论文部分内容阅读
Abstract This study was conducted to investigate the effects of long-term located fertilization on soil phosphorus， the changes of soil available phosphorus （Olsen-P）， the evolution of soil total phosphorus （TP） and the ratio change of Olsen-P to TP （PAC） by 33-year fertilization experiments in winter wheat-summer maize rotation system in Shandong fluvo-aquic soil. Eight treatments were designed as no fertilization （CK）， nitrogen fertilizer （N） ， nitrogen and phosphate fertilizer （NP）， nitrogen and potassium fertilizer （NK）， phosphate and potassium fertilizer （PK）， nitrogen-phosphate-potassium fertilizer （NPK）， reduced NPK fertilizer （N15PK）， and increased NPK fertilizer （N25PK）. Meanwhile， eight organic fertilizer-added treatments were designed based on the application of inorganic fertilizer the same as the above ones. The results showed that TP， Olsen-P and PAC of treatments added with organic fertilizer were higher than those without organic fertilizer， and those of the treatments applied with phosphate fertilizer were higher than those of no phosphate fertilizer. With the increase of years， soil P pool decreased due to crop absorption， nutrient loss and morphological transformation and other causes under the treatments of without and only phosphate fertilizer， while remained stable under the treatments added with organic fertilizer. The PAC values were generally lower in fluvo-aquic soil， and it could be improved by the application of organic fertilizer. On the whole， the application of chemical phosphate fertilizer combined with organic fertilizer could improve the phosphorus content in soil and ensure the supply of phosphorus nutrition. This study would provide scientific basis for fertilization management and soil fertility in fluvo-aquic soil.
Key words Long-term located fertilization； Fluvo-aquic soil； Total phosphorus； Available phosphorus； Phosphorus pool； Activity coefficient
Fluvo-aquic soil is a type of soil with the largest area in Shandong Province， which is 4 666 000 hm2， in which the area of cultivated land is 4 106 000 hm2. The area of fluvo-aquic soil accounts for 38.53% of the total area of soil in the province， and the area of cultivated land in it accounts for 48.12% of the total area of cultivated area in the province. Fluvo-aquic soil is centralized in distribution， and 76.5% of the area of fluvo-aquic soil is distributed in alluvial plain of Yellow River in the northwest of Shandong Province， and 23.5% is distributed in hilly and mountainous area[1]. Fluvo-aquic soil is distributed in flat area with thick soil layer， which is rich in water and heat resources and suitable for many crops. It is the main upland soil in China， which abounds in grain and cotton. However， most fluvo-aquic soil is medium- and low-fertility soil， in addition to drought and flood disasters which happen occasionally， as well as soil salinization， crop yield is not stable. Therefore， rational application of fertilizer should be performed to improve nutrient contents in fluvo-aquic soil， thereby improving crop yield. Long-term located fertilization having the advantages of long term and climate representativeness， and not only could reveal soil fertility evolution and evaluate fertilizer benefit， but also could facilitate the research on the effect of fertilization on sustainable development of farmland ecosystem[2]. There are quite a lot studies on the evolution of phosphorus by long-term located experiments of fluvo-aquic soil in Shandong Province[3-4]， which provides certain theoretical basis for agricultural production and improvement of soil fertility in fluvo-aquic soil areas. Yang et al.[5] discussed the responses of soil available phosphorus （Olsen-P）， total phosphorus （TP） and soil phosphorus profit and loss under long-term fertilization， and concluded that soil phosphorus profit and loss is closely related to combined application of fertilizers. Xin et al.[6] studied the evolution of soil TP and Olsen-P in fluvo-aquic soil under long-term application of organic fertilizer and chemical fertilizer， and the results showed that the evolution of soil TP and Olsen-P is significantly affected by soil phosphorus profit and loss， and the application of organic fertilizer could increase the phosphorus capable of being absorbed and utilized by crops. Yuan et al.[3] found through conventional fertilization and no fertilization treatments that under the no fertilization condition， soil Olsen-P in fluvo-aquic soil decreased by 2.7 mg/kg per 100 kg/hm2 of phosphorus deficit， and under conventional fertilization， soil Olsen-P increased by 1.2 mg/kg per 100 kg/hm2 of phosphorus profit， so variation of soil Olsen-P was in very significant correlation with phosphorus profit and loss.
