Skip to main content
NCBI home page
EPA Author Manuscripts logoLink to EPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Jan 17.
Published in final edited form as: Chemosphere. 2022 Sep 1;308(Pt 2):136288. doi: 10.1016/j.chemosphere.2022.136288

X-ray absorption near edge structure spectroscopy reveals phosphate minerals at surface and agronomic sampling depths in agricultural Ultisols saturated with legacy phosphorus

Emileigh Lucas a,*, Lauren Mosesso b, Taylor Roswall a, Yun-Ya Yang a, Kirk Scheckel c, Amy Shober b, Gurpal S Toor a
PMCID: PMC9843306  NIHMSID: NIHMS1861453  PMID: 36058369

Abstract

Legacy phosphorus (P) soils have received excessive P inputs from historic manure and fertilizer applications and present unique management challenges for protecting water quality as soil P saturation leads to increased soluble P to waterways. We used P K-edge X-ray absorption near edge structure (XANES) spectroscopy to identify and quantify the dominant P minerals in four representative legacy P soils under conventional till and no-till management in Maryland, USA. Various measures of extractable soil P, including water-extractable P (20.6–54.1 mg kg−1 at 1:10 soil-to-water ratio; 52.7–132.2 mg kg−1 at 1:100 soil-to-water ratio), plant available P extracted with Mehlich 3 (692–1139 mg kg−1), and Mehlich 3P saturation ratio (0.54–1.37), were above the environmental threshold values, suggesting the accumulation of legacy P in soils. The quantification of dominant P minerals may provide insights into the potential of legacy P soils to contribute to P release for crop use and soluble P losses. Linear combination fits of XANES spectra identified the presence of four phosphate mineral groups, consisting of (i) calcium-phosphate minerals (11–59%) in the form of fluorapatite, (β-tricalcium phosphate, and brushite, followed by (ii) iron-phosphate minerals (12–49%) in the form of ludlamite, heterosite, P sorbed to ferrihydrite, and amorphous iron phosphates, (iii) aluminum-phosphate minerals (15–33%) in the form of wavellite and P sorbed to aluminum hydroxide, and (iv) other phosphate minerals (5–35%) in the form of copper-phosphate (cornetite, 5–18%) and manganese-phosphate (hureaulite, 25–35%). Organic P consisting of phytic acid was found in most soils (13–24%) and was more pronounced in the surface layer of no-till (21–24%) than in tilled (16%) fields. Of the P forms identified with XANES, we conclude that P sorbed to Fe and Al, and Ca–P in the form of brushite and β-tricalcium phosphate will likely readily contribute to the soil WEP pool as the soil solution P is depleted by crop uptake and lost via runoff and leaching.

Keywords: Phosphorus, Legacy phosphorus soils, Phosphate minerals, XANES, Ultisol, Poultry litter

Graphical Abstract

graphic file with name nihms-1861453-f0001.jpg

1. Introduction

Legacy phosphorus (P) soils are gaining societal and scientific scrutiny due to the concerns of the disproportionate amount of P losses to water bodies and the recognition of P as a finite resource (Rosen, 2020; Roswall et al., 2021). These soils have accumulated excessive amounts of P due to the historic application of P above agronomic needs (Sharpley et al., 2011; Lucas et al., 2021). In regions with concentrated animal production, P-rich grain is imported to feed the animals while the manure is locally land applied (Potter et al., 2010). As soils become saturated with P, a greater amount of P can be lost to surface waters (Roswall et al., 2022; Withers et al., 2019) leading to eutrophication in vulnerable water bodies. In soils, P exists as inorganic P due to the sorption and precipitation reactions and as organic P due to the application of organic wastes and immobilization reactions (Doydora et al., 2020). The type, amount, and availability of P minerals in soils depend on the soil characteristics and associated geology, soil formation factors, and soil management (Chen et al., 2018).

In legacy P soils, soil test P is likely to remain above agronomic and environmental thresholds for decades due to repeated manure or fertilizer applications (McDowell et al., 2020; Lucas et al., 2021). The Chesapeake Bay watershed in the United States, which includes the Eastern Shore of Maryland and parts of Delaware and Virginia, is an example of a region with excessive soil P due to nutrient mass imbalances (Beegle, 2013). The State of Maryland agronomic recommendations define optimum soil test P as ~50–100 mg kg−1 Mehlich 3-P (M3-P), and values above this are considered excessive (UME, 2010). In regions with soil test P above crop requirements, hot spots of legacy P exist from past overapplication of P-containing sources, which may result in increased P losses (Lucas et al., 2021; Roswall et al., 2021). Knowledge of P forms in legacy P soils is fundamental to predicting and managing the risk of P loss.

An effective way to study P forms in legacy P soils is using P K-edge X-ray absorption near edge structure (XANES) spectroscopy, as it allows observation of the soil in situ, without alteration by chemical extractants (Kizewski et al., 2011; Toor et al., 2006). A limitation of XANES studies is the bias toward soils with higher-than-average total P. Minimal XANES data exists for soils with lower TP levels due to a higher signal-to-noise ratio, which reduces data quality. Koch et al. (2018) and Sato et al. (2005) sieved soils to <50 μm and <20 μm, respectively, to concentrate the P present in the clay to medium silt fraction. In non-legacy P soils (i.e., P deficient soils), much of the available P-speciation data have been generated with x-ray diffraction or chemical fractionation techniques, which have their own limitations. For example, X-ray diffraction limits observation to crystalline minerals (Harris and White, 2008) and chemical fractionations may introduce errors because of the interaction with chemicals during the fractionation (Gu et al., 2020). Highly weathered, acidic soils, such as those in the Mid-Atlantic United States, are expected to be dominated by Fe–P and Al–P (Sims and Pierzynski, 2005). In their literature review, Kizewski et al. (2011) noted that the most common inorganic P forms observed in acidic soils and sediments using XANES and 31P nuclear magnetic resonance spectroscopy (31P NMR) were hydroxyapatite, octacalcium phosphate, phosphate (PO4) sorbed to Fe- or Al-oxide minerals, pyrophosphate, monetite, tricalcium phosphate, whereas the noncrystalline Ca–P were often found in alkaline soils.

To understand the differences in the various P forms in legacy P soils compared to soils not impacted by excessive P applications, we first consider that legacy P soils typically have a history of manure, biosolids, or fertilizer application. Different manure and fertilizer treatments could impact the P-species in the soil because of the P-species present in those sources. Analysis of XANES spectra for broiler litter, which is commonly land applied, found dicalcium phosphate (65–76%), aqueous phosphate (13–18%), and phytic acid (7–20%) as dominant P forms (Toor et al., 2005). Alum-amended poultry litter XANES spectra indicated the presence of more Al–P and phytic acid compared to non-amended poultry litter, which had more Ca-bound and organic P (Seiter et al., 2008). Shober et al. (2006) found predominately hydroxyapatite, PO4 sorbed to Al hydroxides, and phytic acid in lime-stabilized biosolids and manures (poultry and dairy). With mixed-species-manure (980–1840 mg TP kg−1) and commercial-fertilizer-only (760–1000 mg TP kg−1) treatments in a Mollisol (World Reference Base; WRB: Phaeozem) and two Inceptisols (WRB: Cambisols) in China, XANES analysis of 0–20 cm samples showed that repeated application of animal manure increased organic P and labile P forms compared to soils with commercial fertilizer only over 20 years (Luo et al., 2017). More phosphate mineral species with decreased crystallinity, suggesting greater P sorption capacity, were found in the top 20 cm of Oxisol (WRB: Ferralsol) soils as pig manure application rate increased over 11 years (Abdala et al., 2020). Abdala et al. (2020) also found that surface application of pig manure did not impact the P species beneath the top 20 cm and that Fe-(hydr) oxide minerals represented the bulk of P accumulations in the topsoil. Treatments of varying commercial and compost P in non-calcareous, tilled Inceptisols (WRB: Cambisols) in Germany did not lead to differences in XANES detected P species in the topsoil (0–30 cm, 630–728 mg TP kg−1) but caused differences in P species in subsoil layers (30–60 cm and 60–90 cm) over 16 years of cropping under the different treatments (Koch et al., 2018). Additionally, in the topsoil of all treatments, PO4 sorbed to Fe- and Al- (hydr)oxides were the dominant forms, followed by a significant amount of Ca–P compounds, whereas the subsoils of compost-amended soils were found to have a significant portion of phytic acid (Koch et al., 2018). A tile-drained sandy loam (Mollisol, WRB: Phaeozem) receiving cattle manure applications for 40 years exhibited a dominance of PO4 sorbed to Fe- and Al- oxides, a significant pool of amorphous Ca–P, and low organic P in the topsoil in both XANES and 31P NMR analysis (Schmieder et al., 2018). Overall, long-term manure application impacts the P-speciation in legacy P soils compared to soils receiving commercial P fertilizer only; however, these changes occur differently according to soil type, manure source, and length of manure history.

