The Rate and Duration of Nitrogen Addition Influence the Response of Soil Heterotrophic Respiration to Nitrogen in Cropping Systems

Abstract

Under today’s climate warming, mitigating the risks of soil organic carbon (SOC) decomposition in cropping systems is critical to maintain carbon sequestration. This study posits that nitrogen addition rate and duration are the key factors influencing the responses of soil heterotrophic respiration (Rh) to climate warming in a cropping system. Based on soil sampled from traditional-agriculture rice-growing regions in the Taihu Lake Basin in eastern China, this study aimed to clarify how nitrogen addition strategies affect soil Rh and its temperature sensitivity (Q₁₀) and to explore the underlying mechanisms of the changes in soil environment that influence carbon emissions under nitrogen addition through the Rh pathway. The results demonstrate that, with the increasing duration of nitrogen addition, soil Rh and its Q₁₀ were initially increased but subsequently suppressed, and the inhibitory effect on soil Rh became apparent after six years of continuous addition. Further analysis revealed that decreases in C/N, pH, and extractable organic nitrogen and increases in mineral nitrogen are the primary factors suppressing soil Rh. These findings indicate that an optimized nitrogen addition strategy tailored to specific crops could achieve profitable crop yields while effectively mitigating the promoting effect of climate warming on SOC decomposition.

1. Introduction

As an important carbon storage reservoir on planet Earth, terrestrial ecosystems‘ carbon release mainly manifests through soil respiration [1]. While agricultural systems serve as the material supply frontline for ensuring human survival, they function not only as a major source of greenhouse gas emissions but also as a substantial carbon sink system [2,3]. Therefore, mitigating the current carbon loss caused by intensive farming and enhancing the carbon sink capacity of agricultural systems are crucial for restoring soil functions, maintaining ecosystem stability, and safeguarding the sustainability of agricultural production [4,5]. Notably, humans interfere with agricultural systems more profoundly than with any other terrestrial ecosystem. This intensive human intervention not only enhances the differentiation of soil respiration intensity across diverse agricultural systems but also endows the soil respiration intensity within agricultural systems with controllability [6,7]. Among the various factors influencing soil respiration, nitrogen input stands out as a crucial factor in altering carbon cycle fluxes via this pathway [8,9]. Additionally, agricultural systems are characterized by annual crop biomass harvests driven by production goals, and nitrogen input from chemical fertilizers is a necessary measure to ensure product yields. Moreover, its input intensity (reaching 120–300 kg ha⁻¹ per single cultivation season for rice, wheat, and oil rapeseed in the Taihu Lake Basin [10]) is significantly higher than that of most other terrestrial ecosystems (their primary exogenous nitrogen input occurs through atmospheric deposition, typically below 30 kg ha⁻¹ [11]). These unique features make the response of soil respiration to various nitrogen addition strategies in agricultural systems distinct from that in other terrestrial ecosystems.

Regarding soil respiration (Rs), its response to nitrogen addition in agricultural systems is the outcome of the interaction between crop roots (an important subject of soil autotrophic respiration) [12] and soil microorganisms (an important subject of soil heterotrophic respiration (Rh)) [13,14]. The influence of nitrogen addition on soil autotrophic respiration via crop roots is regulated by multiple anthropogenic factors, including crop variety characteristics, interannual harvesting frequency, and crop residue disposal methods [12,15,16], thereby exhibiting a high degree of complexity and limited controllability. Soil Rh utilizes soil carbon as its consumption substrate and accounts for more than 60% of total soil respiration [17], which makes it crucial in determining the carbon output of an agricultural system. By altering the abiotic environment of the soil, nitrogen addition may influence multiple aspects of soil microorganisms and, further, influence the soil Rh they mediate. Firstly, the nitrogen addition from agricultural activities may alter the soil organic carbon composition, affecting the accumulation of recalcitrant organic carbon and the content of readily degradable organic carbon, thereby changing the quantity and form of the consumable substrates available to soil microorganisms. Secondly, nitrogen addition may modify the physiochemical conditions of the soil (such as the pH, moisture, and temperature conditions) [18,19] and its carbon and nitrogen contents [14,20,21] and induce morphological changes [9,22], which, in turn, can cause changes in multiple factors including microbial biomass [15] and alterations in the structure of related microbial communities [23,24,25] and enzyme activity intensity [16]. Thirdly, nitrogen addition may also affect nutrient supply availability for soil microorganisms, thereby influencing microbial metabolic activity and their capacity to respond to environmental changes. Regarding soil Rh, its Q₁₀ serves as a key indicator for measuring the risk of carbon loss. In the context of climate warming, studying the soil Rh’s Q₁₀ enables more accurate prediction of Rh intensity in future warmer environments, thereby facilitating improved estimations of carbon loss magnitude in an agricultural system. In summary, it is believed that, in an agricultural system, taking soil Rh and its Q₁₀ as the starting point, exploring the impact of nitrogen addition on carbon decomposition is of greater significance for achieving theoretical carbon regulation via nitrogen.