There are wide studies about phosphorus in fluvo-aquic soil in China， but few systematic studies. Therefore， systematic understanding of the change of soil Olsen-P contents， the evolution of soil phosphorus pool and phosphorus activation coefficient （PAC） in fluvo-aquic soil is of important guiding significance to scientific application of phosphorus in fluvo-aquic soil. In this study， long-term located experiments were carried out on fluvo-aquic soil， so as to investigate the effect of long-term fertilization on soil phosphorus. This study would provide a scientific basis for fertilization management and improvement of soil fertility in fluvo-aquic soil areas.
Materials and Methods
General situation of experimental field
The experiment was carried out in the experimental farm of Shandong Academy of Agricultural Sciences in Jinan City （36°40′ northern latitude， 117°00′ east longitude） with an altitude of 27.5 m. This area is located in the subtropical zone， with a sub-humid warm temperate continental monsoon climate. The annual average temperature is 14.8 ℃， and the accumulated temperature （higher than 10 ℃） is 4 774 ℃； the annual precipitation is 693.4 mm； the annual evaporation is 444.1 mm； the frost-free season is 216.4 d， and the annual sunshine duration is 1 870.9 h. The parent material of the tested fluvo-aquic soil is river sediment in modern times. The clay minerals in the soil are mainly hydromica and chlorite. The long-term located fertilizer monitoring was started in summer of 1982. The initial physical and chemical properties of surface soil （0-20 cm） are shown in Table 1.
Experimental design
The experimental treatments included no fertilization （CK）， nitrogen fertilizer （N）， nitrogen and phosphate fertilizer （NP）， nitrogen and potassium fertilizer （NK）， phosphate and potassium fertilizer （PK）， nitrogen-phosphate-potassium fertilizer （NPK）， reduced NPK fertilizer （N15PK）， increased NPK fertilizer （N25PK）， and each plot had an area of 1 m2， with three replications. Eight treatments added with organic fertilizer were also designed， i.e.， organic fertilizer （CK + M）， nitrogen and organic fertilizer （N + M）， nitrogen and phosphate fertilizer and organic fertilizer （NP + M）， nitrogen and potassium fertilizer and organic fertilizer （NK + M）， phosphate and potassium fertilizer （PK + M）， nitrogen-phosphate-potassium fertilizer and organic fertilizer （NPK + M）， reduced NPK fertilizer and organic fertilizer （N15PK + M）， and increased NPK fertilizer and organic fertilizer （N25PK + M）， without replications， and the application of inorganic fertilizer was the same as former eight treatments. The experiment adopted randomized block design. The application rates of fertilizers for various treatments are shown in Table 2. The organic fertilizer was horse manure， which averagely contained N 4.75 g/kg， P2O5 4.83 g/kg and K2O 9.90 g/kg. Two crops， wheat and maize were yielded a year in rotation. For wheat， nitrogen fertilizer was bottom-applied and topdressed equally， and phosphate and potassium fertilizer was all bottom applied. For maize， nitrogen-phosphate-potassium fertilizer was all bottom applied. On this basis， in the exper-imental area of organic fertilizer， horse manure was applied at a rate of 25 t/hm2 before seeding of wheat in autumn every year. In various treatments， the aboveground part was taken away， and the N， P and K nutrients in straws returning to the field was not reckoned in. As the management measures were the same every year， in consideration of sustainability of pool experiment and nonsignificance of soil nutrient change， soil sampling was performed once every year during 1982-1990， once every 2-3 years during 1991-2010， and once every 5 years from 2011 up to now.
Determination items and methods The analysis of soil sample was performed according to unified method， and the determination method were according to references[7-8]. Total phosphorus （TP） was determined by alkali fusion-Mo-Sb colorimetric method； and rapidly available phosphorus was determined according to Olsen method.
Data processing and analysis
Phosphorus activation coefficient （PAC） （ %） = Olsen P （mg /kg）/[total phosphorous （g/kg） × 1 000]×100.
The experimental data were processed and plotted in Microsoft Excel 2010.
