHG emission and escalates the GHG emission up to a maximum point and afterward emissions drop again. Drying and wetting periods also influenced CO₂ emissions. During rewetting, soil microbes start working from a dormant state in dry soil; therefore, they increase CO₂ development. CH₄ is released into the atmosphere in the anaerobic environment from flooded fields [7]. Soil water content is the key element of CH₄ emission/uptake due to soil diffusivity control by the soil moisture [8]. Although water management is a distinguished alternative to minimize GHG emission, it pays less attention to agricultural lands. So, optimizing irrigation methods and frequency is an effective technology for greenhouse gas reduction.

Currently, water shortages affect traditional irrigation management, and irrigation water shortage is one prominent factor shaping the issue of food security in some areas. The utilization of low-quality water for irrigation, including saline water (SW) and reclaimed water (RW), is a significant strategy for relieving the pressure of water shortages in future agricultural production [9]. Although low-quality water irrigation is a way to ensure a water supply for crop production, it brings salts, trace heavy metals, and nutrients in soils, influencing soil microbial communities and resulting in carbon (C) and nitrogen (N) cycle transformations, thereby indirectly affecting GHG emissions [10]. For instance, Wong et al. [11] found that an increased soil electric conductivity of 0.5 dS m⁻¹ to 10 dS m⁻¹ through the addition of high-concentration salt solutions could inhibit soil microbial activity, thus reducing CO₂ emissions. Zou et al. [12] reported that sewage irrigation significantly increases CH₄ and N₂O emissions from rice paddies in southeast China. Hence, understanding the effects of low-quality water on soil GHG emissions is central to accurately estimating the budget of regional GHG emissions under the context of water shortage aggravation.

The North China Plain (NCP), one of the largest regions of agricultural importance of China, is crucial for China’s food security [6,13]. Groundwater has been severely over-extracted, causing a prominent shortage of water resources in the NCP [5]. To the best of our knowledge, incubation experiments are recommended to test in the short-term the influence of different irrigation water quality on soil environment and GHG emissions compared to real farmland conditions. Therefore, we conducted a short-term incubation experiment to quantify the GHG emissions as affected by the quality of water. We tested two hypotheses: (1) reclaimed water will significantly increase the GHG emissions, whereas saline water will inhibit the GHG emissions, and (2) increasing water added frequency (simulation of irrigation) will increase their accumulation.

2. Materials and Methods

2.1. Soil Sampling and Different Types of Irrigation Water

The incubated soil was collected from a wheat/maize rotation field at the Xinxiang comprehensive experimental station of the Chinese Academy of Agricultural Sciences (35.08° N, 113.45° E, the elevation of 77 m), which is the core crops production area in the NCP. The soil (0–20 cm) was sieved through a 2-mm stainless steel sieve to remove visible roots and organic residue and stored in the cool room (4 °C). The soil type, soil nutrients, and soil physical properties are shown in Jha et al. [14]. Three types of contrasting irrigation water (underground water (UW), RW, and SW) were prepared for this incubation study. UW, the main irrigation water of the NCP, was extracted from the well near the sample site and stored in the cool room. RW was fetched from the Wei River near an outlet pipe of a sewage treatment plant and also stored in the cool room. Based on the chemical properties of the ground SW of this region, SW was adjusted using sea salt and mixed with deionized water (4 g L⁻¹). The properties of the three different types of irrigation water can be seen in

Table 1. The properties of different irrigation water used in this incubation study.

Water Types	pH	EC (μs cm−1)	Silicate (mg L−1)	COD (mg L−1)	SRP (μg L−1)	NH4+ (μg L−1)	NO2− (μg L−1)	NO3− (μg L−1)	DIN (μg L−1)
UW	8.96	150.8	7.44	0.4	3.54	28.92	37.14	30.11	96.18
RW	8.46	594	3.75	3.8	294.79	623.58	161.66	497.62	1282.85
SW	7.26	4580	0.26	0.4	4.15	29.03	3.47	9.98	42.48

The pH: pH of irrigation water; EC: electrical conductivity of irrigation water; Silicate: Silicate in irrigation water; COD: chemical oxygen demand of irrigation water; SRP: soluble reactive phosphorus of irrigation water; NH₄⁺: Ammonium in irrigation water; NO₂⁻: Nitrite in irrigation water; NO₃⁻: Nitrate in irrigation water; DIN: Dissolved inorganic nitrogen of irrigation water; UW: Underground water; RW: Reclaimed water; SW: Saline water.

