Effects of rainfall amount and frequencies on soil net nitrogen mineralization in Gahai wet meadow in the Qinghai-Tibetan Plateau

Global climate change has led to a significant increase in the frequency of extreme rainfall events in the Qinghai-Tibetan Plateau (QTP), thus potentially increasing the annual rainfall amounts and, consequently, affecting the net soil nitrogen (N) mineralization process. However, few studies on the responses of the soil net N mineralization rates to the increases in rainfall amounts and frequencies in alpine wet meadows have been carried out. Therefore, the present study aims to assess the effects of rainfall frequency and amount changes on the N fixation capacity of wet meadow soils by varying the rainfall frequency and amount in the Gahai wet meadow in the northeastern margin of the QTP during the plant-growing season in 2019. The treatment scenarios consisted of ambient rain (CK) and supplementary irrigation at a rate of 25 mm, with different irrigation frequencies, namely weekly (DF1), biweekly (DF2), every three weeks (DF3), and every four weeks (DF4). According to the obtained results, the increased rainfall frequency and amount decreased the soil mineral N stock and increased the aboveground vegetation biomass (AB) amounts and soil water contents in the wet meadows of the QTP. Ammonium (NH4+-N) and nitrate N (NO3–-N) contributed similarly to the mineral N contents. However, the ammonification process played a major role in the soil mineralization process. The effects of increasing rainfall amount and frequency on N mineralization showed seasonal variations. The N mineralization rate showed a single-peaked curve with increasing soil temperature during the rapid vegetation growth phase, reaching the highest value in August. In addition, the N mineralization rates showed significant positive correlations with soil temperatures and NH4+-N contents and a significant negative correlation with AB (P < 0.05). The results of this study demonstrated the key role of low extreme rainfall event frequencies in increasing the net soil N mineralization rates in the vegetation growing season, which is detrimental to soil N accumulation, thereby affecting the effectiveness of soil N contents.


Effects of rainfall amount and frequencies on soil net nitrogen mineralization in Gahai wet meadow in the Qinghai-Tibetan Plateau
Guorong Xu 1,2 , Guang Li 1,2* , Jiangqi Wu 1* , Weiwei Ma 1 , Haiyan Wang 1 , Jianyu Yuan 1 & Xiaodan Li 3 Global climate change has led to a significant increase in the frequency of extreme rainfall events in the Qinghai-Tibetan Plateau (QTP), thus potentially increasing the annual rainfall amounts and, consequently, affecting the net soil nitrogen (N) mineralization process.However, few studies on the responses of the soil net N mineralization rates to the increases in rainfall amounts and frequencies in alpine wet meadows have been carried out.Therefore, the present study aims to assess the effects of rainfall frequency and amount changes on the N fixation capacity of wet meadow soils by varying the rainfall frequency and amount in the Gahai wet meadow in the northeastern margin of the QTP during the plant-growing season in 2019.The treatment scenarios consisted of ambient rain (CK) and supplementary irrigation at a rate of 25 mm, with different irrigation frequencies, namely weekly (DF1), biweekly (DF2), every three weeks (DF3), and every four weeks (DF4).According to the obtained results, the increased rainfall frequency and amount decreased the soil mineral N stock and increased the aboveground vegetation biomass (AB) amounts and soil water contents in the wet meadows of the QTP.Ammonium (NH 4 + -N) and nitrate N (NO 3 --N) contributed similarly to the mineral N contents.However, the ammonification process played a major role in the soil mineralization process.The effects of increasing rainfall amount and frequency on N mineralization showed seasonal variations.The N mineralization rate showed a single-peaked curve with increasing soil temperature during the rapid vegetation growth phase, reaching the highest value in August.In addition, the N mineralization rates showed significant positive correlations with soil temperatures and NH 4 + -N contents and a significant negative correlation with AB (P < 0.05).The results of this study demonstrated the key role of low extreme rainfall event frequencies in increasing the net soil N mineralization rates in the vegetation growing season, which is detrimental to soil N accumulation, thereby affecting the effectiveness of soil N contents.
Nitrogen (N) is one of the main ubiquitous nutrient elements in ecosystems, playing a limiting role in plant growth 1,2 .In terrestrial ecosystems, soil N availability to plants is highly dependent on soil N mineralization, which is the key process controlling the global N cycle 3 .Indeed, soil microorganisms can transform insoluble organic N into soluble organic N through the action of extracellular enzymes, then into inorganic N forms that plants can absorb and utilize 4,5 .Soil N mineralization is the key process transforming organic N into inorganic N, namely ammonium N (NH 4 + -N) and nitrate N (NO 3 --N), through the ammonification and nitrification processes, which are the major internal ecological processes influencing the availability of soil inorganic N to plants and microorganisms 6,7 .Net N mineralization refers to the net accumulation of inorganic N in the soil solution over a specific period 8 .Absorption and chemical fixation are the main processes by which microorganisms fix inorganic N from soils 9 .Soil N availability is limited by inorganic N supply rates and controlled by

