Soil Aggregate Stability Under Freeze–Thaw Cycles in Mollisols as Evidenced by ¹⁵N Distribution

Abstract

Freeze–thaw cycles (FTCs) influence soil nitrogen (N) dynamics and soil aggregate stability. However, the driving mechanism affecting aggregate stability from the combined perspective of N components and N distribution by ¹⁵N tracing technology in both bulk soils and soil aggregates remains worth exploring. This study took the farmland Mollisols in Hailun City, Heilongjiang Province, as the research object, and investigated the variations in soil N components and aggregate stability across five freeze–thaw frequencies (1, 3, 5, 9, and 17 cycles) under three freeze–thaw temperatures (−9 °C/5 °C, −18 °C/5 °C, and −26 °C/5 °C) using ¹⁵N tracing technology. The results demonstrated that freeze–thaw frequency and temperature both influenced aggregate stability. Specifically, with the increase in freeze–thaw frequency, soil aggregate stability was reduced through decreasing the proportion of macroaggregates (2–0.25 mm), increasing the proportion of silt + clay fractions (<0.053 mm), and reducing the total N (TN) content of silt + clay fractions under higher freezing temperature (−9 °C/5 °C). In contrast, for lower freezing temperature (−18 °C/5 °C and −26 °C/5 °C), the increased freeze–thaw frequency enhances soil aggregate stability by decreasing the proportion of silt + clay fractions, increasing the proportion of microaggregates (0.25–0.053 mm), and reducing the TN contents of microaggregates and silt + clay fractions. These findings are essential for developing strategies to mitigate the adverse effects of FTCs on soil quality and ecosystem functions in cold regions.

1. Introduction

Freeze–thaw cycles (FTCs), which are a common phenomenon in cold regions during the winter–spring transition, significantly impact soil structure, nutrient dynamics, and microbial activity, thereby affecting soil quality and ecosystem functions [1,2,3,4,5,6]. Understanding the effects of FTCs on soil properties is crucial for sustainable land management and predicting soil responses to climate change.

Soil total nitrogen (TN) is crucial for maintaining soil fertility, supporting plant growth, and enhancing soil health, which plays a vital role in soil structure, nutrient cycling, microbial activity, and nitrogen (N) accumulation [7]. However, due to the high organic matter content in soil, TN is often unresponsive to short-term agricultural management changes, making it difficult to detect changes over brief periods [8]. Soil labile N components, such as nitrate N (NO₃⁻-N), ammonium N (NH₄⁺-N), and microbial biomass N (MBN), which have short turnover times, can rapidly respond to changes in soil management practices and environmental conditions, and serve as important indicators for assessing soil quality and sustainability [9,10,11,12,13]. The contents of TN and labile N components could be increased, decreased, or show no significant differences, or exhibit a fluctuating tendency due to freeze–thaw action [14,15,16,17]. This demonstrates that FTCs can exert diverse impacts on soil N availability, emphasizing the complexity of N dynamics in soils subjected to FTCs.

Soil aggregates serve as the fundamental units of soil structure, influencing soil porosity, water retention, and resistance to erosion, among other properties [18,19,20,21]. The amount and size of these aggregates are crucial factors that determine the rate and extent of physical processes such as soil erosion and compaction [22]. Soil aggregate stability is also one of the important indicators that reflect the soil structure status, which not only plays an important role in regulating soil fertility and maintaining land productivity, but also has a close relationship with soil erosion and environmental quality [23,24,25]. As an abiotic force, freeze–thaw can change soil water content, water distribution, and soil temperature, and also affect soil microbial activity to a certain extent, thus affecting the physical structure and properties of soils [26,27]. The alternating cycle of freeze–thaw can effectively break the larger aggregates in the soil into smaller aggregates, and the fine particles in the soil tend to aggregate to medium-sized particles [28].

The study of Oztas and Fayetorbay (2003) on clay and sandy soil showed that the content of wet aggregates decreased by 28.6–51.7% when the soil was frozen and thawed nine times, and the increase in water content promoted the decrease in aggregates [29]. However, the content of soil wet aggregates increased gradually at 3–6 times of freezing and thawing, and began to decrease after six FTCs. Edwards (2013) concluded that after freezing and thawing a loam for 15 cycles, the content of macroaggregates (>4.75 mm) decreased by 20–28%, while that of microaggregates (<0.5 mm) increased by 32–50% [30]. The mass fraction of macroaggregates was higher than that of microaggregates, and the overall aggregate content decreased [31]. Due to the increase in porosity, the moisture conductivity of saturated soil after freezing and thawing increases compared with that before freezing, and the increment is most significant when the soil has undergone one freeze–thaw and the fine grain and dry bulk density increase [32]. Compared with the soil that has undergone one freeze–thaw, the moisture conductivity of soil after multiple freeze–thaw increases slightly, but tends to a constant [33]. Therefore, the balance between these effects depends on the frequency and temperature of FTCs, as well as the initial soil conditions.

The application of ¹⁵N-labeled corn straw provides a robust tracer approach to quantify the relative contribution rate, allocation patterns, and transformation efficiency of residue-derived N within soil N fractions [34,35]. Straw N can enter the soil N component after decomposition and transformation in the soil, and the undecomposed straw will remain in the soil to enhance the soil N storage [36,37]. Currently, there is limited research on the distribution and fixation of corn straw N in the bulk soils and aggregates in Mollisols under different freeze–thaw frequencies and intensities. Additionally, studies on the driving mechanisms of soil aggregate stability induced by freeze–thaw action from the combined perspective of N components and N allocation in both bulk soils and soil aggregates are also scarce.

Therefore, we hypothesize that (1) FTCs raise both the N component content and the straw-derived N components in bulk soils, (2) FTCs reduce TN and straw-derived TN contents in >0.25 mm aggregates while increasing them in <0.25 mm aggregates, and (3) aggregate stability declines with increasing FTC frequency regardless of temperature. This study aims to (1) investigate the effects of FTCs on the N components and N distribution in bulk soils and soil aggregates, and (2) quantify soil aggregate stability under different freeze–thaw frequencies and intensities. This knowledge is essential for developing strategies to mitigate the adverse effects of FTCs on soil quality and ecosystem functions in cold regions.

2. Materials and Methods

2.1. Experimental Site

This study selected the typical farmland Mollisols in the black soil belt of Northeast China [38]. The research area is located in Hailun Agricultural Ecological Experiment Station, Chinese Academy of Sciences of Hailun City, Heilongjiang Province (47°27′ N, 126°56′ E). The parent material is Quaternary loeslike parent material, which belongs to the middle and temperate continental monsoon climate, with an annual average temperature of 1.5 °C and annual precipitation of 550 mm [39,40]. In agricultural fields, crops such as soybean (Glycine max Merr.) and corn (Zea mays L.), are grown under a one-crop-per-year system. The soil is a typical Black soil or Mollisols [38]. Soil texture, determined by the hydrometer method [41], is 43% silt, 35% clay, and 22% sand. The basic physicochemical properties of soil are as follows: soil pH of 6.10, soil organic matter (SOM) of 50.6 g kg⁻¹, bulk density (BD) of 1.27 g cm⁻³, TN of 2.6 g kg⁻¹, total phosphorus (TP) of 1.6 g kg⁻¹, and total potassium (TK) of 13.8 g kg⁻¹.

The determination of the soil’s basic physicochemical properties was as follows: Soil pH value was measured at a slurry consisting of 1:2.5 (w/v) soil/distilled water using a pH meter (FiveEasy Plus, Urdorf, Switzerland) [42]. SOM was measured by the potassium dichromate–sulfuric acid oxidation method [43]. Soil BD was determined using the core method by undisturbed soil core of 100 cm³ [42]. Soil TC was conducted using an elemental analyzer (vario PYRO cube, Hanau, Germany) [44]. Soil TP and TK contents were determined using the sodium hydroxide fusion–molybdenum-antimony resistance colorimetric method, and the sodium hydroxide fusion–flame photometry method, respectively [43].

2.2. Preparation of the Straw Used in the Experiment

The straw in the freeze–thaw experiment was corn straw with ¹⁵N-labeled TN content. The soil used for planting corn in the pot experiment was collected from the 0–30 cm depth by the Hailun Agricultural Ecological Experimental Station, Chinese Academy of Sciences. Soil samples were collected using a Dutch auger (AMS type, Shaoxing, China) with a diameter of 50 mm, following an S-shaped pattern for 134 borings, with each pair of borings spaced 5 m apart. The sampling area was approximately 4000 m², and a total of 100 kg of soil was obtained. The soil samples from the 134 borings were combined into one composite sample. Subsequently, the composite soil sample was cleaned to remove gravel, root systems, and other impurities, then air-dried and passed through a 10 mm sieve for subsequent use.