.

The pH of water was determined by a pH meter with a combination electrode; electrical conductivity (EC) was conducted by a conductivity meter (Thermo Scientific Orion, Waltham, WA, USA). Water samples were filtered through a 0.45 μm membrane with a disposable syringe for subsequent analysis. Silicate was measured using the molybdate colorimetry methods. The concentration of the water chemical oxygen demand (COD) was measured using the potassium chromate method [15]. Concentrations of ammonium and soluble reactive phosphorus (SRP) were measured colorimetrically by an ultraviolet spectrophotometer at the wavelength of 420 and 700 nm, respectively (TU-1810D, Beijing, China) [16]. Nitrate and nitrite were analyzed using a Dionex Aquion™ ion chromatography with a Dionex IonPac™ AS11-HC column (250 mm × 4 mm) at 30 °C. The eluent was a 20 mmol/L KOH solution at a flow rate of 1.0 mL/min. DIN was measured with the persulfate digestion method.

2.2. Experiment Design

The influence of the quality and frequency of irrigation water on the soil GHG emissions were investigated using a 3 × 2 factorial design with the following factors: (1) water quality (UW, RW, and SW) and (2) water frequency (two (H) and one (L)) in a laboratory with thirty repetitions. In each sampling time (0, 0.5, 1, 2, 4, 6, 7, 8, 10, and 14 d), the incubated soils (six treatments with triplicate) were measured GHG emissions and destructed for soil properties analysis (i.e., inorganic nitrogen, soil pH and EC). For each bottle (the size of volume: 500 mL), 173.3 g of fresh soil (on an air-dried weight basis) was added to create a field bulk density of 1.55 g cm⁻³, and the bottles were then sealed with a parafilm. A pre-incubation at 25 °C was then applied for 3 days to stabilize the microbial activity. After the pre-incubation, all treatments received 10.5 mL of water on day 0, and then again on day 7 only in high-frequency treatment. The soil was incubated at 25 °C throughout the experiment and not sealed, except when placed in glass jars for gas emission measurements.

2.3. GHG Sampling

At each sampling time, the sample bottles were sealed with a modified cap, which is shown in Figure 1. Gas samples (20 mL) were collected immediately using an injection syringe after sealing and again 30 and 60 min later and stored in 12 mL evacuated septum-sealed glass vials (Exetainer, Labco Ltd., High Wycombe, UK) for GHG analysis. The gas samples were analyzed within 48 h using a gas chromatograph system (Shimadzu 2010 Plus, Shimadzu CO. Ltd., Kyoto, Japan). CO₂ and CH₄ concentrations were measured by the FID detector and an ECD detector for N₂O concentration. Gas concentrations were measured in ppm and converted to emission or production (mass of GHG per mass of soil) through a calibration equation calculated using a standard gas with known gas concentrations. The protocol described by Mehmood et al. [17] was used.

$$\mathrm{J}=\frac{\mathrm{dc}}{\mathrm{dt}}\times \frac{\mathrm{M}}{{\mathrm{V}}_{\mathrm{o}}}\times \frac{\mathrm{P}}{{\mathrm{P}}_{\mathrm{o}}}\times \frac{{\mathrm{T}}_{\mathrm{o}}}{\mathrm{T}}\mathrm{H}$$

(1)

where J is emission flux (mg m⁻² h⁻¹), dc/dt is the slope of the linear regression of gas concentration at the time approaching zero, M is the mole mass of the measured gas (g mol⁻¹), P is the atmospheric pressure (P_a), T is the absolute temperature (K) inside the chamber; V_o, P_o, T_o is the volume (ml), pressure (P_a) and absolute temperature (K) at standard condition, H is chamber height above the soil surface (cm). Cumulative emissions of GHG were measured as follows:

$$\mathrm{CE}=\sum \left[(\frac{{\mathrm{F}}_{\mathrm{i}}+{\mathrm{F}}_{\mathrm{i}+1}}{2})\times {10}^{-3}\times \mathrm{T}\times 24\times 10\right]$$

(2)

where CE is the total cumulative GHG emissions (μg g⁻¹ dry soil), F_i and F_i+1 are the measured fluxes of two consecutive sampling dates (mg m⁻² h⁻¹), and T is sampling interval (days).