Agricultural Biotechnology2018
Results and Analysis
Changing trend of soil TP under long-term fertilization
Phosphorus is a major element essential for plant growth. Application of phosphate could improve yields of agricultural crops， and is an important means for sustaining sustainable agricultural production and ensure food safety， and soil TP content could reflect soil phosphorus storage[9]. As in the long-term located micro-plot experiment， sampling was performed for few times， not every year， correlation regression analysis was adopted to investigate the effects of different fertilization modes on TP contents in fluvo-aquic soil in Shandong Province， as shown in Fig. 1 and Fig. 2. It could be seen from Fig. 1 that under different fertilization modes， soil TP content decreased with the fertilization years increasing， and the three treatments without application of phosphate fertilizer （CK， N and NK） exhibited a very significant decreasing trend of soil TP due to the absorption of soil phosphorus by crops （P<0.01）. After the application of phosphate fertilizer to soil， soil TP exhibited an increasing trend in the first 8 years； and with the fertilization years increasing， soil TP content decreased， but had no big difference from that of the soil before the experiment. Some scholars deem that accumulative phosphorus might turn into apatite state naturally existing in soil when the accumulation time is longer than three years[10].
It could be seen from Fig. 2 that with the fertilization years increasing， the treatments applied with organic fertilizer without phosphate fertilizer （CK + M， N + M and NK + M） exhibited a decreasing trend， while the soil TP of the treatments combining phosphate and organic fertilizer increased at first and then tended to be stable. Compared with tested soil （with total phosphorous content of 1.28 mg/kg）， the total phosphorous contents of treatments NP + M， PK + M， NPK + M， N15PK + M and N25PK + M increased at the time of 20 years by 39.5%， 51.8%， 45.1%， 34.6% and 50.1%， respectively. It could be seen that long-term no application of phosphate would result in the soil state of very strong deficiency of phosphorous； and sole application of phosphate fertilizer or soil application of organic fertilizer would cause decrease of soil phosphorus pool due to crop absorption， running off and state conversation， while after combined application of inorganic phosphate fertilizer and organic fertilizer， soil phosphorous pool is stable， and supply of phosphorus could be ensured.
Changing trend of soil Olsen-P under long-term fertilization
Soil Olsen-P could be directly absorbed by agricultural crops， and soil phosphorus supply level is an important index[11]. The change of soil Olsen-P content in fluvo-aquic soil in Shandong Province under different fertilization modes is shown in Fig. 3 and Fig. 4. Among the treatments without the application of phosphate fertilizer （CK， N and NK）， soil Olsen-P hardly fluctuated with the years increasing， the average content of soil Olsen-P was very low， stably in the range of 2.5-2.8 mg/kg， reaching the very deficient degree， and crop yield was remarkably affected. As to the treatments applied with phosphate fertilizer， soil Olsen-P content exhibited a decreasing trend overall over planting time， and except that treatment NP exhibited significant negative correlation between soil Olsen-P content and planting years （P<0.05）， the soil Olsen-P contents of other treatments applied with phosphate fertilizer were in very significant negative correlation with time （P<0.01）. However， as soil Olsen-P content increased rapidly in the first 8 years， until 2010， soil Olsen-P contents of NP， PK， NPK， N15PK and N25PK increased， respectively， to 9.89， 18.37， 12.66， 12.84 and 20.58 mg/kg， which were all higher than the tested soil （with soil Olsen-P content of 5.90 mg/kg）； and under different fertilization modes， soil Olsen-P content ranked as PK>NP>N15PK>NPK>N25PK>NK>CK>N， which accords with the research results of Xin et al.[6]. It could be seen from Fig. 4 that except treatment NPK + M， other treatments added with organic fertilizer showed soil Olsen-P contents slowly increasing with the planting years increasing； and treatments applied both with phosphate fertilizer and organic fertilizer had overall higher soil Olsen-P contents， soil Olsen-P contents of NP， PK， NPK， N15PK and N25PK in 2010 increased， respectively， to 45.44， 69.76， 47.10， 39.92 and 28.68 mg/kg， with the annual increments of 1.46， 2.37， 1.53， 1.26 and 0.84 mg/kg， respectively， and treatment PK + M exhibited the largest annual increment of Olsen-P， which might be due to the deficiency of nitrogen fertilizer， under which crops absorbed less phosphorus in soil. For treatments added with organic fertilizer， Olsen-P content increased remarkably in 1996-1998， which might be due to relatively higher organic nutrients in these year. The same as the conclusion of soil TP， combined application of inorganic phosphate and organic fertilizer could better improve Olsen-P content. Relationship between Olsen-P and TP under long-term fertilization
PAC could represent the activation capacity of soil phosphorus， and the evolution law of PAC of fluvo-aquic soil in Shandong Province is shown in Fig. 5 and Fig. 6 over time. For treatments without the application of organic fertilizer， the PAC values of treatments without the application of phosphate fertilizer （CK， N and NK） all exhibited a slowly increasing trend over fertilization time， but were slightly lower overall， and the PAC values of treatments CK， N and NK were， respectively， 0.28%， 0.27% and 0.24%， indicating very low soil activation rate. As to treatments applied with phosphate fertilizer （NP， PK， NPK， N15PK and N25PK）， in the range of 3 years to 8 years of fertilization， soil PAC was higher than 2%， indicating that soil TP could be easily converted to Olsen-P； however， over fertilization time， soil PAC value exhibited a decreasing trend， after 20 years of planting， soil TP and Olsen-P both decreased， and due to more remarkable reduction of Olsen-P， PAC value also decreased； and through 28 years of fertilization， the average PAC values were， respectively， 1.87%， 2.10%， 1.71%， 1.69% and 1.65%， which were remarkably higher than those of treatments without application of phosphate fertilizer.