Nitrogen addition strategy encompasses three dimensions: nitrogen form, addition rate, and duration. Previous studies reported the promoting effect of nitrogen addition on soil Rh in grasslands, wetlands, and deserts, but the effect on agricultural systems was uncertain [7,26], with only a significant decrease in soil Rh observed in treatments with a high amount of nitrogen addition in wheat cultivation systems [27]. Moreover, a high nitrogen addition amount has also been reported to inhibit soil Rh in forest systems [28]. Furthermore, a study based on forest soils has found that the effect of nitrogen addition on soil Rs changes with addition duration, with a high addition amount causing the negative effects on soil Rs to occur earlier [28], and structural equation modeling has output the results of the negative effect of nitrogen addition duration on Rh [7,29]. These results all suggest that there may be an interaction effect between the rate and duration of nitrogen addition on soil Rh, but there are currently no relevant studies for agricultural systems. In this context, it is proposed that the relationship between nitrogen addition strategies and soil Rh in agricultural systems represents a promising issue requiring further investigation. It is, therefore, hypothesized that, in an agricultural system, a dynamic and potentially bidirectional response of soil Rh to nitrogen addition may occur depending on the nitrogen addition rate and duration. Meanwhile, in real life, the excessive application of nitrogen chemical fertilizers is a common problem for Chinese farmers nowadays [10]. They (especially conventional ones) tend to apply 50% to over 100% more nitrogen fertilizers than the current season’s crop requirements to secure their yield, rather than mitigating nitrogen losses to enhance fertilizer utilization efficiency [10]. Therefore, the exploration of the appropriate amount of nitrogen fertilizer and its dynamic adjustment on the time scale is key to balancing agricultural output and environmental impact [30]. While addressing the demands of agricultural production, if researchers could explore the impact patterns and main controlling factors of nitrogen addition strategies with respect to carbon decomposition, it would be possible to effectively increase the carbon pool of agricultural systems by adjusting the nitrogen input rate and duration.

Moreover, rice, as one of the most important food crops, plays a dual role of carbon source and carbon sink in the agroecosystem. The global total soil organic carbon (SOC) storage of rice farmlands is 18 Pg C, accounting for 1.2% of the global total soil SOC and 14.2% of the global farmland SOC [2]. China is the world’s largest producer of rice. Reducing carbon emissions from rice systems by controlling the soil Rh pathway, so as to further exploit the carbon sequestration potential of rice soil, plays an indispensable role in mitigating climate warming.

In this regard, this study utilized soil samples from traditional-agriculture rice-growing areas in the Taihu Lake Basin, eastern China, as the experimental subject and selected three test field sites with varying levels of nitrogen fertilizer application (the sites had been carrying out the preliminary exploration of the ‘chemical fertilizer reduction policy’ starting at different time points over the past 15 years). This study investigated the effects of annual-basis nitrogen addition strategies on soil Rh and its Q₁₀ from two dimensions: the application rate and duration of nitrogen addition. By measuring the samples’ soil Rh value under different-temperature experimental conditions, we compared the differences in the effects of different long-term nitrogen addition strategies on soil Rh and its Q₁₀. On this basis, short-term mineral nitrogen addition treatments were introduced to explore the temporal-scale response patterns of carbon release by soil Rh to nitrogen addition. Integrating the changing trends of soil-related factors, this study investigated the mechanistic pathways through which nitrogen addition influences carbon decomposition processes within agricultural cropping systems. The research findings advance our understanding of the mechanism of carbon regulation via nitrogen in agricultural cropping systems and provide a scientific theoretical foundation for both predicting the dynamic carbon-release patterns associated with soil Rh under long-term nitrogen addition conditions and enlarging the carbon sequestration pool in rice systems through optimized strategic nitrogen addition.

2. Materials and Methods

2.1. Field Site Description

Three field sites, located in Changzhou (CZ), Suzhou (SZ), and Wuxi (WX) in the Taihu Lake Basin (eastern China), were selected as the subjects of this study. All field sites experience a northern subtropical humid monsoon climate with a similar annual mean temperature of 15–16 °C and abundant precipitation ranging between 800 and 1100 mm. All field sites have rice soil but with different origins: in the case of CZ and SZ sites, a clay loam soil derived from yellow soil, while in WX, a clay soil derived from lacustrine deposit. The main nutrient properties of the soil are presented in

Table 1. Site information, nitrogen experiment design, and soil (0–20 cm) characteristics.