The legacy P soils have a long history of cropping and historical application of manure or fertilizers. Results of XANES analysis of Chinese Mollisols indicated a conversion of hydroxyapatite to Fe-bound P forms after 27 years of cropping with or without long-term P application, which could also be attributed to acidification of soil from initial pH of 7.6 to <5.8 (Liu et al., 2017). Further, labile P in the form of Ca (H2PO4)2 accumulated in the soils receiving continued applications, and organic P was detected in the form of phytate. Liu et al. (2018) used XANES and other spectroscopic techniques to observe changes in P speciation in Mollisols (WRB: Chernozem) managed for >50 years under different land uses of native and tame grasslands, non-irrigated cropland, and roadside ditches. They found that annual cropland in non-irrigated wheat production had the most variance in P speciation and contained organic P (phytic acid; 67%) at proportions less than grasslands but greater than roadside soils, and tricalcium phosphate (24%) greater than grasslands but less than roadside. Therefore, long-term agricultural land use causes shifts in soil P forms that vary depending on soil conditions and previous land use, even aside from the effects of various P treatments from the studies described.

Gaps in the literature exist in XANES analysis of legacy P forms in highly weathered, acidic soils. Many of the P speciation studies on legacy P soils have been conducted largely on soils (e.g., Mollisols, Inceptisols, and Oxisols) in which P may behave differently than in the Ultisols of the Mid-Atlantic region. Gamble et al. (2020) conducted combined μ-X-ray Fluorescence (μ-XRF) and μ-XANES to identify three Mid-Atlantic Ultisols with greater than 900 mg TP kg−1 and identified Ca–P, Fe–P, Al-sorbed P, and Fe-sorbed P in the samples. While μ-XANES techniques look at specific associations of P minerals in the complex matrix of the soil and are likely to identify more P species in any given sample (Hesterberg et al., 2017), bulk XANES can be useful for characterizing the dominant P species of the soil. Bulk XANES may be more easily applied to the field scale because it captures the most significant pools in the soil; however, shortcomings exist because of the small amount of soil used in the analysis. The variety of minor P species that exist in the soil that may be identified with μ-XANES can cumulatively contribute to P loss, but there is the potential for bias toward hot-spots of P in the soil sample (Lombi and Susini, 2009). Therefore, a bulk XANES approach would provide more useful data to understand the forms and chemistry of legacy P in soils.

Identifying P forms in legacy P soils elucidates how P may be stored, which aids in the prediction of the potential for the future release of P to crops and surface waters. Recent long-term studies observed an initial decline of water-extractable P (WEP) after P applications were ceased, followed by a plateau effect, whereby the rate of decline significantly decreased. This decline and plateau occurred at different timescales in these studies (Dodd and Sharpley, 2016; McDowell et al., 2020; Lucas et al., 2021), but the pattern was similar, meaning that as the labile P pool was exhausted, P became available from other pools. Our previous work in the Mid-Atlantic US soils treated with manure P at four different rates revealed an initial decline in WEP from 0 to 9 years after cessation of P applications.However, no significant declines in WEP occured from 9 to 15 years, at 0.2–0.4 mg kg−1 decrease per year, while initial manure P application effect was still evident (Lucas et al., 2021). In the 0–5 cm layer of the same soils discussed in this study, Roswall et al. (2021) attempted to exhaust the WEP pool by conducting sequential water extractions equivalent to a 1:800 soil-to-water ratio; however, by the end of the final extraction, the remaining WEP pool was still higher than the US EPA recommended limit for regional streams and rivers (0.03125 mg L−1). Speciation of P using XANES may help to explain the P storage mechanisms in legacy P soils. Our objective in this study was to identify various dominant P forms (minerals) in the surface (0–5 cm) and agronomic (0–20 cm) layers of selected Mid-Atlantic Ultisols that may disproportionally contribute P to environmental and agronomic P pools and potentially impact water quality in downstream water bodies, such as the Chesapeake Bay.

2. Materials and methods

2.1. Site characteristics and sampling

We selected four legacy P-impacted soils (hereafter referred to as fields A, B, C, and D) under typical management for the region (with TP of 1280–2235 mg kg−1), and conducted sampling at two depths (0–5 cm and 0–20 cm), except for field D which was only sampled at one depth (0–20 cm). We chose these depths as the top surface layer (0–5 cm) is most prone to surface runoff losses, and the agronomic layer (0–20 cm) is the standard agronomic depth that farmers sample to make fertilizer application decisions and the depth used in local regulatory requirements. Further, the 0–5 cm depth can provide insights in no-till soils, which often exhibit P stratification (Robbins and Voss, 1991). All the fields were located on the Eastern Shore of Maryland, which is part of the Delmarva Peninsula (SI Fig. 1). Field A was an organic vegetable operation with a history of poultry litter application and conventional tillage. Fields B, C, and D were in commodity row-crop rotations common in the Mid-Atlantic, typically corn (Zea mays), soybeans (Glycine max), and wheat (Triticum aestivum L.), have a long-term history of poultry litter application, and only received commercial fertilizer in recent years. All soils were classified as Ultisols (Table 1) (WRB: Gleysol for fields A and B, Acrisol for fields C and D). Detailed manure and P fertilization application history were not available as these fields have been managed by different operators over the last decades. The commodity row-crop sites were managed as no-till at the time of sampling and for at least the prior growing season. Soil pH ranged from 5.97 to 6.46 in fields B, C, and D but was slightly basic for field A (7.25). All the fields were saturated with P above the regional environmental threshold of Mehlich 3–P saturation ratio (M3–PSR) of 0.15 (Sims et al., 2002). Organic matter was 1.87–2.11% in the 0–5 cm and 1.32–1.67% in the 0–20 cm (Table 1).

Table 1.

Characteristics of soils collected from fields A, B,C, and D at 0–5 cm and 0–20 cm depths (site D was sampled from only 0–20 cm).

 
 Field A
 Field B
 Field C
 Field D
0–5 cm 0–20 cm 0–5 cm 0–20 cm 0–5 cm 0–20 cm 0–20 cm
Soil Map Unit QuA QuA QuA QuA AoA AoA UoB
Soil Texture Sandy Loam Sandy Loam Loam Loam Sandy Loam Sandy Loam Sandy Loam
US Soil Classification Fine-loamy, mixed, active, mesic Typic Endoaquults Fine-loamy, mixed, active, mesic Typic Endoaquults Fine-loamy, mixed, active, mesic Typic Endoaquults Fine-loamy, mixed, active, mesic Typic Endoaquults Coarse-loamy, siliceous, semiactive, mesic Aquic Hapludults Coarse-loamy, siliceous,semiactive, mesic Aquic Hapludults Fine-loamy, mixed, semiactive, mesic Typic Hapludults
Type of Operation Organic Vegetable Conventional Till Organic Vegetable Conventional Till Commodity crop No-Till Commodity crop No-Till Commodity crop No-Till Commodity crop No-Till Commodity crop No-Till
pH 7.25 7.25 6.36 6.46 6.22 6.1 5.97
Total P 2235 1882 1401 1304 1405 1281 1328
aM3–P 1127 1139 692 764 785 807 765
M3–K 137 120 122 99 201 195 158
M3–Mg 167 169 162 157 104 82 120
M3–Ca 2662 2608 1605 1560 1204 1016 831
M3–Fe 271 261 317 341 183 189 348
M3–A1 586 666 587 624 1006 1092 1061
M3–Zn 36.6 32.9 28.7 26.5 21.5 17.6 41.2
M3–Cu 26.7 24.9 11.8 11.2 18.5 17.8 7.7
M3-Ca/M3-P 0.423 0.437 0.431 0.490 0.652 0.794 0.920
bOM (%) 1.87 1.67 2.11 1.32 2.08 1.63 1.36
cM3-PSR 1.37 1.25 0.81 0.84 0.62 0.59 0.54
aM3-P/Total P 0.50 0.60 0.49 0.59 0.56 0.63 0.58
dWEPt (1:10) mg kg−1 54.1 20.6 34.7 30.4 29.1 27.7 30.6
dWEPt (1:100) mg kg−1 132.2 117.5 63.0 59.5 52.7 52.8 66.3
Open in a new tab
a

Mehlich 3 (M3). All M3 values are in mg kg−1.

b

Organic Matter (OM).

c

Mehlich 3–Phosphorus Saturation Ratio (PSR) is the molar concentration of P to the sum of the molar concentration of Fe and Al as measured with Mehlich 3 extraction.

d

Total water extractable phosphorus (WEPt).

Soil sampling was conducted in October 2019 by establishing a 4 m × 7 m transect in each field in an unbiased portion of the field away from perimeters, grass strips, and abnormal areas. Ten soil samples from each field were collected from the selected transect using a standard soil corer at 0–5 cm and another 10 samples at 0–20 cm. A composite soil sample was then taken for each depth. Samples were air-dried at 35 °C, ground to pass a 2-mm sieve, and stored in sealed containers at room temperature until XANES and chemical analyses.

