Introduction

Rice (Oryza sativa L.) is one of the most important crops worldwide, but its production is limited by soil phosphorus (P) deficiencies in many parts of the world (Fageria et al. 2011; Hedley et al. 1994; Raghothama 1999). Phosphorus is a critical nutrient for biological activity (Ramaekers et al. 2010). Inorganic P in soil is the primary source of P for crops (McDowell and Stewart 2006), and the majority of bio-available P is water-soluble inorganic P (Toor et al. 2003). Compared with other major nutrients, P is by far the least mobile and least available to plants under most soil conditions (Hinsinger 2001; Schachtman et al. 1998). The orthophosphate taken up by crops originates from the soil solution and solid phase of various inorganic and organic P fractions (Negassa and Leinweber 2009). The practice of straw retention has been reported to affect available P concentrations in the soil and, consequently, crop yield (Beri et al. 1995; Gupta et al. 2007; Lan et al. 2012; Pathak et al. 2006; Yadvinder et al. 2004). Biological and biochemical reactions, such as mineralization and dissolution by phosphatase enzymes, control much of the organic P fractions (McGill and Cole 1981; Murrmann and Peech 1969; Medley et al. 1982), and microbial activity has been shown to play a major role in redistributing P into different forms in the soil. Several studies have demonstrated that multiple years of continuous straw retention reduce the absorption of P in soil and increase the level of P released to the surface soil; compared to the practice of straw burning, straw retention increases the available P content of soil and the crop yield (Pathak et al. 2006). In a northern Indian rice–wheat rotation area, Gupta et al. (2007) observed that after four consecutive years of straw retention, the levels of available P, organic P, and inorganic P in the soil increased to varying degrees, and continuous straw retention provided an equivalent of 13 kg ha−1 year−1 of inorganic P. McLaughlin et al. (1988) found that straw retention in wheat pasture rotations did not have an impact on subsequent crops, although it increased the level of organic P in the soil. Lan et al. (2012) in southern China and Yadvinder et al. (2004) in northern India observed that continuous straw retention coupled with the application of chemical fertilizer increased the content of available P in the soil. Residues also increase the P availability in soil by stimulating the microbial production of organic acid anions and phenolic compounds, which might lead to the mobilization of adsorbed P through competition for P adsorption sites (Xu et al. 2006). The products of organic residue decomposition can also modify the availability of native soil P (Yadvinder et al. 1992; Nwoke et al. 2004). In waterlogged soils under rice, crop residues can increase the availability of indigenous P as a result of intense soil reduction (Singh et al. 1988; Yadvinder et al. 2005). Other studies have shown that straw retention has no significant effect on the available P in soil (Malhi et al. 2011; Prasad et al. 1999; Yadvinder et al. 2008). Beri et al. (1995) observed a decrease in the available P content in soil after 11 consecutive years of straw retention in an Indian rice–wheat rotation cropping area. In the same area, Bhandari et al. (2002) reported that the available P in soil decreased significantly after 10 consecutive years of straw retention after the rice season. This might have been caused by the simple crop rotation, which in the long term might have led to reduced yields and less net return of nutrients to the soil as crop residue. However, the disposal of crop residues by burning is often criticized because it accelerates the loss of soil organic matter and nutrients, increases C emissions, causes intense air pollution, and reduces soil microbial activity (Biederbeck et al. 1980; Kumar and Goh 1999; Rasmussen et al. 1980).

In agricultural production, the nutrient content of the soil solution is an important measurement of soil fertility (Hoagland 1922). P in the soil solution is a direct source of P absorbed by crops (Holford 1997). The purpose of this experiment was to study the effects of straw retention on the available P content in the soil solution and on rice yield. This research will provide a basis for understanding the effects of straw retention on the available P concentration in soil solutions and determining appropriate fertilizer application levels for rice-growing areas in northeast China.

Materials and methods

Experimental site

The study was conducted at XiangFang experimental farm of the Northeast Agricultural University, Harbin (45° 34′–45° 46′N, 126° 22′–126° 50′E, 171.7 m.a.s.l.), which is located in the state of Heilongjiang, China (Fig. 1). The area receives an average annual rainfall of 500–550 mm, of which approximately 80 % occurs between June and September. The region is in a cold temperate climate zone with a continental climate and has a frost-free period of approximately 140 days and a ≥10 °C accumulated temperature of approximately 2,700 °C per year. There is no crop rotation, the system is composed of continuous rice cropping, and the soil used in the field experimental is Mollisol.

