ABSTRACT

In-depth knowledge of total soil organic phosphorus (TP_O) as a potential P source for plants allows for a comprehensive understanding of the adoption of an efficient management system of phosphorus fertilization in forest plantations. Thus, we aimed to compare three TP_O extraction methods (Hedley; Bowman; Bowman and Moir) in different Eucalyptus plantations on strongly weathered soils. The TP_O concentrations obtained by the “Hedley” (mean of 130 mg kg^–1), and “Bowman and Moir” methods (mean of 131 mg kg^–1) were similar. The “Bowman” method extracted less than 50 % of the TP_O content extracted by the other methods. Both “Hedley” and “Bowman and Moir” methods showed similar TP_O extraction efficiency compared to TP determined by acid digestion. However, the “Hedley” method is the most expensive and time-consuming analytical technique. In view of this, the NaOH+Na₂EDTA extractor (Bowman and Moir) would be the most suitable since, in addition to showing high extraction efficiency, it offers fast analysis execution, low analytical error, and lower cost.

basic EDTA extraction; forest soils; phosphorus fractionation; sequential acid-base extraction

Introduction

Soil organic phosphorus (P_O) is of fundamental importance to the maintenance of production systems. The planted forest areas in Brazil are largely located on soils with high phosphorus (P) deficiency, mainly those under Eucalyptus grandis Hill (ex Maiden) and Eucalyptus grandis × E. urophylla S.T. Blake clonal hybrid (Eucalyptus spp.) which represents 76 % of the planted forest area in Brazil (SNIF, 2019). Thus, the quantification of the P_O pool and its potential use as a source of this nutrient makes it possible to reduce the use of phosphate fertilizers, a resource that is limited, since P reserves in the soil are finite and non-renewable in the short term (Laclau et al., 2010Laclau JP, Ranger J, Gonçalves JLM, Maquère V, Krusche AV, M’Bou AT, et al. 2010. Biogeochemical cycles of nutrients in tropical Eucalyptus plantations: main features shown by intensive monitoring in Congo and Brazil. Forest Ecology and Management 259: 1771-1785. https://doi.org/10.1016/j.foreco.2009.06.010
https://doi.org/10.1016/j.foreco.2009.06... ). Soil P_O is an important reserve pool in forest systems in temperate (Prietzel et al., 2022Prietzel J, Krüger J, Kaiser K, Amelung W, Bauke SL, Dippold MA, et al. 2022. Soil phosphorus status and P nutrition strategies of European beech forests on carbonate compared to silicate parent material. Biogeochemistry 158: 39-72. https://doi.org/10.1007/s10533-021-00884-7
https://doi.org/10.1007/s10533-021-00884... ) and tropical climates (Spohn, 2020Spohn M. 2020. Phosphorus and carbon in soil particle size fractions: a synthesis. Biogeochemistry 147: 225-242. https://doi.org/10.1007/s10533-019-00633-x
https://doi.org/10.1007/s10533-019-00633... ). Given this, estimating soil P_O concentrations is relevant to P studies as P_O has high potential for supplying inorganic P (P_I) to forest cover through mineralization processes (Aleixo et al., 2019Aleixo S, Gama-Rodrigues AC, Gama-Rodrigues EF, Schripsema J. 2019. Organic phosphorus of soils under cacao agroforests in the Atlantic coast of Brazil. Geoderma Regional 17: e00220. https://doi.org/10.1016/j.geodrs.2019.e00220
https://doi.org/10.1016/j.geodrs.2019.e0... ; Spohn and Stendahl, 2021Spohn M, Stendahl J. 2021. Carbon, nitrogen, and phosphorus stoichiometry of organic matter in Swedish forest soils and its relationship with climate, tree species, and soil texture. Biogeosciences 19: 2171-2186. https://doi.org/10.5194/bg-2021-346
https://doi.org/10.5194/bg-2021-346... ). However, few studies deal with P_O extraction methods and their efficiency in estimating the real soil contents. Thus, it is necessary to improve the knowledge about the different extraction methods used to determine the total P_O (TP_O) to identify the methodology best suited to soils with high P buffering capacity.