For various treatments added with organic fertilizer， except treatment N25PK + M， soil PAC values of other treatments always had an increasing trend， the PAC values of treatments CK + M， NP + M， PK + M， NPK + M， N15PK + M and N25PK + M were 3.15%， 4.66%， 4.00%， 4.03%， 3.85% and 2.82% after 23 years of planting， respectively， and compared with the PAC value of tested soil （0.46%）， the PAC values of these treatments had annual increasing rates in the range of 0.10%-0.18%.
The PAC values of treatments added with organic fertilizer were higher than those of inorganic fertilizer treatments overall. Huang et al.[12] found that the effectiveness of organic fertilizer treatments was higher than that of inorganic fertilizer treatments. However， PAC value and rising speed were remarkably lower than those of red loam， indicating that PAC value has certain relation with soil type. In addition， lower PAC value also might be related to smaller pool area and not rich community structure， thereby resulting in lower activation rate of Olsen-P.
Conclusions
Phosphorous is an important factor for crop growth and development. No application of phosphate would cause deficiency of soil phosphorus， and the application of phosphate fertilizer， especially the combined application of phosphate fertilizer and organic fertilizer could keep the stability of soil phosphorus storage. Under long-term no application of phosphate fertilizer， Olsen-P in soil decreased to 2.5-2.8 mg/kg， which seriously affected crop growth. The application of inorganic phosphate or organic fertilizer would improve Olsen-P content， but also would change to deficient state with planting years increasing.
Under the same application rate of phosphate fertilizer， due to the effect from crop growth， the soil phosphorous contents of treatments with partial application of phosphate fertilizer （ NP and PK） were higher than balance fertilization treatments （NPK， N15PK and N25PK）.
PAC value of soil without application of fertilizer decreased to 0.24%， while the application of phosphate fertilizer could improve PAC value， and the application of organic fertilizer is more beneficial to the activation of phosphorus in soil.
References
[1] Soil and Fertilizer Station of Shandong Province[M]. Beijing： China Agriculture Press， 1994.
[2] LAFLEN JM， LANE LJ， FOSTER GR. WEPP： A new generation of erosion prediction technology [J]. Journal of Soil & Water Conservation， 1991， 46（1） ： 34-38.
[3] YUAN TY， WANG JZ， JI JH， et al. Changes in soil available phosphorus and its response to phosphorus balance under long-term fertilization in fluvo-aquic soil[J]. Acta Agriculturae Nucleatae Sinica， 2017， 31（1） ： 125-134.
[4] NAN ZW， LIANG B， LIU ST. Effect of Long-term located fertilization on nitrogen mineralization characteristics and crop yields in the fluvo-aquic soil[J]. Journal of Soil and Water Conservation， 2015， 29（6） ：107-112.
[5] YANG J， GAO W， REM SR. Response of soil phosphorus to P balance under long-term fertilization in fluvo-aquic soil[J]. Scientia Agricultura Sinica， 2015， 48（ 23） ： 4738 -4747.
[6] XIN XL， QIN SW， ZHANG JB， et al. Dynamics of phosphorus in Fluvo-aquic soil under long-term fertilization[J]. Journal of Plant Nutrition and Fertilizer， 2015， 21 （6） ：1514-1520.
[7] BAO SD. Soil agrochemical analysis[M]. Beijing： China Agriculture Press， 2000.
[8] Soil Science Society of China. Routine analytical methods for soil and agro-chemistry[M]. Beijing： Science Press， 1983.
[9] YE YS， LIANG XQ， LI L， et al. Effects of different water and nitrogen managements on phosphorus loss via runoff and leaching from paddy fields in Taihu Lake basin[J]. Acta Scientiae Circumstantiae， 2015， 35（4） ： 1125-1135.