The global warming potential (GWP) of GHG emission for 100 years was estimated with the equation described by Yeboah et al. [18].

$$\mathrm{GWP}={\mathrm{GWP}}_{{\mathrm{CO}}_2}+{\mathrm{GWP}}_{{\mathrm{N}}_2\mathrm{O}}+{\mathrm{GWP}}_{{\mathrm{CH}}_4}={\mathrm{CO}}_2+298\times {\mathrm{N}}_2\mathrm{O}+25\times {\mathrm{CH}}_4$$

(3)

2.4. Soil Chemical Analysis

The soil pH and electrical conductivity were determined from soil–water extracts (ratio of 1:2.5) using a pH and conductivity meter (Thermo Scientific Orion, Beverly, MA, USA). Inorganic nitrogen (NH₄⁺ and NO₃⁻) was extracted from the soil samples by adding 80 mL of 1 M potassium chloride (KCl) to 20 g of incubated soil and shaking for 1 h at 25 °C, centrifuged at 4000 rpm for 10 min, and filtered through a Whatman ^#42 filter. The filtered solution was frozen at −20 °C until analysis. The soil extracts were measured using an automated ion flow injection analyzer (FIA, Lachat Instruments, Loveland, CO, USA). The gravimetric soil–water content was determined using an oven drying method at 105 °C for a period of at least 24 h.

2.5. Statistical Analysis

Repeated measures analyses of variance (ANOVA) was applied for testing the effects of the quality and application frequency of irrigation water on the soil GHG emissions based on the sampling date. Such effects on the fluxes of the soil GHG and the cumulative GHG emissions during the incubation period were investigated using a two-way ANOVA and a least standard difference (LSD) test. Significant differences were reported at p < 0.05. We further fitted a piecewise structural equation model (SEM) to estimate the direct and indirect effects of the quality and application frequency of irrigation water on the GWP. Details of the SEM analyses were previously described by Wang et al. [19]. The data were analyzed using SPSS 22.0 (SPSS Inc., Chicago, IL, USA) and plotted using SigmaPlot 12.5 (Systat Software Inc., San Jose, CA, USA), and SEM was applied using Amos 22.0 (SPSS Inc., Chicago, IL, USA).

3. Results

3.1. Effects of Different Irrigation Water Qualities on GHG Emissions

3.1.1. CO₂ Emission

The variations in the CO₂ emission during the incubation period are shown in Figure 2. One clear peak in the CO₂ emission appeared for the single water application treatments, and two peaks appeared for the two water application treatments. Compared with the single water application treatments, the CO₂ emission was significantly increased when more water was added, namely by 25.1% for UW treatment, 48.2% for RW treatment, and 55.1% for SW treatment (

Table 2. The mean emissions of CO₂, CH₄, and N₂O under different water quality and water frequencies in this incubation study (Mean ± SE, n = 3).