Fig. 1
figure 1

Sketch map showing the locations of the long-term experimental sites in China

Full size image

Experimental design

This work includes three experiments: a field experiment, a pot experiment, and an incubation experiment. The field experiment (experiment A) was conducted from 2008 to 2012, with preliminary experiments conducted in 2008 and 2009. The size of the plots were 2 by 2 m, and they were built using cement and filled with testing soil at a depth of approximately 50 cm. The soil basal fertilities were as follows: 23.92 g kg−1 organic matter, 1.48 g kg−1 total N, 0.83 g kg−1 total P, 21.91 g kg−1 total K, 14.62 mg kg−1 NH4 +-N, 30.29 mg kg−1 NO3 −-N, 41.95 mg kg−1 Olsen-P, and 130.17 mg kg−1 available K. The experimental design was a completely randomized block design with two treatments (straw retention (SRT) and straw removal (SRM)) and three replicates. The P content of the straw returned to the soil was 1.80 ± 0.06 g kg−1. The straws were cut into 5-cm-long pieces, and 5 kg (12.5 t ha−1) of straw cuttings was embedded in the soil in each plot. Each year, the plots were plowed (20 cm) on May 20 and then soaked with water on May 25. The seed spacing intra-row was 13 cm and inter-row was 30 cm, with 3 seedlings per hill. The seedlings were transplanted on May 30. Beginning 10 days after transplanting, samples from the soil solution in the plots were collected once every 10 days. In each plot, 120 g urea (N 46 %, 300 kg ha−1), 60 g (NH4)2HPO4 (N 18 %, P2O5 46 %, 150 kg ha−1), and 40 g potassium sulfate (K2O 30 %, 100 kg ha−1) were applied. One half of the urea, P, and potassium were applied as a basal fertilizer, and the other half of the urea was applied at the tillering stage. Prior to transplanting (25–30 May), a water level of 5–7 cm was maintained in the plots. The water level was reduced to 2–3 cm per week after transplanting and then increased to 5–7 cm from the retention green stage to the milky stage and irrigated to 5 cm after the water level dried out.

The pot experiment (experiment B) was conducted in 2012. In this experiment, 30-cm-diameter plastic buckets were filled with 15 kg of soil. To prevent the influence of P on the basal fertility of the P fertilizer treatment, a Mollisol with low Olsen-P was used in the experiment. The soil was from a paddy field sowed with rice for multiple consecutive years. The soil’s basal fertilities were as follows: 17.42 g kg−1 organic matter, 1.53 g kg−1 total N, 0.36 g kg−1 total P, 25.52 g kg−1 total K, 10.55 mg kg−1 NH4 +-N, 42.63 mg kg−1 NO3-N, 14.97 mg kg−1 Olsen-P, and 137.57 mg kg−1 available K. Three seedlings were planted in each hole, with 3 holes pot−1. Five P levels (P0, P1, P2, P3, and P4) were established by applying 0 g (0 kg ha−1), 0.525 g (75 kg ha−1), 1.05 g (150 kg ha−1), 1.575 g (225 kg ha−1), or 2.1 g (300 kg ha−1) Ca(H2PO4)2 (P2O5:61 %), respectively, to the five pots. In each pot, 1.05 g urea (150 kg ha−1) and 0.70 g K2SO4 (100 kg ha−1) were also applied as a basal fertilizer, and 1.05 g urea (150 kg ha−1) was added at the tillering stage. The experiments were divided into two subgroups: SRT, in which 80 g straw (12.5 t ha−1) was added, and SRM. Experiment B was replicated five times, and the crop management was the same as in experiment A.

The incubation experiment (experiment C) included two treatments: straw retention (SRT) and straw removal (SRM). In a 500-ml wide-mouthed bottle, 150 g dry soil (supplemented with 8 g chopped rice straw [P: 1.82 ± 0.01 g kg−1] in the straw retention group) and 250 ml available P solution were added, and the water line was marked on the bottle. The bottle was then sealed with plastic film and incubated at 25 °C in an incubator. After each sampling, distilled water was added to compensate for the lost liquid based on the initial marked water line on the bottle. Two subgroups were divided from this group: subgroup I was treated with 5.00 mg L−1 available P and subgroup II was treated with 10.00 mg L−1 available P. Samples were collected in triplicate at various time points (1d, 5d, 10d, 15d, 20d, 30d, and 40d after inoculation). For sampling, the soil and water were well mixed, the mixture was filtered, and the available P content of the filtrate was determined.