Three methods are commonly used for TP_O determination. Hedley et al. (1992) proposed the sequential P fractionation, widely used in the most diverse studies on P transformations in different soil classes and land uses, both in temperate and tropical ecosystems (Condron and Newman, 2011Condron LM, Newman S. 2011. Revisiting the fundamentals of phosphorus fractionation of sediments and soils. Journal of Soils and Sediments 11: 830-840. https://doi.org/10.1007/s11368-011-0363-2
https://doi.org/10.1007/s11368-011-0363-... ; McDowell and Burkitt, 2022McDowell RW, Burkitt LL. 2022. In recognition of Mike Hedley: fate of fertiliser in soil and mobilisation of recalcitrant nutrients. Nutrient Cycling in Agroecosystems 124: 131-134. https://doi.org/10.1007/s10705-022-10243-z
https://doi.org/10.1007/s10705-022-10243... ). Sequential extraction allows for characterizing different P_I and P_O fractions on a scale of solubility variation based on changes in different pH levels and extracting strength, in decreasing order of soil lability, in which TP_O is the sum of fractions extracted in sodium bicarbonate (NaHCO₃), sodium hydroxide (NaOH) and NaOH+Sonification (Aleixo et al., 2017Aleixo S, Gama-Rodrigues AC, Costa MG, Sales MVS, Gama-Rodrigues EF, Marques JRB. 2017. P transformations in cacao agroforests soils in the Atlantic forest region of Bahia, Brazil. Agroforestry Systems 91: 423-437. https://doi.org/10.1007/s10457-016-9939-6
https://doi.org/10.1007/s10457-016-9939-... ; Viana et al., 2018Viana TO, Gama-Rodrigues AC, Gama-Rodrigues EF, Aleixo S, Moreira RVS, Sales MVS, et al. 2018. Phosphorus transformations in Alfisols and Ultisols under different land uses in the Atlantic Forest region of Brazil. Geoderma Regional 14: e00184. https://doi.org/10.1016/j.geodrs.2018.e00184
https://doi.org/10.1016/j.geodrs.2018.e0... ). Another method is Bowman’s sequential extraction (Bowman, 1989), considered suitable for comparing P levels in different tropical soil classes (Condron et al., 1990Condron LM, Frossard E, Tiessen H, Newman RH, Stewart JWB. 1990. Chemical nature of organic phosphorus in cultivated and uncultivated soils under different environmental conditions. European Journal of Soil Science 41: 41-50. https://doi.org/10.1111/j.1365-2389.1990.tb00043.x
https://doi.org/10.1111/j.1365-2389.1990... ), and especially the soil TP_O dynamics in pastures and forest systems (Cunha et al., 2007Cunha GM, Gama-Rodrigues AC, Costa GS, Velloso ACX. 2007. Organic phosphorus in soils under montane forests, pasture and eucalypt in the North of Rio de Janeiro State, Brazil. Revista Brasileira de Ciência do Solo 31: 667-672 (in Portuguese, with abstract in English). http://doi.org/10.1590/S0100-06832007000400007
http://doi.org/10.1590/S0100-06832007000... ; Zaia et al., 2008Zaia FC, Gama-Rodrigues AC, Gama-Rodrigues EF, Machado RCR. 2008. Organic phosphorus in soils under cocoa agroecosystems. Revista Brasileira de Ciência do Solo 32: 1987-1995 (in Portuguese, with abstract in English). http://doi.org/10.1590/S0100-06832008000500020
http://doi.org/10.1590/S0100-06832008000... , 2012Zaia FC, Gama-Rodrigues AC, Gama-Rodrigues EF, Moço MKS, Fontes AG, Machado RCR, Baligar VC. 2012. Carbon; nitrogen: organic phosphorus microbial biomass and N mineralization in soils under cacao agroforestry systems in Bahia, Brazil. Agroforestry Systems 86: 197-212. http://doi.org/10.1007/s10457-012-9550-4
http://doi.org/10.1007/s10457-012-9550-4... ; Rita et al., 2013Rita JCO, Gama-Rodrigues AC, Gama-Rodrigues EF, Zaia FC, Nunes DAD. 2013. Mineralization of organic phosphorus in soil size fractions under different vegetation covers in the north of Rio de Janeiro. Revista Brasileira de Ciência do Solo 37: 1207-1215. http://dx.doi.org/10.1590/S0100-06832013000500010
http://dx.doi.org/10.1590/S0100-06832013... ). The method proposed by Bowman and Moir (1993)Bowman RA, Moir JO. 1993. Basic EDTA as an extractant for soil organic phosphorus. Soil Science Society of America Journal 57: 1516-1518. https://doi.org/10.2136/sssaj1993.03615995005700060020x
https://doi.org/10.2136/sssaj1993.036159... was developed to obtain an easy and reproducible methodology in a single step using an alkaline source such as sodium hydroxide (NaOH) associated with Na₂EDTA. Under this methodology, P_O associated with soil organic matter is solubilized by the extractor alkaline matrix, and Na₂EDTA helps in the formation of paramagnetic ion complexes present in the soil extract [e.g., aluminium (Al), iron (Fe) and manganese (Mn)] (Turner, 2008Turner BL. 2008. Soil organic phosphorus in tropical forests: An assessment of the NaOH-EDTA extraction procedure for quantitative analysis by solution 31 P NMR spectroscopy. European Journal of Soil Science 59: 453-466. https://doi.org/10.1111/j.1365-2389.2007.00994.x
https://doi.org/10.1111/j.1365-2389.2007... ), which can be precipitated after quick centrifugation, increasing the P recovery percentage during soil extraction.

Soil P_O is not directly determined by colorimetry, being, therefore, obtained by the difference between total phosphorus (TP) and P_I contents. This implies that different methods present different results (Turner et al., 2005Turner BL, Cade-Menun BJ, Condron LM, Newman S. 2005. Extraction of soil organic phosphorus. Talanta 66: 294-306. https://doi.org/10.1016/j.talanta.2004.11.012
https://doi.org/10.1016/j.talanta.2004.1... ). Given this backdrop, this scientific note aims to compare three analytical methods of TP_O extraction to identify which method is best suited to planted Eucalyptus forests.

Materials and Methods

Sites description

The present study was carried out on Eucalyptus plantations [E. grandis, E. urophylla, E. grandis (Parb.), E. grandis (G-232), E. grandis (Suz.)] in nine sites with different soil orders, climatic conditions and management, distributed in seven municipalities in the state of São Paulo, Brazil. The sites are located at altitudes from 570 m to 889 m. Soils were classified according to the Brazilian Soil Classification System (EMBRAPA, 2013): Latossolo Vermelho-Amarelo distrófico [equivalent to Oxisol in USDA Soil Taxonomy (Soil Survey Staff, 2014Soil Survey Staff. 2014. Soil Survey Field and Laboratory Methods Manual: Version 1.0. USDA-NRCS, Washington, DC, USA (Soil Survey Investigations Report, 51).) - Itatinga (ITA) and Paraibuna (PA)], Neossolo Quartzarênico [Entisol – Angatuba (ANG) and Botucatu (BOT)], Latossolo Vermelho distrófico [Oxisol – Agudos (AG) and Capão Bonito 2 (CB2)], Latossolo Amarelo distrófico [Oxisol – Capão Bonito 1 (CB1)], Argissolo Vermelho-Amarelo distrófico [Ultisol – Votorantim (VOT)] and Cambissolo Háplico distrófico [Inceptisol – Capão Bonito 3(CB3)]. The municipality’s climate where the soils were collected was considered Cwa, according to the Köppen classification (Costa et al., 2016Costa MG, Gama-Rodrigues AC, Gonçalves JLM, Gama-Rodrigues EF, Sales MVS, Aleixo S. 2016. Labile and non-labile fractions of phosphorus and its transformations in soil under eucalyptus plantations, Brazil. Forests 7: 15. https://doi.org/10.3390/f7010015
https://doi.org/10.3390/f7010015... ), except for the AG site, which was characterized as Aw. The site’s ages ranged from 1.4 to 11 years with a spacing of 3.0 m × 2.0 m, except for the CB1 (3.0 m × 3.0 m spacing), and the PA (3.0 m × 2.5 m spacing) sites. In all sites, ten simple soil samples were collected in each experimental plot (three plots per site) at a depth of 0-20 cm in diagonal transect to the useful plot areas between the planting lines. The single samples gave rise to a sample composed of a plot which was air-dried, homogenized, crushed and passed through a 2.0 mm sieve. These samples were used for physical-chemical characterization; a detailed description is in the Table 1.

