Phosphorus distribution in Chinese albic black soil affected by land use

In last decades, large area of dry-farming field has been converted to paddy field in the Sanjiang Plain of the Northeastern China. It is assumed that this land-use change might impact the level, chemical forms, and availability of phosphorus in the soils. Five soil cores were collected separately in the paddy field and dry-farming field and were analyzed for pH, soil organic matter (OM), aluminium (Al), iron (Fe), calcium (Ca), potassium (K), total phosphorus (TP), and phosphorus forms. The topsoil of the paddy filed contained lower OM, Fe, Al, organic phosphorus (Por), and residual phosphorus and higher inorganic phosphorus (Pinor) and iron oxides-bound phosphorus (Fe-P) than that of the dry-farming field. However, TP and plant-available phosphorus (AP) levels were generally not affected by the landuse change. The bulk of soil phosphorus mainly exists in Por, Fe and Al-bounded phosphorus, and Cabound phosphorus, while AP was significantly correlated to Pinor, Al-bound phosphorus (Al-P), and FeP but not correlated to Por in the both fields. Therefore, the conversion of dry-farming field to paddy field did not impact TP and AP levels, but changed the phosphorus forms in the albic black topsoil, with increase of Pinor and iron oxides-bound P and decrease of Por.


INTRODUCTION
The Sanjiang Plain is situated in the Northeastern part of the Heilongjiang province in Northeastern China.The most common landscape types are wetland and cultivated land, accounting for 0.9 and 4 million hectares, respectively, as about 27 and 30% of the provincial wetland and cultivated land, respectively (Zhang et al., 2010).The cultivated lands in the Sanjiang Plain are well known for producing large quantity and high quality of rice, soybean, and maize in recent decades in China, with large amount input of N and phosphorus fertilizer (Zheng et al., 2010).The effects of land uses or tillage practices on phosphorus geochemistry in the soils of the Sanjiang Plain are of considerable interest to soil and plant scientists worldwide, but a thorough understanding of them has remained elusive.*Corresponding author.E-mail: c.lin@bnu.edu.cn.Tel: +86 10 58802078.Fax: +86 10 5880397.
A number of studies have focused on the effects of tillage and/or land use on phosphorus content, forms, and availability in various soils.Hedley et al. (1982) investigated the changes in inorganic and organic soil phosphorus fractions induced by cultivation practices, finding that total phosphorus content of the cultivated soil was 29% lower than that of the adjacent permanent pasture and the major loss of phosphorus (74% of total phosphorus lost) was organic phosphorus and residual phosphorus.Negassa and Leinweber (2009) reviewed how the Hedley sequential fractionation reflects impacts of land use and management on soil phosphorus, showing that medium and long-term cultivation without phosphorus application depleted all phosphorus fractions, whereas most phosphorus fractions increased with continuous phosphorus application regardless of the amount and source of phosphorus.Wang et al. (2011) observed that no-till increased total and organic phosphorus, but no available phosphorus as compared to conventional tillage treatments.Pavinato et al. (2009) also found that labile phosphorus fractions in Brazilian cerrado soils were not affected by tillage system.However, Zamuner et al. (2008) investigated organic and inorganic phosphorus in mollisol soil under different tillage practices and documented that the tillage systems did not affect the total phosphorus content, but the distribution of phosphorus among factions changed between no tillage and conventional tillage.Wright (2009) determined phosphorus content and forms in the soils of cropping for 50 years and perennial pasture for 100 years, showing the phosphorus distribution varied between the two land uses.In addition, land-use effects on phosphorus fractions in Cerrado oxisols from central Brazil were investigated by Neufeldt et al. (2000), indicating that fertilization after land-use change only increased levels of inorganic phosphorus forms.
Many studies examined the relationships among phosphorus chemical forms in the native vegetation soils and arable fertilized soils in the temperate, subtropical, and tropical zones, often in connection with particular phosphorus amendments.Bray I and Pi test (a method extracting soil phosphorus with iron-oxide paper) were compared to evaluating plant-available phosphorus from soils treated with different partially acidulated phosphate rocks (Menon et al., 1989), showing Bray I solution extracted more phosphorus than Pi test.Hosseinpur and Sinegani (2009) compared Pi test with Olsen, Bray I, and Mehlich I methods, indicating that available phosphorus (AP) was significantly correlated with phosphorus extracted by Olsen method.The relationships between AP and Olsen, Bray I, Mehlich-III, and phosphorus fractions were investigated for 203 surface soils representing all soil orders, documenting that AP was closely correlated with Olsen-P, Bray I-P, and Mehlich III-P (r 2 of 0.89 to 0.93, P>0.001) (Sharpley, 1991).
However, such information on the effect of land use on phosphorus contents and forms in the albic black soil (argialbolls) is generally lacking.In addition, the relationships between AP and phosphorus chemical forms are not clear for the soil.
The objectives of this study were to compare the chemical forms of phosphorus in the paddy soil and dryfarming soil, with the hypothesis that the change of landuse change might impact the phosphorus forms in the albic black soils.In addition, the relationship between the AP extracted by iron oxide impregnated filter paper strips and phosphorus contents in various fractions was investigated.This knowledge is important to reasonably manage phosphorus in the arable soils.