Materials and methods
Study area.The present study was conducted in the Gahai Zecha National Nature Reserve in the eastern part of the QTP (33°58′12″-34°32′16″N, 102°05′00″-102°47′39″E), which is the intersection of the Yellow and Yangtze rivers.The QTP is a water conservation and important transit area, as well as a habitat and breeding place for North-South migratory birds.Indeed, the study area has been listed among internationally important wetlands.The wet meadow of the study area covers a total area of 57,846 ha, accounting for 88.4% of the total wetland area.The region is characterized by a cold-temperate continental monsoon climate.The average temperature in the hottest (July) and coldest (January) months are 10.5 and − 8.5 °C, respectively, with an average annual temperature of 2.9 °C.There is no absolute frost-free period in the study area.On the other hand, the average annual rainfall is 785 mm, of which 76% of the total annual rainfall amount occurs mainly in the May-September period.In addition, the highest rainfall amounts occur mainly in August.Affected by the climatic conditions of the plateau, mountain types, and the freeze-thaw effect, the soil types in the region consist mainly of meadow, bog, and peat soils.The field plot considered in this study is characterized by meadow soil, with sandy loam and clayey soil textures in the 0-20 and 20-40 cm soil layers, respectively.The main plant species in the study area consist of Carexmeyeriana, Thalictrum aquilegifolium, Artemisia subulata, Potentilla chinensis, Polygonum viviparum L., and Potentilla anserina L 28,30 .
Experimental design.In this study, a typical wet meadow field was selected in 2019 in the Gahai-Zecha wetland.We established a total of 15 experimental plots, covering a total area of 100 m 2 .The experimental design consisted of five treatments with three replications.Each plot covered 4 m 2 (2 m × 2 m), with 2 m spacing as a buffer zone.The extreme precipitation event in the Tibetan Plateau was about 25 mm•d -1 , of which the frequency did not exceed five times per year 31 .However, the frequency of the extreme precipitation event is expected to increase significantly in the future, thus potentially increasing the precipitation amounts 32 .Extreme rainfall events were mainly concentrated during the plant growing season 27,29 .In this study, we used sprinklers to slowly irrigate each plot with the collected rainwater from May to early October 2019, ensuring even infiltration of the rainwater at each plot without generating surface runoff 29 .In total, four different rainfall frequency events were considered in this study to simulate precipitation inputs to the field experiment area based on an irrigation rate of 25 mm, taking into account ambient rainfall as a control (CK) 28,33 .The irrigation frequency treatments with natural rainwater consisted of one watering per week, 2 weeks, 3 weeks, and 4 weeks, corresponding to total irrigation amounts of 475 mm (DF1) (19 times × 25 mm), 225 mm (DF2) (9 times × 25 mm), 150 mm (DF3) (6 times × 25 mm), and 100 mm (DF4) (4 times × 25 mm), respectively.The first irrigation event occurred on May 27, 2019.