		1-Year Site	6-Year Site	11-Year Site
Site
Lat./log.		31°43′24″ N, 119°29′37″ E	31°8′22″ N, 120°31′4″ E	31°16′44″ N, 119°55′10″ E
City		CZ	SZ	WX
Long-term nitrogen fertilizer application rate (kg ha−1)
N0		0	0	0
N1		150	150	150
N2		210	210	200
N3		270	270	270
Short-term nitrogen addition rate (kg ha−1)
SN0		0
SN1		135
SN2		270
Soil (0~20 cm) characteristics
pH (2.5:1 water)	N0	6.32	5.95	5.38
	N1	6.38	5.81	5.38
	N2	6.35	5.74	5.43
	N3	6.37	5.73	5.35
Soil organic carbon (SOC, g kg−1)	N0	17.27	15.90	13.65
	N1	16.97	15.70	12.64
	N2	17.32	15.10	12.51
	N3	17.76	15.70	13.43
Total nitrogen (TN, g kg−1)	N0	1.68	1.36	1.41
	N1	1.74	1.27	1.31
	N2	1.87	1.30	1.32
	N3	1.85	1.34	1.45
Total phosphorus (TP, g kg−1)	N0	0.19	0.25	0.22
	N1	0.20	0.25	0.20
	N2	0.19	0.27	0.24
	N3	0.21	0.27	0.24
Labile organic carbon (LOC, g kg−1)	N0	2.51	3.59	1.98
	N1	2.74	3.39	2.21
	N2	2.73	3.66	2.14
	N3	2.72	3.71	2.12
Mineral nitrogen (MN, mg kg−1)	N0	19.60	12.86	22.42
	N1	20.82	16.59	21.57
	N2	28.12	22.03	23.15
	N3	20.19	24.32	22.99
Extracted organic nitrogen (EON, mg kg−1)	N0	7.01	20.10	16.72
	N1	15.61	21.68	14.85
	N2	20.22	17.89	12.62
	N3	25.01	15.41	9.78
Carbon/nitrogen ratio (C/N)	N0	10.30	11.69	9.67
	N1	9.75	12.36	9.66
	N2	9.28	11.62	9.47
	N3	9.58	11.72	9.24

.

2.2. Experimental Design

2.2.1. Experiment on Long-Term Nitrogen Fertilizer Application

These agricultural fields in the Taihu Lake Basin have been cultivated for decades or even centuries, with common crop-rotation systems including rice–wheat, rice–oilseed rape, and rice–fallow. Rice is transplanted in June and harvested in early November, followed by a winter season of either fallow or non-fallow. The selection of rotation crops is determined by individual farmers’ production strategies, which may be influenced by the guidance of the government’s year-by-year subsidy policies.

In order to minimize the confounding effects of differences in the annual total nitrogen input amount and forms brought about by crop rotation and various fertilizer types, all soil samples were collected from the rice–fallow rotation fields. All three field sites were under a rice–wheat rotation system before 2010. Subsequently, due to policy changes, they gradually adopted a rice–fallow system and remained as such in a yearly rotation. In the rice-growing season each year, 4 different long-term nitrogen addition rates were implemented as major parts of the experimental treatments at all three field sites, with chemical nitrogen fertilizer (urea) used exclusively. They were as follows: the farmers’ average nitrogen addition rate, N3 (270 kg ha⁻¹); the recommended nitrogen addition rate, N2 (200–210 kg ha⁻¹); the reference low nitrogen addition rate, N1 (150 kg ha⁻¹); and no nitrogen, N0 (0 kg ha⁻¹) (Table 1). Nitrogen fertilizers were applied exclusively during the rice-growing season, from June to August. All treatments varied solely in the nitrogen addition rate, whereas the rates and methods for phosphorus and potassium fertilizer remained constant across the 4 treatments.

In all three field sites, 3 or 4 independent replicate plots were established for each nitrogen treatment rate. Following the rice-harvest season in November, five soil samples were collected from the 0–20 cm plough layer of each plot. At the time of sampling, most aboveground biomass had been removed from all sites, and no straw had yet been returned; only crop residues and root remnants remained in the fields. The collected soil samples were thoroughly homogenized and sieved through a 2 mm mesh to remove stones and organic debris. The resulting composite samples, representative of each nitrogen treatment, were then stored at 4 °C for subsequent analysis and short-term nitrogen addition experiments.

By the time all the soil samples were collected in 2022, the N fertilizer experiments in rice growing at the three field sites had been ongoing for different lengths of time. Among the three sites, the 1-year site received nitrogen treatments for only 1 single rice-growing season (initiated in 2022, location: CZ), the 6-year site was subjected to four treatments for 6 rice-growing seasons (initiated in 2017, location: SZ), and the 11-year site had been carrying out experiments for 11 seasons (initiated in 2012, location: WX).

2.2.2. Short-Term Nitrogen Addition

Short-term nitrogen addition was set based on the soil samples with different long-term nitrogen fertilizer application rates and application durations. Short-term nitrogen addition to each soil sample occurred at three levels: SN0 (0 kg ha⁻¹), SN1 (equivalent to 135 kg ha⁻¹), and SN2 (equivalent to 270 kg ha⁻¹). Urea was dissolved in deionized water and applied to soil samples to achieve the designed short-term N addition, and then all soil samples were placed in a well-ventilated location and maintained at 60% of the saturated water content daily for 15 days of pre-incubation at 20 °C.

2.3. Soil Incubation and Soil Rh Analysis

After nitrogen addition and pre-incubation, soil samples, equivalent to 20 g dry weight, were incubated in 150 mL glass bottles under changing temperatures with three replicates. Incubation temperature varied from 4 to 28 °C, with a step width of 4 °C for soil heterotrophic respiration analysis. The rates of soil Rh at different temperatures were determined from the change in CO₂ concentration in headspace. The whole soil incubation period lasted for 8 days. A full description of the incubation system and measuring procedure was detailed by Chen [31].

