Reducing adverse side effects by seasonally lowering nitrate removal in subsurface flow constructed wetlands.

Subsurface flow constructed wetlands with wood chips (SSF-CWs) have proven to effectively reduce the loss of nitrogen (N) from agricultural fields to surface water, however in some cases production of negative side effects such as methane and phosphate occur. We examined if these side effects can be avoided by decreasing the hydraulic retention time (HRT) from on average 82 h to 11 h during summer to autumn in two pilot SSF-CWs. Furthermore, we investigated the potential of the SSF-CWs to reduce phosphorus (P) loss from agricultural drainage systems. The influent and effluent concentration of total N (TN), nitrate-N, total P, phosphate-P, suspended sediment, and sulphate were monitored for five years (2013-2017). Methane concentrations were measured during two periods in 2014 and 2017. Flow was measured continuously by electromagnetic flow-meters. The nitrate-N removal was reduced from 98-100% to 27-32% and the sulphate reduction from 32-53% to 1-2% when decreasing HRT. Concurrently this resulted in a considerable decrease in the difference between the effluent and influent concentration of phosphate-P and methane concentration compared to similar periods in the preceding years. The SSF-CWs retained 67-85% of the annual loading of particulate P, but acted as both a sink and source of phosphate-P, thus further initiatives are therefore required to prevent phosphate-P release from SSF-CWs. Although during the entire monitoring period the SSF-CWs retained 29-33% of the total P loading. In summary, this study stresses how important a holistic approach is when implementing and designing new N mitigation measures.

Heavy Metal Leaching as Affected by Long-Time Organic Waste Fertilizer Application.

The recycling of urban waste products as fertilizers in agriculture may introduce contaminants such as heavy metals into soil that may leach and contaminate groundwater. In the present study, we investigated the leaching of heavy metals from intact soil cores collected in the long-term agricultural field trial CRUCIAL. At the time of sampling, the equivalent of >100 yr of urban waste fertilizers following Danish legislation had been applied. The leaching of Cu was significantly increased in the treatments receiving organic waste products compared with the unfertilized control but remained below the permissible level following Danish drinking water guidelines. The leaching of Cu was controlled primarily by the topsoil Cu content and by the leaching of dissolved organic carbon (DOC) but at the same time significantly correlated with leaching of colloids in soils that had not received fertilizer or had received an organic fertilizer with a low concentration of Cu. The leaching of Zn, Cd, and Co was not significantly increased in urban waste-fertilized treatments. The leaching of Mo was elevated in accelerated waste treatments (both agricultural and urban), and the leaching of Mo was linked to the leaching of DOC. Since leaching of Cr and Pb was strongly linked to the level of colloid leaching, leaching of these metals was reduced in the urban waste treatments. Overall, the results presented should not raise concern regarding the agricultural use of urban waste products in agriculture as long as the relevant guidelines are followed.

Long-term Effects of Organic Waste Fertilizers on Soil Structure, Tracer Transport, and Leaching of Colloids.

Organic waste fertilizers have previously been observed to significantly affect soil organic carbon (SOC) content and soil structure. However, the effect of organic waste fertilizers on colloid dispersibility and leaching of colloids from topsoil has not yet been studied extensively. We investigated how the repeated application of different types of agricultural (liquid cattle slurry and solid cattle manure) and urban waste fertilizers (sewage sludge and composted organic household waste) affected soil physical properties, colloid dispersion from aggregates, tracer transport, and colloid leaching from intact soil cores. Total porosity was positively correlated with SOC content. Yearly applications of sewage sludge increased absolute microporosity (pores <30 µm) and decreased relative macroporosity (pores >30 µm) compared with the unfertilized control, whereas organic household waste compost fertilization increased both total porosity and the absolute porosity in all pore size classes (though not significant for 100-600 µm). Treatments receiving large amounts of organic fertilizers exhibited significantly lower levels of dispersible colloids compared with an unfertilized control and a treatment that had received moderate applications of cattle slurry. The content of water-dispersible colloids could not be explained by a single factor, but differences in SOC content, electrical conductivity, and sodium adsorption ratio were important factors. Moreover, we found that the fertilizer treatments did not significantly affect the solute transport properties of the topsoil. Finally, we found that the leaching of soil colloids was significantly decreased in treatments that had received large amounts of organic waste fertilizers, and we ascribe this primarily to treatment-induced differences in effluent electrical conductivity during leaching.