Measurement of soil nitrogen mineralization.
In-situ incubations can reflect the actual states of soil net N mineralization, nitrification, and ammonification 34,35 .In this study, we used the in-situ closed-top PVC tube incubation method to ensure the same structure and temperature characteristics of the experimental soil as the external soil 36 .At the beginning of the growing season, two PVC pipes with an inner diameter of 5 cm were placed vertically in each square of the soil at a 20 cm depth.In addition, a soil core sample was collected from the experimental field to determine the initial soil NH 4 + -N and NO 3 --N contents, while other soil cores were in-situ incubated after sealing the topsoil with a plastic film to prevent rainwater infiltration and to exclude vegetation roots while allowing gas exchange.The remaining PVC pipes and soil samples were collected after 15 days of in situ incubation.New PVC tubes were placed in each soil square following each growing period to ensure continuous in-situ observations.Soil samples were collected from the 0-10 and 10-20 cm soil layers.Each treatment was repeated 3 times.All soil samples (initial and incubated) were hand sorted and sieved through a 2 mm sieve.The incubated soil samples were processed in the same manner as the initial cores.All initial and incubated soil samples were sent to the laboratory and analyzed within 24 h 35 .

Aboveground vegetation biomass. The aboveground vegetation biomass (AB) was determined in this
study based on the cumulative plant biomass in August 2019, corresponding to the peak of the plant growing season.Three 25 cm × 25 cm squares were first randomly selected from each treatment plot, then plants were harvested from the squares to determine the AB.The collected biomass samples were initially oven-dried at 100 °C for 0.5 h, then over-dried at 60 °C for 72 h until reaching constant weights 35 .The dry weights of the biomass samples were measured using analytical balances.

Soil water contents and soil temperatures.
In this study, soil water contents (SWC) and soil temperatures (ST) at the 0-10 cm and 10-20 cm depths were measured in situ using a soil moisture content analyzer (QS-SFY/RS232), Qiang Sheng Manufacturing Center of Analysis Instruments, Beijing, China) and temperature sensor, respectively, during soil sampling 35 .

Determination of soil carbon and N forms. Determination of soil NH 4 + -N and NO 3 --N contents.
Approximately 10 g of the collected soil samples were first added to 50 mL of 2 M KCl solutions, then mixed for 1 h using a rotational shaker.Soil NH 4 + -N and NO 3 --N were separated by distillation using magnesium oxide (MgO) and Devarda's alloy (50% Al, 45% Cu, and 5% Zn) 28,30,33 .Indeed, parts of the soil solution samples were steam-distilled with MgO using a steam distillation system to separate soil NH 4 + -N, while the remaining parts in the flasks were redistilled after adding Devarda's alloy and 1 mL sulfamic acid to separate soil NO 3 --N.The released NH 3 was trapped using a boric acid solution, then converted into (NH 4 ) 2 SO 4 using 0.05 mol•L -1 H 2 SO 4 solution.The NH 4 + -N and NO 3 --N contents were determined based on the volume of sulfuric acid added [36][37][38] .
Determination of soil dissolved organic N (DON) contents.The DON contents were determined according to the method described by Willett 39 .Specifically, 10 g fresh soil samples were placed into 150 mL Erlenmeyer flasks, then 50 mL of 0.5 M K 2 SO 4 solutions was added and shaken at 200 rpm for 1 h.The mixtures were centrifuged at 12,000 rpm for 10 min.5 mL of the solutions were used to determine the total dissolved N contents using the Semi-Micro Kjeldahl method.The DON content was determined by calculating the difference between the total dissolved N contents and the combined NH 4 + -N and NO 3 --N contents 39 .

Determination of soil microbial biomass N (MBN).
The MBN contents in the soil samples were determined using the chloroform fumigation extraction method 40 .Two groups of 10 g fresh soil samples were placed in a vacuum desiccator.The first group was fumigated with ethanol-free chloroform for 24 h, while the second group was considered the control group (without fumigation).The fumigated and unfumigated soil samples were mixed with a 0.5 M K 2 SO 4 solution and shaken at 180 rpm for 30 min using a shaker.5 mL of the supernatants were extracted and analyzed using the semi-micro Kjeldahl method [41][42][43][44] : where Cfumigated and cnonfumigated denote the MBN contents in the fumigated and unfumigated soil samples, respectively.