Determination methods of total soil organic P

Hedley’s sequential fractionation (Hedley et al., 1982)

Total organic P fraction (TP_O) values were the sum of the concentrations of the three P_O fractions extracted sequentially in NaHCO₃, NaOH and NaOH+Sonification. P_O was calculated by the difference between TP and P_I concentrations in each fraction extract. The extracted TP value was the sum of all P_O, P_I and residual-P fractions. The analytical procedure is detailed by Costa et al. (2016)Costa MG, Gama-Rodrigues AC, Gonçalves JLM, Gama-Rodrigues EF, Sales MVS, Aleixo S. 2016. Labile and non-labile fractions of phosphorus and its transformations in soil under eucalyptus plantations, Brazil. Forests 7: 15. https://doi.org/10.3390/f7010015
https://doi.org/10.3390/f7010015... , who studied the same Eucalyptus sites as the present study.

Bowman’s sequential acid-base extraction (Bowman, 1989)

Total P_O was determined in a sequential extraction with concentrated sulfuric acid and dilute base. For the determination of TP_O, 3.0 mL of 18 mol L¹ H₂SO₄ were added to falcon tubes containing 2.0 g of soil samples (in triplicate). The tube was manually shaken, then subjected to the addition of 4.0 mL of distilled water in 1.0 mL aliquots at a time, followed by manual shaking again, a further addition of 43 mL of distilled water and the tube was shaken again. Subsequently, the tubes were centrifuged at 83.3 Hz for 10 min. The supernatant was filtered with N° 42 Whatman filter paper and transferred to a clean vial. The filter paper used to filter the acid extracts was removed and returned to the same centrifuge tube containing the remaining soil and 40 mL of 0.5 mol L¹ NaOH was added. After adding NaOH, the suspension was taken to the end-over-end shaker at 2 Hz for 30 min. The centrifuge tube was placed in a water bath at 80 °C for 2.0 h. After this period, the centrifuge tube was cooled in running water, then centrifuged and filtered as in the acid extraction procedure.

A 5.0 mL aliquot of each of the acid and alkaline extracts was removed for TP determination with the addition of 1.0 mL of MgCl₂ (at saturation point) plus 1.0 mL of 69 % HClO₄ (mL 100 mL¹) in acid extracts and only 1.0 mL of HClO₄ at 69 % (mol 100 mL¹) in basic extracts. Digestion tubes were manually shaken and then placed in the digester block at 80 °C, with a slow temperature increase to 180 °C. Digestion was terminated when a colorless gel formed at the bottom of the tube. After tube cooling, 5.0 mL of distilled water was added and an aliquot was removed for the quantification of TP. P_I was determined using 20 mL of acid and alkaline extracts after clarification by filtration in 0.5 cm³ of activated carbon (Guerra et al., 1996Guerra JGM, Almeida DL, Santos GA, Fernandes MS. 1996. Organic phosphorus content in soil samples. Pesquisa Agropecuária Brasileira 31: 291-299 (in Portuguese, with abstract in English).). Due to the occurrence of P_I in the filter element (activated carbon), the purification was carried out with successive washings using solutions of HCl 6.0 mol L¹(acid), NaHCO₃ 0.5 mol L¹ and NaOH 0.5 mol L¹ (alkaline). P concentrations in clarified acid extracts were determined at 880 nm following the method of Murphy and Riley (1962)Murphy J, Riley JP. 1962. A modified single solution method for the determination of phosphate in natural waters. Analytica Chimica Acta 27: 31-36. https://doi.org/10.1016/S0003-2670 (00)88444-5
https://doi.org/10.1016/S0003-2670 (00)8... . P concentrations in clarified alkaline extracts were determined at 700 nm following the Dick and Tabatabai method (1977). All extracts from the steps were analyzed by colorimetry for P reactive to ammonium molybdate in SPECORD 210 PLUS. Total P_O (TP_O) concentrations were calculated by summing the differences in concentrations between TP and P_I of all acid and alkaline extracts as follows:

Extraction with NaOH+Na2EDTA by Bowman and Moir (Bowman and Moir, 1993)

Total P_O (TP_O) was determined by extraction with modified NaOH+Na₂EDTA. For this, 1.0 g of soil was used to which 10 mL of solution with 0.25 mol L¹ NaOH plus 0.05 mol L¹ Na₂EDTA was added, followed by end-over-end stirring at 2 Hz for 10 h (25 °C) in 15 mL falcon tubes. Next, the tubes were centrifuged at 83.3 Hz for 15 min, and the supernatant was filtered through 45 μm cellulose membranes (75-80 % nitrate and acetate) and P_I was determined. An aliquot (3.0 mL) of each NaOH+Na₂EDTA soil extract for total P digestion was autoclaved at 103 kPa and 121 °C for 3.0 h with a solution containing 10 mL of (NH₄)₂S₂O₈ at 7.5 % (g 100 mL¹) plus 1.0 mL H₂SO₄ at 24 mol L¹.TP in the extracts was determined at 880 nm by the method of Murphy and Riley (1962)Murphy J, Riley JP. 1962. A modified single solution method for the determination of phosphate in natural waters. Analytica Chimica Acta 27: 31-36. https://doi.org/10.1016/S0003-2670 (00)88444-5
https://doi.org/10.1016/S0003-2670 (00)8... . The P_I in extracts was determined at 700 nm by the method of Dick and Tabatabai (1977)Dick WA, Tabatabai MA. 1977. Determination of orthophosphate in aqueous solutions containing labile organic and inorganic phosphorus compounds. Journal of Environmental Quality 6: 82-85. https://doi.org/10.2134/jeq1977.00472425000600010018x
https://doi.org/10.2134/jeq1977.00472425... . All extracts from the steps were analyzed by colorimetry for P reactive to ammonium molybdate in SPECORD 210 PLUS. The difference between TP and P_I concentrations in each NaOH+Na₂EDTA extract calculated organic P (PO).