Soil solution collection and analysis

The collection device used to sample the soil solution is shown in Fig. 2. A 30-cm-long hard PVC plastic pipe was sealed at one end, and two holes (2 mm in diameter) were drilled 2–3 cm from the sealed end as collection apertures for the soil solution. At the other end of the pipe, a vent hole (5 mm in diameter) was drilled. The collection apertures were wrapped with topical gauze to prevent clogging. Two sampling pipes were placed in each plot, with one pipe for each pot. The pipes were inserted into the soil, so that the collection holes were 10 cm below the soil surface, and the soil solution was collected from a depth of 10 cm. To ensure that the collected soil solution had infiltrated on the same day, the solution that accumulated during the interval between collections was drained 1 day prior to collection, and the pipe was covered with a rubber stopper to prevent rain and debris from filtering in. During sampling, the soil solutions were transferred to plastic bottles and immediately stored in a freezer for subsequent analysis (Diao et al. 2010; Yan et al. 2012).

Fig. 2
figure 2

Soil solution sampling device

Full size image

The available P in the soil solution was determined according to Watanabe and Olsen (1965). Briefly, the filtrate of the soil solution was added to a developing agent consisting of ammonium molybdate, antimony potassium tartrate, and ascorbic acid, and a colorimetric assay was conducted 30 min later.

Statistical analysis

All P data for the collected soil solutions were subjected to a normality test prior to a one-way analysis of variance (ANOVA) using IBM SPSS Statistics 21.0 (SPSS Inc., Shanghai, China). To compare the treatment mean values, Duncan’s multiple range test was used at a significance level of P < 0.05. Graphs were produced using Origin 9.0 software.

Results

Effects of continuous straw retention on the available P concentration in the soil solution

As shown in Fig. 3, the available P concentration in the soil solution gradually decreased with rice growth. From 0 to 20 days after transplanting, the available P concentration in the soil solution was high; the concentration then decreased rapidly, and at 30 days after transplanting, a gradual decrease in the available P concentration in the soil solution was observed. Compared with 2011 and 2012, the available P concentration in the soil solution decreased more slowly in 2010, but the same trend was observed for all 3 years. The dynamics of the available P concentration in the soil solution were consistent in both the SRT and SRM treatments during rice growth. However, the level of available P concentration in the soil solution was lower for the SRT treatment than the SRM treatment. In 2010, 2011, and 2012, the available P concentration in the soil solution of the SRT treatment was lower by 0.086 ± 0.006 mg L−1, 0.022 ± 0.007 mg L−1, and 0.055 ± 0.004 mg L−1, respectively, compared with the SRM treatment.

An analysis of the experimental results is shown in Fig. 3, and the regression equations for the duration after transplanting and available P concentration in the soil solution are shown in Table 1. The results demonstrate that the trends of the available P concentration in the soil solution over time were identical for the SRT and SRM treatments, whereas the available P concentration in the soil solution for the SRT treatment was lower than that of the SRM treatment.

Fig. 3
figure 3

Changes in the available P concentration in the soil solution. SRT straw retained; SRM straw removed

Full size image
Table 1 Relationship between days after transplanting and available P concentration in soil solution
Full size table

The results of experiment C (Table 2) showed that when solutions containing 5 mg L−1 or 10 mg L−1 available P were used to soak the soil for incubation, the available P levels were reduced to 2.250 mg L−1 and 3.143 mg L−1, respectively, after 1 day of incubation. With prolonged incubation time, the available P in the solution continued to decrease, and it was significantly higher in the treatment with SRT than that in SRM, indicating that the water-soluble P in the rice straw was released into the soil solution at the beginning of the trial. After 5 days of incubation, the available P content in the soil solution was significantly lower in the SRT than in the SRM treatment. This trend occurred in both of the indoor simulation experiments, which illustrated that straw retention strengthened the soil’s absorption capability for available P in the soil solution.

Table 2 Changes in the available P in soil solution (mg L−1)
Full size table

Effects of the application of fertilizer on the available P concentration in soil solution

The change in the available P concentration in soil solution under different levels of P application during rice growth is shown in Fig. 4. The available P concentration in the soil solution for the SRT and SRM treatments increased with increasing levels of P fertilizer application. The P fertilizer application increased the available P concentration in the soil solution during early rice growth, but it had little effect on the later growth stages (50 days after transplanting). The changes in the available P concentration in the soil solution followed the same trend as in experiment A.

