lmprovement of soil fertility and rice yield after long-term application of cow manure combined with inorganic fertilizers

GAO Peng,ZHANG Tuo,LEl Xing-yu,CUl Xin-wei,LU Yao-xiong,FAN Peng-feiLONG Shi-ping,HUANG Jing,GAO Ju-sheng,ZHANG Zhen-hua#,ZHANG Hui-min#

1 College of Resources,Hunan Agricultural University,Changsha 410128,P.R.China

2 Institute of Agro-environment and Ecology,Hunan Academy of Agricultural Sciences,Changsha 410125,P.R.China

3 College of Environmental Science and Engineering,China West Normal University,Nanchong 637002,P.R.China

4 Institute of Nuclear Agriculture Science and Space Breeding,Hunan Academy of Agricultural Sciences,Changsha 410125,P.R.China

5 Qiyang Farmland Ecosystem National Observation and Research Station,Institute of Agricultural Resources and Regional Planning,Chinese Academy of Agricultural Sciences,Qiyang 426182,P.R.China

Abstract Fertilization is an effective technique to improve soil fertility and increase crop yield. The long-term effects of different fertilizers on soil considerably vary. Over 38 consecutive years of different fertilization positioning experiments in a double cropping rice field of Qiyang Red Soil Experimental Station,seven different fertilization treatments including CK(no fertilization),NPK (nitrogen,phosphorus,and potassium fertilizer),M (cow manure),NPKM (nitrogen,phosphorus,and potassium with cow manure),NPM (nitrogen and phosphorus with cow manure),NKM (nitrogen and potassium with cow manure),and PKM (phosphorus and potassium with cow manure) were applied to study the effects on rice yield,soil fertility,and nutrient apparent balance in a paddy field. The results showed that the annual average yields of rice in NPKM,NPM,NKM,PKM,M,NPK and CK treatments ranged from 6 214 to 11 562 kg ha-1. Yields under longterm organic and inorganic treatments (NPKM,NPM,NKM and PKM) were 22.58,15.35,10.53 and 4.41%,respectively,greater than under the NPK treatment. Soil organic carbon (SOC),total nitrogen (TN),available nitrogen (AN) and available potassium (AK) concentration with long-term organic and inorganic treatment (NPKM,NPM,NKM and PKM) were significantly higher than in inorganic fertilizer (NPK) treatments. Soil total phosphorus (TP) and available phosphorus (AP) contentration with organic fertilizer combined with inorganic N and P fertilizer treatment (NPKM,NPM and PKM) were significantly higher than with inorganic fertilizer alone (NPK treatments). The average annual rice yield (11 562 kg ha-1),SOC (20.88 g kg-1),TN (2.30 g kg-1),TP (0.95 g kg-1),TK (22.50 g kg-1) and AP (38.94 mg kg-1) concentrations were the highest in the NPKM treatment. The soil AN concentration (152.40 mg kg-1) and AK contentration (151.00 mg kg-1) were the highest in the NKM treatment. N and P application led to a surplus of nitrogen and phosphorus in the soil,but NPKM treatment effectively reduced the surplus compared with other treatments.Soils under all treatments were deficient in potassium. Correlation analysis showed that SOC,TN,AN,TP,and AP contentration was significantly correlated with rice yield;the correlation coefficients were 0.428,0.496,0.518,0.501,and 0.438,respectively. This study showed that the combined application of N,P,and K with cow manure had important effects on rice yield and soil fertility,but balanced application of N,P,and K with cow manure was required.

Keywords: long-term fertilization,rice yield,soil fertility,nutrient balance

1.lntroduction

Rice is an important global food crop. Approximately 3 billion people depend on rice for survival and a high and stable rice yield is an important factor in human food security worldwide. However,global rice production has declined in recent years because of the challenges of global climate change,water shortages,decline in soil fertility,reduction in arable land,and decline in farmers’preference for rice (Denget al.2019).