Determination of total soil P by sulfuric and nitric-perchloric digestions

Total soil P by sulfuric digestion (Hedley et al., 1982Hedley MJ, Stewart JWB, Chauhan BS. 1982. Changes in inorganic and organic soil phosphorus fractions induced by cultivation practices and by laboratory incubations. Soil Science Society of America Journal 46: 970-976. https://doi.org/10.2136/sssaj1982.03615995004600050017x
https://doi.org/10.2136/sssaj1982.036159... ) was determined using 0.1 g of soil macerated in a digestion tube and 1.0 mL of MgCl₂ (at saturation point) plus 2.0 mL of H₂SO₄ at 49 % (mL 100 mL¹) were added. The tubes were taken to the digester block by heating them at 200 °C for 1.0 h, and after cooling 2.0 mL H₂O₂ at 30 % (mL 100 mL¹) was added and heated in the digester block at 100 °C for another 1 h. After cooling, the volume was adjusted to 50 mL with deionized water. TP by nitric-perchloric digestion, described by Guerra et al. (1996)Guerra JGM, Almeida DL, Santos GA, Fernandes MS. 1996. Organic phosphorus content in soil samples. Pesquisa Agropecuária Brasileira 31: 291-299 (in Portuguese, with abstract in English)., was determined using 0.2 g of soil macerated in a digestion tube and 3.0 mL of HNO₃ and HClO₄ in a 2:1 ratio (mixture containing 120 mL of HNO₃ and 60 mL of HClO₄) was added. The tubes were taken to the digester block by slowly heating them to 160 °C, remaining for ~40 min (volume reduced to half). After this period, the temperature was raised to 210 °C to obtain a clear solution, which gave off dense vapors of HClO₄ (~20 min). After cooling, 50 mL of distilled water was added and the sample was transferred to a 250 mL volumetric flask. The extracts were filtered through n° 42 Whatman filter paper and stored for TP determination. Total P (TP) was determined by the colorimetric method after pH adjustment, using p-nitrophenol as an indicator. All extracts were analyzed by colorimetry for P reactive to ammonium molybdate by spectrophotometry at 880 nm (Murphy and Riley, 1962Murphy J, Riley JP. 1962. A modified single solution method for the determination of phosphate in natural waters. Analytica Chimica Acta 27: 31-36. https://doi.org/10.1016/S0003-2670 (00)88444-5
https://doi.org/10.1016/S0003-2670 (00)8... ) in SPECORD 210 PLUS. The TP concentrations from both methods were used to estimate the TP_O recovery capacity following the Hedley, Bowman, and Bowman and Moir extraction methods [(TP_O extracted × 100) / TP by digestion].

Statistical analysis

Each Eucalyptus site was considered a pseudo-replication. The data were submitted to the Lilliefors and Kolmogorov-Smirnov normality test, showing a normal distribution of the analyzed variables. Subsequently, the data were submitted to descriptive analysis and test of means (Tukey) at 5 % probability. The relationships between the three extraction methods of TP_O and TP; between extracted TP and digested TP; between TP_O and clay content; and, TP_O recovery values and clay contents were analyzed using Pearson’s correlation coefficient. Data were analyzed using the R 3.2.1 program (R Core Team, 2016) with the ISwR package (Dalgaard, 2008Dalgaard P. 2008. The R Environment: In Introductory Statistics with R. Statistics and Computing. Springer, New York, NY, USA.) and ExpDes.pt (Ferreira et al., 2014Ferreira EB, Cavalcanti PP, Nogueira DA. 2014. ExpDes: An R package for ANOVA and experimental designs. Applied Mathematics 5: 2952-2958. https://doi.org/10.4236/am.2014.519280
https://doi.org/10.4236/am.2014.519280... ). For the construction of the graphs, the Sigmaplot 12.0 program was used.

Results

The TP_O levels by the three extraction methods evaluated are shown in Table 2. The TP_O contents extracted by the “Hedley” fractionation and by the “Bowman and Moir” extraction (NaOH+Na₂EDTA) were much higher (~63 %) than the contents determined by the “Bowman” method. The mean values of TP_O contents between the “Hedley” and “Bowman and Moir” methods were similar, ranging between 58 and 273 mg kg¹ (Hedley) and 80 and 214 mg kg¹ (Bowman and Moir). While the TP_O extracted by “Bowman” quantified, on average, less than 50 mg kg¹ for almost all forest sites (Table 2; Figure 1). On the other hand, there were no significant differences between the three extraction methods for the inorganic P fraction (Table 2; Figure 1). The highest levels of TP extracted were obtained by the Hedley method, followed by sulfuric digestion (H₂SO₄+H₂O₂+MgCl₂), “Bowman”, “Bowman and Moir” methods (NaOH+Na₂EDTA) and nitric-perchloric digestion (HNO₃+HClO₄) (Table 2; Figure 1). The difference between the TP and the sum of TP_I and TP_O extracted by the “Hedley” method corresponds to the residual P content, a P fraction considered recalcitrant.

Significant positive correlations existed between the three TP_O extraction methods (Figure 2). The average TP_O recovery rate was 34 % for both the “Hedley” method, and the “Bowman and Moir” method (NaOH+Na₂EDTA), and 11 % for the “Bowman” method compared to the PT by sulfuric digestion (Figure 3). On the other hand, when compared to TP by nitric-perchloric digestion, the mean recovery rate of TP_O was 56 % for the “Hedley” method, 56 % for the “Bowman and Moir” method (NaOH+Na₂EDTA), and 18 % for the “Bowman” method (Figure 3). The three extraction methods evaluated were significantly correlated with each other for TP (r = 0.97, p < 0.001, n = 9 for “Hedley vs. Bowman and Moir”; r = 0.91, p < 0.001, n = 9 for “Hedley vs. Bowman”; and r = 0.90, p < 0.001, n = 9 for “Bowman vs. Bowman and Moir”). These extracted TP contents also correlated significantly only with the TP contents determined by nitric-perchloric digestion (r = 0.98, p < 0.001, n = 9 for “Hedley”; r = 0.97, p < 0.001, n = 9 for “Bowman and Moir” and r = 0.96, p < 0.001, n = 9 for “Bowman”). There was no significant correlation between the two methods of TP determination by acid digestion. For TP_I, the significant correlations between the “Bowman and Moir” and “Bowman” methods (r = 0.87, p < 0.01, n = 9), and between the “Bowman” and “Hedley” methods (r = 0.83, p < 0.01, n = 9).