Experiment B (Fig. 4) also revealed that under different applications of P fertilizer, the available P concentration in the soil solution was lower for the SRT treatment than for the SRM treatment. At the P0, P1, P2, P3, and P4 levels, the available P concentrations in the soil solution for the SRT treatment were lower by 0.010 ± 0.004 mg L−1, 0.005 ± 0.002 mg L−1, 0.009 ± 0.003 mg L−1, 0.015 ± 0.005 mg L−1, and 0.009 ± 0.002 mg L−1, respectively, compared with the SRM treatment, indicating that the amount of available P in the soil solution decreased with SRT treatment. The available P concentration in the soil solution gradually decreased with rice growth. However, the trend did not change with the different amounts of P application.

Fig. 4
figure 4

Changes in the available P content in the soil solution. SRT straw retained, SRM straw removed. P 0 0 kg ha−1 Ca(H2PO4)2, P 1 75 kg ha−1 Ca(H2PO4)2, P 2 150 kg ha−1 Ca(H2PO4)2, P 3 225 kg ha−1 Ca(H2PO4)2, and P 4 300 kg ha−1 Ca(H2PO4)2

Full size image

For experiment B, the regression equation for the amount of applied P fertilizer and available P content in the soil solution is shown in Table 3. The available P concentration in the soil solution was positively correlated with the amount of applied P fertilizer. The regression equation is adequate for 10 days (June 10) and 40 days (July 10) after transplanting. As shown in Table 2, the P fertilizer obviously affected the available P concentration in the soil solution 10 days after transplanting, and this effect decreased with time.

Table 3 Relationship between P fertilizer applied and available P concentration in solution
Full size table

Based on the one-dimensional linear regression equation of the available P concentration in the soil solution and the P application amount, the level of available P in the soil solution decreased by 0.021 mg L−1 on average as a result of the SRT treatment. This decline corresponds to 23.33 kg ha−1 Ca (H2PO4)2, which is equivalent to 6.21 kg ha−1 P. This amount of P had to be applied to compensate for the loss of available P resulting from the SRT treatment and achieve the same level of available P content in the soil solution and rice yield as the SRM treatment.

Effect of straw retention and P application rate on rice yield

The rice yield from experiment A is shown in Table 4. The rice yield was higher for the SRT treatment than for the SRM treatment, and the average yield in 2012 differed from the yields in 2010 and 2011. Table 5 shows the trend of the rice yield in experiment B; the rice yield in the SRT and SRM treatments increased with increasing amounts of applied P fertilizer. The rice yield was significantly lower for the SRT treatment than for the SRM treatment at low levels of P (P0 and P1). The yield was slightly higher for the SRT treatment than for the SRM treatment at the P2, P3, and P4 levels, but these differences were not significant. The results demonstrate that straw retention has a negative impact on the rice yield at low levels of P soil fertility, and thus, the amount of P fertilizer application should be adjusted to ensure rice yield.

Table 4 Effect of rice straw amendment on yield of rice (kg m−2)
Full size table
Table 5 Rice yield and phosphorous fertilizer application (g pot−1)
Full size table

Discussion

Changes in the available P concentration in the soil solution during rice growth

In the early stages of plant growth, the P supply is critical for crop production, and a deficiency in the P supply during the first 4–6 weeks of wheat growth has been found to produce significant decreases in tillering and panicle formation in the plant (Grant et al. 2001). After the application of P fertilizer to the soil, the phosphate ions in the soil solution are consumed by organisms or adsorbed and immobilized by the soil over time (Miller et al. 2011). The results of experiments A (Fig. 3) and B (Fig. 4) revealed that the available P concentration in the soil solution was high during the 0- to 20-day period after transplanting but gradually decreased during rice growth, possibly due to the absorption and immobilization of P by rice in the solid phase in the soil. These dynamic changes in the available P concentration suggest that P absorption by rice plants is sufficient during the early growth stages but becomes insufficient during the later stages of growth.

Stevenson and Cole (1999) claimed that straw retention changes the balance of P in the soil due to straw decomposition and consumption by organisms. The results from experiments A and B indicate that the SRT treatment decreased the available P concentration in the soil solution, possibly due to microbial fixation of the available P in the soil solution. The results of Gupta et al. (2007) and Phiri et al. (2001) demonstrated that multiple years of straw retention reduced the soil absorption of P and increased the release of available P in the surface soil, respectively. Yadvinder et al. (2010) also reported a gradual decrease in the relative P content of straw with increasing time after straw retention, indicating a continuous release of P from the decomposing straw. Both the field planting and the pot-planting experiments indicate that most of the P nutrients released from the decomposed straw were absorbed by the soil solid phase and did not enter the soil solution.