Description of study area
The Sanjiang Plain, an area of low relief within the Heilongjiang province of China, is located in the temperate climate zone, characterized by a mean annual temperature of 3°C and annual precipitation ranging from 500 to 600 mm.The arable soil is Yang et al. 5811 generally classified as albic black luvisols (Argialbolls in soil taxonomy), completely frozen from late October to early April.The major soil forming material is quaternary clay sediment.Before 1975, the cultivated land was not fertilized.Between 1975 and 1985, inorganic phosphorus fertilizer was applied at a dosage of ca.36 to 136 kg phosphorus ha -1 year -1 .From 1985 onward, about 136 kg phosphorus ha -1 year was applied in the soil.Wheat, soybean, and maize were grown before 1985, while wheat, soybean, maize, and paddy rice have been grown since 1985, with gradually increase in the area of paddy field.In 2005, the harvested area of paddy rice and other crops (soybean, maize, and wheat) were about 15,330 and 10,000 ha, respectively.

Soil sample collection and analysis
Ten soil cores of 0 to 50 cm depth were collected in the northeastern cultivated land (859 Farm) of the Sanjiang Plain in May 2010 (before growing season) (5 cores from paddy field and 5 cores from dry-farming field) (Figure 1).The paddy field has a land use history of about 25 years" dry-farming (ca.1960 to 1985) and about 25 years" paddy field (ca.1985 to 2010), while the dryfarming field has been used as dry farming for wheat, maize, and soybean since about 1960.The soil cores were sliced into 10 cm slices, leading to 50 samples.The soil samples were transferred to acid-washed dark-colored polyethylene bags and were transported to the laboratory where they were freeze-dried, slightly crushed, passed through a 2 mm sieve and were stored in glass bottles.
The pH value of each soil sample was analyzed in a 1:10 solid/liquid ratio suspension using a combination pH electrode (Orion, USA).Soil organic matter (SOM) concentrations were measured by weight loss on ignition to 400°C (Ben-Dor and Banin, 1989).Portions of the soil samples were digested with HNO3-HF-HClO4, and the concentrations of aluminium (Al), iron (Fe), calcium (Ca), potassium (K), and total phosphorus (TP) in the extracts were measured with ICP-AES (IRIS Intrepid II, Thermo Electron) (Lin et al., 2008).
Organic phosphorus (Por) content was measured by the ignition method (Kuo, 1996), for which phosphorus in the ignited (550°C) and unignited soil samples were extracted by 0.5 M H2SO4.Plant AP was extracted with iron oxide-impregnated filter paper (Kuo, 1996).Selective sequential fractionation of inorganic phosphorus (Pinor) in the soil was performed, using the following stepwise extraction scheme: soluble and loosely bound phosphorus (S/L-P), Al bound phosphorus (Al-P), Fe bound phosphorus (Fe-P), reductant soluble phosphorus (RS-P), and Ca bound phosphorus (Ca-P) were selectively extracted by NH4Cl, NH4F, NaOH, Na2S2O4, and H2SO4, respectively (Kuo, 1996).Phosphorus was measured in the extracted supernatants by employing the Malachite green method (Rao et al., 1997).Residual phosphorus (RES-P) was given by the difference between the total phosphorus content (TP) and the sum of reactive (inorganic) fractions mainly consisted of organic phosphorus.Pinor was given by the difference between the TP and Por.