Soil provides most of the nutrients,including nitrogen,phosphorus,potassium,and all trace elements required for the sustained growth of crops. Soil fertility represented by plant growth and its ability to integrate nutrient and environmental conditions (Rawalet al.2017) is instrumental in achieving a high,stable,and sustainable crop yield (Jahanet al.2021). Fertilization is an effective way of improving soil fertility and increasing crop production. For a long time,the continuous input of fertilizer nutrients has led to a steady increase in crop yields and ensured a continuous and stable food supply.However,incorrect fertilization can lead to a decline in soil quality and crop yield and quality,as well as a series of environmental problems,such as greenhouse gas emissions and agricultural non-point source pollution(Wei and Ma 2015). Therefore,scientific and rational fertilization is essential to maintain soil fertility and stable crop yields. For instance,a recent study has shown that rational fertilization of rice can improve soil fertility and increase farmland production and rice yields (Takamitsuet al.2020). In addition,it can reduce the volume of fertilizer required,improve fertilizer utilization,and reduce environmental risks.

On the one hand,organic fertilizers effectively increase crop yields and improve soil quality,but there are disadvantages,such as their low efficiency,low nutrient content,high application rate,high labour intensity,and no significant effect on increasing yield (Guoet al.2016).On the other hand,inorganic fertilizers rapidly increase yield,have a high nutrient content,and are required in low volumes. However,excessive use of inorganic fertilizers can degrade soil quality and cause serious nonpoint source pollution (Xiaoet al.2021). In contrast to the disadvantages of organic and inorganic fertilizer when applied alone,their combination can not only improve soil fertility but also promote high and stable crop yields (Ruiet al.2015;Tianet al.2022). However,the strategy for balanced and rational fertilization that can maintain and improve soil fertility and promote increased and stable rice yields are issues worth continuous attention.

Numerous studies have focussed on the effects of different long-term fertilization measures,especially the combined application of long-term organic and inorganic fertilizer,on crop yield and soil fertility. Studies show that long-term application of organic fertilizer or organic combined with inorganic fertilizer can better maintain and improve crop yield and soil fertility (Gaiet al.2018;Wanget al.2019;Liu J Aet al.2021). At present,organic fertilizers used in a long-term positioning experiment of rice in a red soil hilly area of South China mainly include pig manure,straw,and green manure (Zhaoet al.2019;Zhouet al.2019;Qaswaret al.2020),while cow manure is less used as an organic fertilizer source. Rice yields in the study are medium to low. Currently,few studies have attempted to explore the nearly 40-year long-term fertility effects of organic fertilizer in medium to low yielding fields on red soil. Based on the existing long-term rice fertilization experiments,the purposes of the present study were:(1) to analyse the effects of long-term application of cow manure,inorganic fertilizer,and combined cow manure and inorganic fertilizer on rice yield and soil nutrients;and (2) to examine the soil nutrient status after long-term fertilization with different fertilizers. It is considered that this work will be of significance for guiding a rational fertilization strategy for farmland and increasing the rice yields in middle to low yield fields in hilly red soil areas.

2.Materials and methods

2.1.Site description and soil physicochemical properties

The long-term experiment was performed from 1982 to 2020 in paddy fields at the Red Soil Experiment Station,Chinese Academy of Agricultural Sciences. The test site is in Guanshanping Village,Wenfu Town,Qiyang County,Hunan Province (26°45′42′′N,111°52′32′′E) (Gaoet al.2021). The soil type is Ferralic cambisol according to World Reference Base for soil resources (WRB),which is also known as red soil according to the Chinese soil classification (Baxter 2007). It is a typical mid-subtropical region with a hilly terrain and a mean temperature of 24.8°C during rice planting and a mean annual rainfall of 1 250 mm. The mean annual temperature (MAT) and mean annual precipitation (MAP) during the experiment period are given in Fig.1. The basic properties of the soil at the start of the experiment were as follows: pH=5.2,soil organic carbon (SOC)=12.0 g kg-1,total nitrogen(TN)=1.1 g kg-1,available nitrogen (AN)=82.8 mg kg-1,total phosphorus (TP)=0.48 g kg-1,available phosphorus(AP)=9.6 mg kg-1,total potassium (TK) 14.2 g kg-1,readily available potassium (AK)=55.0 mg kg-1.