[10] LIU SL， JIE XL， LI YT， et al. Study on bio-availability and transformation of different phosphates in calcareous soils[J]. Journal of Henan Agricultural University， 2002， 36（4） ： 370-373.
[11] QU JF， LI JM， XU MG， et al. Response of typical soil phosphorus evolution to long-term single nitrogen fertilization[J]. Scientia Agricultura Sinica， 2009， 42（11） ： 3933-3939.
[12] HUANG SM， BAO DJ， HUANGPU XR， et al. Effect of long-term fertilization on utilization and accumulation of phosphate nutrient in fluvo-aquic soil[J]. Scientia Agricultura Sinica， 2006， 39（1） ：102-108.
Key words Long-term located fertilization； Fluvo-aquic soil； Total phosphorus； Available phosphorus； Phosphorus pool； Activity coefficient
Fluvo-aquic soil is a type of soil with the largest area in Shandong Province， which is 4 666 000 hm2， in which the area of cultivated land is 4 106 000 hm2. The area of fluvo-aquic soil accounts for 38.53% of the total area of soil in the province， and the area of cultivated land in it accounts for 48.12% of the total area of cultivated area in the province. Fluvo-aquic soil is centralized in distribution， and 76.5% of the area of fluvo-aquic soil is distributed in alluvial plain of Yellow River in the northwest of Shandong Province， and 23.5% is distributed in hilly and mountainous area[1]. Fluvo-aquic soil is distributed in flat area with thick soil layer， which is rich in water and heat resources and suitable for many crops. It is the main upland soil in China， which abounds in grain and cotton. However， most fluvo-aquic soil is medium- and low-fertility soil， in addition to drought and flood disasters which happen occasionally， as well as soil salinization， crop yield is not stable. Therefore， rational application of fertilizer should be performed to improve nutrient contents in fluvo-aquic soil， thereby improving crop yield. Long-term located fertilization having the advantages of long term and climate representativeness， and not only could reveal soil fertility evolution and evaluate fertilizer benefit， but also could facilitate the research on the effect of fertilization on sustainable development of farmland ecosystem[2]. There are quite a lot studies on the evolution of phosphorus by long-term located experiments of fluvo-aquic soil in Shandong Province[3-4]， which provides certain theoretical basis for agricultural production and improvement of soil fertility in fluvo-aquic soil areas. Yang et al.[5] discussed the responses of soil available phosphorus （Olsen-P）， total phosphorus （TP） and soil phosphorus profit and loss under long-term fertilization， and concluded that soil phosphorus profit and loss is closely related to combined application of fertilizers. Xin et al.[6] studied the evolution of soil TP and Olsen-P in fluvo-aquic soil under long-term application of organic fertilizer and chemical fertilizer， and the results showed that the evolution of soil TP and Olsen-P is significantly affected by soil phosphorus profit and loss， and the application of organic fertilizer could increase the phosphorus capable of being absorbed and utilized by crops. Yuan et al.[3] found through conventional fertilization and no fertilization treatments that under the no fertilization condition， soil Olsen-P in fluvo-aquic soil decreased by 2.7 mg/kg per 100 kg/hm2 of phosphorus deficit， and under conventional fertilization， soil Olsen-P increased by 1.2 mg/kg per 100 kg/hm2 of phosphorus profit， so variation of soil Olsen-P was in very significant correlation with phosphorus profit and loss.
There are wide studies about phosphorus in fluvo-aquic soil in China， but few systematic studies. Therefore， systematic understanding of the change of soil Olsen-P contents， the evolution of soil phosphorus pool and phosphorus activation coefficient （PAC） in fluvo-aquic soil is of important guiding significance to scientific application of phosphorus in fluvo-aquic soil. In this study， long-term located experiments were carried out on fluvo-aquic soil， so as to investigate the effect of long-term fertilization on soil phosphorus. This study would provide a scientific basis for fertilization management and improvement of soil fertility in fluvo-aquic soil areas.