Solute Transport Properties of Fen Peat Differing in Organic Matter Content.

There is a limited understanding of solute transport properties of degraded peat soils as compared to mineral substrates. A lower organic matter (OM) content is often the result of peat degradation and mineralization following artificial drainage. In this study, we aimed at deducing changes in solute transport properties of peat soils differing in OM content. Miscible displacement experiments were conducted on 70 undisturbed soil columns with OM contents ranging from 11 to 86% w/w under saturated steady-state conditions using tritium and bromide as conservative tracers. Measured breakthrough curves (BTCs) were subjected to model analysis using three different approaches: single-porosity model (SPM), mobile-immobile model (MIM), and two-flow region model (TFRM). The results indicated that (i) nonequilibrium solute transport processes are common in peat soils; (ii) the TFRM improved predictions of BTCs with heavy tailing or two peaks; (iii) applied tracers, tritium and bromide, were retarded in peat soils with higher OM content; and (iv) pronounced preferential flow mainly occurred in peat soils with lower OM content. This type of strong preferential flow had a small ratio of measured to fitted pore water velocity and a greater ratio of velocities (/ > 3.0) in the fast and slow transport region as obtained from the TFRM. We conclude that shallow groundwater resources are more likely to become polluted in drained and degraded fen peats that are used for agricultural purposes.

Non-equilibrium model for solute transport in differently designed biofilters targeting agricultural drainage water.

Biogeochemical processes in subsurface flow constructed wetlands are influenced by flow direction, degree of saturation and influent loading position. This study presents a simulation tool, which aims to predict the performance of the unit and improve the design. The model was developed using the HYDRUS program, calibrated and verified on previously measured bromide (Br-) pulse tracer tests. Three different hydraulic designs (Horizontal (HF), Vertical upward (VF-up), Vertical downward (VF-down) and two different flow rates: Low (L), and High (H)) were investigated. The model simulated well the Br- transport behaviour and the results underline the importance of the hydraulic design. Calibrated model parameters (longitudinal dispersivity, immobile liquid phase, mass transfer coefficient) showed a common trend for all the designs, for increasing flow rates within the investigated range. The VF-down performed best, i.e. had the highest hydraulic retention time.

Nitrogen Removal in Permeable Woodchip Filters Affected by Hydraulic Loading Rate and Woodchip Ratio.

Unregulated and event-driven agricultural tile drainage discharge poses several challenges that potentially limit the nitrate (NO) removal performance of woodchip-based wetlands constructed to intercept subsurface tile drain flows. Laboratory column tests were conducted to evaluate the biogeochemical response of mixed reactive media (woodchips-seashells and woodchips-Filtralite mixtures) at two woodchip ratios to changes in hydraulic loading rate (HLR). The tests involved continuous loading of aerated artificial drainage water spiked with NO-N and tritium (HO) breakthrough experiments. Flow-normalized NO reduction rates ranged from 0.35 to 3.97 g N m L, corresponding to N removal efficiencies of 5 to 64%, depending on HLR and filter mixtures. At high HLRs, oxic conditions prevailed in the woodchip filters, resulting in reduced N removal. At low HLRs, progressively lower pore-water velocities extended the period for consumption of terminal electron acceptors, increasing N removal. When increasing the content of mineral material, N removal declined, probably due to a lower denitrifying biomass at lower woodchip mass. The effect of woodchip ratios on solute transport characteristics was difficult to assess. However, woodchip media including a mineral fraction of crushed seashells demonstrated the highest N removal rates and efficiencies, most likely due to the alkalizing effect of the seashells. In conclusion, filter mixtures consisting of woodchips and seashells were the most effective material for N removal in subsurface flow-constructed wetlands treating agricultural drainage water.

Long-term phosphorus removal by Ca and Fe-rich drainage filter materials under variable flow and inlet concentrations.