For TP_I, the recovery rates of the three extraction methods were similar to each other for both sulfuric digestion (on average 25 %) and nitric-perchloric digestion (on average 44 %) (Figure 4). The recovery rates of TP extracted by the methods of “Hedley”, “Bowman and Moir”, and “Bowman” were 105 %, 59 % and 39 %, respectively, compared to sulfuric digestion. In turn, the values of recovery rates of TP extracted by the three methods evaluated compared to nitric-perchloric digestion were higher than by sulfuric digestion (Figure 5). The meaning and values of the correlations between TP_O recovery rates and clay contents differed between sulfuric and nitric-perchloric digestion methods (Table 3). The nitric-perchloric digestion method had significant negative correlations for the “Hedley” and “Bowman and Moir” methods. On the other hand, in the sulfuric digestion method, there was only significant positive correlation for the “Bowman” method. The values and direction of correlations for TP_I and TP also varied between extraction and digestion methods. Total organic P (TP_O) contents varied positively and significantly with clay contents for the three extraction methods (Figure 6).Similar results occurred for extracted TP_I [“Hedley” (r = 0.88, p < 0.01, n = 9), “Bowman” (r = 0.96, p < 0.001, n = 9), “Bowman and Moir” (r = 0.78, p < 0.05, n = 9)] and TP [“Hedley” (r = 0.89, p < 0.01, n = 9), “Bowman” (r = 0.95, p < 0.001, n = 9), “Bowman and Moir” (r = 0.83, p < 0.01, n = 9)]. Significant correlations between the clay contents and sulfuric digestion (r = 0.69, p < 0.05, n = 9), as well as the nitric-perchloric digestion (r = 0.93, p < 0.001, n = 9) also varied positively.

Discussion

Residual-P values (Hedley’s method), which were not added to P_O and P_I totals, presented an overall average of 41.5 % of the soil total phosphorus. Studies carried out in tropical soils by Araújo et al. (2004)Araújo MSB, Schaefer CER, Sampaio EVSB. 2004. Soil phosphorus fractions from toposequences of semi-arid Latosols and Luvisols in northeastern Brazil. Geoderma 119: 309-321. https://doi.org/10.1016/j.geoderma.2003.07.002
https://doi.org/10.1016/j.geoderma.2003.... ; Gama-Rodrigues et al. (2014)Gama-Rodrigues AC, Sales MVS, Silva PSD, Comerford NB, Cropper WP, Gama-Rodrigues EF. 2014. An exploratory analysis of phosphorus transformations in tropical soils using structural equation modeling. Biogeochemistry 118: 453-469. https://doi.org/10.1007/s10533-013-9946-x
https://doi.org/10.1007/s10533-013-9946-... ; Viana et al. (2018)Viana TO, Gama-Rodrigues AC, Gama-Rodrigues EF, Aleixo S, Moreira RVS, Sales MVS, et al. 2018. Phosphorus transformations in Alfisols and Ultisols under different land uses in the Atlantic Forest region of Brazil. Geoderma Regional 14: e00184. https://doi.org/10.1016/j.geodrs.2018.e00184
https://doi.org/10.1016/j.geodrs.2018.e0... ; Soltangheisi et al. (2021)Soltangheisi A, Haygarth PM, Pavinato PS, Cherubin MR, Teles APB, Bordonal RO, et al. 2021. Long term sugarcane straw removal affects soil phosphorus dynamics. Soil Tillage Research 208: 104898. https://doi.org/10.1016/j.still.2020.104898
https://doi.org/10.1016/j.still.2020.104... found similar values. As they do not have clear limits between the P compartments in the soil, the exact quantification of TP_O does not always have the desired precision. Successive extractions may overestimate TP levels compared to single-step extractions. Viana et al. (2018)Viana TO, Gama-Rodrigues AC, Gama-Rodrigues EF, Aleixo S, Moreira RVS, Sales MVS, et al. 2018. Phosphorus transformations in Alfisols and Ultisols under different land uses in the Atlantic Forest region of Brazil. Geoderma Regional 14: e00184. https://doi.org/10.1016/j.geodrs.2018.e00184
https://doi.org/10.1016/j.geodrs.2018.e0... showed that concentrations of TP using the “Hedley” fractionation technique can be higher than those found with a single-step extraction such as sulfuric digestion. Among those responsible for this high extraction capacity is the use of an anion exchange resin (RTA) in its initial phase that simulates the extraction of readily available P (solution-P) by the plant root system in situ (Hedley et al., 1982Hedley MJ, Stewart JWB, Chauhan BS. 1982. Changes in inorganic and organic soil phosphorus fractions induced by cultivation practices and by laboratory incubations. Soil Science Society of America Journal 46: 970-976. https://doi.org/10.2136/sssaj1982.03615995004600050017x
https://doi.org/10.2136/sssaj1982.036159... ), and combined with the fact of obtaining P_O extracted by NaHCO₃, NaOH and the P_O resulting from the Sonification (Costa et al., 2016Costa MG, Gama-Rodrigues AC, Gonçalves JLM, Gama-Rodrigues EF, Sales MVS, Aleixo S. 2016. Labile and non-labile fractions of phosphorus and its transformations in soil under eucalyptus plantations, Brazil. Forests 7: 15. https://doi.org/10.3390/f7010015
https://doi.org/10.3390/f7010015... ), where strong acids and bases can induce P_O hydrolysis (Turner et al., 2005Turner BL, Cade-Menun BJ, Condron LM, Newman S. 2005. Extraction of soil organic phosphorus. Talanta 66: 294-306. https://doi.org/10.1016/j.talanta.2004.11.012
https://doi.org/10.1016/j.talanta.2004.1... ).

Thus, during the sequential extraction, the different extractors levels act at different P adsorption sites. For example, in the step of the P_I and P_O fractions at NaOH extracts plus sonication, the rupture of aggregates occurs, making new available P sites, followed by the Residual-P fraction determination on H₂O₄+H₂O₂+MgCl₂ digestion, where there would thus be a greater amount of TP at the end of the analytical run compared to that measured in the incomplete sulfuric digestion in a single step of the entire volume of soil sampled (Aleixo et al., 2017Aleixo S, Gama-Rodrigues AC, Costa MG, Sales MVS, Gama-Rodrigues EF, Marques JRB. 2017. P transformations in cacao agroforests soils in the Atlantic forest region of Bahia, Brazil. Agroforestry Systems 91: 423-437. https://doi.org/10.1007/s10457-016-9939-6
https://doi.org/10.1007/s10457-016-9939-... ).