Treatments	CO2 Emission (μg g−1 Dry Soil h−1)	CH4 Emission (ng g−1 Dry Soil h−1)	N2O Emission (ng g−1 Dry Soil h−1)
UW_H	10.52 ± 0.76 b	−1.15 ± 0.087 b	3.68 ± 0.28 ab
UW_L	8.41 ± 0.93 c	−1.09 ± 0.11 ab	2.28 ± 0.24 c
RW_H	13.00 ± 0.90 a	−0.91 ± 0.12 ab	4.04 ± 0.32 a
RW_L	8.77 ± 0.91 c	−1.05 ± 0.11 ab	3.16 ± 0.24 b
SW_H	7.46 ± 0.61 d	−0.82 ± 0.090 a	2.07 ± 0.16 cd
SW_L	4.81 ± 0.57 e	−0.97 ± 0.042 a	1.49 ± 0.12 d
p-values of RM-ANOVA
Date (D)	<0.001	<0.001	<0.001
Water quality (WQ)	<0.001	0.002	<0.001
Water Frequency (WF)	<0.001	0.29	<0.001
WQ × D	<0.001	0.057	0.001
WF × D	<0.001	0.12	<0.001
WQ × WF	0.001	0.26	0.06
WQ × WF × D	<0.001	0.28	0.096

UW_H: Underground water added with high level; UW_L: Underground water added with low level; RW_H: Reclaimed water added with high level; RW_L: Reclaimed water added with low level; SW_H: Saline water added with high level; SW_L: Saline water added with low level. Different letters denote significant differences (p < 0.05) between treatments by LSD. Bold letter: significantly different.

). The mean soil CO₂ emission after RW_H treatment was 13.00 μg g⁻¹ dry soil h⁻¹, which is 23.6% and 74.3% significantly higher than those after the UW_H and SW_H treatments, respectively; in addition, UW_H treatment showed significantly higher CO₂ emission with 41.0% than SW_H treatment. The lowest CO₂ emission was observed for the SW_L treatment (4.81 μg g⁻¹ dry soil h⁻¹), which were 42.8% and 45.2% significantly lower than those of the UW_L and RW_L treatments, respectively. According to the repeated measure analyses, we found that the water quality, water frequency, sampling time, and interaction among them significantly affected the CO₂ emission (Table 2). The cumulative CO₂ emission was significantly increased by the second water application for all three water quality treatments (all at p < 0.01) (Figure 3 and Figure 4). With a change in the water quality, we observed the lowest cumulative CO₂ emission (2.41 mg g⁻¹ dry soil) for SW_H treatment as compared with UW_H and RW_H treatments (3.54 and 4.50 mg g⁻¹ dry soil) and remarkably higher cumulative CO₂ emission in RW_H than in UW_H treatment. We also found the same relationship in one water application treatment, that the cumulative CO₂ emission in RW_L treatment was significantly higher than UW_L and SW_L treatments, and the cumulative emission of CO₂ in UW was still higher than in SW treatment (Figure 4).

3.1.2. CH₄ Uptake

All treatments were revealed to significantly sink the CH₄ into the soil throughout the incubation period, except on day 7 for RW_H treatment (Figure 2). The CH₄ uptake showed highly irregular patterns during the entire incubation period, and no significant peaks were observed. The maximum CH₄ emission was −0.82 ng CH₄ g⁻¹ dry soil h⁻¹, which was shown for SW_H, whereas the minimum mean CH₄ emission was −1.15 ng CH₄ g⁻¹ dry soil h⁻¹, which occurred in UW_H and is significantly lower than in SW_H and SW_L treatments and not obviously different with RW_H, RW_L and UW_L treatments, respectively. The water quality and sampling date were shown to affect the mean CH₄ uptake significantly (Table 2). Compared with one water application, adding more water to the soil significantly increased the cumulative CH₄ uptake (21.6%) for UW and reduced the cumulative CH₄ uptake by 9.7% and 6.0% for the RW and SW treatments, respectively. The cumulative CH₄ uptake was significantly influenced by the water quality in the high water application treatments only (Figure 3 and Figure 4). Compared with UW_H, RW_H and SW_H significantly decreased the cumulative CH₄ uptake by 25.9% and 35.3%, respectively.