2.2. Phosphorus K-edge X-ray absorption near edge structure (XANES) analysis

Samples were analyzed using XANES spectroscopy at the bending magnet beamline 9-BM at Advanced Photon Source (APS), Argonne National Laboratory, IL, USA. The beamline was equipped with a Si double crystal monochromator which was calibrated to 2149 eV using P2O5 as a calibration standard. Dried and sieved soil samples were pelleted, and double-sided carbon tape was used to adhere them to the sample plate for analysis. The monochromator’s 4-element Vortex silica drift detector collected sample spectra in fluorescence mode along the energy range of 2125–2331 eV, with step sizes of 1.0, 0.12, and 0.05 eV for 2125–2140, 2140–2175, and 2175–2331 eV, respectively. Preparation of 16 of the 23 standards (Supplementary Information, SI Table 1 ) is described in Qin et al. (2018), where the spectra were collected in total electron yield (TEY) mode along the energy range of 2099–2309 eV with step sizes of 2.0, 0.125, and 1.0 eV for 2099–2134, 2134–2179, and 2179–2309 eV, respectively. The remaining seven standards were procured from various sources, and composition was verified using X-ray Fluorescence (XRF) (SI Table 1). Spectra for these new standards were also collected in TEY mode, but along energy range 2125–2331 eV with step sizes 1.0 eV for 2125–2140 eV, 0.12 eV for 2140–2175 eV, and 0.05 for 2175–2331 eV, respectively.

2.3. XANES data analysis

Phosphorus K-edge X-ray absorption data were processed using Athena XAS data processing software (Ravel and Newville, 2005). For each sample, three spectra were collected. The multiple spectra were merged, smoothed using Boxcar Average Kernal Size 3, calibrated, and normalized. The normalization range was approximately 30–160, and the fit range was −5 to 50. No e0 shift was allowed. Linear combination fits (LCF) were used to compare sample spectra with our standard library spectra to identify and quantify the dominant P species present in each soil sample (SI Table 2). All standards used in LCF are described in SI Table 1.

2.4. Chemical analyses

Water-extractable P (WEP) was determined as described in detail by Roswall et al. (2021) at 1:10 and 1:100 soil-to-water ratios (w:v), analyzed on a Lachat QuikChem 8500 Automated Ion Analyzer (Hach, Loveland, CO, USA). Mehlich 3 P in soils was determined following the method reported by Wolf and Beegle (2011) on an ICP–OES. The pH was determined using a 1:1 soil-to-water ratio (Eckert, 2011). Organic matter was determined by loss on ignition (Schulte, 2011). Total P in soils was determined by the EPA 3050B method (USEPA, 1996). The Mehlich 3-P saturation ratio (M3–PSR) for each sample was calculated using the molar ratio P to the sum of Fe and Al (Maguire and Sims, 2002; Nair and Harris, 2004).

3. Results and discussion

Inorganic P species dominated our four legacy P soils (Table 2) at 0–5 cm (76–84%) and 0–20 cm (79–100%) depths based on LCF fits of the XANES spectra, which is in line with the previous studies in soils (Kizewski et al., 2011; Doydora et al., 2020). We observed the presence of organic P as phytic acid and four groups of inorganic P minerals, as discussed in the following sections.

Table 2.

Percent of various phosphorus groups and minerals quantified in four legacy phosphorus soils using X-ray absorption near edge structure linear combination fitting. Measured values (from SI Fig. 2) were converted to a total of 100 percent.

P group P mineral Field A
 Field B
 Field C
 Field D
0–5 cm 0–20 cm 0–5 cm 0–20 cm 0–5 cm 0–20 cm 0–20 cm
Percent———
Organic P Phytic Acid 15.5 14.7 20.8 13.1 23.4 nd a 21.4
Calcium Phosphates Fluorapatite 33.5 41.1 nd nd nd nd nd
Beta-Tricalcium Phosphate 7.0 nd 33.4 22.7 22.4 11.6 32.8
Brushite nd nd nd 28.6 nd nd 26.2
Aluminum Phosphates Wavellite nd nd nd nd nd 14.6 nd
Phosphate sorbed to Aluminum hydroxide 26.6 33.0 17.9 nd 19.2 nd 19.6
Iron Phosphates Ludlamite nd nd nd nd nd 24.0 nd
Heterosite nd nd nd nd nd 24.9 nd
P sorbed to ferrihydrite nd nd 11.5 nd nd nd nd
Amorphous Fe Phosphates nd nd nd 31.1 nd nd nd
Other Phosphates Cornetite 17.5 11.2 16.3 4.6 nd nd nd
Hureaulite nd nd nd nd 35.0 24.9 nd
bLCF R-factor 0.002 0.002 0.003 0.003 0.010 0.003 0.019
Open in a new tab
a

nd: not detected.

b

The R-factor of Linear Combination Fitting (LCF) of standard spectra is a measure of the mean square sum of the misfit (between the standard and the sample) at each data point.

3.1. Presence of organic P in legacy P soils

Inorganic P species were identified in the highest proportions, whereas the organic P in the form of phytic acid was detected at 13.1–23.4%, except for 0–20 cm in field C (Fig. 1, Table 2). In all fields at both depths, phytic acid was greater at 0–5 cm (15.5–23.4%) than 0–20 cm (0–14.7%), whereas in no-till fields B and C, phytic acid was 8–23% greater in the 0–5 cm than 0–20 cm. Phytic acid accumulation has been reported to occur in the surface layer in no-till soils, particularly in fields with a history of corn and rye production and poultry litter application (Koopmans et al., 2007; Cade-Menun et al., 2015). Phytic acid is considered unavailable to plants due to its strong binding with soil minerals and the low phytase content in the soil; however, root excretions may mineralize a small part of the phytic acid (Gerke, 2015). While root exudates, enzymes, and microorganisms have a role in improving phytic acid availability, these may not occur as readily in soils with abundant P in the soil solution (Giles and Cade-Menun, 2014). Phosphorus is present as phytic acid in crop seeds (e.g. corn, soybeans). When crop seeds are fed to poultry, which lack phytase, phytic acid is excreted in poultry litter and thus has been found in litters ranging from 7 to 20% (Toor et al., 2005). In Chinese Mollisols, phytic acid content was higher in both fertilized and unfertilized treatments due to the addition of crop residues and maize roots (Liu et al., 2017). The presence of phytic acid in our soils is expected due to the history of poultry litter application and the presence of phytic acid in commonly grown crops (corn, rye, and soybeans). Low organic P observed in field C at 0–20 cm, where organic P was not detected as a significant form, could be due to the fact that speciation of organic P is not as successful with XANES as compared to other methods such as solution-state 31P NMR (Toor et al., 2006; Cade-Menun and Liu, 2014). Despite this, we have confidence in the detection of phytic acid, representing the organic P pool, at fields A and B, due to the excellent fits achieved with R-factors of 0.002–0.003. An effective method to discern between specific forms of organic P is 31P NMR spectroscopy. Some studies using 31P NMR spectroscopy have found that the addition of manure and other organic P sources does not lead to the accumulation of organic P forms in legacy P soils, but rather the organic P is efficiently mineralized into inorganic forms (Dou et al., 2009; Annaheim et al., 2015). The soils in the Annaheim et al. (2015) study had a total P of 442–1150 mg kg−1, lower than those in our study, and nutrient sources were mineral P fertilizer, dairy manure, green waste compost, and biosolids, whereas our soils had a history of poultry litter application. The soils in the Dou et al. (2009) had high total P in treated soils (808–4866 mg kg−1) but received varying species of manure and compost and were also still actively receiving manure at the time of the study. Therefore, the unique characteristics of these legacy-P Ultisols with a history of poultry litter application but no recent applications and high total P concentrations have likely resulted in significant concentrations of XANES-detectable phytic acid. This could warrant further investigation with other techniques such as 31P NMR spectroscopy to examine the other organic P forms present in these soils.

Fig. 1.

Fig. 1.

Open in a new tab

Percent of phosphate (PO4) mineral associated with calcium, aluminum, iron, and other elements at surface (0–5 cm) and agronomic (0–20 cm) sampling depths of four legacy phosphorus soils determined with X-ray absorption near edge structure (XANES) linear combination fitting (LCF).