Determination of soil bulk density (BD), soil total nitrogen (TN) contents, and soil total phosphorus (TP) contents.
In this study, some undisturbed soil samples were collected using a ring knife, then oven-dried at 105℃ for 24 h to determine the soil BD values 45 .On the other hand, 1 g of the air-dried soil samples were nitrified using concentrated H 2 SO 4 at 400℃ until achieving a milky white color.The solutions were placed then into 100 mL volumetric flasks, then 5 mL and 10 mL of the solutions were analyzed for the TN and TP contents using the Semi-Micro Kjeldahl and molybdenum colorimetric methods, respectively 46,47 .
Determination of soil organic carbon (SOC) contents.The SOC contents in the collected samples were analyzed in this study using the improved Walkley and Black method.Specifically, 0.2 g (accurate to 0.001 g) of the airdried soil samples were digested at 180 °C for 30 min using 5 ml of 0.8 M 1/6K 2 CrO 7 and 5 ml of concentrated H 2 SO 4 .Afterward, 2-3 drops of Ortho phenanthroline indicator were added to the solutions, followed by titration with standardized FeSO 4 .The SOC contents were determined based on the consumed K 2 CrO 7 48 .

Data calculation.
According to the difference in the soil NH 4 + -N and NO 3 --N contents before and after culture, the net ammonification, net nitrification, and net N mineralization rates of the soil were calculated in mg•kg -1 using the following formulas 49-51 : where Δt denotes the time interval before and after culture; Δc (NH 4 + -N) denotes the change in the soil NH 4 + -N contents before and after culture; Δc (NO 3 --N) denotes the change in the soil NO 3 --N contents before and after culture; Ra, Rn, and Rm denote the net ammonification, net nitrification, and net N mineralization rates, respectively, of the soil.

Statistical analysis.
In this study, the one-way ANOVA analysis was performed to determine whether the soil inorganic N contents, Ra, Rn, and Rm were statistically different between different sampling dates and treatments (at the P < 0.05 level).All the collected data were assessed for variance homogeneity and normal distribution.The effects of soil depths and rainfall frequency, as well as their interactions, on the Rm values, were assessed using the two-way ANOVA analysis.In addition, a regression analysis was performed to analyze the relationship between soil water contents, soil temperatures, and soil net N mineralization rates.All statistical analyses were performed using SPSS 20.0 software (SPSS for Windows, Chicago, USA), taking into account the significance level of P < 0.05.Whereas graphs were generated using Origin 2017.

Results
Soil properties and aboveground vegetation biomass.The obtained results showed different effects of rainfall frequencies and amounts on soil nutrients (Table 1).Indeed, the soil BD, TN contents, TP contents, and DON contents decreased significantly with increasing rainfall frequency and amount (P < 0.05).However, significantly lower DON contents were observed in DF1 and DF4 than those in CK (P < 0.05).In contrast, no significant differences in the DON contents were observed between DF2 and DF3 (P > 0.05).The SOC contents in DF2 were significantly higher than those in CK, while the SOC contents in DF3 and DF4 were significantly lower than those in CK (P < 0.05).The soil MBN contents in the DF1 and DF3 treatments were significantly higher than those in CK.In addition, the increase in the rainfall amount and frequency significantly increased the AB amounts in DF1, DF2, and DF3 compared to those in CK (P < 0.05) (Table 1).
Soil temperature and soil water contents.The changes in rainfall amount and frequency showed significant effects on the soil water contents and soil temperatures (Fig. 1).The soil water contents increased significantly with increasing precipitation frequency (P < 0.05).In addition, the soil water contents in DF1 and DF2 showed a decreasing trend with seasonal changes, while those in CK, DF3, and DF4 exhibited increasing-decreasing trends with seasonal changes.On the other hand, the soil temperature values of the five treatments showed consistent seasonal changes.However, significant differences in the soil temperature values were observed between the treatments (P < 0.05).The overall change in the soil temperature values exhibited an "N" --N contents were significantly higher in DF4 and lower in DF2 than those in CK.Furthermore, the interaction between the rainfall frequency treatments and soil depths exhibited a significant effect on the soil inorganic N contents (P < 0.01) (Table 2).The increase in the rainfall amount and frequency reduced significantly the mineral N stock (Fig. 3).In addition, it can be seen that the increase in the rainfall amount and frequency reduced the soil NH 4 + -N and NO 3 --N contents, particularly in the 0-10 cm soil layer.