The plastic pots used for growing corn had an inner diameter of 30 cm and a height of 32 cm. Each pot was filled with 19 kg of soil and fertilized with urea, potassium chloride, and triple superphosphate as the base fertilizers. ¹⁵N-labeled fertilizers were applied at different growth stages of corn to label the N in the straw [45]. Corn was sown in March 2021 and harvested in November 2021. After the corn was harvested, the corn straw was collected. The plants were separated into grains, straw, and roots [46]. The straw was subjected to steaming at 105 °C for 30 min, dried at 65 °C until constant weight, ground, and then passed through a 0.25 mm sieve before being bagged for later use [47].

The basic chemical properties of the experimental corn straw are as follows: pH of 6.31, total carbon (TC) of 457.20 g kg⁻¹, TN of 5.82 g kg⁻¹, the ratio of carbon to N (C/N) of 78.56, and an atom% ¹⁵N value of 3.94. The measurement methods for these indicators were as follows: The pH value of corn straw was measured at a slurry consisting of 1:10 (w/v) corn straw/distilled water using a pH meter (FiveEasy Plus, Urdorf, Switzerland) [42]. TC and TN contents of the corn straw were determined using an elemental analyzer (vario PYRO cube, Hanau, Germany) [44]. C/N of corn straw was calculated by the ratio of TC to TN. The ¹⁵N-labeled TN content of corn straw was determined by an elemental analyzer coupled with an isotope ratio mass spectrometer (Elementar vario PYRO cube-Iso Prime100 Isotope Ratio Mass Spectrometer, Hanau, Germany) [48].

2.3. Soil Column Filling

In this study, the experimental unit was defined as each individual soil column. Each soil column contained the same mass of black soil samples, which had undergone strict pretreatment before the start of the experiment, including removal of impurities, grinding, and sieving, to ensure the consistency of their initial states. Each experimental unit represented an independent observation object for assessing the effects of freeze–thaw cycles on the properties of Mollisols.

The soil columns used in the experiment were cylindrical polyethylene (PE) bottles (Glowmore Express Sdn Bhd, Selangor, Malaysia) with a diameter of 10 cm and a height of 13 cm. Each soil column was filled with 1021 g of soil, and the soil moisture content was adjusted to 25% using distilled water. During the freeze–thaw cycles, the moisture content of the samples was monitored by regular weighing for 8 h and replenishing the water by adding distilled water when necessary.

The experiment included the addition of straw containing ¹⁵N-labeled N. The straw was applied at a rate equivalent to 7500 kg ha⁻¹, which corresponded to adding 3.6750 g of straw per soil column. The straw was mixed evenly and placed into nylon mesh bags with a mesh size of 0.045 mm and dimensions of 7 cm × 7 cm. These bags were buried in the soil columns to ensure that the thickness of the soil above and below the straw was consistent. To mitigate the immobilization of soil N caused by the high C/N ratio of the straw, 0.2529 g urea was applied to each soil column. The caps of the bottles were put back on after all the soil columns were filled.

2.4. Soil Pre-Cultivation, Freeze–Thaw Treatment, and Soil Sampling

Three freeze–thaw intensities, each with three replications, were conducted using different temperature settings for freezing and thawing: −9 °C/5 °C, −18 °C/5 °C, and −26 °C/5 °C. After the 7-day pre-cultivation, the soil columns were placed into a refrigerator for freezing. The timing began when the refrigerator temperature reached the set point of −9 °C/−18 °C/−26 °C. After freezing for 84 h, the soil columns were removed and placed into a refrigerator set at 5 °C for 84 h of thawing, thus constituting a freeze–thaw cycle.

Under the three freeze–thaw temperatures, the soil columns were subjected to 1, 3, 5, 9, and 17 freeze–thaw cycles, respectively. Each treatment of the freeze–thaw cycles had three replications with a completely random arrangement. Therefore, the entire experiment consisted of 45 experimental units (Figure 1). During the incubation period, soil columns corresponding to each freeze–thaw frequency treatment were taken out for destructive sampling to facilitate subsequent experiments. Soil samples from each treatment were collected and divided into three portions. Two portions were air-dried, ground, and sieved through mesh sizes of 10 mm and 0.149 mm to determine soil aggregates and TN content, respectively. The remaining soil was manually disrupted, plant debris removed, sieved through a 2 mm sieve, and stored at −20 °C for the analysis of NO₃⁻-N, NH₄⁺-N, and MBN contents [49].

To minimize the potential impact of non-uniform temperature distribution inside the freezer on the experimental results, the experimental units were randomized before each freeze–thaw cycle. Specifically, the soil columns were randomly placed at different locations within the freezer. Randomization was achieved by using a random number generator to determine the exact position of each container, ensuring that the placement of each experimental unit within the freezer was entirely random. Throughout the experiment, the randomization process was repeated multiple times to further enhance the reliability of the experimental results.

This experiment places great emphasis on the independence between different freeze–thaw cycle levels. The experimental units for each combination of freeze–thaw cycle frequency and temperature were set up independently. That is, each level had its own set of soil sample containers, and these containers did not interfere with each other during the experiment. Throughout the experimental process, the independence between different levels was ensured through a strict experimental procedure and detailed records. Specifically, the placement of experimental units from different levels inside the freezer was completely independent, and during the freeze–thaw cycles, each level of experimental units operated independently according to its corresponding freeze–thaw program. With this design, the impacts of different freeze–thaw cycle frequencies on the properties of Mollisols without interference from other levels were able to accurately assess.

2.5. Determination Indicators and Methods

2.5.1. Soil TN, NO₃⁻-N, and NH₄⁺-N Contents and Their ¹⁵N-Labeled Contents in Bulk Soils

The ¹⁵N-labeled TN content of soil was determined by an elemental analyzer coupled with an isotope ratio mass spectrometer (Elementar vario PYRO cube-Iso Prime100 Isotope Ratio Mass Spectrometer, Hanau, Germany) [48]. Soil NH₄⁺-N and NO₃⁻-N were extracted with 2 mol L⁻¹ KCl solution, and the ratio of soil to KCl solution was 1:10. NH₄⁺-N was measured by indophenol blue colorimetry, and NO₃⁻-N was determined by dual-wavelength ultraviolet spectrophotometry [43]. The measurement of NH₄⁺-¹⁵N and NO₃⁻-¹⁵N contents in soil was conducted using the modified micro-diffusion method [50]. A 20 mL portion of the filtrate was transferred into a 250 mL Schott bottle fitted with a suspended filter paper. Subsequently, 0.1 g of MgO and 20 µL of 1 mol L⁻¹ oxalic acid were added. The bottle was then placed on a shaker and shaken at 25 °C, 140 r min⁻¹ for 24 h to facilitate the diffusion of NH₄⁺-¹⁵N. The filter paper was removed and placed into a desiccator for drying over a period of 24 h. Concurrently, two acidified filter papers were inserted into the Schott bottle, and the diffusion process continued for an additional 48 h to eliminate any residual NH₄⁺-¹⁵N. Following, two more acidified filter papers and 0.1 g of Devarda’s alloy were added, and the diffusion was carried out for another 24 h to complete the diffusion of NO₃⁻-¹⁵N. Thereafter, the filter papers were removed and placed into the desiccator to dry for an additional 24 h. After drying, the filter papers were analyzed for ¹⁵N abundance using an elemental analyzer coupled with an isotope ratio mass spectrometer (Elementar vario PYRO cube-Iso Prime100 Isotope Ratio Mass Spectrometer, Hanau, Germany) [48].

2.5.2. MBN and ¹⁵N-Labeled MBN Contents in Bulk Soils

The soil microbial biomass was determined using the chloroform fumigation with 0.5 mol L⁻¹ K₂SO₄ solution extraction method [45]. The TN content in the extract was measured using the 3% alkaline persulfate oxidation method combined with an AA3 flow injection analyzer (Bran + Luebbe, Norderstedt, Germany). The soil MBN content is the difference in labile N content between fumigated and non-fumigated soil samples, with a correction factor (k_EC) of 0.45 [12,43,49]. Six milliliters of the non-fumigated or fumigated extract (with chloroform removed for fumigated samples) were taken and freeze-dried. The atom% ¹⁵N in the samples were then determined using an elemental analyzer coupled with an isotope ratio mass spectrometer (Elementar vario PYRO cube-Iso Prime100 Isoto Pe Ratio Mass SPectrometer, Langenselbold, Germany) [48].

