Enhancing pedotransfer functions with environmental data for estimating bulk density and effective cation exchange capacity in a data‐sparse situation

Soil bulk density (BD) and effective cation exchange capacity (ECEC) are among the most important soil properties required for crop growth and environmental management. This study aimed to explore the combination of soil and environmental data in developing pedotransfer functions (PTFs) for BD and ECEC. Multiple linear regression (MLR) and random forest model (RFM) were employed in developing PTFs using three different data sets: soil data (PTF‐1), environmental data (PTF‐2) and the combination of soil and environmental data (PTF‐3). In developing the PTFs, three depth increments were also considered: all depth, topsoil (<0.40 m) and subsoil (>0.40 m). Results showed that PTF‐3 (R²; 0.29–0.69) outperformed both PTF‐1 (R²; 0.11–0.18) and PTF‐2 (R²; 0.22–0.59) in BD estimation. However, for ECEC estimation, PTF‐3 (R²; 0.61–0.86) performed comparably as PTF‐1 (R²; 0.58–0.76) with both PTFs out‐performing PTF‐2 (R²; 0.30–0.71). Also, grouping of data into different soil depth increments improves the estimation of BD with PTFs (especially PTF‐2 and PTF‐3) performing better at subsoils than topsoils. Generally, the most important predictors of BD are sand, silt, elevation, rainfall, temperature for estimation at topsoil while EVI, elevation, temperature and clay are the most important BD predictors in the subsoil. Also, clay, sand, pH, rainfall and SOC are the most important predictors of ECEC in the topsoil while pH, sand, clay, temperature and rainfall are the most important predictors of ECEC in the subsoil. Findings are important for overcoming the challenges of building national soil databases for large‐scale modelling in most data‐sparse countries, especially in the sub‐Saharan Africa (SSA).     

Pedotransfer functions for the pool size of slowly mineralizable organic N in sandy arable soils

The major aim of this study was to evaluate how the pool size of slowly mineralizable, ‘old’ soil organic N can be derived from more easily accessible soil and site information via pedotransfer functions (PTF). Besides modeling, this pool size might be of great importance for the identification of soils with high mineralization potential in drinking‐water catchments. From long‐term laboratory incubations (ca. 200 days) at 35 °C, the pool sizes of easily mineralizable organic N (N_fast), mainly in fresh residues, and slowly mineralizable, ‘old’ soil organic N (N_slow) as well as their first‐order rate coefficients were obtained. 90 sandy arable soils from NW Germany served to derive PTFs for N_slow that were evaluated using another 20 soils from the same region. Information on former land‐use and soil type was obtained from topographical, historical, and soil maps (partly from 1780). Pool size N_slow very strongly depends on soil type and former land‐use. Mean pool sizes of N_slow were much lower in old arable lowland (105 mg N kg^–1) than upland soils (175 mg N kg^–1) possibly due to lower clay contents. Within lowlands, mean pool sizes in former grassland soils (245 mg N kg^–1) were 2 to 3 times larger than in old arable soils due to accumulation of mineralizable N. In contrast, mean pool sizes of N_slow were lowest in recently cleared, former heath‐ and woodland (31 mg N kg^–1) as a result of the input of hardly decomposable organic matter. Neither N nor C in the light fraction (density < 1.8 g cm^–3) was adequate to derive pool size N_slow in the studied soils (r² < 0.03). Instead, N_slow can be accurately (r² = 0.55 – 0.83) derived from one or two basic soil characteristics (e.g. organic C, total N, C : N, mineral fraction < 20 μm), provided that sites were grouped by former land‐use. Field mineralization from N_slow during winter (independent data set) can be predicted as well on the basis of N_slow‐values calculated from PTFs that were derived after grouping the soils by former land‐use (r² = 0.51***). In contrast, using the PTF without soil grouping strongly reduced the reliability (r² = 0.16).     