Soil N mineralization.
The average Ra, Rn, and Rm values obtained across the treatment scenarios were -0.099, 0.117, and 0.017 mg•kg −1 •d −1 , respectively.In addition, Ra and Rn were significantly and positively correlated with Rm (P < 0.01).However, the slope of the regression line between Rn and Rm (0.298) was lower than that between Ra and Rm (0.702) (Fig. 6).Therefore, Rm was dominated mainly by the ammonification process in the study site.Indeed, these patterns were consistent with the rainfall frequencies and soil depths and inconsistent with the different stages of the growing season (Table 2; Fig. 4).
The Ra, Rn, and Rm values of the five treatments exhibited obvious seasonal changes.However, although Ra, Rn, and Rm showed consistent seasonal trends, they were significantly different between the treatments (Fig. 4).For example, Rm showed an increasing-decreasing trend with the seasonal change, reaching the highest value in August 2019.Except for the CK treatment, the Rm values were negative in the June-July and September-October periods, indicating that soil N was in a net storage state during the non-growing season.In the DF3 experiment scenario, Ra and Rm in the 0-20 cm soil layer were significantly higher than those in the other experiment scenarios (P < 0.05).The interactions of the rainfall amounts and frequencies with the soil depths showed significant   effects on Ra (P < 0.01), Rm (P < 0.01), and Rn (P < 0.05) (Table 3).Therefore, the obtained results demonstrated the importance of appropriate rainfall increases in improving Rm.

