Simulation of water flow and solute transport in free-drainage lysimeters and field soils with heterogeneous structures

Lysimeters are valuable for studying the fate and transport of chemicals in soil. Large‐scale field lysimeters are used to assess pesticide behaviour and radionuclide transport, and are assumed to represent natural field conditions better than laboratory columns. Field lysimeters are usually characterized by a free‐draining lower boundary. As a result, the hydraulic gradient is disrupted, and leachate cannot be collected until the bottom of the lysimeter becomes saturated. We compared heterogeneously structured, free‐drainage lysimeters and field soils with respect to water flow and solute transport. Numerical simulations were carried out in a two‐dimensional heterogeneous sandy soil under unsaturated water flow conditions with the CHAIN_2D code. Three different soil structures (isotropic, horizontal, and vertical) were generated, and Miller–Miller similitude was used to scale the hydraulic properties of the soil. The results showed that ponding occurs at the bottom of the lysimeter for the three soil structures and that it occurred faster and was more pronounced with the vertical structure (preferential flow effect). Breakthrough curves of a conservative solute (bromide) showed that solutes are moving faster in the field than in the lysimeters. Fewer differences between lysimeters and field soils were found with the horizontal soil structure than with the isotropic and vertical structures.     

Investigation into the effect of tillage on solute movement to drains through a heavy clay soil

Abstract. Eight lysimeters, each with a surface area of 0.5 m² and a length of 60 cm, were taken over mole drains from a Denchworth soil and divided into two groups with either a standard agricultural tilth or a finer, deeper topsoil tilth. They were variously instrumented to measure soil moisture content at three depths and losses of nitrate, a bromide tracer and radiolabelled isoproturon, all of which were followed over a year. Leaching of isoproturon was initiated by artificial irrigation either 1 or 39 days after application. The finer tilth seemed to increase the water-holding capacity of the topsoil, and this resulted in slower wetting of the subsoil, decreased flow volumes from the first events of the season and a delay of approximately four weeks in the time to the maximum concentration of the bromide tracer in leachate. The finer topsoil tilth also decreased maximum concentrations of isoproturon from 29 to 15 μg l⁻¹ following irrigation 1 day after treatment and from 43 to 9 μg l⁻¹ following irrigation 39 days after treatment. Total losses of isoproturon were three times larger with the standard agricultural tilth. Differences were attributed to a decrease in bypass flow through the topsoil with the finer tilth, particularly during events early in the season. There was a small decrease in total losses of nitrate in leachate from the finer tilth compared to that from the standard tilth.     

A laboratory investigation of the release of a conservative tracer and herbicide from topsoil aggregates under varying rainfall intensities

Abstract. This paper describes the results of laboratory-based research which investigated the losses of the herbicide, isoproturon, and a non-adsorbing solute, bromide, from topsoil aggregates under high (7 mm h⁻¹ and low 2.7 mm h⁻¹) intensity simulated rainfall. The structures of the micro- and macro-aggregates from a clay soil of the Denchworth Series (from Wytham, Oxfordshire, UK) were observed using a scanning electron microscope, and the slaking of the topsoil aggregates was also investigated in order to provide a context for interpreting the experimental results. The topsoil at Wytham was found to have a bimodal structure with small microaggregates of 0.71 to 1 mm diameter clustered together to form larger macroaggregates of up to several centimetres diameter. Selected aggregates were air dried and then repacked in Buchner funnels to several centimetres depth in order to remove variability in the experimental results due to the effects of surface microrelief. Under saturated surface conditions high intensity simulated rainfall was less efficient at removing both non-adsorbing and adsorbing solutes from the repacked topsoil aggregates than low intensity rainfall. The observed decrease in the concentration of isoproturon and bromide in the leachate with time could be explained by a transport non-equilibrium effect.     

Investigating preferential flow in a large intact soil block under pasture

Abstract. A large soil block was constructed to determine the importance of preferential flow routes compared with matric flow pathways at a pasture site in mid-Devon. The sandy loam soil was well structured and uniform. The soil block measured 5 m×3 m×1 m and was instrumented with an array of 54 tensiometers, TDR wave guides and suction samplers connected to an in situ chloride analysis system. Four steady state irrigation experiments were conducted with a range of rainfall intensities. During each experiment chloride and nitrate tracers were applied and the patterns of movement were observed. Although the application of tracer was uniform and the soil was relatively homogeneous, there was large variability across the block in terms of time taken to reach the peak concentration (TPC) and the peak concentration itself. About 44 samplers operated at the greatest intensities (10–2 mm h⁻¹) and only 35 at the smallest (1 mm h⁻¹). No relationship was found between TPC and depth. The fastest TPC and largest concentrations were associated with the greatest rainfall intensities. Relative importance of the individual water pathways was a function of soil heterogeneity: parts of the soil block were highly active with several pathways having short TPCs and conductivities in excess of 4 m day⁻¹ whereas other areas had longer TPCs and conductivities of 1–2 m day⁻¹. The pattern was also dynamic, with conductivities of the pathways changing through time, though most of the faster pathways maintained their greater conductivities for more than one year.     