2.2.Experimental design

At the start of the experiment in 1982,six treatments with different combinations of inorganic fertilizers and decomposed cow manure (organic fertilizer) were set up: NPK (nitrogen,phosphorus,and potassium);M (cow manure);NPKM (nitrogen,phosphorus,potassium,and cow manure);NPM (nitrogen,phosphorus,and cow manure);NKM (nitrogen,potassium,and cow manure);and PKM (phosphorus,potassium,and cow manure).In 2000,a control without any fertilizer or cow manure(CK) was started. Each treatment was repeated in three plots,each with an area of 27 m2(1.8 m length×15 m width),arranged randomly and separated by a cement ridge. Decomposed cow manure was used as the organic fertilizer with an average (over the years)N=0.32%,P2O5=0.25%,and K2O=0.15%. The inorganic fertilizers used were urea (N=46%),superphosphate(P2O5=12%),and potassium chloride (K2O=60%),as N,P,and K fertilizers,respectively. Cow manure was taken from the same farm and the contents of nutrient were measured every 3-5 years. All fertilizers were applied as a base fertilizer at one time. The fertilization ratio of early rice and late rice was the same (1:1),and the annual contents of fertilizer used for each treatment are shown in Table 1.

Table 1 Annual fertilizer application rates of different treatments applied in a long-term fertilizer experiment with double cropped rice (kg ha-1)

Fig. 1 Mean annual temperature (MAT) and mean annual precipitation (MAP) during a long-term rice fertilization experiment period.

Fig. 2 Fertilizer effect on yield of double cropped rice. CK,no fertilization;NPK,nitrogen,phosphorus,and potassium fertilizer;M,cow manure;NPKM,nitrogen,phosphorus,and potassium with cow manure;NPM,nitrogen and phosphorus with cow manure;NKM,nitrogen and potassium with cow manure;PKM,phosphorus and potassium with cow manure. The solid lines represent the medianvalues and the dash lines represent the average values. Different letters above the bars indicate significant differences between fertilizer treatments at P＜0.05.

Fig. 3 Soil nutrient concentrations and pH under different long-term fertilization treatments. CK,no fertilization;NPK,nitrogen,phosphorus,and potassium fertilizer;M,cow manure;NPKM,nitrogen,phosphorus,and potassium with cow manure;NPM,nitrogen and phosphorus with cow manure;NKM,nitrogen,potassium with cow manure;PKM,phosphorus and potassium with cow manure. The solid lines represent the median values and the dashed lines represent the average values. Different letters above the boxplots indicate significant difference between fertilizer treatments at P＜0.05.

Fig. 4 Pearson correlation coefficients for relationships between rice yield and soil fertility index. SOC,soil organic carbon;TN,total nitrogen;AN,available nitrogen;TP,total phosphorus;AP,available phosphorus;AK,available potassium.

2.3.Crop management and soil sample collection

The local and commonly used rice varieties,which were replaced every three to five years,were used in this study.In the past five years,Yuliangyou 4156 was used for early rice and Lingyouhuazhan was used for late rice. The planting system followed planting early rice (April-July)followed by late rice (July-October) and leaving the fields idle in winter (October-April). After harvesting,rice and straw were removed from the plots,and stubbles were left in the fields. The grain and straw collected from each plot were air-dried,and the grain quantity was measured to determine the yield.

After harvesting the late rice each year,five samples(0-20 cm soil layer) from five different points of each replicated plot were collected using an S-shaped sampling pattern within the first week. The samples from each plot were thoroughly mixed to make a homogeneous composite sample of each replicate,naturally air-dried,sieved through a 20-mm mesh,and stored in brown glass bottles till further use to determine the physicochemical properties of soil following the methods described in previous studies (Yanget al.2012;Zhanget al.2017).

2.4.Sample analysis

SOC was measured using the oxidation method (Pageset al.1982),and the TN,TP,and TK of the soil were measured using the methods of Black (1965),Murphy and Riley (1964),and Pageset al.(1982),respectively.Nitrogen release,AP,and AK were measured using the alkaline hydrolysis diffusion method (Lu 2000) Olsen method (Olsen 1954),and ammonium acetate leaching and flame photometry (Pageset al.1982),respectively.