2.4. Soil Carbon and Nitrogen Analysis

Two sets of soil samples were used for the detection of soil carbon and nitrogen indicators. Set1 was the soil samples collected from long-term nitrogen fertilizer application plot in the 3 field sites prior to short-term nitrogen addition; Set2 was the soil samples after short-term nitrogen addition and incubation. Soil total nitrogen (TN) and soil organic carbon (SOC) contents were analyzed based on Set1 soil samples. Moreover, soil labile organic carbon (LOC), extractable organic nitrogen (EON), mineral nitrogen (MN), and microbial biomass carbon (C_mic) were analyzed based on Set1 and Set2 soil samples. Parts of soil samples were oven-dried at 105 °C to a constant mass then ground to pass a 0.149 mm sieve for TN and SOC analysis using an elemental analyzer (Thermo Flash EA1112, Waltham, MA, USA) and LOC analysis using a KMnO₄ oxidized method detailed by Weil [32]. Fresh soil samples, stored at 4 °C, were used for C_mic analysis by a fumigation extraction method described by Vance [33]. For MN and EON analysis, fresh soil samples were extracted with 2 mol·L⁻¹ KCl solution; after that, MN (NH₄⁺-N + NO₃⁻-N) and extractable total nitrogen (ETN) in the filtrate were analyzed by an autoanalyzer (Traacs 800, Bran & Luebbe, Tokyo, Japan). EON was determined by the difference between ETN and MN.

2.5. Calculations

The relationship of the rate of soil Rh was described by an exponential function:

Rh₂₀ = a × e^bT

(1)

where Rh₂₀ is soil Rh rate (μg g⁻¹ soil h⁻¹); a is a fitting parameter referring to respiration rate at 20 °C; T is temperature (°C); and b is a parameter defining the temperature dependence of soil respiration. A reference respiration rate of 20 °C (Rh₂₀) was then estimated with fitted Equation (1). The Q₁₀ value, commonly referred to as Q₁₀ of soil Rh, was calculated using Equation (2) [34]:

Q₁₀ = e^10b

(2)

The metabolic quotient of soil microbes, q_mic, was calculated using Equation (3) [35]:

q_mic = Rh₂₀ ÷ C_mic

(3)

where C_mic was soil microbial biomass carbon.

The rate of soil Rh and its Q₁₀ values measured under variable temperature control under laboratory conditions are placed in Tables S1 and S2.

Due to the differences in soil backgrounds among the three rice fields, the effect sizes of Rh₂₀ and its Q₁₀ brought about by nitrogen addition were used for comparison. An N0 and an SN0 treatment were set to indicate the degree of change in Rh₂₀ or Q₁₀ based on the corresponding comparison processing. Effect size of Rh₂₀ or Q₁₀ under different long-term nitrogen fertilizer application levels (N1, N2, or N3) or short-term nitrogen addition levels (SN1 or SN2) was calculated based on the value of Rh₂₀ or Q₁₀ under N0 or SN0 treatments. Effect sizes of pH, C/N, EON, and MN under different long-term nitrogen fertilizer application levels (N1, N2, or N3) were calculated based on the values of pH, C/N, EON, and MN under N0 treatments. Effect sizes of EON, MN, and q_mic under short-term nitrogen addition levels (SN1 or SN2) were calculated based on the values under SN0 treatments.

2.6. Statistics

Significant differences in the changes in Rh₂₀ and its Q₁₀ value and soil environmental factors (pH, C/N, EON, and MN after long-term nitrogen fertilizer application) among the soil samples from each field site were tested using Duncan’s multiple-range test. Additionally, T test was used to test the significant differences between the changes in soil EON, MN, and q_mic after short-term nitrogen addition.

Linear regression analysis was implemented to assess the effects of long-term nitrogen fertilizer application (LN) and short-term nitrogen addition (SN) rates on Rh₂₀ and its Q₁₀ value at each field site using Equation (4):

NI = α1 × LN + α2 × SN + α0

(4)

where NI represents nitrogen addition impact on Rh and its Q₁₀ as affected by two variables (LN and SN). α is estimated coefficients, and α0 is the constant.

Further regression analysis was developed to estimate how Rh and its Q₁₀ would respond to soil environmental factors under different durations of nitrogen addition using Equation (5):

$$\mathrm{SI} = \mathrm{\beta }0 + {\sum }_1^k{\mathrm{\beta }}_k{Controls}_k + \epsilon$$

(5)

where SI represents soil environment impact and ${Controls}_k$ are the various control variables affecting Rh₂₀ and its Q₁₀ under long-term and short-term nitrogen addition conditions. β is estimated coefficients, β0 is the constant, and ε is the residual error. The soil environmental factors C/N, pH, LOC, EON, MN, and C_mic were used to estimate the response of Rh₂₀ to long-term nitrogen fertilizer application; LOC, EON, MN, and C_mic to estimate the response of Rh₂₀ to short-term nitrogen addition; C/N, pH, LOC, EON, MN, C_mic, and q_mic to estimate the response of Q₁₀ to long-term nitrogen fertilizer application; and LOC, EON, MN, C_mic, and q_mic to estimate the response of Q₁₀ to short-term nitrogen addition_.

Statistical analyses were conducted with SPSS 16.0 (SPSS Inc., Chicago, IL, USA), with significance at p < 0.05.