3.2. Identification and quantification of calcium-P minerals

In all fields, three calcium phosphate (Ca–P) minerals, including brushite (CaHPO4·2H2O), β-TCP (β-Ca3(PO4)2), and fluorapatite (Ca5 (PO4)3F) were identified (Table 2). The proportions of these minerals did not follow a pattern based on 0–5 and 0–20 cm depths. Field A had nearly the same proportion in both depths (40–41%), while Ca–P minerals were much higher in the field B at 0–20 cm (51%) than 0–5 cm (33%), and field C conversely had greater P minerals in 0–5 cm (22%) than 0–20 cm (11%) (Fig. 1). In our legacy P soils, we found an abundance of β-TCP, which is a non-crystalline mineral, in all fields and depths, ranging from 7.0 to 32.8% except for field A (0–20 cm). In soil, β-TCP has relatively low solubility and can contain small cations like Mg2+ and Fe2+ which are abundant in the soils (Table 1) that may further stabilize this mineral (e.g., Gregory et al., 1974). Fluorapatite was found only in field A (33.5% in 0–5 cm, 41.1% in 0–20 cm); this mineral is present in most soils and is the least soluble of the Ca–P minerals in the environment (Dixon et al., 1989). It is important to note that field A had a pH of 7.25, which likely resulted in an abundance of fluorapatite as compared to other low pH (5.97–6.46) fields.

In general, acid soils (< pH 7) are not considered to have significant reserves of Ca–P, as Ca is more soluble in low pH soils. However, most soils contain Ca, and further application of manures that contain Ca bind with soluble P present in the soil solution. The abundance of both Ca and P in the soil leads to Ca–P precipitation reactions (Penn and Camberato, 2019). Among three Ca–P minerals, brushite is the most soluble mineral, which was only found in the 0–20 cm layer of field B (28.3%) and field D (0.5%) (Table 2). While β-TCP has relatively low solubility compared to brushite and other lower Ca:P minerals, it is more soluble than fluorapatite and could solubilize once other more soluble forms are exhausted (Diaz et al., 1994; McDowell et al., 2003; Doydora et al., 2020). Thus, it is important to consider the presence and solubility of Ca–P minerals in the acidic legacy P soils, which are dominated by Al and Fe. Increased Ca content in acid soils is a consequence of manure application history (Sharpley et al., 2004) and Ca–P has been observed in other XANES studies of acidic soils after a history of manure application (Abdala et al., 2018; Gamble et al., 2020). Sato et al. (2005) found soluble Ca–P species, dibasic calcium phosphate, and dibasic calcium phosphate dehydrate, while Shober et al. (2006) found hydroxyapatite and β-TCP as dominant P minerals in poultry litter. Acidic, silt loam, Inceptisol (WRB: Cambisol), with short-term litter applications contained Fe–P forms reminiscent of the native forest soil and soluble Ca–P species, whereas soils receiving long-term poultry litter applications were dominated by Ca–P in relatively stable forms such as β-TCP (Sato et al., 2005).

As the soluble P in soil solution is gradually depleted (plant uptake, loss) and equilibrium shifts, more soluble Ca–P minerals may dissolve and contribute to soil solution P soil once fertilizer applications cease, as noted by Liu et al. (2015). The M3-Ca in our legacy P fields was 1204–2662 mg kg−1 at the 0–5 cm depth and 831–2608 mg kg−1 at the 0–20 cm (Table 1), which likely favored precipitation of P with Ca. Further, the M3 Ca:P ratio in soils was 0.42–0.92, which indicates the abundance of soil solid phase with Ca2+. There is a possibility that some of the Ca in the 0–5 cm was precipitated with phytic acid, which was greater in the 0–5 cm than in 0–20 cm of field B. Under oversaturated conditions, precipitation of phytic acid with calcite is possible in soil (Wan et al., 2016). Analysis with XANES has indicated that phytic acid is strongly retained by calcite at pH 6.0 (Prietzel et al., 2016), which was the pH in field B (6.36 and 6.46). More soluble forms of Ca, such as dicalcium phosphate, are added in poultry diets (Toor et al., 2005), which can precipitate into less available forms under Ca over-saturated conditions. Calcite was not included in our XANES standards because the USDA-NRCS soil characterization report for soil series at the fields reports that only trace amounts of the mineral may exist; however, future studies should examine if the historic application of poultry litter can cause calcite precipitation in significant amounts in legacy P soils.

3.3. Aluminum-phosphate minerals presence in legacy soils

Two Al–P minerals, namely PO4 sorbed to Al hydroxide and wavellite, were detected in legacy P soils (Fig. 1, Table 2). PO4 sorbed to Al hydroxide was found at the highest percentage in both depths of field A (26.6–33.0%), which had a history of tillage, whereas in no-till fields (B and C), this mineral was only found at 0–5 cm depth (17.9–19.2%) and field D at 0–20 cm (19.6%), which might be due to the P stratification resulting from the applications of alum-treated poultry litter, which will reside in the surface layer of soils (Huang et al., 2016). The mineral wavellite, which is a less soluble AlP form, was only found in field C at 0–20 cm (18.3%). While wavellite is a rare secondary Phosphate mineral, it has been detected in weathering of phosphatized clays (Nriagu, 1976); however, because XANES analysis is based on linear combination fitting, the statistical software may pick other Al–P minerals and categorize them as wavellite. Regardless, our results show a significant contribution of Al–P as an inorganic P source in legacy-P soils, specifically in the surface 0–5 cm for no-till soils with a history of poultry litter application.

Studies have found that Al-P minerals are significant in the soils, particularly in the form of PO4 sorbed to Al hydroxide in acidic soils (Beauchemin et al., 2003; Koch et al., 2018). In highly weathered acidic soils, such as the Ultisols in this study, Fe and Al are commonly present and have a strong affinity for P sorption (Sims and Pierzynski, 2005). In these soils, P sorption occurs via two main mechanisms, a surface reaction that is easily reversible, and then a long-term reaction where inorganic P is sorbed to Al- and Fe-oxides within soil aggregates (Maguire et al., 2001; Koopmans et al., 2004a). A portion of the PO4 sorbed to Al hydroxide in legacy soils, as found in fields A, B, and C, would likely become available in soil solution as the soil solution pool of P is depleted by crop uptake and leaching or runoff to surface water (Liu et al., 2015; Doydora et al., 2020).

3.4. Iron-phosphate minerals quantification in legacy soils

Four Fe–P minerals, including P sorbed to ferrihydrite, ludlamite, heterosite, and amorphous Fe phosphate were identified in legacy P soils. In fields A and D, Fe–P species were not identified as dominant P species, and in field B, P sorbed to ferrihydrite (11.5%) was found at 0–5 cm and amorphous Fe phosphate (31.1%) at 0–20 cm, whereas in the field C, ludlamite (24.0%) and heterosite (24.9%) were found at 0–20 cm depth.

Similar to PO4 sorbed to Al hydroxide, desorption of P from ferrihydrite could be expected once equilibrium shifts and the soil solution is less saturated with P. Amorphous Fe–P is also accessible to plants and would become available as P decreases in the soil solution (Armstrong et al., 1993). The forms of Fe–P found in field C, ludlamite and heterosite, are secondary minerals and are not likely to contribute P until an equilibrium is reached. Ludlamite was used as a standard in XANES analysis for the purpose of identifying Mn(II)–Fe(II) phosphates, not necessarily the presence of actual ludlamite in the agricultural soil (Ingall et al., 2011; Attanayake et al., 2022). Detection of Mn(II)–Fe(II) phosphates is consistent with identification of another Mn–P mineral at this site, which is discussed in section 3.5. Heterosite indicates the presence of oxidized Fe and Mn in soil. The soil in field C is an Aquic Hapludult, which means the water table is shallow and the soils undergo both oxidation and reduction conditions periodically. Kinetics of precipitation-dissolution reactions for these minerals are complex (Sims and Pierzynski, 2005); however, other P forms are likely to control the equilibrium before dissolution and release of P by secondary minerals due to the high total P content in these soils. Phosphate associated with Fe-minerals is one of the major sedimentary pools of P found in the Chesapeake Bay (Li et al., 2015). Iron phosphate was reported as an intermediate P reserve in legacy P Mollisols soils with either discontinued or continued commercial fertilizer application (Liu et al., 2015). In temperate climates, Al–P tends to have a larger influence than Fe–P; however, XANES analysis has shown a greater influence of Fe–P in tropical and sub-tropical regions (Abdala et al., 2020).

3.5. Existence of other P minerals in legacy soils

Copper phosphate (cornetite) was found in fields A and B, with a higher percentage in 0–5 cm (16.8–17.5%) than 0–20 cm (4.7–11.4%). In a literature review, cornetite was not commonly found in agricultural soils, however, other Cu forms are known to accumulate in soils. For example, Cu is applied as organic-approved fungicides to vegetable and fruit crops and is also a component of many commercial fertilizers, manures, biosolids, and pesticides (He et al., 2005). Field A was an organic vegetable operation and likely used Cu-containing pesticides. Poultry litter application has also been found to increase Cu and Mn concentrations in Mid-Atlantic Coastal Plain soils compared to adjacent forested sites due to their supplementation in poultry feed to promote bird health (Codling et al., 2008). Sandy loams in the piedmont region of Georgia, USA with a history of poultry litter application showed increases in Cu in the top layer in the no-till soils (He et al., 2009). Poultry litter application is common in most legacy P soils in the area (Parker and Li, 2006) and was the major source of crop nutrients for the organic vegetable operation. Hureaulite, a relatively insoluble Mn phosphate, was only detected in field C and was higher in the 0–5 cm (39.0%) than 0–20 cm (25.2%). The detection of hureaulite with XANES does not necessarily signify that hureaulite itself is present, but rather indicates presence of reduced Mn(II) phosphates in the soil at this site, at which reducing conditions are likely to occur because it is an Aquic Hapludult. Similar to Cu, accumulation of Mn was also noted in the surface soil as a result of poultry litter application and no-till management by He et al. (2009).