Materials and Methods
General situation of experimental field
The experiment was carried out in the experimental farm of Shandong Academy of Agricultural Sciences in Jinan City （36°40′ northern latitude， 117°00′ east longitude） with an altitude of 27.5 m. This area is located in the subtropical zone， with a sub-humid warm temperate continental monsoon climate. The annual average temperature is 14.8 ℃， and the accumulated temperature （higher than 10 ℃） is 4 774 ℃； the annual precipitation is 693.4 mm； the annual evaporation is 444.1 mm； the frost-free season is 216.4 d， and the annual sunshine duration is 1 870.9 h. The parent material of the tested fluvo-aquic soil is river sediment in modern times. The clay minerals in the soil are mainly hydromica and chlorite. The long-term located fertilizer monitoring was started in summer of 1982. The initial physical and chemical properties of surface soil （0-20 cm） are shown in Table 1.
Experimental design
The experimental treatments included no fertilization （CK）， nitrogen fertilizer （N）， nitrogen and phosphate fertilizer （NP）， nitrogen and potassium fertilizer （NK）， phosphate and potassium fertilizer （PK）， nitrogen-phosphate-potassium fertilizer （NPK）， reduced NPK fertilizer （N15PK）， increased NPK fertilizer （N25PK）， and each plot had an area of 1 m2， with three replications. Eight treatments added with organic fertilizer were also designed， i.e.， organic fertilizer （CK + M）， nitrogen and organic fertilizer （N + M）， nitrogen and phosphate fertilizer and organic fertilizer （NP + M）， nitrogen and potassium fertilizer and organic fertilizer （NK + M）， phosphate and potassium fertilizer （PK + M）， nitrogen-phosphate-potassium fertilizer and organic fertilizer （NPK + M）， reduced NPK fertilizer and organic fertilizer （N15PK + M）， and increased NPK fertilizer and organic fertilizer （N25PK + M）， without replications， and the application of inorganic fertilizer was the same as former eight treatments. The experiment adopted randomized block design. The application rates of fertilizers for various treatments are shown in Table 2. The organic fertilizer was horse manure， which averagely contained N 4.75 g/kg， P2O5 4.83 g/kg and K2O 9.90 g/kg. Two crops， wheat and maize were yielded a year in rotation. For wheat， nitrogen fertilizer was bottom-applied and topdressed equally， and phosphate and potassium fertilizer was all bottom applied. For maize， nitrogen-phosphate-potassium fertilizer was all bottom applied. On this basis， in the exper-imental area of organic fertilizer， horse manure was applied at a rate of 25 t/hm2 before seeding of wheat in autumn every year. In various treatments， the aboveground part was taken away， and the N， P and K nutrients in straws returning to the field was not reckoned in. As the management measures were the same every year， in consideration of sustainability of pool experiment and nonsignificance of soil nutrient change， soil sampling was performed once every year during 1982-1990， once every 2-3 years during 1991-2010， and once every 5 years from 2011 up to now.
Determination items and methods The analysis of soil sample was performed according to unified method， and the determination method were according to references[7-8]. Total phosphorus （TP） was determined by alkali fusion-Mo-Sb colorimetric method； and rapidly available phosphorus was determined according to Olsen method.
Data processing and analysis
Phosphorus activation coefficient （PAC） （ %） = Olsen P （mg /kg）/[total phosphorous （g/kg） × 1 000]×100.
The experimental data were processed and plotted in Microsoft Excel 2010.
Agricultural Biotechnology2018
Results and Analysis
Changing trend of soil TP under long-term fertilization
Phosphorus is a major element essential for plant growth. Application of phosphate could improve yields of agricultural crops， and is an important means for sustaining sustainable agricultural production and ensure food safety， and soil TP content could reflect soil phosphorus storage[9]. As in the long-term located micro-plot experiment， sampling was performed for few times， not every year， correlation regression analysis was adopted to investigate the effects of different fertilization modes on TP contents in fluvo-aquic soil in Shandong Province， as shown in Fig. 1 and Fig. 2. It could be seen from Fig. 1 that under different fertilization modes， soil TP content decreased with the fertilization years increasing， and the three treatments without application of phosphate fertilizer （CK， N and NK） exhibited a very significant decreasing trend of soil TP due to the absorption of soil phosphorus by crops （P<0.01）. After the application of phosphate fertilizer to soil， soil TP exhibited an increasing trend in the first 8 years； and with the fertilization years increasing， soil TP content decreased， but had no big difference from that of the soil before the experiment. Some scholars deem that accumulative phosphorus might turn into apatite state naturally existing in soil when the accumulation time is longer than three years[10].