General characteristics of the soil cores
General chemical properties of the soil samples are as shown in Figure 2. The pH ranged from 5.4 to 6.5, indicating that the soil is slightly acidic.Average pH for the dry-farming soil cores gradually increased from 5.8 in the top horizon to 6.1 in the bottom horizon, while pH peaked at 20 to 30 cm depth (6.1) for the paddy soil cores.The pH in the topsoil (plow layer) of the paddy field was higher than that of the dry-farming field, while the pH in the subsoil (under plow layer) might be similar for the paddy field and dry-farming field.The content of SOM ranged from 2.6 to 9.7% and generally decreased with depth.In addition, the OM content in the topsoil of the dry-farming field was higher than that of the paddy field.Average contents of Al, Fe, and K were higher in the subsoil than in the topsoil, while average content of Ca generally remained stable over the depth, indicating that Al, Fe, and K accumulate in the lower horizons, while Ca dose not accumulated there due to acid condition.In addition, the Al and Fe contents in the topsoil of the paddy field were lower than that of the dry-farming field.

Contents, fractions, and availability of phosphorus in the soil cores
In general, the contents of TP, AP, Por, and Pinor decreased downward along soil profiles (Figure 3).For example, the average TP content generally decreased from about 980 mg kg -1 in the surface layer to about 550 mg kg -1 in the bottom layer.The TP and AP content profiles for the paddy field were similar to those for the dry-farming field, while the Por and Pinor content profiles were different.The topsoil of the paddy field contained lower Por and slightly higher Pinor than the topsoil of the upper field.This may be ascribed to the higher OM content in the topsoil of the dry-farming field than in the Yang et al. proportion in the subsoil of the dry-farming field.In addition, higher Por proportion and lower Pinor proportion were observed in the topsoil of dry-farming filed than in the paddy field.The AP proportion generally decreased from about 6% in the surface layer to <3% in the bottom layer for both paddy field and dry-farming field.Previous researches reported 0.4 to 4.8% of AP to TP ratio for various soils over the world, with the minimal 0.4% for the topsoil of volcanic andosol in Chile and acid ferrasol in Thailand and the maximal 4.8% for the topsoil of noncalcareous loamy sandy soils in the Netherlands (Sharpley, 1991;Aigner et al., 2002;Koopmans et al., 2004).
The changes of the S/L-P, Al-P, Fe-P, RS-P, Ca-P, and RES-P proportions out of TP are as shown in Figure 5.The S/L-P proportion was usually less than 0.6%, with high variability.The Al-P proportion profiles were similar for the paddy field and dry-farming field, with higher Al-P proportion in the topsoil than in the subsoil.The Fe-P and RS-P proportions in the topsoil of the paddy field might be slightly higher than those of the dry-farming field, but in the subsoil, they were similar.Noting that Fe content in the topsoil of the paddy field was lower than that of the dry-farming field, the higher proportions of Fe-P and RS-P in the topsoil of the paddy field might be due to the higher reactivity of iron oxides in the soil.For example, anaerobic and aerobic cycle in paddy field generally favors formation of lepidocrocite in cool-temperature climate condition (Bigham et al., 2002), thus favors formation of strengite and vivianite in the soils (Lindsay, 1979).The Ca-P proportion generally increased with depth and was similar for the both paddy field and dryfarming field.The RES-P proportion profiles were similar to Por proportion profiles, because RES-P is in essence Por.
Most (50 to 75%) TP in mineral soils is inorganic in nature and is predominantly associated with Al and Fe in acidic soils (Sims and Pierzyski, 2005).The acidic albic black soils in the Sanjiang Plain contained relatively high Por in the topsoil, due to high content of SOM in the soils.In addition, Pinor was mostly bound to Fe, due to high content of Fe in the soils, because iron oxides can absorb  phosphate anion (Zhang et al., 2011).In general, if residue phosphorus content is >30% of TP, mineralization of Por is likely (Tisdale et al., 1993).Thus, mineralization of Por in the cultivated soils of the Sanjiang Plain can be expected, because RES-P content was usually more than 45% of TP.