3.6. Relating wet chemistry extracted phosphorus pools with XANES analysis

Soil WEP was greater in 0–5 cm than 0–20 cm for all fields at 1:10 soil-to-water ratio and in fields A and B at 1:100 soil-to-water ratio. Soil WEP (1:100) was two-times higher in field A (117.5–132.2 mg kg−1) than other fields (52.7–63.0 mg kg−1), as reported by Roswall et al. (2021; Table 1). Even multiple sequential WEP (equivalent to 1:800) extractions in legacy P soils were not able to fully exhaust the WEP pool, suggesting accumulation and presence of a large reservoir of soluble P that can continue releasing P to soil solution pool over a long time (Roswall et al., 2021). The XANES data provided insight into the composition of the reservoir of P that will replenish soluble P pools and become available as WEP is depleted. Solubility of the inorganic P forms varies based on conditions in the soil, especially pH (Penn and Camberato, 2019); however, brushite is the most soluble Ca–P mineral found in two sites of these legacy soils at 0–20 cm depth (Table 2). Beta-TCP could also contribute to the WEP pool once more soluble Ca–P is exhausted, as it was found at all sites (Table 2) and is a form found in impacted stream waters (Diaz et al., 1994). A portion of the P bound with Al and Fe through sorption reactions (such as PO4 sorbed to Al hydroxide, PO4 sorbed to ferrihydrite, and amorphous Fe phosphate) will desorb when the equilibrium in the soil shifts and less P is in the soil solution (Liu et al., 2015; Doydora et al., 2020).

Mehlich 3-P test has been correlated with crop response in several states in the United States to estimate plant-available P and guide agronomic soil P recommendations in the 0–20 cm depth, where M3-P of 50–100 mg kg−1 is considered optimum for plant growth. In our legacy P soils, M3-P concentrations were similar in the 0–20 cm (764–1139 mg kg−1) and 0–5 cm (692–1127 mg kg−1) across all fields, indicating P accumulation and saturation in these soils (Table 1). Mehlich 3 extractions liberate chemical P forms that would be extracted by water (i.e., WEP), P sorbed to Fe, Al, or Ca, and organic P predominately present as phytic acid (Cade-Menun et al., 2018). Our observation of a significant portion of phytic acid (16–24% at 0–5 cm depth) in all fields (Table 2), which is considered minimally available to plants but is likely extracted by M3, also indicates that M3-P may inaccurately represent the plant-available P in legacy P soils.

The Ca:P ratio based on M3 was highest (0.920) in field D (0–20 cm), which also had the highest percent of Ca–P minerals (59%). In fields A and B, Ca:P was 0.423–0.437, with 33–51% Ca–P minerals. Even though in field C, Ca:P was 0.652–0.794, the Ca–P minerals were least in abundance (11–22%). Instead, field C had a large presence of hureaulite (25–35%), an Mn–P, which was unique from other fields. This could explain our observation of fewer Ca–P minerals in field C despite the high Ca:P ratio because Mn2+, is interchangeable with Ca2+, and has a smaller radius than Ca2+ (Schmidt and Husted, 2019) so it may favorably precipitate with P under certain conditions. Precipitation of Mn minerals is favored in soils that are subject to waterlogged conditions (Johnson et al., 2016), and field C is considered aquic, whereas all other soils in this study are typic according to the US soil classification system (Table 1).

The M3-PSR, which is a molar ratio of Mehlich 3 extracted P to Fe and Al, is used in acidic soils to determine the degree of P saturation. This ratio was highest in field B (>0.81), followed by C (0.59–0.62) and D (0.54) (Table 1). The combined proportion of Al–P and Fe–P minerals was 30–31% in field B, 19–64% in field C, and 20% in field D. The average M3-PSR was 0.68 for all samples, excluding field A (pH > 7), indicating oversaturation of P at Fe and Al binding sites (Table 1). All soil samples were well above the M3-PSR threshold change point of 0.15 for runoff and leaching, above which the risk of eutrophication of waterways increases (Sims et al., 2002). In field A, pH was in the basic range (7.25), where the use of M3-PSR using M3-Fe and M3-Al as the denominator is ineffective (Kleinman and Sharpley, 2002; Ige et al., 2005); however, this soil contained 27–33% of Al–P mineral. The high pH (7.25) and excessive Ca concentrations (M3-Ca: 2608–2662 mg kg−1) in field A likely led to precipitation of fluorapatite, which was only found in this field at 34–41%, indicating that Ca likely plays a more significant role in controlling P solubility than Fe and Al (Ige et al., 2005).

3.7. Management implications of chemical analyses and XANES

Our WEP and M3-PSR data suggest that a large pool of legacy P present in these soils will continue to contribute to the soluble P pool for decades as P is used by plants and lost from the fields (Lucas et al., 2021; Roswall et al., 2021). The abundance of various minerals, namely Fe–P, Al–P, and Ca–P, will likely control soil solution P when the WEP pool is slowly exhausted over a multi-decade timescale. The large reserves of P will provide a source of P for crops and will negate the need for applying commercial P fertilizers. On the other hand, cessation of P application alone is not likely to be sufficient in mitigating the risk of P loss to surface water. Other management strategies must be considered in legacy P soils saturated with long-term application of P inputs.

In the acidic Mid-Atlantic soils, Fe and Al dominate soils, which is indicated in the abundance of Fe–P and Al–P minerals obseived in legacy P soils. The rate of P desorption and the equilibrium kinetics are challenging to quantify in the soils (Koopmans et al., 2004b). Among these minerals, the occurrence of reducing conditions, such as shallow depth to groundwater, which is common in the region where soil samples were collected, can lead to a reduction of Fe2+ to Fe3+, resulting in a large release of P associated with Fe and further contribute to P release and loss (Ruiz et al., 1997). Steps should be taken to avoid waterlogged soil conditions in these legacy P fields. In addition, the use of M3-PSR, common for acidic soils, is not accurate for determining the risk of P loss in basic soils or soils that contain abundant Ca (Kleinman and Sharpley, 2002; Ige et al., 2005), so other ways to determine P saturation in legacy P soils that consider Ca along with Al and Fe should be explored.

Mehlich 3-P was 49–56% of TP in 0–5 cm and 58–63% in 0–20 cm, leaving a large proportion of P present in other pools (Table 1). Mehlich 3-P extraction and analysis with ICP includes both inorganic and organic P, including phytic acid (Cade-Menun et al., 2018). The XANES data showed that phytic acid was 13–23% in soils, which was likely extracted with M3-P. Liu et al. (2015) analyzed sediment P in the Chesapeake Bay and found high amounts of organic P in surface sediments. While a larger sample size and more controlled studies are needed to make a firm assessment about P speciation in the surface (0–5 cm) and agronomic depth (0–20 cm), our observation of greater phytic acid in the surface layer of no-till fields is consistent with 31P NMR findings by Cade-Menun et al. (2015). Mineralization of P from phytic acid may occur more readily than previously thought and therefore more research is needed to discern if changes in tillage practices, such as intermittent tillage, would be beneficial in preventing P stratification and reducing P loss. An in-depth analysis of the P and C cycling on the surface may be useful in assessing the organic P pool and its impact on P availability.

Lastly, the LCF of XANES data only detects the dominant P minerals in the soil and may miss some of the minor P forms that could influence the P availability (Lombi and Susini, 2009). Data processing, including normalization and smoothing of data, could lead to slightly different outcomes. Similarly, energy shift, quality of standards, and similarities in the spectral features between the standards could lead to differences in identified P species during LCF (Kelly et al., 2008). Therefore, identified P forms in our legacy P soils have been discussed broadly. Pairing XANES with additional techniques, such as solution-state 31P NMR, would provide a complete evaluation of organic P forms. Other XANES studies have used sequential chemical P fractionations to pair with results; however, these have merits and pitfalls as well (Gu and Margenot, 2021). Edge-of-field monitoring to determine the concentration-discharge relationship of soluble P forms transported from the legacy P soils would be a key to investigating the kinetics of P dissolution and developing solutions to mitigate diffuse P loss to water bodies.