2.5.3. Soil Aggregate Fractionation Methods, and TN and ¹⁵N-Labeled TN Contents in Soil Aggregates

A soil aggregate analyzer (TTF-100 type, Suzhou, China) was utilized to separate the soil aggregates through the wet-sieving method [19]. A total of 50 g of air-dried soil samples passed through a 10 mm sieve were uniformly spread across the nested sieves with mesh sizes of 2 mm, 1 mm, 0.5 mm, 0.25 mm, and 0.053 mm to separate the soil into six different aggregate sizes. The nest was positioned at the peak of the oscillating cylinder, with distilled water being added to each cylinder until it reached above the highest sieve. Before proceeding with wet-sieving, the soils were immersed in distilled water for 10 min. Throughout the fractionating process, the stroke length in the vertical direction stayed constant at 4 cm, the oscillation time at 10 min, and the frequency at 900 cycles per hour. According to the procedures mentioned above, the soil aggregates were classified into six different sizes: >2 mm, 2–1 mm, 1–0.5 mm, 0.5–0.25 mm, 0.25–0.053 mm, and <0.053 mm.

Subsequently, the aggregates of different particle sizes after fractionation were collected separately in dry containers and then dried to constant weight at 60 °C. The mass of each aggregate fraction was accurately measured using an electronic balance, and the data were recorded. The recovery rate of aggregate fractionation was calculated by comparing the total mass of fractionated aggregates with the original soil sample mass. In this study, the recovery rate is 91.2–93.5%, which is within the acceptable recovery rate range [51].

2.6. Data Calculation

2.6.1. Soil Aggregate Stability

The indicators of soil aggregate stability included a proportion of aggregates with a size larger than 0.25 mm in diameter (R_0.25), mean weight diameter (MWD), geometric mean diameter (GMD), and fractal dimension (D). These four indicators were computed by Equations (1)–(4), based on the methods described by Zhou et al. (2020), Bavel (1950), Kemper and Rosenau (1986), and Yang et al. (1993), respectively [19,52,53,54].

$${\mathrm{R}}_{0.25}={\mathrm{M}}_{\mathrm{r}>0.25}/{\mathrm{M}}_{\mathrm{T}}$$

(1)

$$\mathrm{MWD}={\sum }_{\mathrm{i}=1}^{\mathrm{n}}\left({\overline{\mathrm{X}}}_{\mathrm{i}}{\mathrm{P}}_{\mathrm{i}}\right)$$

(2)

$$\mathrm{GMD}=\exp \left({\displaystyle \frac{{\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{P}}_{\mathrm{i}}\lg {\overline{\mathrm{X}}}_{\mathrm{i}}}{{\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{P}}_{\mathrm{i}}}}\right)$$

(3)

$$\left(3-\mathrm{D}\right)\lg \left({\displaystyle \frac{{\overline{\mathrm{X}}}_{\mathrm{i}}}{{\mathrm{X}}_{\max }}}\right)=\lg \left({\displaystyle \frac{{\mathrm{W}}_{\left(\delta <{\overline{\mathrm{X}}}_{\mathrm{i}}\right)}}{\mathrm{W}}}\right)$$

(4)

where M_r>0.25 indicates the mass of aggregates that are larger than 0.25 mm, and M_T means the total mass of the aggregates. $\overline{{\mathrm{X}}_{\mathrm{i}}}$ refers to the mean diameter of each size (mm), calculated as the average of the minimum and maximum diameters within each aggregate. P_i represents the mass fraction of each size, X_max indicates the soil particles with a diameter of 10 mm, and ${\mathrm{W}}_{\left(\delta <{\overline{\mathrm{X}}}_{\mathrm{i}}\right)}$ signifies the sum of soil weights with size < $\overline{{\mathrm{X}}_{\mathrm{i}}}$.

2.6.2. Relative Contribution Rate, Allocation Amount, and Allocation Rate

Regarding the soil N components of TN, NO₃⁻-N, and NH₄⁺-N, the relative contribution rate (f, %), allocation amount (C_s-straw, mg kg⁻¹) and allocation ratio (P, %) of straw N in soil N components were calculated by Equations (5)–(7), which was according to the methods described by Liang et al. (2011), De Troyer et al. (2011), and Chen et al. (2018), respectively [9,12,55].

f = (atom% ¹⁵N_i − atom% ¹⁵N_CK)/(atom% ¹⁵N_R − atom% ¹⁵N_CK) × 100

(5)

C_s-straw = C_s × f/100

(6)

P = C_s-straw/C′_R × 100

(7)

where atom% ¹⁵N_i and atom% ¹⁵N_CK represent the atom% ¹⁵N in soil N components (TN, NO₃⁻-N, and NH₄⁺-N) after incubation with and without added straw, respectively. atom% ¹⁵N_R represents the initial atom% ¹⁵N of the corn straw before incubation. C_s denotes the content of soil N components in the straw-added treatment after incubation (mg kg⁻¹ soil), and C′_R represents the initial TN content of the corn straw (g kg⁻¹ soil).

Regarding the MBN, the relative contribution rate (∆f, %), allocation amount (C_s-straw-MBN, mg kg⁻¹), and allocation ratio (P_MBN, %) of straw N in soil N components were calculated by Equations (8)–(12).

f₁ = (atom% ¹⁵N_fumigated − atom% ¹⁵N_CK)/(atom% ¹⁵N_R − atom% ¹⁵N_CK) × 100

(8)

f₂ = (atom% ¹⁵N_{non-fumigated} − atom% ¹⁵N_CK)/(atom% ¹⁵N_R − atom% ¹⁵N_CK) × 100

(9)

∆f = f₁ − f₂

(10)

C_s-straw-MBN = C_s-MBN × ∆f/100

(11)

P_MBN = C_s-straw-MBN/C′_R × 100

(12)

where atom% ¹⁵N_fumigated and atom% ¹⁵N_{non-fumigated} represent the atom% ¹⁵N in soil MBN after incubation with added straw for fumigated and non-fumigated samples, respectively. C_s-MBN denotes the content of soil MBN in the straw-added treatment after incubation (mg kg⁻¹ soil).

2.7. Statistical Analysis

A one-way ANOVA was conducted to compare the contents of soil N components (TN, NO₃⁻-N, NH₄⁺-N, and MBN), and relative contribution rate, allocation amount, and ratio, as well as the distribution of each soil aggregate size and soil aggregate stability indicators, across the different freeze–thaw frequencies. The mean treatment effects were compared by the test of Tukey’s honestly significant difference at p < 0.05. The Pearson correlation analysis was conducted by SPSS software version 20.0 (IBM, Armonk, NY, USA). The plots showing the contents of N components, and relative contribution rate, allocation amount, and ratio in bulk soils and soil aggregates, as well as the distribution of each soil aggregate size, and soil aggregate stability indicators were generated using Origin software 2021 (OriginLab, Northampton, MA, USA).

In addition, a two-way ANOVA was utilized to simultaneously examine the effects of two factors (freeze–thaw frequency and freeze–thaw temperature) on the dependent variable and to test for any interaction between these factors. The dependent variable included soil N components (TN, NO₃⁻-N, NH₄⁺-N, and MBN), relative contribution rate, allocation amount, and ratio both in bulk soils and soil aggregates, as well as soil aggregate stability. This approach enabled us to determine if the impact of one factor on the dependent variable varied depending on the level of the other factor. This analysis was conducted using SPSS software version 20.0 (IBM, Armonk, NY, USA), and the results were evaluated at the significance levels of p < 0.05, p < 0.01, and p < 0.001. In instances where a significant interaction was detected, additional Tukey’s HSD post hoc tests with a significance level of 0.05 were performed to identify which specific group differences contributed to the interaction effect.

3. Results

3.1. Effects of Freeze–Thaw Frequency and Temperature on Soil Nitrogen Components in Bulk Soils

As the frequency of freeze–thaw cycles grows, there was a noticeable increase in the contents of soil N components, including TN (Figure 2a), NO₃⁻-N (Figure 2b), NH₄⁺-N (Figure 2c), and MBN (Figure 2d), within the range of freezing temperatures studied. Specifically, compared with 1 freeze–thaw cycle, the 17 cycles significantly (p < 0.05) increased soil TN, NO₃⁻-N, NH₄⁺-N, and MBN contents by 19.84%, 66.71%, 201.07%, and 6.49-fold for −9 °C freezing temperature, by 20.04%, 51.28%, 62.66%, and 1.36-fold for −18 °C freezing temperature, and by 18.87%, 58.60%, 221.20%, and 4.73-fold for −26 °C freezing temperature, respectively (Figure 2a–d).