Relationships between soil mineralization rates and environmental factors.
In this study, the principal component analysis (PCA) was performed to reveal the relationships between the soil physicochemical characteristics and Ra, Rn, and Rm.According to the obtained results, the Rm values were significantly and positively correlated with BD, NH 4 + -N, NO 3 --N, and soil temperature (P < 0.01) (Figs. 5 and 6).However, Ra, Rn, and Rm exhibited significant negative correlation coefficients with AB of -0.559, -0.826, and -0.662 (P < 0.01), respectively.Furthermore, Ra and Rm showed significant negative correlation coefficients with SWC of − 0.358 and − 0.235, respectively (P < 0.01) (Fig. 6).-N contents decreased significantly with increasing rainfall frequency (Fig. 3), which is consistent with the results of rainfall experiments conducted in semi-arid steppes of Patagonia 52 .This finding might be due to the positive roles of rainfall frequency and amount increases in increasing soil moisture contents (Fig. 1) and community species richness 53 , resulting in significantly high AB amounts (Table 1) and, consequently, enhancing the uptake of soil NH 4 + -N.Ammonium N (NH 4 + -N) is the main N form absorbed by plants due to its high solubility in the soil, increasing the chlorophyll-a content of plants and playing a key role in the leaf functional traits, thus increasing the AB amounts 54 .In addition, it was shown that rainfall variability significantly affects plant communities and their growth characteristics, as well as species compositions and diversity 55 .Indeed, precipitation frequency and amount increases can increase plant species richness within a plant community and result in ecological-niche divergence 55 .The positive correlation between species diversity and productivity can be explained by resource complementarity and sampling effects.Resource complementation effects suggest that species ecological niche differentiation and interspecific facilitation among plant communities with high diversity can improve resource utilization efficiency and, consequently, enhance community productivity 52 .Whereas the sampling effect indicates that the increase in the species diversity of a plant community results in a higher likelihood of including more productive species, thereby leading to higher community productivity and greater uptake of soil inorganic N and resulting in lower soil inorganic N contents 52 .On the other hand, the significant decreases in the soil NO 3 --N contents with increasing rainfall frequency might be due to the occurrence of the leaching process, which can be promoted with increasing rainfall frequency.Indeed, NO 3 --N cannot be adsorbed on the negatively charged soil colloids and mineral surfaces, resulting in a significant decrease in the soil NO 3 --N contents 56,57 .In addition, the soil BD and soil water contents decreased and increased, respectively, with increasing rainfall frequency and amount (Table 1).The changes in the soil BD and soil water contents might inhibit soil microbial activity and, consequently, reduce the soil inorganic N contents 58 .In addition, the results of the present study highlighted the significant effects of the interaction between soil depths and rainfall frequencies on the soil NH 4 + -N and NO 3 --N contents (Table 2).This finding might be due to the abundance of plant roots and microbial activity in the 0-20 cm soil layer in the wet meadow 59 , supplying more organic N into the soil through root decomposition.
Effects of soil moisture regulation on soil N dynamics.Rainfall can significantly affect the N cycle in terrestrial ecosystems and physically affect the microbial-mediated N mineralization process 11 .On the one hand, rainfall changes can directly affect soil water infiltration and N volatilization 60 .On the other hand, rainfall can affect N transformations, root uptakes, and microbial utilization, thereby directly or indirectly affecting the soil N cycle process 61 .In this study, Ra, Rn, and Rm were negative in all the treatment scenarios from June to July and from September to October, indicating N immobilization by soil microorganisms.In addition, NH 4 + -N might be stored by N-limited microorganisms or converted into NO 3 --N by nitrifying bacteria.This also implies that microbial growth in the wet meadow soils may not be carbon-limited.The soluble organic N stock in the soil is not conducive to microbial growth and respiration 62,63 .In contrast, Ra, Rn, and Rm were positive in August, which might be due to the gradual increase in the soil temperature, reaching the highest value in August (Fig. 1).In this period, soil microorganisms can exhibit high activity and strong competition for carbon.As a result, the availability of carbon may become a limiting factor for soil microbial growth.Indeed, soil microorganisms can utilize soluble organic N to meet their growth requirements, resulting in a net release of N.This suggestion is consistent with the high and significant positive correlation between the soil temperature and Rm values observed in this study (Fig. 6).In early July, the Ra, Rn, and Rm values of CK were significantly higher than those observed in the other treatment scenarios.This result was probably due to the first applied irrigation rate in late May when high soil water contents and low soil temperatures were observed following rainfall events, thereby resulting in a faster increase in soil temperature and higher Rm in CK compared to the other treatments.This study also found significantly higher Ra values in the DF3 treatment scenario than those in the    remaining treatments, which might be due to the increased soil soluble nutrient contents and accelerated soil microbial activity with increasing rainfall and frequency 63,64 .Indeed, the increase in soil microbial activity can promote the decomposition of organic matter, thus increasing the Ra values.However, high soil water contents exceeding the threshold can enhance the anaerobic conditions in the soil, thereby preventing soil microbial oxidation, which is detrimental to litter decomposition and, consequently, reducing Ra 65 .Ammonium (NH 4 + -N) availability and oxygen concentrations are the main factors affecting the nitrification process 66 .The results of the present study revealed a decrease in Rn with increasing rainfall frequency, which may be due to two reasons: (1) the decrease in the soil NH 4 + -N contents with increasing rainfall was not conducive to NH 4 + -N migration to nitrifying bacteria, thus limiting the availability of soil NH 4 + -N; (2) High soil water contents can lead to low dissolved oxygen concentrations in the soil solution 67,68 , thereby inhibiting the nitrifying bacterial activity and leading to a decrease in Rn.In this study, low Rm values were observed during the growing season, suggesting limited soil organic mineralization and potentially representing a fraction of the effective natural flux of organic matter.Indeed, since the plant root system and mineralization action of mycorrhizal fungi need to be considered when using the buried PVC pipe method, Rm can be underestimated in the presence of plant root systems in soils.In addition, the increase in the soil water content within a certain range can promote N mineralization.However, the appropriate soil water content range for soil N mineralization varies among ecosystems, depending on several factors, including soil type, soil texture, and microbial community 69,70 .In this study, the Rm values of the DF3 treatment scenario were significantly higher than those observed in the remaining treatment scenarios.This result might be due to the decreased soil permeability and microbial metabolic activity with increasing rainfall amount and frequency, resulting in reduced Rm values 71 .Moreover, the low Rm values might be due to slightly acidic rainwater, which can decrease the soil pH values and, consequently, increase the fungi-to-bacteria ratio, as fungi are characterized by a higher tolerance to high H + concentrations than bacteria due to their thick and interconnected peptidoglycan cell walls 72 .Soil Rm can only be enhanced under relatively suitable soil moisture conditions.