基于土壤传递函数的土壤持水特性间接估计及其空间变异性研究

A total of 107 soil samples were taken from the city of Qingdao,Shandong Province,China.Soil water retention data at 2.5,6,10,33,100,300,and 1 500 kPa matric potentials were measured using a pressure membrane apparatus.Multiple linear regression (MLR) was used to develop pedotransfer functions (PTFs) for single point estimation and van Genuchten parameter estimation based on readily measurable soil properties,i.e.,MLR-based point (MLRP) PTF and MLR-based parametric (MLRV) PTF.The double cross-validation method was used to evaluate the accuracy of PTF estimates and the stability of the PTFs developed in this study.The performance of MLRP and MLRV PTFs in estimating water contents at matric potentials of 10,33,and 1 500 kPa was compared with that of two existing PTFs,the Rawls PTF and the Vereecken PTF.In addition,geostatistical analyses were conducted to assess the capabilities of these PTFs in describing the spatial variability of soil water retention characteristics.Results showed that among all PTFs only the Vereecken PTF failed to accurately estimate water retention characteristics.Although the MLRP PTF can be used to predict retention characteristics through traditional statistical analyses,it failed to describe the spatial variability of soil water retention characteristics.Although the MLRV and Rawls PTFs failed to describe the spatial variability of water contents at a matric potential of 10 kPa,they can be used to quantify the spatial variability of water contents at matric potentials of 33 and 1 500 kPa.     

Availability of soil boron fractions to M.26 apple rootstock

The availability of various boron (B) fractions in soil to M.26 apple (Malus spp.) rootstock was examined. The study was carried out in a greenhouse on soils with diverse chemical and physical properties. The following B fractions were determined: (i) B in soil solution, (ii) B non‐specifically adsorbed on soil surface, (iii) B specifically adsorbed on soil colloid surfaces, (iv) B occluded in Mn oxyhydroxides, (v) B occluded in noncrystalline aluminum (Al) and iron (Fe) oxides, (vi) B occluded in crystalline Al and Fe oxides, (vii) B fixed with soil silicates, and (viii) total soil B. In the studied soils there were: 0.07–0.17 mg kg^‐1 B in soil solution, 0.01–0.03 mg kg^‐1 B non‐specifically adsorbed on soil surface, 0.04–0.08 mg kg^‐1 B specifically adsorbed on soil colloid surfaces, 0.28–0.67 mg kg^‐1 B occluded in manganese (Mn) oxides, 4.03–17.22 mg kg^‐1 B occluded in noncrystalline Al and Fe oxides, 8.93–50.62 mg kg^‐1 B occluded in crystalline Al and Fe oxides, 12.2–42.5 mg kg^‐1 B fixed with soil silicate, and 52.9–82.2 mg kg^‐1 total B. Simple correlation analysis showed positive correlation between B contents in M.26 apple rootstocks and amounts of B in soil solution (r=0.77), B non‐specifically adsorbed on soil colloid surfaces (r=0.65), B specifically adsorbed on soil surface (r=0.76) and B occluded in Mn oxyhydroxides (r=0.77). No relation was found between plant B contents and amounts of B occluded in non‐crystalline and crystalline Al and Fe oxides, B fixed with soil silicates and total B. The results indicated that extraction of B by 0.1 M NH₂OH HCl solution adequately represented amounts of B in soil solution, B non‐specifically and specifically adsorbed on soil compound surfaces and B occluded in Mn oxyhydroxides to assess availability of B to apple trees.     

Estimating the bulk density in 0– 20 cm of tilled soils in China’s Loess Plateau using support vector machine modeling

ABSTRACT

Pedotransfer functions (PTFs), as an indirect forecasting method, offer an alternative for labor-intensive bulk density (BD) measurements. In order to improve the forecasting accuracies, support vector machine (SVM) method was first used to develop PTFs for predicting BD. Cross-validation and grid-search methods were used to automatically determine the SVM parameters in the forecasting process. Soil texture and organic matter content were selected as input variables based on results of predecessors, coupled with gray correlation theory. And additional properties were added as inputs for improving PTF's accuracy and reliability. The performance of the PTF established by SVM method was compared with artificial neural network (ANN) method and published PTFs using two indexes: root-mean-square error (RMSE) and coefficient of determination(R²). Results showed that the average RMSE of published PTFs was 0.1053, and the R² was 0.4558. The RMSE of ANN–PTF was 0.0638, and the R² was 0.7235. The RMSE of SVM–PTF was 0.0558, and the R² was 0.7658. Apparently, the SVM–PTF had better performance, followed by ANN–PTF. Additionally, performances could be improved when accumulated receiving water was added as predictor variable. Therefore, the first application of SVM data mining techniques in the prediction of soil BD was successful, improved the accuracy of predictions, and enhanced the function of soil PTFs. The idea of developing PTFs using SVM method for predicting soil BD in the study area could provide a reference for other areas.     