3. Results

3.1. Effect of Nitrogen Addition on Soil Rh₂₀ and Its Q₁₀

Long-term nitrogen fertilizer application: The direction of the effect of nitrogen fertilizer application in rice field systems on soil Rh is affected by the duration of continuous application (Figure 1). When the nitrogen fertilization treatment was maintained for only 1 year, the high nitrogen application treatment (N3) resulted in a significantly increased soil Rh₂₀ compared to the treatment without nitrogen fertilizer (N0). When the nitrogen fertilizer application lasted for 6 years, the response of the soil Rh to differences in the nitrogen application level was not significant. When the nitrogen fertilizer application lasted for 11 years, the nitrogen application treatment reduced the soil Rh₂₀ by 6% to 14% compared with the treatment without nitrogen fertilizer (N0), and the effect was significant when the nitrogen addition was greater than 200 kg ha⁻¹ (N2 and N3 treatments).

In addition, differences in the duration of nitrogen fertilizer application also produced different responses in the Q₁₀ of soil Rh in rice field systems. When nitrogen fertilizer was applied for 6 years, the Q₁₀ value decreased by 7% compared with the treatment without nitrogen fertilizer (N0); when nitrogen fertilizer was applied for 11 years, the Q₁₀ value decreased by 12%. The Q₁₀ of the 200 kg ha⁻¹ annual nitrogen addition treatment (N2) reached a significant decrease of 10% after 11 years of nitrogen fertilizer application.

Short-term nitrogen addition: Short-term nitrogen addition increased the rate of soil Rh and reduced Q₁₀ to a certain extent (Figure 2). Soil Rh₂₀ and Q₁₀ values responded more significantly to the short-term high-intensity nitrogen addition (SN2) condition than to the short-term low-intensity nitrogen addition (SN1) condition. Moreover, among the different long-term nitrogen fertilizer application treatments, short-term nitrogen addition had a more significant effect on the soil Rh₂₀ and Q₁₀ values of the soil in treatment without nitrogen fertilizer (Rh₂₀ increased by 2–10% and the Q₁₀ value decreased by 4–12%), while it had a relatively small effect on the soil Rh₂₀ and Q₁₀ values of the soil in the nitrogen application treatment, especially the higher nitrogen application treatment (Rh₂₀ changed by −3–6% and the Q₁₀ value changed by −7–2%, Figure 2).

In addition, the duration of nitrogen fertilizer application also significantly influenced the responses of soil Rh₂₀ and Q₁₀ values to short-term nitrogen addition. When the nitrogen fertilizer treatment lasted for 1 year, short-term nitrogen addition increased the soil Rh₂₀ by 3–10% and reduced the Q₁₀ value by 2~12% (Figure 2a,b). Soil Rh₂₀ and Q₁₀ values were sensitive to short-term nitrogen addition. Among them, the Rh₂₀ and Q₁₀ values of the soil in the long-term no-nitrogen application treatment condition (N0) and in the annual-average-nitrogen application condition of less than 200 kg ha⁻¹ (N1) responded significantly to short-term nitrogen addition. As the duration of nitrogen fertilizer application increased, the soil originally treated with nitrogen fertilizer gradually became less responsive to short-term nitrogen addition. When the nitrogen fertilizer treatment lasted for 6 years, only the soil Rh₂₀ and Q₁₀ values of the no-nitrogen treatment (N0) showed a significant increase of 4–8% and a significant decrease of 8–9%, respectively, due to the short-term nitrogen addition. The soils with nitrogen treatments greater than 200 kg ha⁻¹ (N2, N3) even showed the opposite response (Figure 2c,d). When the nitrogen fertilizer treatment lasted for 11 years (WX), short-term nitrogen addition had no significant effect on soil Rh₂₀ or Q₁₀ values in the fertilization treatment conditions (Figure 2e,f).

3.2. Response of Soil Rh₂₀ and Its Q₁₀ to the Rate and Duration of Nitrogen Addition

Nitrogen addition rate had a significant effect on the rate of soil Rh, but the duration of nitrogen fertilizer application affected the direction of the effect of nitrogen addition on soil Rh (

Table 2. Response of soil Rh₂₀ and Q₁₀ value to the rate and duration of nitrogen addition.

	Rh20 × 105			Q10 × 103
	1 Year	6 Years	11 Years	1 Year	6 Years	11 Years
LN	2.47	−1.75	−0.710	0.74	0.79	0.24
	(0.50) ***	(0.40) ***	(0.09) ***	(0.23) **	(0.25) **	(0.23)
SN	3.78	0.34	0.06	−1.85	−0.94	0.28
	(1.30) **	(1.11)	(0.32)	(0.57) **	(0.68)	(0.70)
CONT	5638.20	4307.30	1247.84	3324.44	3656.43	2672.3
	(118.10) ***	(100.82) ***	(25.52) ***	(51.51) ***	(60.61) ***	(63.65) **
N	36	36	36	36	36	36
F	15.17	8.59	19.41	10.49	5.845	0.42
R-squared	0.48 ***	0.34 **	0.54 ***	0.39 ***	0.262 **	0.03
P for normality S-W test	0.01	0.23	0.09	0.86	0.11	0.22
P for Levene homogeneity of variance	0.36	0.16	0.16	0.80	0.21	0.58