After air-drying,the grain and straws were oven-dried at 105°C for 30 min,and then heated at a temperature of 70°C to determine dry matter weight and nutrient content.Subsequently,the oven-dried straw and grain were ground and digested in H2SO4-H2O2at 260-270°C. The N,P and K contents were measured following the semimicro Kjeldahl method and vanadomolybdophosphoric acid method (Jackson 1969;Pageset al.1982).

2.5.Calculation of apparent balance

Soil apparent balance refers to the difference between the nutrients applied to the soil in the form of fertilizer and the nutrients absorbed by the rice in a certain period (OECD 2013). Apparent balance (AB) was determined using the following formula: AB=IF-Ut,where IF is the total amount of fertilizer applied to the soil (kg ha-1) and Ut is the total amount of nutrients absorbed by the shoot (kg ha-1).The amount of nutrients entering the soil through other channels and lost through evaporation and leaching was not considered (Ouyanget al.2017).

2.6.Data analysis

The differences among fertilization treatments for different parameters were analysed by one-way ANOVA followed by Tukey’s HSD test withP=0.05 considered as a statistically significant difference using SPSS 19.0.Pearson correlation analysis was performed for the relationships between grain yield,SOC,TN,AN,TP,AP,AK,and pH using the SPSS 19.0,andP＜0.01 was considered statistically significant. All the figures and tables were drawn using Excel 2010 and Origin 2021.

3.Results

3.1.Effects of different long-term fertilization on rice yield

The annual rice yields under various long-term fertilization treatments differed significantly (Fig.2). The rice yield in the no-fertilizer control group was significantly lower than that in the fertilization treatments. The highest rice yield was achieved with NPKM treatment. The increases in yield with NPKM,NPM,NKM,PKM,and M treatments compared with NPK were 22.58,15.35,10.53,4.41,and 3.74%,respectively. These results showed that the rice yields after NPKM,NPM,and NKM treatments were in order NPKM＞NPM＞NKM and higher than those after PKM,M,and NPK treatments (P＜0.05). However,the yields with PKM,M,and NPK treatments did not differ significantly,indicating that nitrogen is an important factor affecting rice yield,and the application of inorganic phosphorus and potassium fertilizers together with the organic fertilizer does not effectively increase rice yield.Furthermore,under the same nutrient input,there was no significant difference in the rice yield with CK and NPK treatments. The effect of combining inorganic nitrogen and phosphate fertilizer with organic fertilizer was better than that of the combination of inorganic nitrogen and potassium fertilizer with organic fertilizer.

3.2.Effects of long-term fertilization on soil fertility

After 38 consecutive years of rice planting and different fertilization treatments,the pH and total soil potassium between the treatments did not significantly differ. The other fertility factors showed significant differences between treatments (Fig.3). As shown in Fig.3-B,the SOC content after long-term NPK treatment was 15.59 g kg-1,which indicated an average increase of 10.14%compared with those after CK and NPK treatment(P＜0.05). Compared with NPK treatment,long-term organic and combination of organic and inorganic fertilizer treatments (M,NPKM,NPM,NKM,and PKM) significantly increased SOC content by 21.91,35.26,37.32,33.88 and 33.63%,respectively. The NPKM treatment,which combines organic and inorganic fertilizers,caused the greatest increase in SOC.

As shown in Fig.3-C,compared with NPK treatment,all the long-term organic fertilizer treatments significantly increased TN content by 24.88,28.74,26.4,25.38,and 18.65%,respectively. In particular,NPKM treatment,which combines organic and inorganic fertilizers,produced the greatest increase in TN.

As shown in Fig.3-D,the average AN content in each fertilization treatment from low to high was CK＜NPK＜NPM＜M＜NPKM＜PKM＜NKM,with values of 111.6,120.5,145.2,149.3,144.9,152.4,and 150.4 mg kg-1,respectively. Compared with CK,long-term NPK treatment increased AN content by 7.96% on average,whereas the other fertilizer treatments significantly increased soil AN content by 20.49,23.90,20.25,26.47,and 24.81%,respectively,when compared with traditional NPK treatment. NKM treatment caused the greatest increase in AN content.