The influence of iron oxides on phosphate adsorption by soil

Soils from Denmark and Tanzania were extracted with ammonium acetate (controls), EDTA to dissolve amorphous iron oxides, and dithionite-EDTA (DE) to dissolve crystalline iron oxides. The phosphate adsorption capacities of the extracted soils were taken as the maximum quantity of phosphate adsorbed computed from the Langmuir equation. The decreases in the phosphate adsorption capacity following EDTA extraction and DE extraction were attributed to the removal of iron oxides. Close correlations (P<0.001) were found (i) between EDTA-extractable iron (amorphous iron oxides) and the decrease in phosphate adsorption capacity following EDTA extraction, and (ii) between the difference between DE-extractable iron and EDTA-extractable iron (crystalline iron oxides) and the further decrease in phosphate adsorption capacity following DE extraction. The phosphate adsorption capacity, estimated to be approximately 2.5 μmol P m^?2, was in good agreement with the capacity of various synthetic iron oxides. The calculated phosphate adsorption capacity of soil iron oxides, obtained from the contents and specific surfaces of amorphous and crystalline iron oxides together with the phosphate adsorption capacity per m² for synthetic iron oxides, compared favourably with the measured phosphate adsorption capacity.     

Adsorption of cobalt by soil iron oxides at low solution concentration

Abstract

Eleven selected soils from Denmark and Tanzania were treated with ammonium acetate (controls), EDTA, and dithionite‐EDTA (DE) to fractionate iron and (manganese) oxides. The amounts of cobalt adsorbed were determined from a 3 μM equilibrium cobalt solution, corresponding to the cobalt level in natural soil solutions using sodium nitrate (0.2 M) to suppress non‐specific adsorption, by the extracted soils as well as by two synthetic iron oxides.

No significant correlations were found between cobalt adsorption and the contents of organic matter and extractable manganese, presumably due to their low contents in the soils investigated. Close correlations were, however, found between the amounts of cobalt adsorbed, especially fractions thereof, and the contents of iron oxides.

The amounts of cobalt adsorbed by the DE‐ex‐tracted soils, void of iron (and manganese) oxides, were attributed to the clay silicates. The remaining cobalt adsorption, i.e. the difference between cobalt adsorbed by acetate‐extracted and DE‐extracted samples, was attributed to the iron oxides. This portion of adsorbed cobalt was well described by considering soil iron oxides composed of only two fractions, an EDTA‐extractable fraction of high reactivity and a less reactive fraction corresponding to the difference between DE‐extractable iron and EDTA‐extractable iron.

The amounts of cobalt adsorbed by the soil iron oxides were well predicted from the contents and specific surface of the two iron oxide fractions in soil together with the amount of cobalt adsorbed by the synthetic iron oxides.     

Soil properties affecting solid–liquid distribution of As(V) in soils

The prediction of the mobility of arsenic (As) is crucial for predicting risks in soils contaminated with As. The objective of this study is to predict the distribution of As between solid and solution in soils based on soil properties and the fraction of As in soil that is reversibly adsorbed. We studied adsorption of As(V) in suspensions at radiotrace concentrations for 30 uncontaminated soils (pH 4.4–6.6). The solid–liquid distribution coefficient of As (K_d) varied from 14 to 4430 l kg^?1. The logarithm of the concentration of oxalate‐extractable Fe explained 63% of the variation in log K_d; by introducing the logarithm of the concentration of oxalate‐extractable P in the regression model, 85% of the variation in log K_d is explained. Double labelling experiments with ⁷³As(V) and ³²P(V) showed that the As to P adsorption selectivity coefficient decreased from 3.1 to 0.2 with increasing degree of P saturation of the amorphous oxides. The addition of As(V) (0–6 mmol kg^?1) reduced the K_d of ⁷³As up to 17‐fold, whereas corresponding additions of P(V) had smaller effects. These studies suggest that As(V) is adsorbed to amorphous oxides in soils and that sites of adsorption vary in their selectivity in respect of As and P. The concentration of isotopically exchangeable As in 27 contaminated soils (total As 13–1080 mg kg^?1) was between 1.2 and 19% (mean 8.2%) of its total concentration, illustrating that a major fraction of As is fixed. We propose a two‐site model of competitive As(V)–P(V) sorption in which amorphous Fe and Al oxides represent the site capacity and the isotopically exchangeable As represents the adsorbed phase. This model is fitted to ⁷³As adsorption data of uncontaminated soils and explains 69% of the variation of log K_d in these soils. The log K_d in contaminated soils predicted using this two‐site model correlated well with the observed log K_d (r = 0.75). We conclude that solubility of As is related to the available binding sites on amorphous oxides and to the fraction of As that is fixed.     