*** p < 0.001; ** p < 0.01. The value in parentheses is standard error.

). Under the conditions of only 1 year of nitrogen fertilizer application, the nitrogen addition rate was significantly positively correlated with soil Rh₂₀; that is, the greater the nitrogen addition rate, the stronger the soil Rh. However, when the nitrogen fertilizer application treatment reached more than 6 years, the nitrogen addition rate and the soil Rh₂₀ showed a significant negative correlation; that is, the nitrogen addition rate inhibited soil Rh. In addition, the rate of long-term nitrogen fertilizer application showed a positive correlation with Q₁₀ values, but as the duration of nitrogen fertilizer application increased, the response of Q₁₀ to the nitrogen addition rate gradually weakened (correlation in the 11-year condition was not significant). The effects of short-term nitrogen addition on soil Rh and its Q₁₀ were significant only when nitrogen fertilizer was applied for 1 year; that is, short-term nitrogen addition promoted soil Rh₂₀ and reduced the Q₁₀ value (Table 2). As the duration of nitrogen fertilizer application increased, soil Rh and Q₁₀ became less sensitive to short-term nitrogen addition.

3.3. Response of Soil Rh₂₀ and Its Q₁₀ to the Soil Environmental Factors

Under long-term nitrogen fertilizer application conditions, soil C/N, pH, EON, and MN are the main factors affecting the rate of soil Rh; soil C/N and MN are also related to the Q₁₀ of soil Rh (

Table 3. Response of soil Rh₂₀ and Q₁₀ value to the soil environmental factors under different nitrogen addition conditions.

	Rh20 × 104		Q10 × 102
	LN	LN × SN	LN	LN × SN
C/N	16.24		17.80
	(9.05) *		(6.86) *
pH	495.16		81.48
	(24.95) ***		(45.93)
LOC	−17.87	0.00	−14.91	0.00
	28.37	(0.00)	(19.77)	(0.00)
EON	3.15	13.37	0.03	−0.11
	(1.05) **	(3.37) ***	(0.74)	(0.49)
MN	−2.62	−1.87	2.69	1.59
	(1.24) *	(1.28)	(0.87) **	(0.17) ***
Cmic	−0.01	0.52	−0.02	−0.01
	(0.08)	(0.29)	(0.11)	(0.04)
qmic			143.34	661.30
			(335.87)	(52.62) ***
CONT	−2762.05	7.98	−376.07	−208.58
	(155.25) ***	(135.52)	(234.62)	(19.21) **
N	36	108	36	108
F	418.62	5.14	36.89	55.34
R-squared	0.99 ***	0.41 **	0.90 ***	0.73 ***

*** p < 0.001; ** p < 0.01; * p < 0.05. The value in parentheses is standard error.

). Under the combined short-term nitrogen addition conditions, EON was the main factor affecting the rate of soil Rh, while the Q₁₀ of soil Rh was mainly associated with soil MN and q_mic.

Long-term nitrogen fertilizer application significantly increased soil MN content, and the magnitude of the change became more significant with increasing nitrogen fertilizer application intensity and a longer treatment duration (Figure 3d). The EON content was increased by the nitrogen fertilizer application rate only in the 1-year planting condition, while it decreased with the increase in the application rate in the 6-year and 11-year planting conditions (Figure 3c). The soil C/N fluctuation caused by long-term nitrogen fertilizer application showed a downward trend in the 1-year condition and tended to be stable as the years of continuous nitrogen fertilizer application treatment were prolonged (Figure 3b). The pH value of rice field soil varied from −3% to 1% under long-term nitrogen fertilizer application and showed a decrease only in the 6-year condition, but it did not reach a significant level, and there was no significant difference between different nitrogen fertilizer application rates (Figure 3a).

The response of soil q_mic to short-term nitrogen addition was mainly affected by the nitrogen fertilizer application rate and the duration of treatment (Figure 4a). Short-term nitrogen addition had a tendency to reduce soil q_mic, but there was no significant difference among the different short-term nitrogen addition rates. In the no-nitrogen fertilizer application condition, the reduction effect of short-term nitrogen addition on soil q_mic was more significant. In addition, as the duration of nitrogen fertilizer application increased, the impact of short-term nitrogen addition on soil q_mic gradually weakened, and the differences in soil background caused by long-term nitrogen fertilizer application also gradually narrowed.

4. Discussion

4.1. Nitrogen Addition Altering Soil Environmental Factors and Soil Rh

In this study, under conditions of long-term nitrogen fertilizer application and short-term nitrogen addition, soil C/N, pH, EON, and MN all exhibited significant correlations with soil Rh rates (Table 3). Although in this study, all treatments involved only the application of chemical nitrogen fertilizers, both soil MN and EON (related to nitrogen addition) demonstrated a significant correlation with Rh in the regression analysis.