It could be seen from Fig. 2 that with the fertilization years increasing， the treatments applied with organic fertilizer without phosphate fertilizer （CK + M， N + M and NK + M） exhibited a decreasing trend， while the soil TP of the treatments combining phosphate and organic fertilizer increased at first and then tended to be stable. Compared with tested soil （with total phosphorous content of 1.28 mg/kg）， the total phosphorous contents of treatments NP + M， PK + M， NPK + M， N15PK + M and N25PK + M increased at the time of 20 years by 39.5%， 51.8%， 45.1%， 34.6% and 50.1%， respectively. It could be seen that long-term no application of phosphate would result in the soil state of very strong deficiency of phosphorous； and sole application of phosphate fertilizer or soil application of organic fertilizer would cause decrease of soil phosphorus pool due to crop absorption， running off and state conversation， while after combined application of inorganic phosphate fertilizer and organic fertilizer， soil phosphorous pool is stable， and supply of phosphorus could be ensured.
Changing trend of soil Olsen-P under long-term fertilization
Soil Olsen-P could be directly absorbed by agricultural crops， and soil phosphorus supply level is an important index[11]. The change of soil Olsen-P content in fluvo-aquic soil in Shandong Province under different fertilization modes is shown in Fig. 3 and Fig. 4. Among the treatments without the application of phosphate fertilizer （CK， N and NK）， soil Olsen-P hardly fluctuated with the years increasing， the average content of soil Olsen-P was very low， stably in the range of 2.5-2.8 mg/kg， reaching the very deficient degree， and crop yield was remarkably affected. As to the treatments applied with phosphate fertilizer， soil Olsen-P content exhibited a decreasing trend overall over planting time， and except that treatment NP exhibited significant negative correlation between soil Olsen-P content and planting years （P<0.05）， the soil Olsen-P contents of other treatments applied with phosphate fertilizer were in very significant negative correlation with time （P<0.01）. However， as soil Olsen-P content increased rapidly in the first 8 years， until 2010， soil Olsen-P contents of NP， PK， NPK， N15PK and N25PK increased， respectively， to 9.89， 18.37， 12.66， 12.84 and 20.58 mg/kg， which were all higher than the tested soil （with soil Olsen-P content of 5.90 mg/kg）； and under different fertilization modes， soil Olsen-P content ranked as PK>NP>N15PK>NPK>N25PK>NK>CK>N， which accords with the research results of Xin et al.[6]. It could be seen from Fig. 4 that except treatment NPK + M， other treatments added with organic fertilizer showed soil Olsen-P contents slowly increasing with the planting years increasing； and treatments applied both with phosphate fertilizer and organic fertilizer had overall higher soil Olsen-P contents， soil Olsen-P contents of NP， PK， NPK， N15PK and N25PK in 2010 increased， respectively， to 45.44， 69.76， 47.10， 39.92 and 28.68 mg/kg， with the annual increments of 1.46， 2.37， 1.53， 1.26 and 0.84 mg/kg， respectively， and treatment PK + M exhibited the largest annual increment of Olsen-P， which might be due to the deficiency of nitrogen fertilizer， under which crops absorbed less phosphorus in soil. For treatments added with organic fertilizer， Olsen-P content increased remarkably in 1996-1998， which might be due to relatively higher organic nutrients in these year. The same as the conclusion of soil TP， combined application of inorganic phosphate and organic fertilizer could better improve Olsen-P content. Relationship between Olsen-P and TP under long-term fertilization
PAC could represent the activation capacity of soil phosphorus， and the evolution law of PAC of fluvo-aquic soil in Shandong Province is shown in Fig. 5 and Fig. 6 over time. For treatments without the application of organic fertilizer， the PAC values of treatments without the application of phosphate fertilizer （CK， N and NK） all exhibited a slowly increasing trend over fertilization time， but were slightly lower overall， and the PAC values of treatments CK， N and NK were， respectively， 0.28%， 0.27% and 0.24%， indicating very low soil activation rate. As to treatments applied with phosphate fertilizer （NP， PK， NPK， N15PK and N25PK）， in the range of 3 years to 8 years of fertilization， soil PAC was higher than 2%， indicating that soil TP could be easily converted to Olsen-P； however， over fertilization time， soil PAC value exhibited a decreasing trend， after 20 years of planting， soil TP and Olsen-P both decreased， and due to more remarkable reduction of Olsen-P， PAC value also decreased； and through 28 years of fertilization， the average PAC values were， respectively， 1.87%， 2.10%， 1.71%， 1.69% and 1.65%， which were remarkably higher than those of treatments without application of phosphate fertilizer.