Influence of organic matter on phosphate adsorption by aluminium and iron oxides in sandy soils

The phosphate adsorption capacity (P_max) of samples from various horizons of five Danish podzolized soils were investigated before and after organic matter removal. Removal of organic matter had no direct influence on P_max suggesting that organic matter did not compete with phosphate for adsorption sites. In the soils investigated aluminium and iron oxides were the main phosphate adsorbents. Thus, more than 96% of the variation in P_max could be accounted for by poorly crystalline aluminium and iron oxides (extractable by oxalate) and by well-crystallized iron oxides (taken as the difference between dithionite-citrate-bicarbonate-extractable iron and oxalate-extractable iron). Organic matter affected phosphate adsorption indirectly by inhibiting aluminium oxide crystallization. The resulting poorly crystalline oxides had high P_max. In contrast, the influence of organic matter on the crystallinity of the iron oxides, and therefore on their capacity to adsorb phosphate, seemed limited.     

Performance of pedotransfer functions in predicting soil water characteristics of soils in Norway

Abstract

Pedotransfer functions (PTFs), predicting the soil water retention curve (SWRC) from basic soil physical properties, need to be validated on arable soils in Norway. In this study we compared the performance of PTFs developed by Riley (1996), Rawls and Brakensiek (1989), Vereecken et al. (1989), Wösten et al. (1999) and Schaap et al. (2001). We compared SWRCs calculated using textural composition, organic matter content (SOM) and bulk density as input to these PTFs to pairs of measured water content and matric potential. The measured SWRCs and PTF input data were from 540 soil horizons on agricultural land in Norway. We used various statistical indicators to evaluate the PTFs, including an integrated index by Donatelli et al. (2004). The Riley PTFs showed good overall performance. The soil specific version of Riley is preferred over the layer specific, as the latter may introduce a negative change in water content with increasing matric potential (h). Among the parameter PTFs, Wösten's continuous PTF showed the overall best performance, closely followed by Rawls&B and Vereecken. The ANN-based continuous PTF of Schaap showed poorer performance than its regression based counterparts. Systematic errors related to both particle size and SOM caused the class PTFs to perform poorly; these PTFs do not use SOM as input, and are therefore inappropriate for soils in Norway, being highly variable in SOM. The PTF performance showed little difference between soil groups. Water contents in the dry range of the SWRC were generally better predicted than water contents in the wet range. Pedotransfer functions that included both SOM and measured bulk density as input, i.e. Wösten, Vereecken and Rawls&B, performed best in the wet range.     

磷酸盐吸附对可变电荷土壤正负电荷的影响

本文研究了华南地区不同类型的可变电荷土壤，并对磷酸盐的吸附量和吸附磷后土壤的正、负电荷的变化，以及ｐＨ和游离氧化铁对这种变化的影响进行了研究。结果表明，土壤吸磷量与土壤游离氧化铁含量成良好的正相关。土壤吸磷后正电荷减少，负电荷增加，土壤电荷量与吸磷量之间呈抛物线状相关。吸附１摩尔磷酸盐对土壤净负电荷的贡献在０．３－１．０摩尔之间。土壤中的游离氧化铁使吸附的磷对土壤负电荷的贡献减少。     

Changes in distribution of inorganic soil phosphorus forms with phosphate desorption by iron oxide‐impregnated paper strips

Abstract

The release of soil phosphorus (P) to solution has been described by extraction of soil with iron (Fe)‐oxide coated paper strips. Little information is available, however, on where this P is coming from. The effect of removal of reversibly adsorbed soil P on the distribution of inorganic P forms was investigated for 12 Italian soils. Phosphate was removed from these soils by Fe‐oxide strips after incubation with P (0 and 100 mg P kg^‐1) for 90 days. With no applied P, 3 to 17% of the total soil active P [saloid‐P, aluminum‐phosphate (Al‐P), iron‐phosphate (Fe‐P), and calcium‐phosphate (Ca‐P) was removed by the Fe‐oxide strips. The change in strip‐P following P addition (100 mg kg^‐1 soil), ranged from 12.9 to 53.5 mg P kg^‐1, with P coming almost entirely from the active P fractions. A close relationship between the changes in desorbed strip‐P after P equilibration and soil P sorption index (SI) was found for the studied soils (r²=0.96). Thus, the release of soil P for plant uptake or transport in runoff was a function of the amount of “actively”; sorbed P and an estimate of P sorption.     

