As described by Anderson et al. [12] , the current study used two struvite sources. One was a commercially available, CPST material (i.e., trade name Crystal Green, Ostara Nutrient Recovery Technologies, Inc., Vancouver, Canada), which was produced from a raw, municipal wastewater source from near Atlanta, GA. The second was an ECST material produced by researchers in the Department of Chemical Engineering at the University of Arkansas from synthetic wastewater [24] . The ECST material was electrochemically precipitated from a synthetic solution of known N and P concentrations with Mg supplied from a Mg anode as an electrical current was applied to the solution [24] . Along with the two struvite sources, four commercially available P-fertilizers were used in this study, MAP, DAP, TSP, and RP, along with a UC treatment. The CPST, MAP, DAP, and TSP materials were in pellet form, the RP was in powder form, and the ECST was in crystalline-flake form. Table 1 summarizes properties of the various fertilizer-P sources.

As described by Anderson et al. [12] , four soils were collected from the top 10 - 15 cm from historic agricultural management throughout Arkansas for use in the current study. A Dardanelle silty clay loam (SiCL; fine-silty, mixed, superactive, thermic Typic Argiudolls) and a Roxana loam (L; coarse-silty, mixed, superactive, nonacid, thermic Typic Udifluvents) were collected at the University of Arkansas Division of Agriculture’s Vegetable Research Station in Kibler, AR [25] . A Calloway silt loam (SiL 1; fine-silty, mixed, active, thermic Aquic Fraglossudalfs) was collected at the Cotton Branch Experiment Station near Marianna, AR. A Henry silt loam (SiL 2; coarse-silty, mixed, active, thermic Typic Fragiaqualfs) was collected at the Pine Tree Branch Experiment Station near Colt, AR [25] . All soils were sieved field-moist to pass through a 7-mm mesh screen and air-dried for two weeks in a greenhouse at ~21˚C with periodic manual mixing to achieve uniform drying. Additional details of the soils were described in Anderson et al. [12] . Table 2 summarizes initial properties among the four soils.

As described by Anderson et al. [12] , moist-soil incubation was conducted for 9 months between 6 December, 2018 and 15 August, 2019. Approximately 150 g of air-dry soil were added to small plastic soil cups, 10.5-cm diameter by 4.5-cm tall. Two replicate soil cups were used for each soil-fertilizer combination for each of six sampling intervals: 0.5, 1, 2, 4, 6, and 9 months. Each soil cup replicate in each soil-fertilizer combination had one of seven fertilizer treatments applied to the cup, including a UC treatment that received no fertilizer addition. A total of 336 soil cups were prepared.

Based on the initial M3-extractable soil-test P concentrations, fertilizer-P sources were applied at a single rate recommended for row-crop production in Arkansas, which was determined to be 56 kg P₂O₅ ha⁻¹ (24.5 kg∙P∙ha⁻¹). After fertilizer was added to soil cups containing air-dry soil, each soil cup was manually shaken for 10 seconds to mix the fertilizer material into the soil. After mixing, soil cups were randomly placed on wooden shelving units located in a laboratory with lids with multiple holes in them placed loosely on top of each soil cup. Soils cups were rotated among shelves every two weeks during the incubation period. The average room air temperature and relative humidity throughout the 9-mo incubation were 21.6˚C and 56.5%.

Soil cups were watered periodically using identical methods as described in Anderson et al. [9] , in which a target gravimetric water content of 0.20 g∙g⁻¹ was used to determine the target mass for each soil after watering. Soil cups were initially watered on the day of fertilizer application and were gravimetrically rewatered to the same target mass every two weeks using tap water.