Lower freezing temperature led to greater reduction in soil TN and NO₃⁻-N contents, within the range of the numbers of freeze–thaw cycles studied (Figure 2a,b). The contents of soil NH₄⁺-N changed inconsistently with the decrease in freezing temperature under the five freeze–thaw frequencies (Figure 2c). Specifically, as freezing temperature decreases, the content of soil NH₄⁺-N in the 1 and 3 freeze–thaw cycles both increased, while that in the 5, 9, and 17 freeze–thaw cycles decreased first and then increased (Figure 2c). In addition, soil MBN content in the five freeze–thaw frequencies all increased first and then decreased with the decrease in freezing temperature (Figure 2d).

3.2. Effects of Freeze–Thaw Frequency and Temperature on Relative Contribution Rate, Allocation Amount, and Allocation Ratio of Corn Straw Nitrogen in Soil Nitrogen Components for Bulk Soils

In general, the relative contribution rate, allocation amount, and allocation ratio of corn straw N in soil TN, NO₃⁻-N, NH₄⁺-N, and MBN all enhanced as freeze–thaw frequency increases at any freezing temperature (Figure 3). Specifically, the relative contribution rate of corn straw N in soil TN significantly (p < 0.05) increased by 21.11% and 19.82% under 17 freeze–thaw cycles compared to 1 cycle at freezing temperatures of −9 °C and −18 °C (Figure 3a). Simultaneously, compared with 1 freeze–thaw cycle, the relative contribution rate of corn straw N in soil NO₃⁻-N under 5, 9, and 17 cycles were significantly increased (p < 0.05) by 43.1–139.51% for −9 °C, by 55.03–164.24% for −18 °C, and by 34.62–112.93% for −26 °C freezing temperature (Figure 3b). In regard to the relative contribution rate of corn straw N in soil NH₄⁺-N, only 17 freeze–thaw cycles increased by 64.36% to 1 cycle at −18 °C freezing temperature, other freeze–thaw cycles at the three freezing temperatures showed no significant (p ≥ 0.05) difference (Figure 3c). With respect to the relative contribution rate of corn straw N in soil MBN, compared to 1 freeze–thaw cycle, the 17 freeze–thaw cycles increased by 33.02% at −9 °C freezing temperature, and 9 and 17 cycles enhanced by 23.59% and 35.06% at −26 °C freezing temperature, respectively, while no obvious difference (p ≥ 0.05) was observed among the five freeze–thaw frequencies at −18 °C freezing temperature (Figure 3d).

A similar tendency was observed between the allocation amount and allocation ratio of corn straw N in soil N components (Figure 3e–l). In comparison with one freeze–thaw cycle, the other four cycles both increased allocation amount and ratio of corn straw N in soil TN, NO₃⁻-N, NH₄⁺-N, and MBN to varying degrees across the three freezing temperatures (Figure 3e–l). The increments for allocation amount ranged from 29.58 to 69.58% (Figure 3e), 89.82–621.61% (Figure 3f), 2.1–4.49-fold (Figure 3g), and 4.74–13.54-fold (Figure 3h) correspondingly, and allocation ratio showed the same increments (Figure 3i–l).

The labile N components, including NO₃⁻-N, NH₄⁺-N, and MBN, showed noticeable changes under the same freeze–thaw frequency among the three freezing temperatures (Figure 3a–i). With the decrease in freezing temperature (the decrease in freezing temperature), the relative contribution rate, allocation amount and ratio of corn straw N in soil NO₃⁻-N all first increased and then slowly decreased (Figure 3b,f,j). The allocation amount and ratio of soil TN and NH₄⁺-N showed a decreasing trend (Figure 3a,c,e,g,i,k), and that of soil MBN increased first and then fell sharply (Figure 3d,h,l).

3.3. Effects of Freeze–Thaw Frequency and Temperature on Soil Total Nitrogen Content in Soil Aggregates

With the increase in freeze–thaw frequency, TN contents in each soil aggregate size tended to decrease at any freezing temperature (Figure 4a–c). Detailedly, at −9 °C freezing temperature, compared to 1 freeze–thaw cycle, the other 4 cycles all significantly (p < 0.05) reduced TN content by 11.47–24.90% for >2 mm aggregate size, and by 14.52–22.81% for 2–1 mm aggregate size; only 17 cycles obviously (p < 0.05) decreased TN content by 9.24% for 0.25–0.053 mm aggregate size; the 5, 9, and 17 cycles significantly (p < 0.05) reduced TN content by 15.36%, 15.81%, and 16.27% correspondingly (Figure 4a). At −18 °C freezing temperature, in comparison with 1 freeze–thaw cycle, the 5, 9, and 17 cycles all significantly (p < 0.05) decreased TN content by 13.10–18.91%, 5.49–11.39%, 9.53–22.64%, and 15.25–24.01% for >2 mm, 0.5–0.25 mm, 0.25–0.053 mm, and <0.053 mm aggregate sizes, respectively; 17 cycles alone obviously (p < 0.05) reduced TN content by 11.01% for 1–0.5 mm aggregate size (Figure 4b). At −26 °C freezing temperature, compared to 1 freeze–thaw cycle, the 3, 5, and 9 cycles significantly (p < 0.05) decreased TN content by 4.99–9.19% for >2 mm aggregate size; the 5, 9, and 17 cycles all obviously (p < 0.05) reduced TN content with the range of 5.24–16.92% across the 2–1 mm, 0.25–0.053 mm, and 0.053 mm aggregate sizes; while only 17 cycles significantly (p < 0.05) increased TN content by 6.74% for 1–0.5 mm aggregate size (Figure 4c). Furthermore, with the decrease in soil aggregate sizes, the TN content at each freezing temperature and frequency showed a trend of decreasing-increasing-decreasing, and the maximum value basically appeared in the aggregate size of 0.25–0.053 mm (Figure 4).

3.4. Effects of Freeze–Thaw Frequency and Temperature on Relative Contribution Rate, Allocation Amount and Allocation Ratio of Corn Straw Nitrogen in Total Nitrogen for Soil Aggregates

With the increase in the freeze–thaw frequency, at −9 and −18 °C, the most obvious changes in relative contribution rate were 0.5–0.25 mm aggregate size, and the other four freeze–thaw cycles reduced relative contribution rate by 24.64–43.55% than one cycle (Figure 5a,b). At −26 °C, the relative contribution rate of <0.053 mm aggregate size exhibited the largest variation, showing a trend of first rising and then falling (Figure 5c). Compared with 1 freeze–thaw cycle, the relative contribution rate of 3 cycles significantly (p < 0.05) increased by 33.46%, while that of 5, 9, and 17 cycles significantly (p < 0.05) decreased by 27.34%, 17.33% and 24.52%, respectively (Figure 5c).

Regarding allocation amount, as the freeze–thaw frequency increases, the largest changes occurred in the 1–0.5 mm and 0.5–0.25 mm aggregate sizes at −9 °C freezing temperature (Figure 5d). Compared with 1 freeze–thaw cycle, the other four cycles reduced allocation amount by 19.59–43.18% and 31.74–41.46% in the 1–0.5 mm and 0.5–0.25 mm aggregate sizes correspondingly (Figure 5d). At −18 °C, only the 0.5–0.25 mm size showed the largest change (Figure 5e). Compared with the first freeze–thaw cycle, the other four freeze–thaw cycles significantly (p < 0.05) reduced allocation amount by 52.68–58.77% than one cycle for 0.5–0.25 mm aggregate size (Figure 5e). At −26 °C, the biggest change was found in the <0.053 mm size, which showed a trend of first increasing and then decreasing (Figure 5f). Compared with 1 freeze–thaw cycle, the allocation amount was increased by 43.17% in the 3 cycles, while it was decreased significantly in the 5, 9, and 17 cycles by 42.49%, 32.38% and 40.38%, respectively, for the <0.053 mm size (Figure 5f). Simultaneously, the variation tendency of the allocation ratio was consistent with the allocation amount (Figure 5g–i).