MN serves as a nutrient that can be directly absorbed and utilized by plants and microorganisms for growth. Theoretically, an increase in the MN content in the soil environment could promote the growth of soil microorganisms and their carbon decomposition activities [23,24,36]. However, in this study, MN and the rate of soil Rh show a significant negative correlation under long-term nitrogen addition conditions. Conversely, under short-term nitrogen addition conditions, a significant positive correlation is observed. Combined with the changes in the other soil properties (Figure 3), partially due to the possible direct inhibition of the respiration rate by mineral nitrogen in the soil [29,37], the increase in soil MN content caused by long-term nitrogen addition may change the microbial community or activity [16,18,28], the shifts of which then inhibit soil Rh in the presence of LOC and nutrients [36]. By contrast, short-term nitrogen addition increases soil Rh to a certain extent, and this increase is most evident in the non-fertilized treatment in the 1-year planting condition (Figure 2a). These findings show that when the native soil MN content is relatively low, nitrogen addition could stimulate carbon decomposition in the short term; however, this effect may not be significant under soil native conditions subjected to continuous nitrogen addition over multiple years. Therefore, it is evident that the behavioral feedback of microorganisms regarding carbon fixation or decomposition depends on the nutrient status of the soil [38].

On another note, under both long-term and short-term nitrogen addition conditions, the soil EON content exhibited a strong positive correlation with the soil Rh. Only in the one-year condition did the EON content increase significantly with the increase in both the long-term and short-term nitrogen addition rates. Regarding the increased EON in this part, it is highly probable that it stems from the decomposition of biomass detritus by microorganisms. The application of nitrogen fertilizers augments the amount of biomass detritus entering the soil, thus leading to an elevation in the EON content [25,36]. Conversely, in the 6-year and 11-year conditions, the EON content decreased in response to an increase in the long-term nitrogen addition rate yet exhibited no marked response to short-term nitrogen addition (Figure 3c). These findings indicate that the maintenance of soil EON content is, to a large extent, reliant on the addition of organic nitrogen [9], whereas the long-term application of chemical nitrogen fertilizer alone could result in the depletion of EON.

As relatively stable physical and chemical properties of soil, pH, and C/N did not show a significant response to nitrogen addition rates, even under long-term nitrogen fertilization conditions. Be that as it may, this study still finds that soil C/N and pH show a positive correlation with Rh (Table 3, Figure 3). Regarding soil pH, with urea serving as the main nitrogen fertilizer, both the nitrification process of nitrogen fertilizers and the uptake of NH₄⁺-N by the rice plant are accompanied by the release of protons, thus leading to soil acidification and inhibiting microbial biomass and activity [18]. Therefore, the appropriateness of the nitrogen addition rate of chemical fertilizers (in line with crop growth) is crucial for determining the soil acidification rate. Long-term high-rate nitrogen addition is likely to result in a simultaneous decline in both soil pH and Rh [28].

When it comes to C/N, this study shows a downward trend in C/N with long-term nitrogen addition, especially when the addition rate reached more than 200 kg ha⁻¹ (Figure 3). Although during one year of nitrogen addition, the decline in C/N is accompanied by an increase in the soil Rh rate, when extended to six continuous years or more, a depressing effect on the soil Rh rate is observed (Figure 3a,c), which indicates a positive correlation between the soil C/N and the rate of soil Rh (Table 3). Numerous studies suggest that a lower C/N condition is more favorable for soil Rh [7,20]. Nevertheless, the positive correlation between soil C/N and the respiration rate typically emerges under circumstances where the inherent soil properties are at a relatively high level, as exemplified by forest soils [35]. In this context, nitrogen supplementation or the addition of low-C/N substances could alleviate the competition for nitrogen between microorganisms and crops [7]. It is worth noting that, as the cropping systems in this study are relatively abundant in nitrogen, the soil C/N ranges from 9 to 13. The synchronous changes observed between C/N and the rate of soil Rh, to a certain extent, indicate that even when C/N increases, nitrogen is still abundant and does not constitute a nitrogen-limited state to limit the carbon decomposition by microorganisms [35,37]. In addition, as carbon serves as the substrate for soil Rh, its depletion could become a decisive factor influencing the Rh response [21,37]. However, in this study, although continuous multi-year cultivation and varying nitrogen application rates resulted in yield differences, the implementation of straw return did not lead to significant changes in soil SOC or LOC (Table 1). Therefore, it could be inferred that the inhibition of soil Rh caused by long-term nitrogen fertilizer application (exceeding 6 years) is unlikely to be attributable to the reduction in substrate availability.

4.2. Nitrogen Addition and Q₁₀ of Soil Rh

In this study, soil MN showed a significant positive correlation with the Q₁₀ of Rh under both long-term nitrogen fertilizer application and short-term nitrogen addition conditions. Under short-term nitrogen addition conditions, the positive effect of q_mic on Q₁₀ was also significant (Table 2), implying that nitrogen increased Rh’s Q₁₀ through the increase in substrate availability and better specific microbial activity [25,39] under short-term nitrogen addition conditions. A previous study suggests that, in agricultural systems, the response of soil microbial communities to nitrogen addition enhances their activity in decomposing organic carbon under a warmer climate [39]. The fate or management pathway of carbon and nitrogen in biological residues is a key factor determining the sensitivity of soil Rh to nitrogen addition across ecosystems [7,21]. In this study, affected by the continuous return of straw back to the field, the contents of soil SOC and LOC remained relatively stable and sufficient across different nitrogen strategies (Table 1). Under these conditions, after 11 years of nitrogen addition, the effect of nitrogen on Rh’s Q₁₀ is no longer statistically significant (Table 2). This is an interesting phenomenon, suggesting that for long-term rice-cultivation systems, nitrogen addition may weaken the impact of climate warming on SOC decomposition. Therefore, compared with the nitrogen addition rate, the addition duration seemed to contribute more to the Q₁₀ of Rh.