Estimating the water retention curve from soil properties: comparison of linear, nonlinear and concomitant variable methods

The unsaturated soil hydraulic functions involving the soil–water retention curve (SWRC) and the hydraulic conductivity provide useful integrated indices of soil quality. Existing and newly devised methods were used to formulate pedotransfer functions (PTFs) that predict the SWRC from readily available soil data. The PTFs were calibrated using a large soils database from Hungary. The database contains measured soil–water retention data, the dry bulk density, sand, silt and clay percentages, and the organic matter content of 305 soil layers from some 80 soil profiles. A three-parameter van Genuchten type function was fitted to the measured retention data to obtain SWRC parameters for each soil sample in the database. Using a quasi-random procedure, the database was divided into “evaluation” (EVAL) and “test” (TEST) parts containing 225 and 80 soil samples, respectively. Linear PTFs for the SWRC parameters were calculated for the EVAL database. The PTFs used for this purpose particle-size percentages, dry bulk density, organic matter content, and the sand/silt ratio, as well as simple transforms (such as logarithms and products) of these independent variables. Of the various independent variables, the eight most significant were used to calculate the different PTFs. A nonlinear (NL) predictive method was obtained by substituting the linear PTFs directly into the SWRC equation, and subsequently adjusting the PTF parameters to all retention data of the EVAL database. The estimation error (SSQ) and efficiency (EE) were used to compare the effectiveness of the linear and nonlinearly adjusted PTFs. We found that EE of the EVAL and the TEST databases increased by 4 and 7%, respectively, using the second nonlinear optimization approach. To further increase EE, one measured retention data point was used as an additional (concomitant) variable in the PTFs. Using the 20 kPa water retention data point in the linear PTFs improved the EE by about 25% for the TEST data set. Nonlinear adjustment of the concomitant variable PTF using the 20 kPa retention data point as concomitant variable produced the best PTF. This PTF produced EE values of 93 and 88% for the EVAL and TEST soil data sets, respectively.     

Potential Iron and Phosphate Mobilization During Flooding of Soil Material

This study shows that mobilization of phosphate from soils under anaerobic conditions can be intimately coupled with reductive dissolution of iron from iron oxides. Among four soil samples from the reclaimed Skjernå estuary in Denmark incubated anaerobically and amended with glucose, 28–39% of the dithionite-citrate-bicarbonate-extractable iron and 10–25% of the oxalate-extractable phosphorus (P_ox) were released to the soil solution after 31 days. Significant correlation (r = 0.992**) between the molar ratio P_ox/(Fe_ox + Al_ox) for the aerobic samples and (P_{P sol}/Fe_sol) (the molar ratio between phosphate and iron in solution during anaerobic incubation), indicates that the phosphate saturation status of the soil is an important determinant of the amount of phosphate released during flooding of moderately acid soils.     

Calcium‐ and iron‐related phosphorus in calcareous and calcareous marsh soils: Sequential chemical fractionation and 31p nuclear magnetic resonance study