Retarded leaching of nitrate measured in monolith lysimeters in south-east Nigeria

At Onne in South-east Nigeria, drainage water was collected from four monolith lysimeters and analysed for nitrate. The lysimeters contained an acid sandy loam. At the start of the first rainy season two lysimeters received urea labelled with ¹⁵NO₃^– and two received no nitrogen fertilizer; all four were uncropped in the first year.
The peak concentrations of ¹⁵NO₃^– and of unlabelled (soil) NO₃^– were found after 2.5 pore volumes of water had passed through the lysimeters. Using the same soil in the laboratory after fine sieving, the peak concentration of tritiated water was found at 1 pore volume whereas nitrate leaching was retarded. The pattern of nitrate leaching was well described by miscible and immiscible models which included an adsorption coefficient for nitrate. Over the 2 years 81.4% of the ¹⁵N added at the start of the first rainy season was recovered in the drainage water.     

Net nitrogen mineralization and nitrification rates in forest soil in northeastern Taiwan

We used a laboratory incubation approach to measure rates of net N mineralization and nitrification in forest soils from Fu-shan Experimental Forest WS1 in northern Taiwan. Net mineralization rates in the O horizon ranged from 4.0 to 13.8 mg N kg⁻¹ day⁻¹, and net nitrification rates ranged from 2.2 to 11.6 mg N kg⁻¹ day⁻¹. For mineral (10–20 cm depth) soil, net mineralization ranged from 0.06 to 2.8 mg N kg⁻¹ day⁻¹ and net nitrification rates ranged from 0.02 to 2.8 mg N kg⁻¹ day⁻¹. We did not find any consistent differences in N mineralization or nitrification rates in soils from the upper and lower part of the watershed. We compared the rates of these processes in three soil horizons (to a soil depth of 30 cm) on a single sampling date and found a large decrease in both net N mineralization and nitrification with depth. We estimated that the soil total N pool was 6,909 kg N ha⁻¹. The present study demonstrates the importance of the stock of mineral soil N in WS1, mostly organic N, which can be transformed to inorganic N and potentially exported to surface and ground water from this watershed. Additional studies quantifying the rates of soil N cycling, particularly multi-site comparisons within Taiwan and the East Asia–Pacific region, will greatly improve our understanding of regional patterns in nitrogen cycling.     

Water quality implications of dairy slurry applied to cut pastures in the northeast USA

Abstract. Nitrate nitrogen (NO₃-N) leaching from animal production systems in the northeast USA is a major non-point source of pollution in the Chesapeake Bay. We conducted a study to measure NO₃-N leaching from dairy slurry applied to orchardgrass ( Dactylis glomerata L., cv. Pennlate) using large drainage lysimeters to measure the direct impact of four rates of slurry (urine and faeces) N application (0, 168, 336, 672 kg N ha⁻¹ yr⁻¹) on NO₃-N leaching on three soil types. We then used experimentally-based relationships developed earlier between stocking density and NO₃-N leaching loss and leachate NO₃-N concentration to estimate the added impact of animal grazing. Nitrate N leaching losses from only dairy slurry applied at the 0, 158, 336, and 672 kg N ha⁻¹ yr⁻¹ rates were 5.85, 8.26, 8.83, and 12.1 kg N ha⁻¹ yr⁻¹, respectively with corresponding NO₃-N concentrations of 1.60, 2.30, 2.46, and 3.48 mg l⁻¹. These NO₃-N concentrations met the 10 mg l⁻¹ US EPA drinking water standard. However, when a scenario was constructed to include the effect of NO₃-N leaching caused by animal grazing, the NO₃-N drinking water standard was calculated to be exceeded.     

Monitoring of water erosion on arable farms in England and Wales. 1990–94

Abstract. The incidence of soil water erosion was monitored in 12 erosion-susceptible arable catchments ( c . 80 fields) in England and Wales between 1990 and 1994. Factors associated with the initiation of erosion were recorded, and the extent of rills and gullies measured. Approximately 80% of the erosion events were on land cropped to winter cereals. In 30% of cases, the initiation of erosion was linked to valley floor features, which concentrated runoff. Poor crop cover, wheelings and tramlines were also assessed as contributory factors in 22%, 19% and 14% of cases, respectively. In c . 95% of cases rainfall events causing erosion were ≥10 mm day⁻¹ and c . 80% were >15 mm day⁻¹. Erosion was also associated with maximum rainfall intensities of >4 mm h⁻¹ for c . 90% of cases and >10 mm h⁻¹ for c . 20%. Mean net soil erosion rates were approximately 4 t ha⁻¹ per annum (median value 0.41 t ha⁻¹ per annum) and associated mean P losses 3.4 kg ha⁻¹.     