Over the 9-month period, soil cups were destructively sampled at the six sampling periods (i.e., 0.5, 1, 2, 4, 6, and 9 mo). When sampled, soil was removed from the cups, oven dried at 70˚C for at least 48 hours, mechanically crushed, and sieved through a 2-mm mesh screen. Soil subsamples were analyzed for WS and M3-extractable K, S, Na, Mn, Zn, Cu, and B. To determine WS concentrations, a 1:10 soil mass/water volume ratio was used. Mixtures were stirred for 1 hr before being filtered through a 0.45-μm filter and analyzed using inductively coupled, argon-plasma spectrometry (ICAPS; Spectro Arcos ICP, Spectro Analytical Instruments) [26] . To determine M3-extractable concentrations, a M3-extraction solution, 0.2 M acetic acid, 0.25 M ammonium nitrate, 0.015 M ammonium fluoride, 0.013 M nitric acid, and 0.001 M ethylenediaminetetraacetic acid [26] , with a 1:10 soil mass/extractant volume ratio was used before being analyzed by ICAPS [27] .

To determine changes in nutrient concentrations from their initial levels, each nutrient’s initial mean concentration in the soil was subtracted on a cup-by-cup basis from that measured at each sampling period. More details of the incubation experiment are described in Anderson et al. [12] .

Similar to Anderson et al. [12] , based on a split-split-plot, complete factorial, completely randomized design, a three-factor analysis of variance (ANOVA) was conducted to evaluate the effects of soil (i.e., L, SiCL, SiL 1, and SiL 2), fertilizer treatment (i.e., MAP, DAP, TSP, RP, CPST, ECST, and UC), sampling time (i.e., 0.5, 1, 2, 4, 6, and 9 mo), and their interactions on the change in WS and M3-extractable soil K, S, Na, Mn, Zn, Cu, and B from their initial values. The whole-plot factor was soil, the split-plot factor was fertilizer treatment, and split-split-plot factor was sampling time. Soil, fertilizer treatment, and sampling time were fixed effects. Replication was considered as a random effect. All data was analyzed according to a normal distribution. Significance was judged at P ≤ 0.05. When appropriate, means were separated by least significant difference at the 0.05 level.

The changes in all WS nutrient concentrations compared to the initial value were affected by soil, fertilizer treatment, and/or sampling time ( Table 3 ). The change in WS-S concentration from the initial differed (P < 0.01) among fertilizer treatments ( Table 3 ). The change in WS-K, -Mn, and -B concentrations from the initial differed (P < 0.01) among fertilizer treatments within soils ( Table 3 ). The change in WS-K, -S, -Na, -Mn, -Cu, and -B concentrations from the initial differed (P < 0.01) among soils over time ( Table 3 ). The change in WS-Mn and -Cu concentrations from the initial differed (P ≤ 0.04) among fertilizer treatments over time ( Table 3 ). The change in WS-Zn concentration from the initial differed (P = 0.02) among soil-fertilizer treatment combinations over time ( Table 3 ).

^†Monoammonium phosphate (MAP), diammonium phosphate (DAP), triple superphosphate (TSP), chemically precipitated struvite (CPST), electrochemically precipitated struvite (ECST), and rock phosphate (RP); ^‡Means in a row followed by different letters are different at P < 0.05; *An asterisk (*) indicates mean change is different than zero (P < 0.05).

Averaged across soils and sampling times, WS-S concentrations increased (P < 0.05) from the initial in all fertilizer treatments ( Table 4 ). The change in WS-S concentration was most positive from DAP and was least positive from RP, which did not differ from CPST, ECST, and the UC ( Table 4 ). The change in WS-S concentration did not differ between MAP and TSP, but both were intermediate and differed from DAP and CPST, ECST, RP, and the UC ( Table 4 ).