Phosphorus (P) losses from tile-drained agricultural fields may degrade surface water quality by accelerating eutrophication. Among the different edge-of-field technologies, compact filter systems using different filter materials have been identified as potentially effective solutions for removing P from drainage water before discharge downstream. This study investigated the long-term (>696 days) P removal efficiency of 5 different filter materials in a column setup, using artificial drainage water (pH 6). Filter materials included two iron-based granulates (calcinated diatomaceous earth (CDE), ferric hydroxide granules (CFH)), and three calcium-based granulates (seashells, limestone, calcinated silicate/calcium oxide (Filtralite-P)). Experiments were performed under variable flow rates (0.037 and 1.52 L h-1; hydraulic retention time of 26-43 min and 18-30 h) and inlet P concentrations (0.14 and 0.7 mg L-1). An overall analysis revealed that the Fe-based materials achieved higher P retention than Ca-based materials. In particular, CFH was capable of retaining 99 and 98 % of the high and low inlet P concentrations, respectively. Conversely, limestone retained only 25 % of the high P load. CDE performed moderately well, independently of the inlet P concentration. Filtralite-P and Seashells performed well at high inlet P concentration but relatively poorly at low P concentration. The sensitivity of filter material P removal efficiency to variations in P loading was generally lowest for CFH and highest for limestone.

Vivianite precipitation and phosphate sorption following iron reduction in anoxic soils.

Phosphorus retention in lowland soils depends on redox conditions. The aim of this study was to evaluate how the Fe(III) reduction degree affects phosphate adsorption and precipitation. Two similarly P-saturated, ferric Fe-rich lowland soils, a sandy and a peat soil, were incubated under anaerobic conditions. Mössbauer spectroscopy demonstrated that Fe(III) in the sandy soil was present as goethite and phyllosilicates, whereas Fe(III) in the peat soil was mainly present as polynuclear, Fe-humic complexes. Following anoxic incubation, extensive formation of Fe(II) in the solids occurred. After 100 d, the Fe(II) production reached its maximum and 34% of the citrate-bicarbonate-dithionite extractable Fe (Fe(CBD)) was reduced to Fe(II) in the sandy soil. The peat soil showed a much faster reduction of Fe(III) and the maximum reduction of 89% of Fe(CBD) was reached after 200 d. Neoformation of a metavivianite/vivianite phase under anoxic conditions was identified by X-ray diffraction in the peat. The sandy soil exhibited small changes in the point of zero net sorption (EPC0) and P(i) desorption with increasing Fe(III) reduction, whereas in the peat soil P desorption increased from 80 to 3100 µmol kg⁻¹ and EPC0 increased from 1.7 to 83 µM, after 322 d of anoxic incubation. The fast Fe(III) reduction made the peat soils particularly vulnerable to changes in redox conditions. However, the precipitation of vivianite/metavivianite minerals may control soluble P(i) concentrations to between 2 and 3 µM in the long term if the soil is not disturbed.

Interactions between soil texture and placement of dairy slurry application: I. Flow characteristics and leaching of nonreactive components.

Land application of manure can exacerbate nutrient and contaminant transfers to the aquatic environment. This study examined the effect of injecting a dairy cattle (Bostaurus L.) manure slurry on mobilization and leaching of dissolved, nonreactive slurry components across a range of agricultural soils. We compared leaching of slurry-applied bromide through intact soil columns (20 cm diam., 20 cm high) of differing textures following surface application or injection of slurry. The volumetric fraction of soil pores >30 microm ranged from 43% in a loamy sand to 28% in a sandy loam and 15% in a loam-textured soil. Smaller active flow volumes and higher proportions of preferential flow were observed with increasing soil clay content. Injection of slurry in the loam soil significantly enhanced diffusion of applied bromide into the large fraction of small pores compared with surface application. The resulting physical protection against leaching of bromide was reflected by 60.2% of the bromide tracer was recovered in the effluent after injection, compared with 80.6% recovery after surface application. No effect of slurry injection was observed in the loamy sand and sandy loam soils. Our findings point to soil texture as an important factor influencing leaching of dissolved, nonreactive slurry components in soils amended with manure slurry.

Interactions between soil texture and placement of dairy slurry application: II. Leaching of phosphorus forms.