SOC,TN,and AN contents after CK and NPK treatments were significantly lower than those after treatments that included organic fertilizer (M,NPKM,NPM,NKM,and PKM),with no significant difference between the latter treatments. Notably,although the nitrogen application rate of the organic fertilizer-only treatment (M) was only 72.5 kg ha-1,it was significantly higher than that of the NPK treatment with the same nitrogen application rate,and the effect on soil TN and AN was almost the same as that of the high-nitrogen fertilizer.

Following the long-term applications of different fertilizers,the average TP content under each treatment differed significantly (Fig.3-E). The treatments can be divided into three distinct groups. The CK,M,and NKM treatments,which do not contain inorganic phosphorus fertilizer,had lower soil TP content,with averages of 0.61,0.63,and 0.61 g kg-1,respectively. The average soil TP content in the NPK treatment containing inorganic phosphorus fertilizer but no organic fertilizer was 0.80 g kg-1,which was significantly higher than that in the CK,M,and NKM treatments. In the treatments that combined inorganic phosphorus fertilizer with organic fertilizer(NPKM,NPM,and PKM),soil TP content was higher,reaching 0.95,0.89,and 0.90 g kg-1,respectively,which was 55.35,46.48,and 48.23% higher than that after CK treatment,respectively. Under different long-term fertilizer treatments,the differences between soil AP and TP were similar (Fig.3-F). The AP content after NPK treatment was significantly higher than those after CK,M,and NKM treatments. Furthermore,the AP content after NPKM,NPM,and PKM treatments was significantly higher than that after the NPK treatment,and NPKM treatment led to the highest increase of 34.60% compared with that after CK.

The different long-term fertilizer treatments did not affect soil TK content (Fig.3-G),but they had a substantial impact on soil AK (Fig.3-H). NPK and M treatments had no obvious effect on improving soil AK content,but NPKM,NKM,and PKM treatments significantly increased soil AK content compared with CK,NPK,and M treatments.

3.3.Soil nutrient apparent balance under different fertilizer treatments

As shown in Table 2,the average annual nitrogen deficit in the soil with CK treatment was 99.9 kg. However,when the nitrogen application rate was 145 kg ha-1,the average annual nitrogen deficit after the PKM,NPK,and M treatments was 6.0-15.3 kg ha-1. When the nitrogen application rate was increased to 290 kg ha-1,a nitrogen surplus occurred in soils with the NKM,NPM,and NPKM treatments. After NPKM treatment,the nitrogen surplus in the soil was 55.2% lower than that after NKM treatment and 19.0% lower than that after NPM treatment. The average annual phosphorus deficit in the soils treated with CK was 38.2 kg ha-1yr-1. When phosphorus application was 56.3 kg ha-1,it still did not meet the demand of rice for phosphorus. The average phosphorus deficit in the soils treated with NKM,M,and NPK was 0.9-6.8 kg ha-1.After doubling the application rate of phosphorus fertilizer,the average annual phosphorus surplus in the soil treated with PKM,NPM,and NPKM was 39.6-51.0 kg ha-1.The phosphorus surplus in the soil treated with NPKM decreased by 15.1% compared with PKM and 13.6%compared with NPM. Unlike nitrogen and phosphorus,regardless of whether potassium was applied,there was a significant potassium deficiency under all treatments.

Table 2 Apparent balance of soil N,P2O5 and K2O relative to fertilization treatment (1982-2020,kg ha-1 yr-1)

3.4.The relationship between soil fertility and rice yield under the different fertilizer treatments

To better understand the relationship between the physical and chemical properties of soil and rice yields,we conducted correlation analyses of rice yield and fertility indicators over the years. To exclude the influence of climate and other factors on yield,a correlation analysis between variation of nutrient indicators and yield across the 38 years was conducted. Correlation analyses between each nutrient indicator and yield for each year revealed significant positive correlations between rice yield and SOC,TN,AN,TP,and AP,with correlation coefficients of 0.428,0.496,0.518,0.501,and 0.438,respectively (Fig.4). SOC had significant positive correlations with TN,AN,and AP,with correlation coefficients of 0.956,0.929,and 0.434,respectively. TN had significant positive correlations with AN and AP,with correlation coefficients of 0.943 and 0.472,respectively.