Relationships between AP and phosphorus chemical forms
The relationships between AP and phosphorus chemical forms are as shown in Figure 6.The results indicated that correlation coefficient for AP versus Pinor was 0.868 and 0.641 for dry-farming field and paddy field, respectively, much higher than that for AP versus Por (<0.243).Iron oxide-impregnated paper strips remove primarily physically-bound extractable phosphorus from soils (Sharpley, 1991), thus, AP is correlated with Pinor but not correlated with Por.AP in the dry-farming field was closely correlated to S/L-P, Al-P, Fe-P, and Ca-P with correlation coefficients of 0.598, 0.914, 0.661, and 0.561, respectively; while AP in the paddy field was correlated to Al-P and Fe-P with correlation coefficients of 0.792 and 0.398, respectively.These results demonstrated that Al-P and Fe-P were plant-available phosphorus for both dry-farming field and paddy field.In addition, Ca-P was also plant-available phosphorus in the dry-farming field.In the soils (Regosol, Fluvisol, Luvisol, Cambisol, Lithosol) of the Southern Latium area in Italy, however, AP content was correlated only with Al-P (Indiati and Sharpley, 1998).The reason for that difference might be that the Sanjiang Plain soil is acidic and contained high Fe-P, while the soils of Southern Latium area in Italy is alkaline and contained high Ca-P (Indiati and Sharpley, 1998).
Therefore, land-use change in the Sanjiang Plain did not impact TP and AP levels, it changed phosphorus forms in the topsoils as well as relationships among phosphorus forms in the soils.In general, the bulk of soil phosphorus exists in the three groups of compounds, namely, organic phosphorus, iron-or aluminum-bound inorganic phosphorus, and calcium-bound inorganic phosphorus in the albic black soils of the Sanjiang Plain.All the three groups of compounds slowly contribute phosphorus to the soil solution for plant use, but Al-P and Fe-P may be used as an indicator of AP, due to their high correlation to AP in the albic black soils.Future researches are needed to further clarify the reasons of changes in the phosphorus and mineralization of Por in the soils.

Conclusions
The land-use change in the Sanjiang Plain generally impact physicochemical properties and phosphorus chemical forms in the topsoil, but not in subsoil.The topsoil of the paddy field was characterized with higher pH and lower OM, Al, and Fe contents than that of the dry-farming field,whereas the land-use change did not impact the TP and AP, it changed chemical forms of phosphorus in the topsoil.The topsoil of the paddy field contained higher Pinor and iron oxides-bound phosphorus and lower Por and residual phosphorus than that of the dry-farming field.AP was significantly correlated to Pinor, but not correlated to Por.In details, AP was closely correlated to Al-P and Fe-P in both fields, inferring that iron oxide-impregnated filter paper might mainly extracted phosphorus associated with iron and aluminum oxides as part of "active" phosphorus pool in the albic black soil, whereas the three groups of Yang et al. 5817 compounds, namely, organic phosphorus, iron-or aluminum-bound inorganic phosphorus, and calciumbound inorganic phosphorus in the soils slowly contribute phosphorus to the soil solution for plant use, but Al-P and Fe-P may be used as an indicator of AP in the agricultural management of phosphorus, due to their high correlation to AP in the albic black soils.

Figure 1 .
Figure 1.Schematic graph of sampling sites in the northeastern cultivated soil (859 Farm) of the Sanjiang Plain, China (1 to 5 from dry-farming filed, 6 to 10 from paddy field).

Figure 2 .Figure 3 .
Figure 2. Change of pH and OM, Fe, Al, Ca, and K contents with depth in soil cores.

Figure 4 .Figure 5 .
Figure 4. Change of Por, Pinor, and AP proportions with depth in the soil cores.