Averaged across sampling time, the change in WS-K concentrations from the initial was most positive for the loam soil with DAP, while the change in WS-K concentration was most negative for the SiCL soil in the UC ( Figure 1 ). The change in WS-K concentration did not differ from a change of zero for the loam soil with CPST, MAP, and ECST, the SiCL soil with CPST and MAP, the SiL 1 soil with CPST, MAP, and ECST, and the SiL 2 soil with RP, ECST, TSP, and the UC ( Figure 1 ). In contrast, the change in WS-Mn concentrations from the initial was most positive for the SiL 1 with DAP, while the change in WS-Mn concentration was numerically least positive for the SiL 2 soil with RP ( Figure 1 ). The change in WS-Mn concentration did not differ from a change of zero for the SiCL 2 soil with RP and the UC ( Figure 1 ). The change in WS-B concentrations from the initial was least negative for the SiL 2 soil in the UC, while the change in WS-B concentration was most negative for the loam soil with DAP ( Figure 1 ).

Averaged across fertilizer treatments, the change in WS-K concentrations from the initial was most positive for the SiCL soil after 9 months, while the change in WS-K concentration was most negative for the SiCL after 2 months of incubation ( Table 5 ). The change in WS-K concentration did not differ from a change of zero for the loam soil after 4, 6, and 9 months, the SiL 1 soil after 4 months, and the SiL 2 soil after 0.5 and 1 month of incubation ( Table 5 ). The change in WS-S concentrations from the initial was most positive for the SiL 1 after 9 months, while the change in WS-S concentration was least positive for the SiL 2 soil after 1 month of incubation ( Table 5 ). The change in WS-Na concentrations from the initial was most positive for the SiL 2 soil after 9 months, while the change in WS-Na concentration was least positive for the SiL 1 soil after 0.5 months of incubation ( Table 5 ). The change in WS-Mn concentrations from the initial was most positive for the SiL 1 soil after 9 months, while the change in WS-Mn concentration was least positive for the SiCL soil after 4 months of incubation ( Table 5 ). The change in WS-Mn concentration did not differ from a change of zero for the loam soil after 0.5, 1, and 2 months, the SiCL soil after 0.5, 1, 2, and 6 months, and the SiL 2 soil after 0.5, 1, 2, and 6 months of incubation ( Table 5 ). The change in WS-Cu concentrations from the initial was most positive for the loam, SiCL, SiL 1, and SiL 2 soils after 4 and 6 months, while the change in WS-Cu concentration was most negative for the loam, SiL 1, and SiL 2 soils after 2 months of incubation ( Table 5 ). The change in WS-Cu concentration did not differ from a change of zero for the loam soil after 9 months and the SiCL soil after 9 months of incubation ( Table 5 ). The change in WS-B concentrations from the initial was numerically most positive for the SiL 1 soil after 1 month and the SiL 2 soil after 0.5, 1, 2, 4, and 6 months, while the change in WS-B concentration was numerically most negative for the loam soil after 9 months of incubation ( Table 5 ). The change in WS-B concentration did not differ from a change of zero for the loam soil after 4, 6, and 9 months, the SiL 1 soil after 4 months, and the SiL 2 soil after 0.5 and 1 month of incubation ( Table 5 ).

Averaged across soils, the change in WS-Mn concentrations from the initial was most positive for DAP after 9 months, while the change in WS-Mn concentration was numerically least positive for the UC after 1 month of incubation ( Figure 2 ). The change in WS-Mn concentration did not differ from a change of zero for CPST, RP, ECST, TSP, and the UC after 0.5, 1, and 2 months or for DAP and MAP after 0.5 and 1 months of incubation ( Figure 2 ). The change in WS-Cu concentrations from the initial was the most positive for CPST after 6 months, while the change in WS-Cu concentration was most negative for MAP after 2 months of incubation ( Figure 2 ). The change in WS-Cu concentration did not differ from a change of zero for CPST, MAP, RP, TSP, and the UC after 9 months or for DAP after 1 and 9 months of incubation ( Figure 2 ).

^†Loam (L), silty clay loam (SiCL), silt loam (SiL); ^‡Means in a row followed by different letters are different at P < 0.05; *An asterisk (*) indicates mean change is different than zero (P < 0.05).