3.5. Effects of Freeze–Thaw Frequency and Temperature on Soil Aggregate Distribution

As the freeze–thaw frequency increases, the proportion of each aggregate size changed significantly, but its variation trend was different under the three freezing temperatures (Figure 6). At the freezing temperature of −9 °C, with the increase in freeze–thaw frequency, the proportions of >2 mm and <0.053 mm sizes tended to increase with the increments of 120.42% and 150.96%, respectively, while those of 2–1 mm, 1–0.5 mm and 0.5–0.25 mm sizes decreased, and the proportion of 0.25–0.053 mm size remained unchanged (Figure 6a).

Different from the treatment at −9 °C, with the increase in freeze–thaw frequency at −18 °C and −26 °C, the proportions of >2 mm and 2–1 mm aggregate sizes showed an “M” shaped trend of increasing–decreasing–increasing (Figure 6b,c). The proportion of 0.5–0.25 mm size at −18 °C and 0.25–0.053 mm size at −26 °C significantly (p < 0.05) increased as freeze–thaw frequency increases, and the increments were 71.54% and 93.53% after 17 freeze–thaw cycles compared with 1 cycle, respectively (Figure 6b,c). The proportion of <0.053 mm size decreased as the frequency of freeze–thaw increased. Specifically, it decreased by 22.59% at −18 °C and by 23.13% at −26 °C after 17 freeze–thaw cycles compared to just 1 cycle (Figure 6b,c).

3.6. Effects of Freeze–Thaw Frequency and Temperature on Soil Aggregate Stability

At −9 °C freezing temperature, with the increase in freeze–thaw frequency, the values of R_0.25, MWD, and GMD gradually decreased, but the value of fractal dimension gradually increased, indicating the decreasing aggregate stability (Figure 7a–d). At −18 °C and −26 °C, the situation was basically opposite to that at −9 °C. MWD and GMD showed an M-shaped trend (Figure 7b,c). The fractal dimension of soil aggregates tended to be stable after three freeze–thaw cycles under three different freezing temperatures (Figure 7d). In comparison with 1 cycle, the 17 cycles significantly (p < 0.05) reduced the value of fractal dimension by 26.51%, 3.56%, and 3.10% at −9 °C, −18 °C, and −26 °C, correspondingly (Figure 7d).

3.7. Relationships of Soil Properties in Bulk Soils and Soil Aggregate Sizes Under Freeze–Thaw Action

Two-factor analysis of variance revealed that the freeze–thaw frequency, freeze–thaw temperature, and their interaction showed significant effects on the MBN content, as well as the relative contribution rate, allocation amount, and allocation ratio of NO₃⁻-N in the bulk soil, and all of them have reached the extremely significant level (

Table 1. The p values of two-factor variance analysis of the freeze–thaw frequency and freeze–thaw temperature on soil properties.

Soil Types	Soil Indicators	Nitrogen Components/Aggregate Sizes	Factors
			Freeze–Thaw Frequency	Freeze–Thaw Temperature	Freeze–Thaw Frequency × Freeze-Thaw Temperature
Bulk soils	Soil nitrogen component contents	TN	<0.001 ***	<0.001 ***	0.401
		NO3−-N	<0.001 ***	0.001 **	0.593
		NH4+-N	<0.001 ***	0.108	0.005 **
		MBN	<0.001 ***	<0.001 ***	0.001 **
	Relative contribution rate	TN	<0.001 ***	0.753	0.859
		NO3−-N	<0.001 ***	<0.001 ***	<0.001 ***
		NH4+-N	<0.001 ***	0.294	0.333
		MBN	0.042 *	0.983	0.851
	Allocation amount	TN	<0.001 ***	0.034	0.807
		NO3−-N	<0.001 ***	<0.001 ***	<0.001 ***
		NH4+-N	<0.001 ***	0.214	0.079
		MBN	<0.001 ***	0.001 **	0.457
	Allocation ratio	TN	<0.001 ***	0.034 *	0.807
		NO3−-N	<0.001 ***	<0.001 ***	<0.001 ***
		NH4+-N	<0.001 ***	0.213	0.078
		MBN	<0.001 ***	0.001 **	0.457
Soil aggregates	TN contents	>2 mm	<0.001 ***	<0.001 ***	0.040 *
		2–1 mm	<0.001 ***	<0.001 ***	<0.001 ***
		1–0.5 mm	0.062	<0.001 ***	<0.001 ***
		0.5–0.25 mm	0.007 **	<0.001 ***	0.006 **
		0.25–0.053 mm	<0.001 ***	0.296	<0.001 ***
		<0.053 mm	<0.001 ***	<0.001 ***	0.061
	Relative contribution rate	>2 mm	0.151	0.563	0.034 *
		2–1 mm	0.273	0.431	0.001 ***
		1–0.5 mm	<0.001 ***	0.004 **	<0.001 ***
		0.5–0.25 mm	<0.001 ***	<0.001 ***	<0.001 ***
		0.25–0.053 mm	<0.001 ***	<0.001 ***	<0.001 ***
		<0.053 mm	<0.001 ***	<0.001 ***	<0.001 ***
	Allocation amount	>2 mm	0.003 **	<0.001 ***	0.003 **
		2–1 mm	0.384	0.038	0.005 **
		1–0.5 mm	<0.001 ***	0.001 **	<0.001 ***
		0.5–0.25 mm	<0.001 ***	<0.001 ***	<0.001 ***
		0.25–0.053 mm	<0.001 ***	<0.001 ***	<0.001 ***
		<0.053 mm	<0.001 ***	<0.001 ***	<0.001 ***
	Allocation ratio	>2 mm	0.003 **	<0.001 ***	0.003 **
		2–1 mm	0.384	0.038 *	0.005 **
		1–0.5 mm	<0.001 ***	0.001 **	<0.001 ***
		0.5–0.25 mm	<0.001 ***	<0.001 ***	<0.001 ***
		0.25–0.053 mm	<0.001 ***	<0.001 ***	<0.001 ***
		<0.053 mm	<0.001 ***	<0.001 ***	<0.001 ***
Soil aggregate stability	R0.25		<0.001 ***	<0.001 ***	<0.001 ***
	MWD		<0.001 ***	<0.001 ***	<0.001 ***
	GMD		<0.001 ***	<0.001 ***	<0.001 ***
	D		<0.001 ***	<0.001 ***	<0.001 ***

*, **, and *** indicate the significant differences at p < 0.05, p < 0.01, and p < 0.001, respectively. n = 45. TN signifies total nitrogen, NO₃⁻-N and NH₄⁺-N indicate nitrate nitrogen and ammonium nitrogen, respectively, and MBN means microbial biomass nitrogen. R_0.25 indicates the distribution of aggregates larger than 0.25 mm in diameter. MWD and GMD represent mean weight diameter and geometric mean diameter, respectively. D indicates fractal dimension.

). For soil aggregates, freeze–thaw frequency, freeze–thaw temperature, and their interaction presented significant effects on the relative contribution rate of 1–0.5 mm, 0.5–0.25 mm, 0.25–0.053 mm, and <0.053 mm sizes (Table 1). Simultaneously, they showed significant influence on the allocation amount and ratio of >2 mm, 1–0.5 mm, 0.5–0.25 mm, 0.25–0.053 mm, and <0.053 mm sizes (Table 1). Moreover, the freeze–thaw frequency, freeze–thaw temperature, and their interaction had extremely significant effects on the four indexes of aggregate stability (Table 1).

Pearson correlation analysis demonstrated that the contents of NO₃⁻-N and MBN in bulk soils, as well as the proportion of 2–1 mm, 0.5–0.25 mm, and <0.053 mm aggregate sizes, and the TN content of 2–1 mm aggregate size, all had a significant correlation with fractal dimension (

Table 2. Pearson correlation analysis between soil properties and fractal dimension under the three freeze–thaw intensities.