Abstract

Phosphorus (P) forms in soils determine the amount of P available for crops and the potential for this element to be released to water. Sequential chemical fractionation can provide some information about major P forms in soils, and allow one to distinguish iron (Fe)‐related phosphorus from calcium (Ca)‐bound P. The ³¹P nuclear magnetic resonance (NMR) spectroscopy has been used in the identification of organic P, precipitated Ca‐phosphates, and aluminum (Al)‐related P in acid soils. Three calcareous soils and four calcareous marsh soils were used in this study. These two types of soils differ in the nature of iron oxides, which are the main P sorbent surfaces. The ratio of low crystalline to high crystalline iron oxides is higher in marsh soils than in calcareous soils as a consequence of the special genesis and conditions of the soil (reduction‐oxidation cycles). Such a ratio is related to the proportion of occluded P in low crystalline oxides relative to that of high crystalline oxides. Citrate‐bicarbonate extractable P (CB‐P) in the fractionation schemes can be ascribed to adsorbed P and high soluble calcium phosphates. CB‐P is correlated with the sum of P fractions in all the soils, thus indicating that the amount of the P that can be easily released is related to the rate of P enrichment of the soil. The ³¹P NMR spectral data reveal that hydroxyapatite is the dominant P form in the soils studied. This is consistent with the fractionation data, where acid‐extractable P is the main P fraction. The spectra also provide some information about the amount of total inorganic P and Ca‐phosphates in calcareous soils.     

Assessment of the usefulness of particle size distribution measured by laser diffraction for soil water retention modelling

Particle size distribution (PSD) is a major soil characteristic, which is essential and commonly used for the development of pedotransfer functions (PTFs) to estimate the water retention of soils. The laser diffraction method (LDM) became a popular alternative to the standard sieve‐hydrometer method (SHM) of PSD measurement. Unfortunately, PSDs determined with LDM and SHM methods differ sometimes substantially. Moreover, it is claimed that the laser diffraction method underestimates finer fractions in favor of coarser fractions. Several authors have tried to elaborate on methods to recalculate LDM PSD into its SHM counterparts, but no universal methodology has been developed to this date. In this paper, we use PSD determined by LDM directly for PTF development and compare it with the classical PTF approach based on PSD measured by SHM. Four different PTF models based on LDM particle size distribution data were developed, with different PSD characteristics taken as the models' input variables. The possibility of using alternative PSD characteristics, such as deciles, area moment mean and volume moment mean, for PTF development was examined. The accuracy of PTF models constructed on the basis of LDM‐measured PSD was comparable with that of the developed models using texture data obtained from SHM, giving approximately the same RMSE and R² values. Our study shows that LDM‐measured particle size distribution may be directly used for PTF developments without any recalculations to their sieve‐hydrometer counterparts.     

Soil properties related to phosphate buffering in calcareous soils

Abstract

In a group of 24 related calcareous soils, derived from Jurassic oolitic limestone, there was marked variability (13‐fold) in phosphate buffering when expressed as the maximum buffer capacity. This variability was most closely related to the iron content and pH of the soils, and these together accounted for 72% of the variance. This percentage was not increased by including CaC0₃ content or organic matter, which were also correlated with the maximum buffer capacity. A high correlation with specific surface area of CaCO₃ was probably an indirect effect due to the high correlation between this variable and the Fe and pH of the soils.

The equilibrium buffer capacity, which is the traditional measure of phosphate buffering, was less variable but quite unrelated to all the soil properties measured except the soil surface area. However the maximum buffer capacity and quantity of adsorbed P together accounted for 63% of the variance in this parameter.     

Dissolution of Poorly Crystalline Iron Oxides in Soils by EDTA and Oxalate

Poorly crystalline iron oxides in soils are often estimated by 2 hours oxalate extraction at pH 3 and less often by 3–7 months EDTA extraction at pH 7.5–10.5. Calculated solubility products (Ksp) of iron oxides in equilibrium with EDTA and oxalate showed EDTA to dissolve only iron oxides with Ksp > 10^?40-10^?41 at pH > 10, whereas at pH 3 oxalate (and EDTA) should theoretically dissolve all iron oxides. The different pHs could largely account for the great difference in extraction speed between the two methods. Although EDTA and oxalate seem to act by surface complexation, where the adsorbed ligand by attenuating lattice Fe-O bonds causes iron detachment, the mechanisms are considered to be different. Possibly EDTA forms tetranuclear surface complexes, which are considered to inhibit dissolution of well crystallized but not poorly crystallized iron oxides due to differences in bond strengths. Oxalate forming binuclear and mononuclear surface complexes can probably also act as an electron bridge between iron(II) in solution and surface iron(III) leading to iron(II) catalyzed dissolution of iron oxides. This mechanism is obviously of particular importance in the dissolution of magnetite and maghemite. Despite the great theoretical differences the published methods with EDTA and oxalate dissolve comparable amounts of iron from many soils and the dissolved iron corresponds to poorly crystalline (highly reactive) iron oxides, mainly ferrihydrite.