The combined effects of soil and fertilizer treatments over time on the change in WS-Zn concentrations from the initial were complex ( Table 6 ). The change in WS-Zn concentrations ranged from a low of -0.46 in the SiL 2-TSP after 9 months to a 1.27 in the SiCL-RP soil-fertilizer combination after 4 months of incubation. Changes in WS-Zn concentration with a magnitude of ± 0.19 mg∙kg⁻¹ were significantly different from a change of zero. After 0.5 months of incubation, all fertilizer treatments in the loam and all but TSP in the SiCL soil had WS-Zn concentrations that increased from the initial, while all fertilizers in the SiL 1 soil and all but ECST in the SiL 2 soil, which significantly decreased, had WS-Zn concentrations that did not change from the initial ( Table 6 ). After 9 months of incubation, all fertilizer treatments in the loam had WS-Zn concentrations that did not change from the initial, except for TSP and the UC which decreased from the initial, while all fertilizer treatments in the SiCL, SiL 1, and SiL 2 soils had WS-Zn concentrations that had decreased from the initial soil-Zn concentration ( Table 6 ).

Treatment effects on WS nutrients likely occurred for a variety of reasons. Averaged across soils and sampling times, WS-S concentrations increased for all fertilizer treatments relative to the initial concentration ( Table 4 ). The greater increases in WS-S concentration from the initial from MAP, DAP, and TSP may have been due to the large WS-S concentrations present in MAP, DAP, and TSP ( Table 1 ) [28] .

^†Fertilizer treatments (Treat) used in this incubation included monoammonium phosphate (MAP), diammonium phosphate (DAP), triple superphosphate (TSP), rock phosphate (RP), chemically precipitated struvite (CPST), electrochemically precipitated struvite (ECST), and an unamended control (UC); ^‡All means for soil-fertilizer treatment combinations followed by different letters are different at P < 0.05. Due to the large number of treatment combinations, letter notations started with a set of capital letters and continued to a set of lower-case letters if necessary; *An asterisk (*) indicates mean change is different than zero (P < 0.05).

Averaged across sampling times, WS-K concentrations varied among soil-fertilizer combinations. The increase from the initial in all soils with DAP ( Figure 1 ), may have been due to DAP having the largest initial total WS-K concentration among fertilizers ( Table 1 ). The greater decreases of WS-K concentrations from the initial in loam and SiCL soils with RP and TSP fertilizer treatments may have been due to loam and SiCL soils having the greatest initial K concentrations ( Table 2 ). The greater increase in WS-Mn concentrations from the initial for all fertilizer treatments in the SiL 1 soil was likely due to the SiL 1 soil having the largest initial WS-Mn concentration among soils ( Table 2 ), which could be due to the aquic properties of SiL 1. Water-soluble-B concentrations decreased from the initial for all soil-fertilizer combinations ( Figure 1 ), which may have been attributed to the increased WS-Ca concentrations for all soil-fertilizer combinations observed by Anderson et al. [12] , causing the formation of calcium borate complexes, which may have lowered the availability of B in the soils [29] .