4.Discussion

4.1.Effects of long-term fertilization on rice yields and soil nutrients

The results of the 38-year experiment demonstrated that compared with traditional NPK fertilization,long-term fertilization treatments by a combination of organic and inorganic fertilizers significantly increased rice yields. In contrast,yields under the organic fertilizer and inorganic fertilizer-only treatments were relatively low,with no significant difference. NPKM treatment that combined organic and inorganic fertilizers increased the amount of nutrient input,improved the balance and availability of nutrients,and thereby,improved the yield. Moreover,the inorganic fertilizers quickly release nutrients in a short time to provide the necessary nutrients for the critical period of rice growth,but the slow release of organic fertilizers ensures a continuous supply of nutrients during the period of most efficient rice growth (Zhang Jet al.2018;Wang Qet al.2021). These explain the complementary effects of NPK and M in increasing rice yield. The study showed that the average annual rice yield of each treatment from high to low was as follows:NPKM＞NPM＞NKM＞PKM＞M＞NPK＞CK. These resultsindicated that applying nitrogen fertilizer together with organic fertilizer led to a notably higher rice yield than the treatments without nitrogen fertilizer. Under equal nitrogen applications,the effect of increasing phosphorus and potassium fertilizers at the same time was better than that of using inorganic fertilizer only or organic fertilizer only treatment. The application of phosphate fertilizer had a greater impact on increasing rice yields than the application of potassium fertilizer.

Many studies have shown that SOC,nitrogen,and phosphorus in soil are key to determining crop yield(Zhang Met al.2018;Maazet al.2021). The results of this experiment show that,compared with the inorganic fertilizer-only treatment,the long-term application of organic fertilizer has a more notable effect on the improvement of SOC. This could be because organic fertilizer increases the supply of soil organic matter and directly increases the amount of organic carbon in the soil. Many studies have shown that combining organic and inorganic fertilizers promotes rice growth and soil nutrient balance (Xinet al.2017;Zhaoet al.2021;Li Qet al.2022;Wuet al.2022). A high degree of coupling between SOC and TN in soil has also been reported;an appropriate application rate of nitrogen fertilizer can increase SOC (Hossainet al.2021).

Long-term fertilization can increase the content and accumulation of nitrogen,phosphorus,and potassium in the soil (Herenciaet al.2007;Xuet al.2009;Eltelibet al.2020). The results of this study show that the long-term application of organic fertilizer caused a large increase in soil TN content,and the combination of organic and inorganic fertilizers had the greatest effect on increasing TN. Because of the high temperatures and rain in paddy fields in southern China,nitrogen is rapidly converted into NH4+and NO3-in the soil and is easily lost through ammonia volatilisation and nitrate leaching. In contrast,the mineralisation of organic nitrogen is slow,with a lower rate of loss and higher retention in the soil,which improves the fertilizer efficiency. Therefore,organic fertilizers are more conducive to soil nitrogen accumulation than inorganic fertilizers and increase soil nitrogen storage capacity (Liu L Yet al.2021;Tuet al.2021). In this study,the change trends of soil AN and TN content under each treatment were the same;however,the effects of different treatments varied. The effect of applying organic fertilizer on increasing AN in soil was significantly better than that of the inorganic fertilizer-only treatment,whereas the increase was the largest after applying a combination of organic and inorganic fertilizers.Furthermore,our results showed that,compared with the organic fertilizer,the application of inorganic phosphorus fertilizer had a superior effect on improving TP and AP in the soil with a higher accumulation of phosphorus.This could be because inorganic fertilizers are rich in inorganic phosphorus,which is easily fixed in the soil,so it accumulates in large quantities following longterm application. In contrast,the phosphorus in organic fertilizer is mainly in the organic form,which is released as AP with the decomposition of organic matter in the soil and is absorbed and utilized by crops. This process is relatively slow,and some studies have found that the main form of phosphorus lost in the soil is the organic fraction. Although the effect of the organic fertilizeronly treatment on soil TP and AP content was limited,experiments have shown that the utilization rate of phosphorus in organic fertilizer is significantly higher than that of phosphorus in inorganic fertilizer. This could be because organic phosphorus is more widely distributed in soil and has a larger contact area with crop roots,making it easier for crops to absorb and utilize it (Dibyenduet al.2021). Furthermore,the application of organic fertilizers to soil increases the organic matter content,which in turn reduces the fixation of inorganic phosphorus in the soil and promotes the dissolution of inorganic phosphorus(Xinet al.2017). As a result,applying an organic fertilizer together with inorganic phosphorus fertilizer has a more significant effect on improving soil AP,which is consistent with the results of this experiment.