Conclusions
In this study, we investigated the effects of rainfall amount and frequency on soil mineral N and net N mineralization during the growing season in the Gahai wet meadow.Our findings demonstrated the significant effect of rainfall on driving the soil N turnover process in the alpine wet meadow.According to the obtained results, the increased rainfall reduced soil mineral N stocks.However, NH 4 + -N was the dominant N mineral form in the soil.The highest Rm value was observed in the DF3 treatment scenario during the vegetation growing season.In addition, the Rm values were significantly and positively correlated with soil temperatures and negatively correlated with soil water contents and AB amounts.The results of the present study provide further insights into the N cycle responses to changes in rainfall amount and frequency, as well as into the underlying mechanisms of the rainfall amount and frequency effects on plant and ecosystem functions.However, soil N turnover processes, such as plant root N uptakes and soil gaseous N emissions, were not considered in this paper.We suggest, therefore, further related studies in the future, taking into account plant root N uptake and gaseous N emissions.

4 +-N and NO 3 --+
https://doi.org/10.1038/s41598-023-39267-3www.nature.com/scientificreports/shape over the entire vegetation growing season, showing the highest and lowest values in August and September, respectively.Soil mineral-N dynamics.The results indicated different effects of rainfall frequencies and amounts on the soil NH N contents in the Gahai wet meadows (Fig. 2).According to average values across the treatment scenarios, the NH 4 + -N and NO 3 --N contents accounted for 59.4 and 40.6% of the total soil mineral N pool.These proportions were consistent with those observed in the soil layers.The NH 4 accounted for 59.1 and 40.9% of the total soil mineral N pool in the 0-10 cm soil layer, respectively.While in the 10-20 cm soil layer, the NH 4 + -N and NO 3 --N contents accounted for 59.6 and 40.4% of the total soil mineral N pool, respectively.The increase in the rainfall amount and frequency decreased the soil NH 4 + -N contents.The NO 3

Table 1 .
The main physicochemical indicators of the soil under different treatments.The data were presented as Mean ± Standard Errors.Different letters indicate significant difference among treatments (P < 0.05).DF1 weekly, DF2 biweekly, DF3 every 3 weeks, DF4 every 4 weeks and CK ambient rain, TP Total phosphorous, TN Total nitrogen, SOC Soil organic carbon, DON Dissolved organic nitrogen, MBN Microbial biomass nitrogen, AB Aboveground biomass, BD Soil bulk density.The same below.
www.nature.com/scientificreports/Discussion Soil mineral-N pool.Soil inorganic N forms (NH 4 + -N and NO 3 --N) are derived mainly from the decomposition and transformation of soluble organic N by microorganisms.These N forms can be directly absorbed and utilized by plants.In this study, the NH 4 + -N and NO 3 -

Figure 3 .Figure 4 .
Figure 3. Rainfall amount and frequency on NH + 4 -N, NO - 3 -N and mineral N stock.Bars and error bars represent means and SD (n = 3).Different letters indicate significant differences in the mean value between different treatments.

Figure 5 .
Figure 5. Principal component analysis of soil mineralization rate and physical and chemical properties of soil, and correlation between soil mineralization rate and soil physical and chemical properties.*Indicates significant at 0.05 level, **indicates significant at 0.01 level.

Figure 6 .
Figure 6.Relationship between soil mineralization rate and environmental factors.

Table 2 .
Effects of rainfall amount and frequency and soil depth on soil inorganic nitrogen.

Table 3 .
Effects of rainfall amount and frequency and soil depth on soil mineralization rate.