A portion of the P fertilizer that was applied to the crop was consumed by the plants, and the remainder entered the soil exchange sites and precipitated due to soil adsorption (Muhammad et al. 2013). In experiment B, augmentation with P fertilizer significantly increased the levels of available P in the soil solution, but the increment diminished 60 days after transplanting (July 30), indicating that the application of P fertilizer could effectively increase the available P content of the soil solution and could maintain that level for up to 60 days after transplantation (Table 3). The availability of P in the soil is a limiting factor in rice production (Dobermann et al. 1998; Pheav et al. 2003). Experiment B demonstrated that the decrease in available P concentration in the soil solution in the SRM treatment had little impact on the rice yield at high P levels but did influence the rice yield at low P levels; therefore, the applied amount of P fertilizer should be increased when straw is retained.

In the A, B, and C experiments, the available P content in the soil solution for the SRT treatment was lower than that for the SRM treatment; this result was caused by increased levels of carbon sources in the soil related to straw retention, which led to increased microbial biomass and microbial activities that promoted the conversion and retention of inorganic P in the soil (McLaughlin et al. 1988). Beri et al. (1995) also claimed that when straw with high ratios of C:P was mixed with soil, the straw enhanced P adsorption in the soil within the same growth season (McGill and Cole 1981; Murrmann and Peech 1969; Medley et al. 1982). Microbial activity has been shown to play a major role in redistributing P into different forms in the soil, and straw retention has been shown to affect the balance of P in the soil (McGill and Cole 1981; Murrmann and Peech 1969).

Effects of straw retention and P fertilizer application on rice yield

Beckwith (1965) and Fox and Kamprath (1970) determined that a P concentration of 0.2 mg L−1 in the soil solution was sufficient to provide adequate P for many crops. Mehdi et al. (2007) also observed that when the concentration of P in a soil solution reached 0.252 mg L−1, a rice yield of 95 % was obtained; beyond this concentration, increased P levels did not affect the yield of the crop, even when P was increased to 0.50 mg L−1. In experiment A in this study, the available P concentration in the soil solution 10–20 days after transplanting was higher than 0.252 mg L−1 for both the treatments. The rice yield was slightly higher in the SRT treatment than in the SRM treatment, and only the yield in 2012 was significantly different among the 3 years of experiments.

At the P0 level in experiment B, the available P concentration in the soil solution was less than 0.252 mg L−1 (Mehdi et al. 2007) throughout the entire growth stage in both the SRT and SRM treatments. At the P1 level, during the growth period 0–30 days after transplanting, the available P content in the soil solution was close to the threshold value in both the straw retention and straw removal treatments, whereas the values were higher than the threshold value at the P2, P3, and P4 levels 30 days after transplanting. The available P content in the soil solution was higher than the threshold value at 40 days after transplanting only for the P4 level.

At the P0 and P1 levels, the yield was lower in the SRT treatment than in the SRM treatment, and there was no significant difference in rice yield between the treatments at the P2, P3, and P4 levels. These results demonstrate that the straw retention reduced the available P in the soil solution in experiments A and B at the P2, P3, and P4 levels, but had no significant influence on the rice yield; thus, additional P fertilizer was not required. However, straw retention should be accompanied by an increase in P fertilizer application to achieve good yields at low P levels.

Studies have been conducted to investigate the relationship between straw retention, available P content in the soil solution and yield (Beri et al. 1995; Gupta et al. 2007; Lan et al. 2012; McLaughlin et al. 1988; Pathak et al. 2006; Yadvinder et al. 2004), and the results showed that straw retention had a significant impact on the available P content in the soil solution and rice yield, although the results were not consistent. In this study, straw retention had no significant impact on yield; however, the test results should be verified in experiments conducted over longer time periods.

Conclusions

Straw retention decreased the available P concentration in the soil solution, although it did not alter the trend of available P concentration in the soil solution during plant growth. In addition, in the 10–20-day period after transplantation, the available P concentration in the soil solution was high, although it decreased thereafter. The available P concentration in the soil solution increased with the amount of applied P fertilizer, and the rice yield also increased with increasing applications of P fertilizer. The results of experiments A and B showed that straw retention had no significant impact on the rice yield; however, years of continuous observation are required to verify these results. In this study, P (inorganic and organic form) content in the soil did not produce a significant effect on the rice yield. Application of P may be absorbed by the soil and result in increased P concentrations in the soil. Therefore, continuous experiments should be conducted at this experimental site to observe the long-term effects.