The study also demonstrated that different longterm fertilization treatments had no significant effect on soil TK content,which may be due to a large reserve of potassium in the soil. However,compared with the inorganic fertilizer-only treatment,the organic fertilizeronly treatment considerably increased the AK content of the soil. It has been shown that the effective conversion rate of potassium in organic fertilizers is higher than that of inorganic potassium fertilizers. Moreover,readily available potassium and slowly available potassium in organic fertilizers are significantly reduced by soil fixation.In summary,these results provide evidence that the longterm application of a combination of organic and inorganic fertilizers can improve the overall fertility of the soil.

4.2.Analysis of key factors affecting rice yield

The nutrient content of soil directly affects the yield and quality of rice,but different nutrients have different influences on yield,and the mutual influences of nutrients also differ. Therefore,their specific impacts require further exploration. To better understand the influence of nutrients on rice yield under long-term fertilizer treatments and the degree of mutual influence between various nutrients,we conducted correlation analysis using the seven variables of SOC,TN,TP,AN,AP,AK,and pH. Our results showed that in the soil tested in this experiment,SOC,TN,AN,TP,and AP were significantly correlated with rice yield,and the correlation coefficients reached 0.428,0.496,0.518,0.501 and 0.438,respectively. These results indicated that soil carbon,nitrogen,and phosphorus are key elements affecting rice yield in this study,which is consistent with the results of other studies (Qaswaret al.2019;Liu J Aet al.2021;Liuet al.2022). Several studies have shown that organic matter can improve soil structure;coordinate soil water,fertilizer,air,and heat;enhance soil aeration and water permeability (Wang X Yet al.2021;Xuet al.2021;Li X Yet al.2022);and increase its ability to retain and supply water and fertilizer. Nitrogen is the main component of protein. It promotes the growth of stems and leaves and the development of tiller primordia,helping maintain and regulate the physiological functions of rice and thus improving the yield and quality of rice.Phosphorus promotes rice root growth and increases its tillering capacity. It enhances the stress resistance of rice,promotes the early maturation of rice,and,therefore,contributes to increased yield. Long-term fertilization can increase the TN,SOC,and TP content in the soil of paddy fields in southern China,so it helps stabilise and increase yields (Wanget al.2019). SOC and TN contents are highly correlated and show significant correlations with AN and AP. Therefore,combining organic and inorganic fertilizers can ensure relatively balanced nutrients in the soil and directly increase SOC,TN,and TP content,which in turn can promote the availability of other nutrients and improve the overall rice yield.

4.3.Effects of various fertilization treatments on soil nutrient balance

The nutrient balance of farmlands under various fertilizer treatments over many years of experimental conditions,soil nutrient content,and crop fertilizer requirements play crucial roles in determining the rice yield. In addition,selecting the appropriate fertilizer type and applying the correct volume of fertilizer help achieve balanced fertilization of farmland and provide a theoretical basis for efficiently utilising nutrients in a double-cropping rice system. The present study demonstrated that a lack of nitrogen fertilizer resulted in a nitrogen deficiency in the soil,and the application of nitrogen resulted in a nitrogen surplus,which increased with increasing rate of nitrogen application. The nitrogen uptake in M treatment was higher than that in NPK treatment,indicating that the application of organic fertilizer indirectly reduces the nitrogen surplus in the soil by promoting nitrogen uptake by the rice. Nevertheless,M treatment also led to a nitrogen deficit;therefore,applying a combination of organic (M) and inorganic fertilizers (NPK) in a reasonable ratio is necessary to maintain balanced nitrogen in the soil. The nitrogen budget in NPK treatment was negative because the nitrogen application rate of this treatment was half of the conventional application rate (with a cumulative nitrogen application rate for early and late rice of only 145 kg ha-1,far lower than the conventional nitrogen fertilizer application level). A long-term low input of nitrogen can easily lead to insufficient nitrogen in the soil and insufficient nitrogen uptake by plants. On the contrary,the nitrogen application rate in the doublecropping rice system reached 290 kg ha-1,and the nitrogen uptake by rice and nitrogen surplus in the soil increased significantly. This can easily lead to nitrogen loss through ammonia volatilisation and nitrogen leaching,which considerably impacts the security of the agricultural eco-environment (Zhaoet al.2012).