The highest PT values extracted by the “Bowman and Moir” method, comparing directly with the “Bowman” methodology are also shown in Table 2. Among the few studies available comparing TP_O methods in tropical soils, Condron et al. (1990)Condron LM, Frossard E, Tiessen H, Newman RH, Stewart JWB. 1990. Chemical nature of organic phosphorus in cultivated and uncultivated soils under different environmental conditions. European Journal of Soil Science 41: 41-50. https://doi.org/10.1111/j.1365-2389.1990.tb00043.x
https://doi.org/10.1111/j.1365-2389.1990... analyzed 23 soils, 20 of which were Brazilian. They obtained differences in P_O determinations between the tested methods, concluding that using an alkaline substance, such as NaOH, allows the P_O fractions to be quickly accessed in strongly weathered tropical soils. The superior extraction efficiency by adding EDTA, mainly in soils with higher organic matter contents, was reported by Bowman and Moir (1993)Bowman RA, Moir JO. 1993. Basic EDTA as an extractant for soil organic phosphorus. Soil Science Society of America Journal 57: 1516-1518. https://doi.org/10.2136/sssaj1993.03615995005700060020x
https://doi.org/10.2136/sssaj1993.036159... . This differs from the method initially proposed by “Bowman”, where only the inorganic orthophosphate was measured by colorimetry.

The alkaline solutions (e.g. those containing NaOH preferentially) are highly efficient in obtaining P_O from the soil (Turner et al., 2005Turner BL, Cade-Menun BJ, Condron LM, Newman S. 2005. Extraction of soil organic phosphorus. Talanta 66: 294-306. https://doi.org/10.1016/j.talanta.2004.11.012
https://doi.org/10.1016/j.talanta.2004.1... ). As observed, the “Bowman” and “Bowman and Moir” methods are positively correlated in all evaluated methodologies (Figure 2). This is explained by the fact that extraction with the addition of Na₂EDTA was a proposal initiated to facilitate and expand the extraction of TP_O, since certain methods proposed to evaluate soil P_O compartments are poorly reproducible and require multiple steps, which increases the possibility of analytical errors in the determinations. Thus, the inclusion of NaOH+Na₂EDTA helps in the complexing of the metals of the colloid functional group through the complexing of cations (i.e., Fe and Al paramagnetic ions), and NaOH promotes the cations hydrolysis with P being released from the soil compound (i.e., due to the addition of the hydroxyl anion); thus P_O associated with soil organic matter (SOM) is available for detection in these extracts (Bowman and Moir, 1993Bowman RA, Moir JO. 1993. Basic EDTA as an extractant for soil organic phosphorus. Soil Science Society of America Journal 57: 1516-1518. https://doi.org/10.2136/sssaj1993.03615995005700060020x
https://doi.org/10.2136/sssaj1993.036159... ; Turner et al., 2005Turner BL, Cade-Menun BJ, Condron LM, Newman S. 2005. Extraction of soil organic phosphorus. Talanta 66: 294-306. https://doi.org/10.1016/j.talanta.2004.11.012
https://doi.org/10.1016/j.talanta.2004.1... ). Furthermore, the “Bowman and Moir” method presented values (extraction efficiency and soil TP_O recovery) statistically similar to the “Hedley” fractionation.

These analyzed soils present distinct physical and chemical properties with textures varying from medium to very clayey, being characterized as very acidic and low fertility in all sites (Table 1). In this study, it was observed that sulfuric digestion presented a significant positive correlation with the evaluated methods, in contrast to nitric-perchloric digestion, which presented negative correlation with the methods compared to their clay contents. It is believed that this fact was due to incomplete and/or inefficient digestion using nitric acid, not dissolving the total P content in the soil. A significant limitation associated with acid extractions is the presence of chemically functional P_I species that are not detected by colorimetric methods (e.g., P complexes can form with ammonium molybdate), which can affect the actual concentration of TP_O form present in soils (Turner et al., 2005Turner BL, Cade-Menun BJ, Condron LM, Newman S. 2005. Extraction of soil organic phosphorus. Talanta 66: 294-306. https://doi.org/10.1016/j.talanta.2004.11.012
https://doi.org/10.1016/j.talanta.2004.1... ). On the other hand, the option of sulfuric digestion to extract TP from the soil can promote an acidic dissolution, where the extracting solution containing H₂SO₄ partially breaks the soil inorganic colloids, which allows access to the P_O and P_I forms that were protected (Aleixo et al., 2017Aleixo S, Gama-Rodrigues AC, Costa MG, Sales MVS, Gama-Rodrigues EF, Marques JRB. 2017. P transformations in cacao agroforests soils in the Atlantic forest region of Bahia, Brazil. Agroforestry Systems 91: 423-437. https://doi.org/10.1007/s10457-016-9939-6
https://doi.org/10.1007/s10457-016-9939-... ). Levels of TP obtained by sulfuric digestion can be very similar to total phosphorus in different soil classes (i.e., Inceptisol and Ultisol), where the extractions step with NaOH plus Sonication was omitted (Szott and Melendez, 2001Szott LT, Melendez G. 2001. Phosphorus availability under annual cropping: alley cropping and multistrata agroforestry systems. Agroforestry Systems 53: 125-132. https://doi.org/10.1023/A:1013316318380
https://doi.org/10.1023/A:1013316318380... ). Thus, due to its superiority in the P content digesting, it is expected that its correlation with the other methods will be positive.