Managing phosphorus (P) losses in soil leachate folllowing land application of manure is key to curbing eutrophication in many regions. We compared P leaching from columns of variably textured, intact soils (20 cm diam., 20 cm high) subjected to surface application or injection of dairy cattle (Bos taurus L.) manure slurry. Surface application of slurry increased P leaching losses relative to baseline losses, but losses declined with increasing active flow volume. After elution of one pore volume, leaching averaged 0.54 kg P ha(-1) from the loam, 0.38 kg P ha(-1) from the sandy loam, and 0.22 kg P ha(-1) from the loamy sand following surface application. Injection decreased leaching of all P forms compared with surface application by an average of 0.26 kg P ha(-1) in loam and 0.23 kg P ha(-1) in sandy loam, but only by 0.03 kg P ha(-1) in loamy sand. Lower leaching losses were attributed to physical retention of particulate P and dissolved organic P, caused by placing slurry away from active flow paths in the fine-textured soil columns, as well as to chemical retention of dissolved inorganic P, caused by better contact between slurry P and soil adsorption sites. Dissolved organic P was less retained in soil after slurry application than other P forms. On these soils with low to intermediate P status, slurry injection lowered P leaching losses from clay-rich soil, but not from the sandy soils, highlighting the importance of soil texture in manageing P losses following slurry application.

A comparative study of phosphate sorption in lowland soils under oxic and anoxic conditions.

Phosphate (P(i)) release due to Fe(III) oxide dissolution is well documented for soils undergoing reduction. The P(i) sorption properties of soils in anoxic conditions are, however, still under consideration. In this investigation, P(i) sorption to strictly anoxic soils was compared with oxic conditions to assess the potential of lowland soils to function as traps for P(i) when flooded with drainage water. Batch sorption experiments were performed on seven minerogenic soils. Sorption to the anoxic soils was conducted after anoxic incubation, resulting in reduction of 36 to 93% of the dithionite-extractable Fe(III) (Fe(BD)). Langmuir fitted P(i) sorption isotherms showed a P(i) release of up to 1.1 mmol kg(-1) in six soils when P(i) concentrations in the matrix (P(sol)) were lower than 10 microM. Phosphate desorption was attributed to dissolution of amorphous iron oxides, and higher pH under anoxic conditions. The point of zero net sorption (EPC(0)) increased 2- to 10-fold on reduction. Five soils showed higher P(i) sorption capacities in the anoxic than in the oxic state at higher P(sol) concentrations. Solubility calculations indicated that precipitation of vivianite or similar Fe(II) phosphates may have caused the higher sorption capacities. Use of maximum sorption capacity (S(max)) is therefore misleading as a measure of P(i) sorption at low P(sol) concentrations. The results demonstrate that none of the strongly anoxic soils, irrespective of the initial Fe(III) oxide content, the P saturation, and the degree of Fe(III) oxide reduction, could retain P(i) at natural P(sol) concentrations in agricultural drainage water.

Phosphorus retention in riparian buffers: review of their efficiency.

Ground water and surface water interactions are of fundamental importance for the biogeochemical processes governing phosphorus (P) dynamics in riparian buffers. The four most important conceptual hydrological pathways for P losses from and P retention in riparian buffers are reviewed in this paper: (i) The diffuse flow path with ground water flow through the riparian aquifer, (ii) the overland flow path across the riparian buffer with water coming from adjacent agricultural fields, (iii) irrigation of the riparian buffer with tile drainage water from agricultural fields where disconnected tile drains irrigate the riparian buffer, and (iv) inundation of the riparian buffer (floodplain) with river water during short or longer periods. We have examined how the different flow paths in the riparian buffer influence P retention mechanisms theoretically and from empirical evidence. The different hydrological flow paths determine where and how water-borne P compounds meet and interact with iron and aluminum oxides or other minerals in the geochemical cycling of P in the complex and dynamic environment that constitutes a riparian buffer. The main physical process in the riparian buffer-sedimentation-is active along several flow paths and may account for P retention rates of up to 128 kg P ha(-1) yr(-1), while plant uptake may temporarily immobilize up to 15 kg P ha(-1) yr(-1). Retention of dissolved P in riparian buffers is not as pronounced as retention of particulate P and is often below 0.5 kg P ha(-1) yr(-1). Several studies show significant release of dissolved P (i.e., up to 8 kg P ha(-1) yr(-1)).