In the one-year condition, the soil Rh’s Q₁₀ responded significantly to short-term nitrogen addition (Table 2), but when the duration was extended to 6 and 11 years, short-term nitrogen addition did not significantly change Rh’s Q₁₀. There is no doubt that the effect of short-term nitrogen addition on the dynamic behavior of microbial carbon decomposition plays a crucial role in determining the direction of Rh’s changes [7,25]. Yet, soil microbial biomass was not a key factor affecting Rh or its Q₁₀ in this study (Table 3), and MN and q_mic are positively correlated with Q₁₀ under short-term nitrogen addition conditions. This shows that, for rice field soils with a C/N of around 10, the changes in mineral nitrogen brought about by short-term (14 days) nitrogen addition are not sufficient to significantly change the total biomass of the microbial community. Considering that Rh’s Q₁₀ is mainly determined by specific microorganisms [24], the correlation between q_mic and the Q₁₀ of Rh could indicate that short-term nitrogen addition changes the response strategy of related functional bacteria to the intensity of carbon decomposition under variable temperature environments [16,35]. The correlation between microbial indicators and the Q₁₀ of Rh was found only in the short-term nitrogen addition conditions and not in the long-term conditions (Figure 4a). It also means that, in a soil environment with relatively abundant total nitrogen, the effect of nitrogen addition on microbial carbon decomposition behavior and its sensitivity is not mainly reflected in microbial biomass. Instead, the structure of the microbial community might be the key factor influencing the carbon decomposition efficiency of relevant microorganisms, and this area is expected to become an important focus of future research.

4.3. Implications for Adjusting Nitrogen Addition Based on Soil Rh and Its Q₁₀

In agroecosystems, the rate of nitrogen fertilizer application is a critical determinant of how nitrogen addition influences soil Rh and its Q₁₀, a finding consistently supported by this study and numerous prior investigations [7,24]. As for the rice systems in the Taihu Lake Basin involved in the current study, when the annual nitrogen addition rate is set to greater than 200 kg ha⁻¹, the inhibition of the soil Rh rate is more significant, and only when the annual addition rate is set to less than 200 kg ha⁻¹ is the decrease in the soil Rh’s Q₁₀ significant (Figure 1). Interestingly, from the perspective of nitrogen management, the annual nitrogen addition rate of approximately 200 kg ha⁻¹ during rice season precisely aligns with the recommended nitrogen addition rate for rice fields in the Taihu Lake Basin in recent years, which aims to achieve green production. This indicates that an annual nitrogen application rate of 200 kg ha⁻¹ represents an optimal strategy that maintains crop productivity while balancing nitrogen use efficiency with nitrogen emission control [30]. When carbon emissions reduction (or the prevention of emission increases) is integrated into agricultural production objectives, a nitrogen fertilizer application strategy tailored to soil properties and crop yield intensity could simultaneously mitigate both soil carbon decomposition via Rh pathways and Q₁₀. Although short-term nitrogen addition may lead to an increase in the soil Rh rate, appropriate nitrogen addition rates and extended addition durations could effectively mitigate the response of soil Rh to such short-term nitrogen addition and enhance the stability of carbon decomposition by the soil Rh process. This indicates that, on a fixed field, long-term and stable nitrogen fertilizer application in rice cultivation would be conducive to reducing carbon emissions from soil carbon decomposition and mitigating temperature-induced fluctuations in decomposition rates, thereby promoting the development of a more stable soil organic carbon pool within agroecosystems.

5. Conclusions

In conclusion, our study demonstrates that, in the rice system, as the duration of continuous nitrogen fertilizer application increases, the rate of soil Rh and its Q₁₀ are initially promoted and then inhibited. A high nitrogen application rate could effectively inhibit soil Rh; however, it also keeps the Q₁₀ value at a relatively high level. In the sample rice systems of our study, the inhibitory effect of nitrogen fertilizer application on soil Rh became evident after more than 6 years of continuous nitrogen fertilizer application. The decrease in C/N, pH, and EON and the increase in MN resulting from long-term nitrogen fertilizer application are the primary factors inhibiting soil Rh, while Q₁₀ exhibits a more notable response to soil C/N and MN. Short-term (14 days) nitrogen addition could increase soil Rh and attenuate its Q₁₀. In nitrogen-rich rice soils, although short-term nitrogen addition does not notably alter microbial biomass, it may reduce q_mic as a result of the increased content of soil mineral nitrogen, thereby influencing the response strategies of related functional bacteria regarding carbon decomposition intensity in a fluctuating-temperature environment.

Based on the responses of soil Rh and its Q₁₀ to the nitrogen addition rate and duration, this study reveals that appropriate nitrogen application strategy could gradually achieve balanced crop yields, suppress the soil carbon decomposition rate, and alleviate carbon pool fluctuations caused by global warming over a long period of time.