In all evaluated sites, soil P recovery rates compared to sulfuric digestion showed a ~40 % of the recovery using the “Bowman” method, which is within the values observed in the literature, which ranges from 30 to 169 % (Condron et al., 1990Condron LM, Frossard E, Tiessen H, Newman RH, Stewart JWB. 1990. Chemical nature of organic phosphorus in cultivated and uncultivated soils under different environmental conditions. European Journal of Soil Science 41: 41-50. https://doi.org/10.1111/j.1365-2389.1990.tb00043.x
https://doi.org/10.1111/j.1365-2389.1990... ; Guerra et al., 1996Guerra JGM, Almeida DL, Santos GA, Fernandes MS. 1996. Organic phosphorus content in soil samples. Pesquisa Agropecuária Brasileira 31: 291-299 (in Portuguese, with abstract in English).; Cunha et al., 2007Cunha GM, Gama-Rodrigues AC, Costa GS, Velloso ACX. 2007. Organic phosphorus in soils under montane forests, pasture and eucalypt in the North of Rio de Janeiro State, Brazil. Revista Brasileira de Ciência do Solo 31: 667-672 (in Portuguese, with abstract in English). http://doi.org/10.1590/S0100-06832007000400007
http://doi.org/10.1590/S0100-06832007000... ; Zaia et al., 2008Zaia FC, Gama-Rodrigues AC, Gama-Rodrigues EF, Machado RCR. 2008. Organic phosphorus in soils under cocoa agroecosystems. Revista Brasileira de Ciência do Solo 32: 1987-1995 (in Portuguese, with abstract in English). http://doi.org/10.1590/S0100-06832008000500020
http://doi.org/10.1590/S0100-06832008000... ; Oliveira et al., 2014Oliveira RIB, Gama-Rodrigues AC, Gama-Rodrigues EF, Zaia FC, Pereira MG, Fontana A. 2014. Organic phosphorus in diagnostic surface horizons of different Brazilian soil orders. Revista Brasileira de Ciência do Solo 38: 1411-1420. https://doi.org/10.1590/S0100-06832014000500006
https://doi.org/10.1590/S0100-0683201400... P_I levels predominate in studies with strongly weathered tropical soils compared to P_O forms using the “Bowman” method (Guerra et al., 1996Guerra JGM, Almeida DL, Santos GA, Fernandes MS. 1996. Organic phosphorus content in soil samples. Pesquisa Agropecuária Brasileira 31: 291-299 (in Portuguese, with abstract in English).; Rita et al., 2013Rita JCO, Gama-Rodrigues AC, Gama-Rodrigues EF, Zaia FC, Nunes DAD. 2013. Mineralization of organic phosphorus in soil size fractions under different vegetation covers in the north of Rio de Janeiro. Revista Brasileira de Ciência do Solo 37: 1207-1215. http://dx.doi.org/10.1590/S0100-06832013000500010
http://dx.doi.org/10.1590/S0100-06832013... ; Oliveira et al., 2014Oliveira RIB, Gama-Rodrigues AC, Gama-Rodrigues EF, Zaia FC, Pereira MG, Fontana A. 2014. Organic phosphorus in diagnostic surface horizons of different Brazilian soil orders. Revista Brasileira de Ciência do Solo 38: 1411-1420. https://doi.org/10.1590/S0100-06832014000500006
https://doi.org/10.1590/S0100-0683201400... ). Thus, P_O values with this method showed relatively low responsive values (on average 27 %) compared to TP. Similar results were observed by Rita et al. (2013)Rita JCO, Gama-Rodrigues AC, Gama-Rodrigues EF, Zaia FC, Nunes DAD. 2013. Mineralization of organic phosphorus in soil size fractions under different vegetation covers in the north of Rio de Janeiro. Revista Brasileira de Ciência do Solo 37: 1207-1215. http://dx.doi.org/10.1590/S0100-06832013000500010
http://dx.doi.org/10.1590/S0100-06832013... , who found P_O levels between 11 % and 32.5 %, and Oliveira et al. (2014)Oliveira RIB, Gama-Rodrigues AC, Gama-Rodrigues EF, Zaia FC, Pereira MG, Fontana A. 2014. Organic phosphorus in diagnostic surface horizons of different Brazilian soil orders. Revista Brasileira de Ciência do Solo 38: 1411-1420. https://doi.org/10.1590/S0100-06832014000500006
https://doi.org/10.1590/S0100-0683201400... , with contents ranging from 36 to 46 %.

Although the different methods display significant efficiency in the P extraction, the soil nature must be considered as it can influence the percentage of recovery in the P compartments (Turner et al., 2005Turner BL, Cade-Menun BJ, Condron LM, Newman S. 2005. Extraction of soil organic phosphorus. Talanta 66: 294-306. https://doi.org/10.1016/j.talanta.2004.11.012
https://doi.org/10.1016/j.talanta.2004.1... ). Therefore, the choice between the methods of “Hedley”, “Bowman”, “Bowman and Moir”, sulfuric and/or nitric-perchloric digestion must be made based on the observation of two soil characteristics under study: (i) chemical solubility of the P forms present in SOM, which may have different quality depending on the vegetation cover inputs. The strong soil acidity present in tropical ecosystems could restrict the SOM decomposition (Aleixo et al., 2017Aleixo S, Gama-Rodrigues AC, Costa MG, Sales MVS, Gama-Rodrigues EF, Marques JRB. 2017. P transformations in cacao agroforests soils in the Atlantic forest region of Bahia, Brazil. Agroforestry Systems 91: 423-437. https://doi.org/10.1007/s10457-016-9939-6
https://doi.org/10.1007/s10457-016-9939-... ), and counterbalance the conditions of high humidity and temperature that favor this process, or even that a high continuous input of senescent plant material can strongly contribute to the maintenance of P_O stocks in the soil (Fontes et al., 2014; Lima et al., 2023Lima M, Vicente LC, Gama-Rodrigues EF, Gama-Rodrigues AC, Lisbôa FM, Aleixo S. 2023. Carbon functional groups of leaf litter in cacao and rubber agroforestry systems in Southern Bahia, Brazil. Agroforestry Systems 97: 249-260. https://doi.org/10.1007/s10457-023-00802-w
https://doi.org/10.1007/s10457-023-00802... ); (ii) and characteristics of different clay fraction contents in different soil orders (Turner et al., 2005Turner BL, Cade-Menun BJ, Condron LM, Newman S. 2005. Extraction of soil organic phosphorus. Talanta 66: 294-306. https://doi.org/10.1016/j.talanta.2004.11.012
https://doi.org/10.1016/j.talanta.2004.1... ; Costa et al., 2016Costa MG, Gama-Rodrigues AC, Gonçalves JLM, Gama-Rodrigues EF, Sales MVS, Aleixo S. 2016. Labile and non-labile fractions of phosphorus and its transformations in soil under eucalyptus plantations, Brazil. Forests 7: 15. https://doi.org/10.3390/f7010015
https://doi.org/10.3390/f7010015... ; Viana et al., 2018Viana TO, Gama-Rodrigues AC, Gama-Rodrigues EF, Aleixo S, Moreira RVS, Sales MVS, et al. 2018. Phosphorus transformations in Alfisols and Ultisols under different land uses in the Atlantic Forest region of Brazil. Geoderma Regional 14: e00184. https://doi.org/10.1016/j.geodrs.2018.e00184
https://doi.org/10.1016/j.geodrs.2018.e0... ).