4. Conclusions

A detailed understanding of the P forms in legacy P soils is critical for managing impacts on water quality. In the surface layer (0–5 cm), the dominant forms of P identified in four Mid-Atlantic Coastal Plain legacy P soils using XANES LCF were phytic acid (16–24%), Ca–P (22–40%), Al–P (18–27%), Fe–P (0–12%), Cu-, and Mn–P (16–35%). At the agronomic sampling depth (0–20 cm), we also found phytic acid (0–21%), Ca–P (11–59%), Al–P (0–33%), Fe–P (0–49%), Cu- and Mn–P (0–25%). Specifically, significant amounts of phytic acid, PO4 sorbed to Al hydroxide, and β-TCP were found as the dominant P-species in all fields in at least one of the sampling depths. The increased Ca–P presence is likely a result of historic applications of poultry litter, and a large amount of Ca–P found in these soils may warrant further exploration of P saturation calculations for legacy P soils, which typically only consider Fe and Al. Phytic acid was more prominent in 0–5 cm in no-till soils than in 0–20 cm. Some of the sorbed P forms detected with XANES are likely candidates to contribute to the soluble P pool once the soil solution P pool is exhausted and appropriate equilibrium is established—these P forms may include PO4 sorbed to Al hydroxides, PO4 sorbed to ferrihydrite, amorphous Fe–P, and brushite (a relatively soluble form of Ca–P). A portion of β-TCP would likely become available as well once soluble P and other pools in equilibrium are exhausted. Previous research has shown that cessation of P applications is not sufficient to prevent water quality impacts and our data demonstrates that legacy P soils are stockpiled with P forms that will likely become available over time as the labile P pools are exhausted (crop uptake, loss). This research highlights the need to further explore, with more fields and detailed management history, if standard agronomic and environmental soil P tests for acidic soils accurately assess the availability and risk for P loss from legacy P soils. A combination of chemical and spectroscopic techniques and field studies will provide the most informative determinations of P species to guide agricultural and environmental decision-making to combat water quality deterioration in agricultural watersheds.

Supplementary Material

Supplementary Material

HIGHLIGHTS.

  • Phosphate minerals in legacy P soils were investigated using XANES spectroscopy.

  • Calcium- (11–59%), iron- (12–49%), and aluminum- (15–33%) phosphate minerals were dominant in all soils.

  • Phytic acid was the major organic P compound (13–24%) identified in most soils.

  • Some P minerals will likely contribute to the soluble P pool after depletion of soil solution P.

Acknowledgments

The use of the Advanced Photon Source was supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under proposal number GUP-58956. We are grateful to Tianpin Wu and George Sterbinsky of Argonne National Lab for their assistance in operations at Beamline 9-BM. We express deep gratitude to the farmers who allowed us to explore the P-species in their legacy P fields. Although EPA contributed to this article, the research presented was not performed by or funded by EPA and was not subject to EPA’s quality system requirements. Gurpal S. Toor received funding for this work from Delmarva Land Grant University Seed Grant, USDA-NIFA Hatch project 1014496, and USDA-AFRI competitive grant 2018-09093.

Footnotes

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Credit author statement

Emileigh Lucas: Writing - Original Draft, Conceptualization, Formal analysis, Investigation, Data curation Lauren Mosesso: Formal analysis, Data curation Taylor Roswall: Investigation, Writing – review & editing Yun-Ya Yang: Writing - Review & Editing, Formal analysis, Investigation Kirk Scheckel: Formal analysis, Data curation Amy Shober: Writing - Review & Editing, Resources Gurpal S. Toor: Writing - Review & Editing, Funding acquisition, Conceptualization, Resources, Supervision.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.chemosphere.2022.136288.

Data availability

Data will be made available on request.