For various treatments added with organic fertilizer， except treatment N25PK + M， soil PAC values of other treatments always had an increasing trend， the PAC values of treatments CK + M， NP + M， PK + M， NPK + M， N15PK + M and N25PK + M were 3.15%， 4.66%， 4.00%， 4.03%， 3.85% and 2.82% after 23 years of planting， respectively， and compared with the PAC value of tested soil （0.46%）， the PAC values of these treatments had annual increasing rates in the range of 0.10%-0.18%.
The PAC values of treatments added with organic fertilizer were higher than those of inorganic fertilizer treatments overall. Huang et al.[12] found that the effectiveness of organic fertilizer treatments was higher than that of inorganic fertilizer treatments. However， PAC value and rising speed were remarkably lower than those of red loam， indicating that PAC value has certain relation with soil type. In addition， lower PAC value also might be related to smaller pool area and not rich community structure， thereby resulting in lower activation rate of Olsen-P.
Conclusions
Phosphorous is an important factor for crop growth and development. No application of phosphate would cause deficiency of soil phosphorus， and the application of phosphate fertilizer， especially the combined application of phosphate fertilizer and organic fertilizer could keep the stability of soil phosphorus storage. Under long-term no application of phosphate fertilizer， Olsen-P in soil decreased to 2.5-2.8 mg/kg， which seriously affected crop growth. The application of inorganic phosphate or organic fertilizer would improve Olsen-P content， but also would change to deficient state with planting years increasing.
Under the same application rate of phosphate fertilizer， due to the effect from crop growth， the soil phosphorous contents of treatments with partial application of phosphate fertilizer （ NP and PK） were higher than balance fertilization treatments （NPK， N15PK and N25PK）.
PAC value of soil without application of fertilizer decreased to 0.24%， while the application of phosphate fertilizer could improve PAC value， and the application of organic fertilizer is more beneficial to the activation of phosphorus in soil.
References
[1] Soil and Fertilizer Station of Shandong Province[M]. Beijing： China Agriculture Press， 1994.
[2] LAFLEN JM， LANE LJ， FOSTER GR. WEPP： A new generation of erosion prediction technology [J]. Journal of Soil & Water Conservation， 1991， 46（1） ： 34-38.
[3] YUAN TY， WANG JZ， JI JH， et al. Changes in soil available phosphorus and its response to phosphorus balance under long-term fertilization in fluvo-aquic soil[J]. Acta Agriculturae Nucleatae Sinica， 2017， 31（1） ： 125-134.
[4] NAN ZW， LIANG B， LIU ST. Effect of Long-term located fertilization on nitrogen mineralization characteristics and crop yields in the fluvo-aquic soil[J]. Journal of Soil and Water Conservation， 2015， 29（6） ：107-112.
[5] YANG J， GAO W， REM SR. Response of soil phosphorus to P balance under long-term fertilization in fluvo-aquic soil[J]. Scientia Agricultura Sinica， 2015， 48（ 23） ： 4738 -4747.
[6] XIN XL， QIN SW， ZHANG JB， et al. Dynamics of phosphorus in Fluvo-aquic soil under long-term fertilization[J]. Journal of Plant Nutrition and Fertilizer， 2015， 21 （6） ：1514-1520.
[7] BAO SD. Soil agrochemical analysis[M]. Beijing： China Agriculture Press， 2000.
[8] Soil Science Society of China. Routine analytical methods for soil and agro-chemistry[M]. Beijing： Science Press， 1983.
[9] YE YS， LIANG XQ， LI L， et al. Effects of different water and nitrogen managements on phosphorus loss via runoff and leaching from paddy fields in Taihu Lake basin[J]. Acta Scientiae Circumstantiae， 2015， 35（4） ： 1125-1135.
[10] LIU SL， JIE XL， LI YT， et al. Study on bio-availability and transformation of different phosphates in calcareous soils[J]. Journal of Henan Agricultural University， 2002， 36（4） ： 370-373.
[11] QU JF， LI JM， XU MG， et al. Response of typical soil phosphorus evolution to long-term single nitrogen fertilization[J]. Scientia Agricultura Sinica， 2009， 42（11） ： 3933-3939.
[12] HUANG SM， BAO DJ， HUANGPU XR， et al. Effect of long-term fertilization on utilization and accumulation of phosphate nutrient in fluvo-aquic soil[J]. Scientia Agricultura Sinica， 2006， 39（1） ：102-108.