Averaged across fertilizer treatments, WS-K concentrations had generally decreased from the initial after 4 months of incubation, at which point, for all soils, except the SiCL soil, WS-K concentration had increased from the initial, and, for all soils, WS-K concentrations continued to increase and had increased from the initial after 9 months of incubation ( Table 5 ). The behavior of WS-K may have been due to soil microbes taking up excess K at the beginning of the incubation period before dying and slowly releasing K back into the soil as microbes decomposed later during the incubation period. WS-S concentrations generally increased in all soils over time ( Table 5 ). This could be due to phosphate anions added from fertilizers replacing sulfate anions at anion adsorption sites, thus releasing sulfate ions into the soil solution [30] . The loam having a significantly lower WS-S concentration than the other three soils ( Table 5 ) may have been due to the loam having the lowest initial WS-S concentration ( Table 2 ). WS-Na and -Mn concentrations generally increased in all soils over time ( Table 5 ). The general increase in WS-Na and -Mn concentrations over time was likely due to the influx of cations from the dissolving fertilizers, which replaced Na and Mn ions on soil exchange sites [31] . WS-Cu concentrations had generally decreased slightly from the initial after 4 months of incubation, at which point WS-Cu concentrations had increased from the initial in all soils before decreasing slightly by after 9 months of incubation ( Table 5 ). At 9 months of incubation, WS-Cu concentrations did not differ from a change of zero in the loam or SiCL soils ( Table 5 ). There was no clear explanation for the behavior of WS-Cu over the 9-month period. The general decrease in WS-B concentrations ( Table 5 ) may have been due to the formation of calcium borate complexes associated with the increase in WS-Ca concentrations observed in all soil-fertilizer combinations [12] , which may have reduced soil-B availability [29] .

Averaged across soils, WS-Mn concentrations generally increased for all fertilizer treatments over time ( Figure 2 ). The larger increase in WS-Mn concentrations from the initial in the DAP treatment at most sampling times was unexpected since DAP had one of the lowest initial WS-Mn concentrations ( Table 1 ). Water-soluble-Cu concentrations had generally decreased from the initial after 4 months of incubation, at which point WS-Cu concentrations increased from the initial for all fertilizer treatments before generally decreasing after 9 months of incubation ( Figure 2 ). After 9 months of incubation, WS-Cu concentrations did not differ from a change of zero for all fertilizer treatments, except for ECST, which had increased from the initial ( Figure 2 ). There was no clear explanation for the behavior of WS-Cu over the 9-month period.

In general, though not formally compared, M3-extractable soil nutrient concentrations were greater than their respective WS nutrient concentrations, but the magnitude of M3 concentration changes from the initial was similar to those for WS nutrients. Similar to WS nutrients, the change in all M3-extractable nutrients evaluated in this study was affected by soil, fertilizer treatment, and/or sampling time ( Table 3 ). The change in M3-Na concentration from the initial differed (P = 0.03) among soils ( Table 3 ). The change in M3-S, -Mn, and -Cu concentrations from the initial differed (P ≤ 0.01) among fertilizer treatments ( Table 3 ). The change in M3-Na and -Zn concentrations from the initial differed (P ≤ 0.04) over time ( Table 3 ). The change in M3-Zn concentrations from the initial differed (P = 0.03) among fertilizer treatments within soils ( Table 3 ). The change in M3-K, -S, -Mn, -Cu, and -B concentration from the initial differed (P ≤ 0.03) among soils over time ( Table 3 ).

Averaged across soils and sampling times, M3-S, -Mn, and -Cu concentrations increased (P < 0.05) from the initial in all fertilizer treatments ( Table 4 ). The change in M3-S concentration was most positive from DAP (23.4) and was least positive from RP (11.5), which did not differ from CPST (11.6), ECST (13.0) and the UC (11.6; Table 4 ). The change in M3-S concentration did not differ between MAP (21.2) and TSP (20.9), but both were intermediate and differed from DAP and CPST, ECST, RP, and the UC ( Table 4 ). The change in M3-Mn concentration was most positive from DAP (70.6), which did not differ from RP (67.7) and the UC (68.0), and was least positive from ECST (64.4) and TSP (64.4), which did not differ from MAP (66.6), CPST (65.9), and RP ( Table 4 ). The change in M3-Cu concentration was most positive from RP (0.5) and was least positive from MAP (0.4), which did not differ from DAP (0.4), TSP (0.5), and ECST (0.5; Table 4 ). The change in M3-Cu concentration did not differ between CPST (0.5) and the UC (0.5), but both were intermediate and CPST differed from MAP and RP, while the UC differed from MAP, DAP, and RP ( Table 4 ).