Soil Types	Soil Properties	Soil Indicators/Aggregate Sizes	Freezing Temperatures
			−9 °C	−18 °C	−26 °C
Bulk soils	TN	Nitrogen content	0.521 *	−0.440	−0.609 *
		Relative contribution rate	0.433	−0.369	−0.604 *
		Allocation amount	0.466	−0.379	−0.680 **
		Allocation ratio	0.466	−0.379	−0.680 **
	NO3−-N	Nitrogen content	0.723 **	−0.532 *	−0.726 **
		Relative contribution rate	0.624 *	−0.418	−0.711 **
		Allocation amount	0.530 *	−0.356	−0.752 **
		Allocation ratio	0.530 *	−0.356	−0.752 **
	NH4+-N	Nitrogen content	0.694 **	−0.474	−0.531 *
		Relative contribution rate	0.489	−0.403	−0.532 *
		Allocation amount	0.558 *	−0.388	−0.595 *
		Allocation ratio	0.558 *	−0.388	−0.595 *
	MBN	Nitrogen content	0.767 **	−0.589 *	−0.650 **
		Relative contribution rate	0.593 *	0.019	−0.582 *
		Allocation amount	0.632 *	−0.248	−0.689 **
		Allocation ratio	0.632 *	−0.248	−0.689 **
Soil aggregates	Proportion	>2 mm	0.061	0.007	0.007
		2–1 mm	−0.868 **	0.597 *	0.597 *
		1–0.5 mm	−0.971 **	0.098	0.098
		0.5–0.25 mm	−0.602 *	−0.622 *	−0.622 *
		0.25–0.053 mm	0.112	−0.533 *	−0.533 *
		<0.053 mm	0.978 **	0.861 **	0.861 **
	TN content	>2 mm	−0.817 **	0.469	0.469
		2–1 mm	−0.917 **	0.599 *	0.599 *
		1–0.5 mm	−0.348	−0.434	−0.434
		0.5–0.25 mm	−0.100	0.205	0.205
		0.25–0.053 mm	−0.631 *	0.461	0.461
		<0.053 mm	−0.651 **	0.435	0.435
	Relative contribution rate	>2 mm	−0.007	−0.535 *	−0.535 *
		2–1 mm	0.524 *	0.247	0.247
		1–0.5 mm	−0.702 **	−0.029	−0.029
		0.5–0.25 mm	−0.653 **	−0.469	−0.469
		0.25–0.053 mm	−0.379	0.156	0.156
		<0.053 mm	0.247	0.509	0.509
	Allocation amount	>2 mm	−0.575 *	−0.419	−0.419
		2–1 mm	−0.227	0.354	0.354
		1–0.5 mm	−0.734 **	−0.103	−0.103
		0.5–0.25 mm	−0.698 **	−0.400	−0.400
		0.25–0.053 mm	−0.429	0.360	0.360
		<0.053 mm	−0.131	0.541 *	0.541 *
	Allocation ratio	>2 mm	−0.575 *	−0.420	−0.420
		2–1 mm	−0.227	0.355	0.355
		1–0.5 mm	−0.734 **	−0.103	−0.103
		0.5–0.25 mm	−0.698 **	−0.400	−0.400
		0.25–0.053 mm	−0.429	0.360	0.360
		<0.053 mm	−0.131	0.541 *	0.541 *

* and ** indicate the significant differences at p < 0.5 and p < 0.1, respectively. n = 15. TN signifies total nitrogen, NO₃⁻-N and NH₄⁺-N indicate nitrate nitrogen and ammonium nitrogen, respectively, and MBN means microbial biomass nitrogen.

). There were negative and significant correlations between the fractal dimension and the relative contribution rate, allocation amount, and allocation ratio in bulk soils under the freeze–thaw action of −26 °C/5 °C. However, as for the aggregates, no consensus was observed between the fractal dimension and the relative contribution rate, allocation amount, and allocation ratio for each freeze–thaw temperature (Table 2).

4. Discussion

4.1. Dynamics of Nitrogen Components in Bulk Soils Due to Freeze–Thaw Action

This study found that with increasing freeze–thaw frequency (number of cycles), the content of N components (TN, NO₃⁻-N, NH₄⁺-N, and MBN), as well as their relative contribution rate, allocation amount, and allocation ratio in bulk soils, all showed an increasing trend under three freeze–thaw intensities (Figure 2 and Figure 3), which was consistent with our first hypothesis and the previous research finding [56]. This phenomenon may be explained by the following aspects. Firstly, freeze–thaw action will destroy the soil aggregate structure, rupture soil microbial cells, release mineral N, and then release a large amount of nutrients accumulated during the freezing period of soil, increasing the content of inorganic N nutrients in soils [29,57]. Secondly, frequent freeze–thaw alternations during the thawing of frozen soil increase soil water content, enhance microbial activity, accelerate the nitrification and denitrification processes of soil N [58,59,60], accelerate the mineralization and decomposition rate of litter, and further increase the effective resources in soils. Thirdly, multiple freeze–thaw alternations can cause the death of soil microorganisms, so that the residual microorganisms in the soil have enough matrix, stimulate microbial activity, and stabilize and utilize the labile nutrients in the soil [61].

With decreasing freezing temperature, the TN and NO₃⁻-N contents in bulk soils tended to decrease for each freeze–thaw frequency (Figure 2a,b). Freeze–thaw cycles can enhance N transformation processes, leading to increased N losses through leaching and denitrification. Research has indicated that freeze–thaw cycles enhance the leaching of NO₃⁻-N, especially in soils with high initial nitrate levels, which is probably intensified with more intense freeze–thaw activity, leading to a decrease in soil NO₃⁻-N content [62].

The relative contribution rate, as well as the allocation amount and ratio of N compounds, are also influenced by the temperature of freeze–thaw cycles. The initial increase in MBN indicators followed by a sharp decline in bulk soils with lower freezing temperature (Figure 3d,h,l) aligns with the observation that microbial activity is initially enhanced by freeze–thaw processes, promoting N mineralization and microbial biomass accumulation [63]. Nevertheless, as the freezing temperature decreases, the extreme temperature changes can impose significant stress on microbial communities, resulting in a reduction in microbial biomass N [64]. Furthermore, as freezing temperature decreases, the relative contribution rate and the allocation amount and ratio of NH₄⁺-N decrease (Figure 3c,g,k), while NO₃⁻-N shows an initial increase followed by a gradual decline (Figure 3b,f,j). This trend is likely due to the differential sensitivity of NH₄⁺-N and NO₃⁻-N to the freeze–thaw processes. NH₄⁺-N is more tightly bound to soil particles due to its positive charge, which allows it to be adsorbed onto negatively charged soil surfaces and makes it less susceptible to leaching.

Freeze–thaw cycles lead to alterations in soil pH, redox potential, and soil structure, thereby influencing N dynamics [26,65]. Freezing can cause soil compaction and reduce pore size, limiting oxygen availability, and promoting denitrification. In contrast, thawing can improve soil aeration, enhancing nitrification [26]. These opposing processes can result in complex changes in N forms and their availability.

As a consequence, the observed variations in N components, as well as their relative contribution rate, allocation amount, and ratio with increasing freeze–thaw frequency and decreasing freezing temperature, are probably due to the combined effects of N transformation and loss, microbial dynamics, differential sensitivity of N forms, and soil chemical and physical changes [66]. These mechanisms together shape the observed tendencies in N cycling and availability.

4.2. Dynamics of Nitrogen Components in Soil Aggregates Due to Freeze–Thaw Action

The consistent decline in TN across all aggregate sizes as the number of freeze–thaw cycles increased (Figure 4) strongly supports the phenomenon that physical disintegration and accelerated microbial processing act synergistically to deplete soil N [67,68]. Our findings corroborate the conceptual model proposed by Suh et al. (2022) [69], who argued that each freeze–thaw event expands pre-existing micro-fissures through ice-lens growth. The magnitude of TN reduction observed (−4.99 to −24.90%) in this study is comparable to the 10–30% losses reported in alpine meadows after 10 freeze–thaw cycles [68] and exceeds the 5–15% losses found in boreal forest floor material [70]. The higher sensitivity of macro-aggregates (>2 mm) at −9 °C and −18 °C suggests that ice nucleation preferentially occurs in large intra-aggregate pores, propagating cracks along weaker organic binding points [71]. Consequently, macro-aggregates lose their capacity to physically protect particulate organic N, thereby amplifying TN depletion.

Although the freezing temperature of −26 °C induced overall TN losses generally, the effect was markedly attenuated compared with that of −18 °C, and even reversed (+6.74%) for the 1–0.5 mm aggregate size after 17 freeze–thaw cycles (Figure 4), which was inconsistent with our second hypothesis. Three non-mutually exclusive mechanisms may explain this paradox. Firstly, at −26 °C, the rate of ice-crystal growth is so rapid that micro-pockets of unfrozen water become viscous and glass-like, limiting hydraulic redistribution and reducing mechanical stress [72]. Secondly, extreme cold suppresses microbial activity to a greater extent than physical disruption, lowering gross N mineralization during thaw [73]. Thirdly, the observed TN gain in 1–0.5 mm aggregate size after 17 cycles at −26 °C may reflect cryo-concentration of dissolved organic N (DON) into this size window, a process analogous to the “freeze-fractionation” described by Lipson et al. (1999) [74] in snow-covered soils. This hypothesis is supported by the concomitant reduction in TN in adjacent size fractions (2–1 mm, 0.25–0.053 mm) during the same treatment, implying a net transfer rather than de novo N synthesis.