References

  1. Abdala DB, Moore PA, Rodrigues M, Herrera WF, Pavinato PS, 2018. Long-term effects of alum-treated litter, untreated litter and NH4NO3 application on phosphorus speciation, distribution and reactivity in soils using K-edge XANES and chemical fractionation. J. Environ. Manag 213, 206–216. 10.1016/j.jenvman.2018.02.007. [DOI] [PubMed] [Google Scholar]
  2. Abdala DB, Gatiboni LC, Schmitt DE, Mumbach GL, Dall’Orsoletta DJ,Bonfada EB, Veiga M, 2020. Phosphorus speciation and iron mineralogy in an oxisol after 11 years of pig slurry application. Sci. Total Environ 743, 1–10. 10.1016/j.scitotenv.2020.140487. [DOI] [PubMed] [Google Scholar]
  3. Annaheim KE, Doolette AL, Smernik RJ, Mayer J, Oberson A, Frossard E, Bünemann EK, 2015. Long-term addition of organic fertilizers has little effect on soil organic phosphorus as characterized by 31P NMR spectroscopy and enzyme additions. Geoderma 257–258, 67–77. 10.1016/j.geoderma.2015.01.014. [DOI] [Google Scholar]
  4. Armstrong RD, Helyar KR, Prangnell R, 1993. Direct assessment of mineral phosphorus availability to tropical crops using 32P labelled compounds. Plant Soil 150, 279–287. [Google Scholar]
  5. Attanayake CP, Kumaragamage D, Amarawansha G, Hettiarachchi GM, Indraratne SP, Goltz DM, 2022. Phosphorus release and speciation in manganese (IV) oxide and zeolite-amended flooded soils. Environ. Sci. Technol 56, 8082–8093. 10.1021/acs.est.2c01185. [DOI] [PubMed] [Google Scholar]
  6. Beauchemin S, Hesterberg D, Chou J, Beauchemin M, Simard RR, Sayers DE, 2003. Speciation of phosphorus in phosphorus-enriched agricultural soils using X-ray absorption near-edge structure spectroscopy and chemical fractionation. J. Environ. Qual 32, 1809–1819. 10.2134/jeq2003.1809. [DOI] [PubMed] [Google Scholar]
  7. Beegle D, 2013. Nutrient management and the Chesapeake Bay. J. Contemp. Water Res. Educ 151, 3–8. 10.1111/j.1936-704X.2013.03146.x. [DOI] [Google Scholar]
  8. Cade-Menun BJ, 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 Sci. Soc. Am. J 78, 19–37. 10.2136/sssaj2013.05.0187dgs. [DOI] [Google Scholar]
  9. Cade-Menun BJ, He Z, Zhang H, Endale DM, Schomberg HH, Liu CW, 2015. Stratification of phosphorus forms from long-term conservation tillage and poultry litter application. Soil Sci. Soc. Am. J 79, 504–516. 10.2136/sssaj2014.08.0310. [DOI] [Google Scholar]
  10. Cade-Menun BJ, Elkin KR, Liu CW, Bryant RB, Kleinman PJA, Moore PA Jr., 2018. Characterizing the phosphorus forms extracted from soil by the Mehlich III soil test. Geochem. Trans 19, 7. 10.1186/s12932-018-0052-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Chen D, Shen H, Hu M, Wang J, Zhang Y, Dahlgren RA, 2018. Legacy nutrient dynamics at the watershed scale: principles, modeling, and implications. Adv. Agron 149, 237–313. 10.1016/bs.agron.2018.01.005. [DOI] [Google Scholar]
  12. Codling EE, Chaney RL, Mulchi CL, 2008. Effects of broiler litter management practices on phosphorus, copper, zinc, manganese, and arsenic concentrations in Maryland Coastal Plain soils. Commun. Soil Sci. Plant Anal 39, 1193–1205. 10.1080/00103620801925901. [DOI] [Google Scholar]
  13. Diaz OA, Reddy KR, Moore PA, 1994. Solubility of inorganic phosphorus in stream water as influenced by pH and calcium concentration. Water Res. 28, 1755–1763. 10.1016/0043-1354(94)90248-8. [DOI] [Google Scholar]
  14. Dixon JB, Weed SB, Dinauer RC, America, S.S.S.o., 1989. Minerals in Soil Environments. Soil Science Society of America, ISBN 9780891187875. [Google Scholar]
  15. Dodd RJ, Sharpley AN, 2016. Conservation practice effectiveness and adoption: unintended consequences and implications for sustainable phosphorus management. Nutrient Cycl. Agroecosyst 104, 373–392. 10.1007/s10705-015-9748-8. [DOI] [Google Scholar]
  16. Dou Z, Ramberg CF, Toth JD, Wang Y, Sharpley AN, Boyd SE, Chen CR, Williams D, Xu ZH, 2009. Phosphorus speciation and sorption-desorption characteristics in heavily manured soils. Soil Sci. Soc. Am. J 73, 93–101. 10.2136/sssaj2007.0416. [DOI] [Google Scholar]
  17. Doydora S, Gatiboni L, Grieger K, Hesterberg D, Jones JL, McLamore ES, Peters R, Sozzani R, Van den Broeck L, Duckworth OW, 2020. Accessing legacy phosphorus in soils. Soil Syst. 4, 74. 10.3390/soilsystems4040074. [DOI] [Google Scholar]
  18. Eckert D a.J.T.S., 2011. Recommended soil pH and lime requirement tests. In: Wolf J.T.S.a.A. (Ed.), Recommended Soil Testing Procedures for the Northeastern United States, Agricultural Experiment Station. University of Delaware, Newark, DE, pp. 19–25. [Google Scholar]
  19. Gamble AV, Northrup PA, Sparks DL, 2020. Elucidation of soil phosphorus speciation in mid-Atlantic soils using synchrotron-based microspectroscopic techniques. J. Environ. Qual 49, 184–193. 10.1002/jeq2.20027. [DOI] [PubMed] [Google Scholar]
  20. Gerke J, 2015. Phytate (inositol hexakisphosphate) in soil and phosphate acquisition from inositol phosphates by higher plants. Rev. Plant. (Basel) 4, 253–266. 10.3390/plants4020253. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Giles CD, Cade-Menun BJ, 2014. Phytate in Animal Manure and Soils: Abundance, Cycling and Bioavailability. Applied Manure and Nutrient Chemistry for Sustainable Agriculture and Environment. Springer, pp. 163–190. [Google Scholar]
  22. Gregory TM, Moreno EC, Patel JM, Brown WE, 1974. Solubility of β-Ca(3)(PO(4))(2) in the system Ca(OH)(2)-H(3)PO(4)-H(2)O at 5, 15, 25, and 37 °C. J. Res. Natl. Bur. Stand A Phys. Chem 78A, 667–674. 10.6028/jres.078A.042. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Gu C, Margenot AJ, 2021. Navigating limitations and opportunities of soil phosphorus fractionation. Plant Soil 459, 13–17. 10.1007/s11104-020-04552-x. [DOI] [Google Scholar]
  24. Gu C, Dam T, Hart SC, Turner BL, Chadwick OA, Berhe AA, Hu Y, Zhu M, 2020. Quantifying uncertainties in sequential chemical extraction of soil phosphorus using XANES spectroscopy. Environ. Sci. Technol 54, 2257–2267. 10.1021/acs.est.9b05278. [DOI] [PubMed] [Google Scholar]
  25. Harris W, White NG, 2008. X-Ray diffraction techniques for soil mineral identification. Methods Soil Anal. Part 5—Mineral. Methods 81–115. [Google Scholar]
  26. He ZL, Yang XE, Stoffella PJ, 2005. Trace elements in agroecosystems and impacts on the environment. J. Trace Elem. Med. Biol 19, 125–140. 10.1016/j.jtemb.2005.02.010. [DOI] [PubMed] [Google Scholar]
  27. He Z, Endale DM, Schomberg HH, Jenkins MB, 2009. Total phosphorus, zinc, copper, and manganese concentrations in cecil soil through 10 Years of poultry litter application. Soil Sci. 174, 687–695. 10.1097/SS.0b013e3181c30821. [DOI] [Google Scholar]
  28. Hesterberg D, McNulty I, Thieme J, 2017. Speciation of soil phosphorus assessed by XANES spectroscopy at different spatial scales. J. Environ. Qual 46, 1190–1197. 10.2134/jeq2016.11.0431. [DOI] [PubMed] [Google Scholar]
  29. Huang L, Moore PA Jr., Kleinman PJ, Elkin KR, Savin MC, Pote DH, Edwards DR, 2016. Reducing phosphorus runoff and leaching from poultry litter with alum: twenty-year small plot and paired-watershed studies. J. Environ. Qual 45, 1413–1420. [DOI] [PubMed] [Google Scholar]
  30. Ige D, Akinremi O, Flaten D, 2005. Environmental index for estimating the risk of phosphorus loss in calcareous soils of Manitoba. J. Environ. Qual 34, 1944–1951. [DOI] [PubMed] [Google Scholar]
  31. Ingall ED, Brandes JA, Diaz JM, de Jonge MD, Paterson D, McNulty I, Elliott WC, Northrup P, 2011. Phosphorus K-edge XANES spectroscopy of mineral standards. J. Synchrotron Radiat 18, 189–197. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Johnson JE, Savalia P, Davis R, Kocar BD, Webb SM, Nealson KH, Fischer WW, 2016. Real-time manganese phase dynamics during biological and abiotic manganese oxide reduction. Environ. Sci. Technol 50, 4248–4258. 10.1021/acs.est.5b04834. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Kelly S, Hesterberg D, Ravel B, 2008. Analysis of soils and minerals using X-ray absorption spectroscopy. Methods Soil Anal, part 5—Mineral. Methods 5, 387–463. [Google Scholar]
  34. Kizewski F, Liu YT, Morris A, Hesterberg D, 2011. Spectroscopic approaches for phosphorus speciation in soils and other environmental systems. J. Environ. Qual 40, 751–766. 10.2134/jeq2010.0169. [DOI] [PubMed] [Google Scholar]
  35. Kleinman PJA, Sharpley AN, 2002. Estimating soil phosphorus sorption saturation from Mehlich-3 data. Commun. Soil Sci. Plant Anal 33, 1825–1839. 10.1081/CSS-120004825. [DOI] [Google Scholar]
  36. Koch M, Kruse J, Eichler-Löbermann B, Zimmer D, Willbold S, Leinweber P, Siebers N, 2018. Phosphorus stocks and speciation in soil profiles of a long-term fertilizer experiment: evidence from sequential fractionation, P K-edge XANES, and 31P NMR spectroscopy. Geoderma 316, 115–126. 10.1016/j.geoderma.2017.12.003. [DOI] [Google Scholar]
  37. Koopmans GF, Chardon WJ, Ehlert PA, Dolfing J, Suurs RA, Oenema O, van Riemsdijk WH, 2004a. Phosphorus availability for plant uptake in a phosphorus-enriched noncalcareous sandy soil. J. Environ. Qual 33, 965–975. 10.2134/jeq2004.0965. [DOI] [PubMed] [Google Scholar]
  38. Koopmans GF, Chardon WJ, de Willigen P, van Riemsdijk WH, 2004b. Phosphorus desorption dynamics in soil and the link to a dynamic concept of bioavailability. J. Environ. Qual 33, 1393–1402. 10.2134/jeq2004.1393. [DOI] [PubMed] [Google Scholar]
  39. Koopmans GF, Chardon WJ, McDowell RW, 2007. Phosphorus movement and speciation in a sandy soil profile after long-term animal manure applications. J. Environ. Qual 36, 305–315. 10.2134/jeq2006.0131. [DOI] [PubMed] [Google Scholar]