3.1.3. N₂O Emission

We observed that the variation in the N₂O emission did not show a clear pattern as compared with the CO₂ emission (Figure 2). The mean soil N₂O emission for UW_H, RW_H, and SW_H was 3.68, 4.04, and 2.07 ng g⁻¹ dry soil h⁻¹, which are 38.0%, 22%, and 28.0% higher than those for UW_L, RW_L, and SW_L, respectively (Table 2). Compared with UW_H, the mean N₂O emission of SW_H significantly decreased by 43.8%, and those of RW_H showed an unclear increase of 9.8%. The mean N₂O emission was 2.28 ng g⁻¹ dry soil h⁻¹ for UW_L treatment, which was significantly lower than that for RW_L treatment (27.8%) and remarkably higher than that for SW_L treatment (53.0%). In addition, water quality, irrigation frequency, sampling date, and their interactions affected N₂O emission as shown in Table 2. The water quality and irrigation frequency significantly affected the cumulative soil N₂O emission during the incubation period (Figure 3 and Figure 4). Compared with the low water application treatment, high water treatment can significantly stimulate an increase in N₂O emissions, namely by 66.0% in UW and 33.2% in RW. Compared with UW treatment, RW treatment significantly increased the cumulative soil N₂O emission by 17.0% for two water applications and by 45.8% for one water application, whereas SW significantly decreased the cumulative soil N₂O emissions by 36.8% for the former.

3.2. Changes in GWP

The maximum GWP was 4.92 CO₂-eq, which appeared in RW_H, and the minimum was 1.47 CO₂-eq, which was shown in SW_L. The high water application treatment significantly increased the GWP for all types of water (Figure 4). Compared with the addition of UW, the GWP significantly increased by 26.48% and 14.5% in RW_H and RW_L treatments and decreased by 32.13% and 43.9% in SW_H and SW_L treatments, respectively.

We conducted SEM analyses to understand the relationship between the GWP and environment factors (Figure 5). The results of the SEM analysis showed that the water quality significantly influenced the inorganic nitrogen content of the soil (R² = 0.34, p < 0.001 for the NH₄⁺ concentration, and R² = −0.20, p < 0.05 for the soil NO₃⁻ concentration) and EC (R² = 0.67, p < 0.001), whereas the soil moisture (R² = 0.36, p < 0.001) increased significantly based on the water application period (Figure 5). The soil pH showed a remarkable positive relationship with the soil moisture (R² = 0.23, p < 0.01) and NH₄⁺ concentration (R² = 0.19, p < 0.05) and a significantly negative relationship with the EC (R² = −0.41, p < 0.001). The GWP showed a significantly positive relationship with the inorganic nitrogen content of the soil (R² = 0.10, p < 0.05 for NH₄⁺ and R² = 0.16, p < 0.01 for NO₃⁻). The soil moisture was the key positive driver of the GWP (R² = 0.84, p < 0.001), which was also negatively affected by the soil EC (R² = −0.37, p < 0.001).

4. Discussion

4.1. Effects of Irrigation Water Quality and Frequency on GHG Emission

4.1.1. CO₂ Emission

Compared with UW treatment, RW significantly increased the CO₂ emission of the soil. Several studies concluded that there are benefits of using treated wastewater on irrigation such as increasing production, nutrients on soil, and crop yield [20,21]. Treated wastewater had at least two times more nutrients than drinking water [22]. Guo et al. [23] showed that the nutrients contained in RW can provide soil microorganisms not only an appropriate water environment but also abundant substrates for microbial metabolism and promote microbial reproduction, thereby increasing soil respiration. In addition, a suitable soil moisture content can create a befitting growth environment for microbial growth, promote microbial metabolism and reproduction, and significantly stimulate microorganism respiration [24]. Irrigated more water can reduce soil WFPS, lead to trace gas accumulating in the soil, and then result in huge CO₂ emission [25]. Muhr et al. [26] showed that drying–wetting alternation can promote CO₂ emissions from the soil. Our study found that such CO₂ emissions after twice water applications were significantly higher than after one water application. The significant positive correlation between cumulative CO₂ emissions and soil moisture was found in all water quality treatments. Salinity is one of the most important factors in affecting gas production [27]. The significant negative relationship between CO₂ emission and soil salinity in incubated soil was found in this study. The addition of SW reduces the soil CO₂ emissions because of the inhibition effects of the salt ions on the microbial activity of the soil [28,29]. The adverse effects of salinity, such as ion toxicity (Na⁺ specifically) [30], osmotic stress [31,32], or their cooperation [10] could inhibit the growth and activity of heterotrophic soil microorganisms and thus reduce CO₂ emission. The extracellular enzyme activity data which we measured from the saline water irrigated soil also support this view by Liang et al. [33].