Averaged across soils and fertilizer treatments, M3-Na concentration from the initial increased (P < 0.05) from the initial after 0.5, 6, and 9 months and decreased after 1 month of incubation ( Figure 3 ). The change in M3-Na concentration from the initial was most positive after 9 months and was most negative after 1 month of incubation ( Figure 3 ). The change in M3-Na concentrations differed from each other at each sampling time ( Figure 3 ). The change in M3-Zn concentration decreased from the initial (P < 0.05) at all sampling times ( Figure 3 ). The change in M3-Zn concentration was least negative after 6 months, which did not differ from 0.5, 2, 4, and 9 months, and was most negative after 1 month, which did not differ from 2 and 4 months of incubation ( Figure 3 ).

Averaged across sampling time, M3-Zn concentrations decreased (P > 0.05) from the initial only in the SiCL soil for all fertilizer treatments except RP ( Figure 1 ). The change in M3-Zn concentration was numerically most positive in the loam soil with ECST and was most negative in the SiCL soil with MAP ( Figure 1 ).

Averaged across fertilizer treatments, M3-K concentrations decreased (P > 0.05) from the initial for all soil-time combinations except for the loam soil after 0.5, 2, and 6 months of incubation ( Table 7 ). The change in M3-K concentration was numerically least negative for the loam soil after 6 months and was most negative for the SiCL soil after 4 months of incubation ( Table 7 ). The change in M3-S concentrations increased (P > 0.05) for all soil-time combinations ( Table 7 ). The change in M3-S concentration was most positive for the SiL 1 soil after 9 months and was least positive for the SiCL soil after 1 month of incubation ( Table 7 ). Similar to Mehlich-3-S, the change in M3-Mn concentrations increased (P > 0.05) for all soil-time combinations ( Table 7 ). The change in M3-Mn concentration was most positive for the SiL 2 soil after 9 months and was least positive for the loam soil after 1 month of incubation ( Table 7 ). The change in M3-Cu concentration was most positive for the SiCL soil after 6 months and was most negative for the SiL 2 soil after 1 and 9 months of incubation ( Table 7 ). The change in M3-Cu concentration did not differ from a change of zero for the loam soil after 0.5, 4, and 6 months, the SiL 1 soil after 0.5 and 6 months, and the SiL 2 soil after 0.5 months of incubation ( Table 7 ). In contrast to M3-S and -Mn, the change in M3-B concentrations decreased (P > 0.05) for all soil-time combinations ( Table 7 ). The change in M3-B concentration was least negative for the loam soil after 9 months and was most negative for the SiCL soil after 0.5, 1, and 2 months of incubation ( Table 7 ).

^†Loam (L), silty clay loam (SiCL), silt loam (SiL); ^‡Means in a soil property followed by different letters are different at P < 0.05; *An asterisk (*) indicates mean change is different than zero (P < 0.05).

Similar to WS nutrients, treatment effects on M3 nutrients occurred for a variety of reasons. Averaged across fertilizer treatments and sampling times, the greater increase in M3-Na concentration from the initial in the loam was unexpected since the initial M3-Na concentration was lowest in the loam ( Table 2 ).

Averaged across soils and sampling times, M3-S concentrations behaved similarly to WS-S concentrations ( Table 4 ). The large changes in M3-S concentration from the initial from MAP, DAP, and TSP may have been due to the large initial M3-S concentrations present in MAP, DAP, and TSP ( Table 1 ; Camberato et al., 2023). Similar to S, M3-Mn concentrations increased from the initial for all fertilizer treatments ( Table 4 ). The increase was greatest for DAP, which was unexpected since DAP had a relatively low initial M3-Mn concentration ( Table 1 ). This behavior of M3-Mn was similar to that of WS-Mn among fertilizer treatments over time ( Figure 2 ). Similar to S and Mn, M3-Cu concentrations increased from the initial for all fertilizer treatments ( Table 4 ). The increase was greatest for RP, which was also unexpected since RP had a relatively low initial M3-Cu concentration ( Table 1 ).