  40. Li W, Joshi SR, Hou G, Burdige DJ, Sparks DL, Jaisi DP, 2015. Characterizing phosphorus speciation of Chesapeake Bay sediments using chemical extraction, 31P NMR, and X-ray absorption fine structure spectroscopy. Environ. Sci. Technol 49, 203–211. 10.1021/es504648d. [DOI] [PubMed] [Google Scholar]
  41. Liu J, Hu Y, Yang J, Abdi D, Cade-Menun BJ, 2015. Investigation of soil legacy phosphorus transformation in long-term agricultural fields using sequential fractionation, P K-edge XANES and solution P NMR spectroscopy. Environ. Sci. Technol 49, 168–176. 10.1021/es504420n. [DOI] [PubMed] [Google Scholar]
  42. Liu J, Yang J, Cade-Menun BJ, Hu Y, Li J, Peng C, Ma Y, 2017. Molecular speciation and transformation of soil legacy phosphorus with and without long-term phosphorus fertilization: insights from bulk and microprobe spectroscopy. Sci. Rep 7 10.1038/s41598-017-13498-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Liu J, Cade-Menun BJ, Yang J, Hu Y, Liu CW, Tremblay J, LaForge K, Schellenberg M, Hamel C, Bainard LD, 2018. Long-term land use affects phosphorus speciation and the composition of phosphorus cycling genes in agricultural soils. Front. Microbiol 9 10.3389/fmicb.2018.01643. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Lombi E, Susini J, 2009. Synchrotron-based techniques for plant and soil science: opportunities, challenges and future perspectives. Plant Soil 320, 1–35. 10.1007/s11104-008-9876-x. [DOI] [Google Scholar]
  45. Lucas ER, Toor GS, McGrath JM, 2021. Agronomic and environmental phosphorus decline in coastal plain soils after cessation of manure application. Agric. Ecosyst. Environ 311 10.1016/j.agee.2021.107337. [DOI] [Google Scholar]
  46. Luo L, Ma Y, Sanders RL, Xu C, Li J, Myneni SCB, 2017. Phosphorus speciation and transformation in long-term fertilized soil: evidence from chemical fractionation and P K-edge XANES spectroscopy. Nutrient Cycl. Agroecosyst 107, 215–226. 10.1007/s10705-017-9830-5. [DOI] [Google Scholar]
  47. Maguire RO, Sims JT, 2002. Measuring agronomic and environmental soil phosphorus saturation and predicting phosphorus leaching with Mehlich 3. Soil Sci. Soc. Am. J 66, 2033–2039. 10.2136/sssaj2002.2033. [DOI] [Google Scholar]
  48. Maguire RO, Foy RH, Bailey JS, Sims JT, 2001. Estimation of the phosphorus sorption capacity of acidic soils in Ireland. Eur. J. Soil Sci 52, 479–487. 10.1046/j.1365-2389.2001.00394.x. [DOI] [Google Scholar]
  49. McDowell RW, Mahieu N, Brookes PC, Poulton PR, 2003. Mechanisms of phosphorus solubilisation in a limed soil as a function of pH. Chemosphere 51, 685–692. 10.1016/S0045-6535(03)00107-3. [DOI] [PubMed] [Google Scholar]
  50. McDowell R, Dodd R, Pletnyakov P, Noble A, 2020. The ability to reduce soil legacy phosphorus at a country scale. Front. Environ. Sci 8 10.3389/fenvs.2020.00006. [DOI] [Google Scholar]
  51. Nair VD, Harris WG, 2004. A capacity factor as an alternative to soil test phosphorus in phosphorus risk assessment. N. Z. J. Agric. Res 47, 491–497. 10.1080/00288233.2004.9513616. [DOI] [Google Scholar]
  52. Nriagu JO, 1976. Phosphate clay mineral relations in soils and sediments. Can. J. Earth Sci 13, 717. [Google Scholar]
  53. Parker D, Li Q, 2006. Poultry Litter Use and Transport in Caroline, Queen Anne’s Somerset and Wicomico Counties in Maryland: A Summary Report. Mid-Atl. Reg. Water Progr. MAWP0601, 1–12. [Google Scholar]
  54. Penn CJ, Camberato JJ, 2019. A critical review on soil chemical processes that control how soil pH affects phosphorus availability to plants. Agriculture 9, 120. 10.3390/agriculture9060120. [DOI] [Google Scholar]
  55. Potter P, Ramankutty N, Bennett EM, Donner SD, 2010. Characterizing the spatial patterns of global fertilizer application and manure production. Earth Interact. 14, 1–22. 10.1175/2009EI288.1. [DOI] [Google Scholar]
  56. Prietzel J, Harrington G, Hausler W, Heister K, Werner F, Klysubun W, 2016. Reference spectra of important adsorbed organic and inorganic phosphate binding forms for soil P speciation using synchrotron-based K-edge XANES spectroscopy. J. Synchrotron Radiat 23, 532–544. 10.1107/S1600577515023085. [DOI] [PubMed] [Google Scholar]
  57. Qin Z, et al. , 2018. Mechanisms of phosphorus removal by phosphorus sorbing materials. J. Environ. Qual 47, 1232–1241. 10.2134/jeq2018.02.0064. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Ravel B, Newville M, 2005. ATHENA, artemis, hephaestus: data analysis for X-ray absorption spectroscopy using IFEFFIT. J. Synchrotron Radiat 12, 537–541. 10.1107/S0909049505012719. [DOI] [PubMed] [Google Scholar]
  59. Robbins SG, Voss RD, 1991. Phosphorus and potassium stratification in conservation tillage systems. J. Soil Water Conserv 46, 298–300. [Google Scholar]
  60. Rosen J, 2020. Farmers Are Facing a Phosphorus Crisis. The Solution Starts with Soil. Science News. National Geographic, https://www.nationalgeographic.com/science/article/farmers-are-facing-a-phosphorus-crisis-the-solution-starts-with-soil. [Google Scholar]
  61. Roswall T, et al. , 2021. Hotspots of legacy phosphorus in agricultural landscapes: revisiting water-extractable phosphorus pools in soils. Water 13, 1006. 10.3390/W13081006. [DOI] [Google Scholar]
  62. Roswall T, Haggard B, Toor GS, 2022. Fate and transformations of dissolved phosphorus forms in runoff: Effect of poultry litter and products extracted with variable water extraction ratios. Chemosphere, 136220. 10.1016/j.chemosphere.2022.136220. In this issue. [DOI] [PubMed] [Google Scholar]
  63. Ruiz JM, Delgado A, Torrent J, 1997. Iron-related phosphorus in overfertilized European soils. J. Environ. Qual 26, 1548–1554. 10.2134/jeq1997.00472425002600060014x. [DOI] [Google Scholar]
  64. Sato S, Solomon D, Hyland C, Ketterings Q, Lehmann J, Brookhaven National Laboratory National Synchrotron Light, S., 2005. Phosphorus speciation in manure and manure-amended soils using XANES spectroscopy. Environ. Sci. Technol 39 10.1021/es0503130. [DOI] [PubMed] [Google Scholar]
  65. Schmidt SB, Husted S, 2019. The biochemical properties of manganese in plants. Plants 8, 381. 10.3390/plants8100381. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Schmieder F, Bergström L, Riddle M, Gustafsson J-P, Klysubun W, Zehetner F, Condron L, Kirchmann H, 2018. Phosphorus speciation in a long-term manure-amended soil profile – evidence from wet chemical extraction, 31P-NMR and P K-edge XANES spectroscopy. Geoderma 322, 19–27. 10.1016/j.geoderma.2018.01.026. [DOI] [Google Scholar]
  67. Schulte EE a.B.H., 2011. Recommended soil organic matter tests. In: Wolf J.T.S.a.A. (Ed.), Recommended Soil Testing Procedures for the Northeastern United States, Agricultural Experiment Station. University of Delaware, Newark, DE, pp. 63–74. [Google Scholar]
  68. Seiter JM, Staats-Borda KE, Ginder-Vogel M, Sparks DL, 2008. XANES spectroscopic analysis of phosphorus speciation in alum-amended poultry litter. J. Environ. Qual 37, 477–485. 10.2134/jeq2007.0285. [DOI] [PubMed] [Google Scholar]
  69. Sharpley AN, McDowell RW, Kleinman PJA, 2004. Amounts, forms, and solubility of phosphorus in soils receiving manure. Soil Sci. Soc. Am. J 68, 2048–2057. 10.2136/sssaj2004.2048. [DOI] [Google Scholar]
  70. Sharpley AN, Kleinman PJA, Flaten DN, Buda AR, 2011. Critical source area management of agricultural phosphorus: experiences, challenges and opportunities. Water Sci. Technol 64, 945–952. 10.2166/wst.2011.712. [DOI] [PubMed] [Google Scholar]
  71. Shober AL, Hesterberg DL, Sims JT, Gardner S, 2006. Characterization of phosphorus species in biosolids and manures using XANES spectroscopy. J. Environ. Qual 35, 1983–1993. [DOI] [PubMed] [Google Scholar]
  72. Sims TJ, Pierzynski GM, 2005. Chemistry of Phosphorus in Soils. Chemical Processes in Soils, pp. 151–192. [Google Scholar]
  73. Sims JT, Maguire RO, Leytem AB, Gartley KL, Pautler MC, 2002. Evaluation of Mehlich 3 as an agri-environmental soil phosphorus test for the mid-atlantic United States of America. Soil Sci. Soc. Am. J 66 10.2136/sssaj2002.2016, 2016. [DOI] [Google Scholar]
  74. Toor GS, Peak JD, Sims JT, 2005. Phosphorus speciation in broiler litter and Turkey manure produced from modified diets. J. Environ. Qual 34, 687–697. 10.2134/jeq2005.0687. [DOI] [PubMed] [Google Scholar]
  75. Toor GS, Hunger S, Peak JD, Sims JT, Sparks DL, 2006. Advances in the characterization of phosphorus in organic wastes: environmental and agronomic applications. Adv. Agron 89, 1–72. 10.1016/S0065-2113(05)89001-7. [DOI] [Google Scholar]
  76. UME, 2010. Soil Fertility Management: Agronomic Crop Nutrient Recommendations Based on Soil Tests and Yield Goals. University of Maryland Extension Publications, pp. 1–12. [Google Scholar]
  77. USEPA, 1996. Method 3050B: Acid Digestion of Sediments, Sludges, Soils. Revision 2. USEPA. [Google Scholar]
  78. Wan B, Yan Y, Liu F, Tan W, Chen X, Feng X, 2016. Surface adsorption and precipitation of inositol hexakisphosphate on calcite: a comparison with orthophosphate. Chem. Geol 421, 103–111. 10.1016/j.chemgeo.2015.12.004. [DOI] [Google Scholar]
  79. Withers PJ, Vadas PA, Uusitalo R, Forber KJ, Hart M, Foy RH, Delgado A, Dougherty W, Lilja H, Burkitt LL, 2019. A global perspective on integrated strategies to manage soil phosphorus status for eutrophication control without limiting land productivity. J. Environ. Qual 48, 1234–1246. 10.2134/jeq2019.03.0131. [DOI] [PubMed] [Google Scholar]
  80. Wolf AM, Beegle DB, 2011. Recommended soil tests for macronutrients. In: Wolf J.T.S.a.A. (Ed.), Recommended Soil Testing Procedures for the Northeastern United States, Agricultural Experiment Station. University of Delaware, Newark, DE, pp. 39–47. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Material
NIHMS1861453-supplement-Supplementary_Material.pdf (518.4KB, pdf)

Data Availability Statement

Data will be made available on request.

RESOURCES