4.1.2. CH₄ Uptake

The results indicated that non-flooded agricultural soil shows a general sinking of the CH₄. The roles of the soil microbial properties are the least understood factors controlling CH₄ uptake [34]. Numerous previous studies have demonstrated the influences of high saline-alkaline levels on CH₄ production, oxidation, and transport [28,35,36]. The inhibition of CH₄ uptake of soil with a higher saline amount was shown to be principally due to the low specific activity of a single methanotroph species such as Methylocella [28,37]. In our study, the increased soil salinity inhibited the CH₄ uptake. We also found that reclaimed water can reduce soil CH₄ uptake, although not statistically significantly, compared with UW_H and UW_L treatments. Reclaimed water added also brings some disadvantages for microorganisms associated with CH₄ production and consumption, such as slight soil salinization and high concentration of inorganic nitrogen ion. This may be due to types of ions inhibiting the microbial activity, which influenced the CH₄ production or consumption [24].

4.1.3. N₂O Emission

With the increase in soil moisture, the N₂O in the soil produced through nitrification and denitrification will increase [38,39]. A significant positive correlation between N₂O emissions and soil moisture was observed in our study, which is consistent with most previous research results [17,39]. Drying–wetting alternation is another major factor affecting the N₂O emissions [40]. Some research results showed that drying–wetting alternation increased the amount of microbial death, destroyed the interaction between soil environment and organic matter, and increased the rate of nitrogen mineralization in soil, and the amount of nitrification and denitrification in soil was significantly higher than that in long-term moist or dry soil [41]. Although we found that two water applications increased the N₂O emissions of the soil compared with one water application. our experiments should be repeated with alternate wetting and drying in future research for better explaining the effects of drying–wetting alternation on N₂O emission. We found that the addition of SW had a significantly negative effect on N₂O emissions in the soil. The addition of exogenous salts into the soil affects the microbial activity, leading to a suppression of the N₂O production in the soil through nitrification and denitrification [28,42]. The addition of RW can increase the N₂O emissions of the soil, which may occur because the reclaimed water can carry organic substances and mineral elements into the soil, provide substrates for a microbial metabolism and nitrogen transform of the soil, and promote microbial activity [43,44].

4.2. Relationship between Environmental Factors and GWP

Soil moisture showed a significant positive correlation with the GWP in this incubation experiment (Figure 5). A significant negative correlation between the GWP and soil salinity was observed with the addition of exogenous salt treatment. We believe that moderate SW irrigation can not only produce reliable yields [45] but also partly reduce GHG emissions [46]. It was indicated that the change in the amount of irrigation water might be one of the key drivers of the changes in the GHG emissions in the soil in the NCP, whereas a change in the quality of the irrigation water results in a positive/negative impact on the GHG emissions.

5. Conclusions

This study was carried out to compare the effects of different qualities of water on GHG emissions in the NCP using a short-term incubation experiment. The soil incubation experiment is used to evaluate the response of observation indexes under controlled conditions for providing theoretical support to validate the same treatments under real farmland conditions. An accurate assessment of the changes in soil GHG emissions after a change in water quality in this region will be helpful under the current conditions of worsening water scarcity. We found that water qualities have an impact on soil GHG emissions. The frequency of water application also has a clear influence on the emissions of CO₂ and N₂O under all three types of water application but showed no significant impact on CH₄ uptake. Compared with UW treatment, RW treatment significantly increased the CO₂ and N₂O emissions and reduced the CH₄ uptake, whereas SW treatment significantly reduced the CO₂ and N₂O emissions and the CH₄ uptake. We believe that moderate saline water irrigation can partly reduce soil GHG emissions. The results help us to improve the knowledge on the effects of poor-quality water application such as saline water on GHG emissions and also to evaluate the environmental effect on inferior water applications.