The “decreasing–increasing–decreasing” TN pattern across the aggregate size spectrum, with a peak in the 0.25–0.053 mm range (Figure 4), aligns with the hierarchical aggregate model [75]. Macro-aggregates (>2 mm) are bound primarily by transient, labile polysaccharides that are highly susceptible to freeze–thaw disturbance, thus TN in this size declines rapidly. Conversely, micro-aggregates (<0.053 mm) are cemented by persistent organo-mineral complexes that resist physical disruption but possess high surface reactivity, promoting N adsorption and thus maintaining moderate TN levels. The 0.25–0.053 mm aggregates represent a transitional zone where physical protection is still operative, yet surface area is sufficiently large to retain microbial by-products [76].

The intricate interplay between freeze–thaw frequency, temperature, and soil aggregate size demonstrably governs the fate of corn straw-derived N, significantly altering its relative contribution, allocation, and distribution within the soil matrix. Our findings reveal that these dynamics are not uniform but exhibit pronounced aggregate size-class specificity and temperature dependency, highlighting the complexity of freeze–thaw impacts on N cycling in structured soils.

The pronounced sensitivity of the 0.5–0.25 mm aggregate size for relative contribution rate to increasing freeze–thaw frequency at −9 °C and −18 °C freezing temperature (Figure 5) underscores a critical vulnerability of this specific size class. This size likely represents a transition zone between macro- and micro-aggregates, potentially possessing structural characteristics (e.g., pore size distribution, binding strength) that make it particularly susceptible to the physical disruptive forces of repeated ice lens formation and melting [77]. Each freeze–thaw cycle subjects soil aggregates to internal stresses from ice crystallization and hydraulic pressures during thawing, progressively weakening bonds and fracturing particles [78]. The 0.5–0.25 mm aggregates may experience optimal conditions for this disruptive mechanism at these temperatures, leading to their breakdown into smaller fractions or disruption of their protective architecture. Consequently, N associated with or physically protected within these aggregates becomes more exposed, potentially facilitating its mineralization, leaching, or redistribution into other size classes, thereby reducing its relative contribution and allocated amount within this specific fraction.

The shift in the most responsive aggregate size class from 0.5 to 0.25 mm at −9 °C/−18 °C to <0.053 mm at −26 °C (Figure 5) represents a fundamental shift in FT impact mechanisms driven by temperature intensity. At milder sub-zero temperatures (−9 °C, −18 °C), the physical disruption likely dominates, preferentially affecting vulnerable mid-sized aggregates. In contrast, the extreme cold of −26 °C appears to profoundly influence the finest fraction (<0.053 mm). The initial increase in relative contribution rate and allocation amount of N in the <0.053 mm fraction after three cycles at −26 °C is particularly intriguing. This suggests that initial FT cycles at this severe temperature may promote the fragmentation of larger aggregates or the stabilization of N onto newly exposed mineral surfaces within the clay/silt fraction, effectively concentrating N here. Cryoturbation or enhanced physical protection within newly formed micro-aggregates under extreme cold could also play a role [79]. However, the subsequent significant decline with further cycling (5–17 cycles) indicates that prolonged or intense FT stress at −26 °C eventually overcomes any initial protective or concentrating mechanisms. This could involve the physical disintegration of micro-aggregates themselves, chemical degradation of organic–mineral complexes protecting N, or the mortality of microbial communities responsible for stabilizing N in this fraction, leading to N loss or redistribution. This non-monotonic response (peak at three cycles) highlights a threshold effect at −26 °C, where initial cycles induce changes beneficial to N retention in fines, but excessive cycling proves detrimental.

4.3. Variations in Soil Aggregate Stability Induced by Freeze–Thaw Action

As freeze–thaw frequency increases, higher freezing temperature (−9 °C/5 °C) are likely to decrease the proportion of macroaggregates (2–1 mm, 1–0.5 mm, and 0.5–0.25 mm) while increasing the proportion of silt + clay fractions (<0.053 mm) (Figure 6a), resulting in reduced soil aggregate stability (Figure 7). This reduction in macroaggregates is mainly due to the mechanical disruption caused by the expansion of water during freezing, which damages the physical structure of these aggregates [80]. Repeated freeze–thaw cycles at this temperature cause larger aggregates to break down into smaller particles, making them more vulnerable to erosion and loss of stability [77]. Moreover, the relatively higher freezing temperature might not be strong enough to trigger substantial re-aggregation during thawing, leading to an overall decline in aggregate stability [16]. The rise in silt + clay fractions is probably because finer particles are more resilient to the freeze–thaw process or are released from macroaggregates during breakdown [77]. Additionally, as freeze–thaw frequency increases, higher freezing temperatures reduce the TN content in silt + clay fractions (<0.053 mm) (Figure 4a). This likely results from the mechanical disruption of macroaggregates during freezing, which releases N compounds and causes their redistribution into smaller fractions [81]. The decrease in TN content in silt + clay fractions may also be due to these finer particles being more susceptible to leaching and loss during freeze–thaw cycles [14,82].

In contrast, as freeze–thaw frequency increases, lower freezing temperature (−18 °C/5 °C and −26 °C/5 °C) tend to increase the proportion of microaggregates (0.25–0.053 mm) while decreasing the proportion of silt + clay fractions (<0.053 mm) (Figure 6b,c), thereby enhancing soil aggregate stability, which was inconsistent with our third hypothesis (Figure 7). The increase in microaggregates could be attributed to the more intense mechanical action of freezing and thawing, which may cause smaller particles to bind together more tightly [83].

Additionally, the reduction in the silt + clay fractions suggests that these particles are either being incorporated into macroaggregates or are being lost through processes like leaching or erosion [16]. Meanwhile, with the increase in freeze–thaw frequency, the TN content tends to decline, induced by lower freezing temperature, both in microaggregates (0.25–0.053 mm) and in silt + clay fractions (<0.053 mm) (Figure 4b,c). This indicates that higher intensity freeze–thaw cycles can induce substantial alterations in the physical structure of smaller aggregates, resulting in the release and potential loss of N compounds. The decline may also be associated with increased microbial activity and N mineralization processes that take place during intense freeze–thaw cycles [16,84].

Furthermore, the mechanical pressure exerted by the expansion of water during freezing causes soil aggregates to break down. At −9 °C, the mechanical action is strong enough to break apart macroaggregates, but it is not intense enough to facilitate significant re-aggregation during thawing. In contrast, at −18 °C and −26 °C, the more extreme temperature changes result in greater mechanical stress, which causes soil particles to reorganize and form more stable aggregates during thawing [85]. At the same time, freeze–thaw cycles can boost microbial activity, especially during thawing. When temperatures are low (−18 °C and −26 °C), extreme freeze–thaw cycles can increase microbial activity, which results in the creation of organic compounds that serve as binding agents for soil particles. These organic compounds enhance aggregate stability by facilitating the development of larger and more stable aggregates [85]. In addition, the stability of soil aggregates is also affected by soil pore structure and water content. At −9 °C, the relatively higher freezing temperature may not be strong enough to cause substantial alterations in soil pore structure, leading to less stable soil aggregates. In contrast, at −18 °C and −26 °C, the more intense freeze–thaw cycles can modify soil pore structure, resulting in improved water retention and movement within the soil. This alteration in soil structure can further boost aggregate stability [77].

The influence of freeze–thaw cycles on soil aggregate stability and N dynamics is strongly influenced by the temperature and frequency of these cycles. Higher freezing temperature cycles often decrease aggregate stability by fragmenting macroaggregates and increasing the proportion of silt and clay complexes, whereas lower freezing temperature cycles improve aggregate stability by facilitating the formation of microaggregates.

Consequently, understanding these dynamics is essential for forecasting soil behavior in areas with frequent freeze–thaw cycles, especially in high-latitude and high-altitude regions. Future research should concentrate on clarifying the specific mechanisms by which freeze–thaw cycles impact soil aggregate stability and N dynamics, including the roles of soil physicochemical properties and microbial communities.