All soil P extraction methods, including its P_O and P_I forms, obtained significant correlations with clay contents in the present study. The methodology proposed by Hedley stands out in relation to the correlations found, even with a small variation in its analytical course. In this sense, the search to evaluate the sensitivity between the TP_O extraction methods (as well as TP_I and TP) showed that the different sites with different soil classes and clay contents have a strong relationship associated with the efficiency of the extractors studied since the soil clay contents can interfere with the P solubilization in the soil and its extraction (Corrêa et al., 2005Corrêa RM, Nascimento CWA, Sá Souza SK, Freire FJ, Silva GB. 2005. Gafsa rock phosphate and triple superphosphate for dry matter production and P uptake by corn. Scientia Agricola 62: 159-164. https://doi.org/10.1590/S0103-90162005000200011
https://doi.org/10.1590/S0103-9016200500... ). Studies considering P molybdate reactive showed that extractors present susceptible differences compared to soil clay contents (Cajuste and Kussow, 1974Cajuste LJ, Kussow WR. 1974. Use and limitations of the North Carolina method to predict available phosphorus in some Oxisols. Tropical Agriculture 51: 246-252.; Lins and Cox, 1989Lins IDG, Cox FR. 1989. Effect of extractant and selected soil properties on predicting the correct phosphorus fertilization of soybean. Soil Science Society of America Journal 53: 813-816. https://doi.org/10.2136/sssaj1989.03615995005300030031x
https://doi.org/10.2136/sssaj1989.036159... ). Soils with high clay contents tend to strongly adsorb P, especially those with high contents of Fe and Al oxyhydroxides (Novais and Smyth, 1999Novais RF, Smyth TJ. 1999. Phosphorus in Soil and Plant in Tropical Conditions = Fósforo em Solo e Planta em Condições Tropicais. Universidade Federal de Viçosa, Viçosa, MG, Brazil (in Portuguese).). Positive correlations between labile P_O (i.e., organic phosphorus forms retained with lower energy in the soil) and the soil clay fraction suggest that soil texture may be closely linked to stabilization and increased P_O contents in the soil minerals (Oliveira et al., 2014Oliveira RIB, Gama-Rodrigues AC, Gama-Rodrigues EF, Zaia FC, Pereira MG, Fontana A. 2014. Organic phosphorus in diagnostic surface horizons of different Brazilian soil orders. Revista Brasileira de Ciência do Solo 38: 1411-1420. https://doi.org/10.1590/S0100-06832014000500006
https://doi.org/10.1590/S0100-0683201400... ).

Given this background, modern techniques for analyzing the composition of chemically functional soil P_O forms are developed and applied together with certain extraction methods evaluated in the present study. The extraction with NaOH+Na₂EDTA (Bowman and Moir, 1993Bowman RA, Moir JO. 1993. Basic EDTA as an extractant for soil organic phosphorus. Soil Science Society of America Journal 57: 1516-1518. https://doi.org/10.2136/sssaj1993.03615995005700060020x
https://doi.org/10.2136/sssaj1993.036159... ), in order to compose the ³¹P nuclear magnetic resonance spectroscopy solution (³¹P NMR), is the most widely used for understanding the soil P cycle (Cade-Menun and Liu, 2014Cade-Menun B, Liu CW. 2014. Solution Phosphorus-31 nuclear magnetic resonance spectroscopy of soils from 2005 to 2013: a review of sample preparation and experimental parameters. Soil Science Society of America Journal. 78: 19-37. https://doi.org/10.2136/sssaj2013.05.0187dgs
https://doi.org/10.2136/sssaj2013.05.018... ). Furthermore, according to Turner et al. (2005)Turner BL, Cade-Menun BJ, Condron LM, Newman S. 2005. Extraction of soil organic phosphorus. Talanta 66: 294-306. https://doi.org/10.1016/j.talanta.2004.11.012
https://doi.org/10.1016/j.talanta.2004.1... , a joint use of quantification methods, such as sequential “Hedley” extraction, with ³¹P NMR can collaborate to elucidate the transformations in discrete pools of the organic and inorganic phosphorus cycle in soils under Eucalyptus plantations (Rinaldi et al., 2021Rinaldi LCB, Aleixo S, Silva EC, Gama-Rodrigues AC, Gama-Rodrigues EF, Gonçalves JLM, et al. 2021. 31 P NMR spectroscopy and structural models of soil organic phosphorus under Eucalyptus. Nutrient Cycling in Agroecosystems 120: 83-97. https://doi.org/10.1007/s10705-021-10139-4
https://doi.org/10.1007/s10705-021-10139... ) and leguminous trees (Aleixo et al., 2020Aleixo S, Gama-Rodrigues AC, Gama-Rodrigues EF, Campello EFC, Silva EC, Schripsema J. 2020. Can soil phosphorus availability in tropical forest systems be increased by nitrogen-fixing leguminous trees? Science of The Total Environment 712: 136405. https://doi.org/10.1016/j.scitotenv.2019.136405
https://doi.org/10.1016/j.scitotenv.2019... ). However, ³¹P NMR may have limited use due to its relatively high cost, and it is currently impossible to recommend it as a routine technique used in Brazilian accredited soil analysis laboratories.

The different analytical methods of extracting soil total P_O forms showed significant differences compared to the methods of analyzing total P from the soil. The “Bowman and Moir” method appeared as a simple and quick technique, which covered a smaller number of manipulations, which reduced the possibility of interference in the results due to errors in the execution of the analytical sequence. Furthermore, fractionation techniques can be costly regarding time and financial resources. Another advantage is that, through extraction with NaOH+Na₂EDTA, there is the future possibility of associating the results with those obtained with the ³¹P NMR analysis, in order to understand and maximize the sustainability of systems, and establish the potential of the labile P_O fractions to provide P_I to the Eucalyptus plantations. The importance of further studies to verify the composition of the P organic forms will allow for an improved understanding of the contribution of this fraction through its soil compounds, and make it possible to evaluate its effects in tropical soils under the Brazilian Eucalyptus planted forests.

Acknowledgments

We thank the Conselho Nacional de Desenvolvimento Científico e Tecnológico – (CNPq) (Grant 302784/2017-9) for granting a research grant to the third author, the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) (Grant 88882.14549/2019-01, Postdoctoral Scholarship; Finance Code 001, Master Scholarship). We would also like to thank the Instituto de Pesquisas e Estudos Florestais (IPEF) for its technical support. The authors thank the anonymous reviewers for their relevant comments and suggestions on the manuscript. Any f trade, firm, or product name used is for descriptive purposes only.

References

Edited by

Publication Dates

History