The nitrogen surplus after NPKM treatment reached 83.9 kg ha-1,which was significantly lower than that after the NKM and NPM treatments with the same nitrogen application rate. This shows that to reduce the soil nitrogen surplus and control the amount of nitrogen fertilizer,it is necessary to maintain a balance between nitrogen,phosphorus,and potassium levels. In other words,balanced fertilization under a fixed nitrogen application rate can promote rice growth and increase absorption of nitrogen,thereby reducing the nitrogen surplus.

In this study,the principles underlying the apparent balance of phosphorus and nitrogen in each treatment were largely similar. Soil treated with phosphorus had a surplus of phosphorus,and phosphorus has poor mobility and is easily fixed in the soil. A long-term surplus can lead to a gradual accumulation of phosphorus in the soil,which then enters groundwater through leaching and runoff,thus posing a threat to the water environment.Like the principles governing nitrogen surplus,under the same phosphorus application rate,the soil phosphorus surplus after NPKM treatment was lower than after PKM and NPM treatments. These results support that the combined application of organic and inorganic fertilizers is a suitable fertilization method to reduce nitrogen and phosphorus surpluses in paddy fields (Luet al.2021).

Under current farmland fertilization scenarios,in terms of the balance of the three elements in the soil,potassium may be in deficit to varying degrees (Chenet al.2020),which was also observed in our experiment.Due to the imbalance of potassium,the consumption of soil potassium intensified,resulting in deepening of the potassium deficiency. Therefore,to improve soil potassium fertility and improve the yield and quality of rice,attention should be paid to maintaining the balance of soil potassium. Understanding the basic characteristics of soil potassium supply,harnessing the ability of paddy soil to supply potassium,and then determining a more appropriate potassium application rate is the key to maintaining high rice yields and a potassium balance in paddy soil. This study showed that a total potassium input of 100-150 kg ha-1is required to maintain the potassium balance of soil for two seasons of rice each year.

5.Conclusion

The 38-year-long fertilization experiment performed under this study unravelled the potential of long-term fertilization to increase the yield of rice. Compared with long-term application of inorganic fertilizer,the application of cow manure combined with inorganic fertilizer (NPKM) improved the physical and chemical properties of soil,enhanced soil fertility,promoted soil nutrient balance,and ultimately significantly increased rice yield. Long-term application of organic fertilizer can contribute to accumulation of SOC,TN,AN,and AP in soil. In addition,compared with single application of organic fertilizer,single application of inorganic fertilizer significantly increased the contents of TP and AP in soil,while combined application of inorganic fertilizer with organic fertilizer significantly increased the contents of TP and AP in the soil in comparison with a single application of inorganic fertilizer. Each fertilization treatment had a negligible effect on TK in the soil,but combined application of organic and inorganic fertilizer significantly increased the content of available potassium in the soil compared with single application of inorganic fertilizer. In agricultural production,combined organic and inorganic fertilizers can best improve soil fertility,improve rice yield,and effectively reduce soil nutrient surplus,which is an ideal fertilization mode. However,it is necessary to consider the balance of nitrogen,phosphorus,and potassium nutrients,to achieve appropriate reduction in the application of nitrogen and phosphate fertilizer.

Acknowledgements

We appreciate all the managers for the long-term experiment. This study was financially supported by the earmarked fund for China Agriculture Research System(CARS-01-02A),the Natural Science Foundation of Hunan Province,China (2022NK2009),the Science and Technology Innovation Program of Hunan Province,China (2021RC2081),the Natural Science Foundation of Sichuan Province,China (2022NSFSC1059) and the Development Program of Hunan Province,China(2021NK2029).

Declaration of competing interest

The authors declare that they have no conflict of interest.