4.4. The Interactive Impacts of Freeze–Thaw Frequency and Temperature on Soil Nitrogen Dynamics and Aggregate Stability

The results of the two-way ANOVA highlight the significant impacts of freeze–thaw frequency, freeze–thaw temperature, and their interaction on various soil properties, including MBN content, NO₃⁻-N dynamics, and aggregate stability (Table 1). These findings underscore the complexity of soil responses to freeze–thaw cycles, which are influenced by both the frequency and intensity of these cycles.

In regard to bulk soils, the interaction between freeze–thaw frequency and temperature likely influences soil N dynamics through several mechanisms. Higher freeze–thaw frequencies may enhance microbial activity and soil moisture dynamics, which in turn affect N mineralization and nitrification processes [86,87]. The temperature at which these cycles occur can modulate the rate and extent of these processes, leading to differential effects on MBN and NO₃⁻-N content. For instance, lower temperatures may slow microbial activity, reducing N mineralization, while higher temperatures may accelerate these processes, leading to increased NO₃⁻-N availability [88].

In terms of soil aggregate stability, the interaction between freeze–thaw frequency and temperature can influence soil structure through changes in soil moisture content and freeze–thaw-induced mechanical stress. Higher frequencies of freeze–thaw cycles may lead to more frequent expansion and contraction of soil pores, which can enhance aggregate formation and stability [89]. However, the temperature at which these cycles occur can modulate the extent of these mechanical stresses, leading to differential effects on aggregate stability.

The Pearson correlation analysis revealed significant relationships between soil N content, aggregate size distribution, and fractal dimension, indicating that these properties are interconnected (Table 2). The significant correlation between NO₃⁻-N and MBN content and fractal dimension suggests that soil N dynamics are closely related to soil structural complexity. The negative correlations between fractal dimension and N dynamics under specific freeze–thaw conditions (−26 °C/5 °C) indicate that higher fractal dimensions may be associated with lower N availability, possibly due to increased immobilization of N within more complex soil aggregates [90]. However, for soil aggregates, the lack of consensus between fractal dimension and N dynamics across different freeze–thaw temperatures suggests that the relationship between soil structure and N cycling is highly dependent on the specific environmental conditions. This highlights the need for a nuanced understanding of how freeze–thaw cycles influence soil properties, as the effects can vary significantly depending on the temperature and frequency of these cycles.

Therefore, the conceptual diagram illustrating the response mechanism of soil N components to aggregate stability under varying freeze–thaw frequencies and intensities was illustrated in Figure 8. The findings from this study have important implications for soil management practices, particularly in regions experiencing frequent freeze–thaw cycles. Understanding the interactive effects of freeze–thaw frequency and temperature on soil properties can inform strategies to optimize soil health and N use efficiency. For instance, managing soil moisture and organic matter content can help mitigate the negative impacts of freeze–thaw cycles on soil structure and N availability.

4.5. Implications and Limitations

The freeze–thaw cycle durations (84 h freeze + 84 h thaw) and temperature ranges were selected in the experimental design to be consistent with field conditions, and the experimental conditions were precisely controlled to thoroughly investigate the mechanisms by which freeze–thaw cycles affect soil properties. Although laboratory experiments cannot fully replicate the complex natural environment in the field, the experiment was positioned as process-oriented. Through precise control and repeated experiments, the mechanisms by which freeze–thaw cycles affect the functions of the soil ecosystem were deeply revealed. This process-oriented research method not only helps to understand the freeze–thaw processes of field soil but also provides important theoretical support for soil management and ecological protection. In addition, the long-term monitoring programs will be implemented to track soil N changes over multiple seasons and years. This approach will provide a more comprehensive understanding of the temporal dynamics and potential cumulative effects of thawing cycles.

In this study, the ¹⁵N recovery rate was relatively low (around 65%), which may be related to the loss of DON and gaseous N [91]. During freeze–thaw cycles, the formation of ice lenses in the soil and subsequent thawing enhance the mobility of water and solutes (including DON), which may lead to the leaching of DON from the soil into groundwater or surface water bodies, thereby reducing the ¹⁵N recovery rate. Concurrently, microbial activity significantly increases during the thawing process, potentially breaking down DON and converting it into other forms of N, such as ammonium (NH₄⁺) and nitrate (NO₃⁻). A portion of the ¹⁵N may also be lost through the volatilization of gaseous N (e.g., N₂O and N₂). These processes are more likely to occur under conditions of high humidity and temperature [92]. The low recovery rate suggests that our experimental design and measurement methods did not fully capture the changes in all N pools, resulting in an incomplete understanding of N cycling processes. Future research should improve experimental design, measure multiple N pools simultaneously, conduct long-term monitoring, and integrate model simulations to more comprehensively assess the pathways and mechanisms of N loss during N cycling, thereby providing a more scientific basis for N management strategies.

The use of chemically labeled fertilizers and plant residues allowed us to focus on the N dynamics related to plant residues, which is a critical component of soil N cycling. This approach provided valuable insights into how N is partitioned among different aggregate fractions after freeze–thaw cycles. However, we acknowledge that this method also has limitations. For instance, the use of pre-labeled plant residues may not fully capture the dynamic interactions between N from different sources in a natural soil environment. Future research could explore the simultaneous application of labeled fertilizers and plant residues to provide a more comprehensive understanding of N cycling in soil systems.

Mollisols are a diverse group of soils characterized by their high organic matter content and dark surface horizons. These soils are found under various land use types, including agricultural fields, grasslands, and forested areas [93,94]. The diversity in land use and management practices significantly influences aggregate stability and N distribution within these soils. Therefore, it is essential to avoid overgeneralization when discussing the properties and behavior of Mollisols. In our study, we focused on a specific type of typical Mollisol, characterized by its properties and management practices. Our experimental site is representative of agricultural Mollisols in the region, with a long history of cultivation and specific management practices that influence soil structure and nutrient cycling. While our findings provide valuable insights into the N dynamics and aggregate stability under these specific conditions, it is important to recognize the potential variability among different Mollisol types and land use practices.

In the context of climate change, where freeze–thaw cycles are expected to become more frequent and intense in many regions, these findings highlight the need for adaptive management strategies. Practices such as cover cropping, reduced tillage, and organic amendments can enhance soil aggregate stability and improve soil resilience to freeze–thaw cycles, thereby maintaining soil fertility and reducing N losses.

5. Conclusions

In terms of bulk soils, with the increase in freeze–thaw frequency, the contents of N components (TN, NO₃⁻-N, NH₄⁺-N, and MBN), as well as their relative contribution rate, allocation amount, and ratio, all showed an increasing trend under the three freeze–thaw intensities. Simultaneously, decreased freezing temperature reduced the contents of TN and NO₃⁻-N regardless of freeze–thaw frequency. Additionally, as freezing temperature decreases, the relative contribution rate, allocation amount, and ratio of TN and NH₄⁺-N all showed a decreasing trend, while those of NO₃⁻-N increased first and then slowly decreased, and those of MBN increased first and then rapidly decreased under the five freeze–thaw frequencies.

For soil aggregates, the freeze–thaw action significantly affects both the proportion of soil aggregates and the N content within them. With the increase in freeze–thaw frequency, the proportion of macroaggregates (2–1 mm, 1–0.5 mm, and 0.5–0.25 mm) and silk + clay fractions (<0.053 mm) were decreased under the higher freezing temperature (−9 °C/5 °C). However, lower freezing temperature (−18 °C/5 °C and −26 °C/5 °C) increased the proportion of microaggregates (0.25–0.053 mm) and decreased that of silk + clay fractions (<0.053 mm). Both higher and lower freezing temperatures reduced the TN content in each aggregate size to varying degrees.

Furthermore, both the number of freeze–thaw cycles and freeze–thaw temperature are important factors affecting the stability of soil aggregates. As freeze–thaw frequency increases, higher freezing temperatures reduced the stability of soil aggregates, while lower freezing temperatures increased the stability in Mollisols. The factors that showed a significant relationship with soil aggregate stability included the contents of NO₃⁻-N and MBN in bulk soils, the proportions of 2–1 mm, 0.5–0.25 mm, and <0.053 mm aggregate sizes, and the TN content of 2–1 mm size, regardless of the freeze–thaw frequency and temperature in Mollisols. Therefore, integrating cover crops, reduced-tillage, and organic amendments reinforces aggregate stability, mitigates frost-induced structural breakdown, and simultaneously curbs N losses, which can provide a readily adoptable, winter-resilient management package for the long-term fertility of Mollisols under freeze–thaw cycles.