Averaged across soils and fertilizer treatments, M3-Zn concentrations decreased from the initial at all sampling times ( Figure 3 ). Mehlich-3-Zn concentrations remained relatively constant and did not differ at any sampling time, except for after 1 month, at which point the change in M3-Zn concentrations was more negative than at any other sampling time ( Figure 3 ). This decrease could be caused by the formation of zinc phosphate, which is relatively insoluble and could reduce the availability of Zn. Averaged across sampling times, the significant decrease in M3-Zn concentration in the SiCL soil ( Figure 1 ) may have been due to the SiCL soil having the greatest initial M3-Zn concentration ( Table 2 ).

Averaged across fertilizer treatments, M3-K concentrations decreased from the initial in all soils over time, where M3-K concentrations decreased from the initial after 0.5 months of incubation, and then remained somewhat constant, with no obvious temporal trend thereafter ( Table 7 ). The decrease in M3-K concentration was greatest in the SiCL, which may have been due to SiCL having the greatest initial M3-K concentration ( Table 2 ). Mehlich-3-S concentrations generally increased from the initial in all soils over time ( Table 7 ), which may have been due to phosphate ions from fertilizers replacing sulfate ions at soil adsorption sites. By 9 months of incubation, the change in M3-S concentrations was 4.5 to 6 times greater than at 0.5 months of incubation ( Table 7 ). Mehlich-3-Mn concentrations generally increased in all soils over time ( Table 7 ). By 9 months of incubation, the change in M3-Mn concentrations was approximately two times greater than at 0.5 months of incubation ( Table 7 ). The change in M3-Cu concentrations from the initial differed between soils over time ( Table 7 ). In general, M3-Cu concentrations decreased from the initial in the loam, SiL 1, and SiL 2 soils, but increased from the initial in the SiCL ( Table 7 ). The general decrease in M3-B concentrations ( Table 7 ) may be due to the formation of calcium borate complexes associated with the increase in Ca concentrations observed in all soil-fertilizer combinations [12] , which likely reduced the soil-B availability [29] .

Regarding the use of ECST as an alternative P-fertilizer and despite the experiment being conducted in the absence of plants, it is unlikely that there would be any negative effects on soil quality or crop growth from unintended additions and accumulation of micronutrients and potassium. The results of this study indicate that ECST had similar effects on soil micronutrients and potassium as commercially available CPST and traditional fertilizer-P sources (i.e., DAP, MAP, and TSP) in multiple soil textures with a history of agricultural management. The results of this study indicated that struvite fertilization would not create any additional concern for runoff water quality or nutrient leaching due to the similar effects of struvite fertilization on soil nutrient concentrations compared to other fertilizer-P sources. Results supported conclusions of Anderson et al. [12] and others [13] [17] - [23] that struvite materials could be a viable alternative fertilizer-P source. In addition, results of this study and those of Anderson et al. [12] further support the idea that struvite recovery could be an effective option to reduce P and N concentrations and recycle nutrients from wastewater effluent that would otherwise be released into surface waters, thus decreasing eutrophication. The benefits of struvite recovery seem to be extensive, but further research will be needed to understand how plants respond to struvite fertilization, and research needs to be done at a larger scale to understand the behavior of struvite fertilizers in a less-controlled environment.

Though a detailed assessment of the potential environmental fate of soil micronutrients associated with CPST and ESCT has not been conducted, Simms et al. [31] evaluate the leaching potential of P, N, Ca, Mg, and Fe across various soils. Simms et al. [31] concluded that macronutrient leachate concentrations from the top 10 cm varied among soil textures, fertilizer-P sources, and initial soil pH, but that temporal effects were complex. It stands to reason that, in lower concentrations than macronutrients, micronutrients added or rendered more mobile by the addition of struvite-P fertilizer materials likely pose little environmental threat.