Leaching of carbon and nitrogen from a sandy soil substrate: A comparison between suction plates and ion exchange resins

To determine boundary effects on leaching, we investigated (1) how filter materials affect the concentrations of dissolved organic carbon (DOC) and nitrate (NO₃‐N) in soil percolates and (2) whether ion exchange resins and suction plates are equally suited to capture NO₃‐N. DOC leaching was higher with PE suction plates and plate material did not affect NO₃‐N leachate concentrations. Cumulative NO₃‐N leaching was similar for glass suction plates and ion exchange resins.     

Comparison of soil solution chemistry assessment using zero-tension lysimeters or centrifugation

The composition of soil solutions obtained from the field varies with the method of extraction. Variations in sampling methods and the difficulties in extracting representative samples from soils in space and time, can explain divergent results. In this study we compared soil solutions from a forest soil in northern Sweden obtained by a centrifuge drainage technique and by zero-tension monolith lysimeters. Zero-tension lysimeters were destructively sampled, and centrifuge solutions from this soil were compared with that from soil outside. In our study we found three major differences in the solute composition between the centrifugate and the lysimeter leachate: (i) larger concentrations of most solutes in the mor layer centrifugate than in the mor layer leachate, (ii) accumulation of nitrate in the lysimeters, and (iii) larger concentrations of base cations in the zero-tension lysimeters below 0.3 m depth. Water contents within the lysimeters were up to 3.5 times greater than under natural conditions and the water yields from the lysimeters indicate that water residence time ranged from < 1 to >5 years. This study shows that differences in results from the two methods are due to inherent differences in the methods themselves and not just to the collection of different soil waters. The hydrological anomaly and disturbance induced by the zero-tension lysimeters affects the solute chemistry and thus the applicability of the results to field conditions.     

Recovery of 15N-labelled fertilizer in crops, drainage water and soil using monolith lysimeters in south-east Nigeria

Labelled urea was applied to monolith lysimeters in the 1st year of a 2-year experiment at Onne in south-east Nigeria. On eight lysimeters maize and rice were grown in each of the 2 years. Four lysimeters were similarly cropped in the 2nd year after being uncropped in the 1st year. Measurements were made over the 2-year period of labelled and unlabelled mineral nitrogen in the drainage water, and labelled and unlabelled nitrogen in the crops. At the end of the experiment, weeds and the soil were also analysed for labelled and unlabelled nitrogen.
The total recovery of ¹⁵N in crop, soil and leachate varied between 70 and 93%. It was lowest when applied to the second season rice crop, which recovered only 15%, and highest when it was leached in the 1st year or was taken up by the maize crop. The highest crop uptake was 31%. The results indicate that, depending on the treatment, between 10 and 30% of the ¹⁵N was immobilized in the soil, lysimeters cropped in the 1st year lost between 22 and 29% of the ¹⁵N in drainage water, and between 7 and 30% was lost by denitrification. The accuracy of these figures is discussed.     

The estimation of mineralization, immobilization and nitrification in nitrogen-15 field experiments using computer simulation

A combination of mathematical analysis and computer simulation, using parameters readily measured in a nitrogen-15 field experiment, is employed to determine rates of mineralization, immobilization and nitrification under a growing crop. The procedure also yields the proportion of crop nitrogen uptake occurring as ammonium and nitrate.
When applied to -results from grass lysimeters receiving 250 or 900 kg N ha^–1 a^–1 as ammonium nitrate, the analysis suggested that at 250 kgN ha^–1 a^–1 64–66% of crop nitrogen uptake was as ammonium; at 900 kg N ha^–1 a^–1 the figure was 43–49%. Nitrification at 250kgNha^–1 was only 13–19kgN ha^–1 over 160d while at 900 kg N ha^–1 between 191 and 232 kg N ha^–1 were nitrified.
The results suggested that the apparent inhibition of nitrification in grassland soils may simply reflect poor substrate competition by nitrifying bacteria. Finally, there was a suggestion that mineralization/immobilization was lower at the high fertilizer rate.     

Comparison of field techniques for sampling soil solution in an upland peatland

Abstract. Results from recent studies of peatland biogeochemistry suggest that appropriate soil water sampling techniques are required in order to advance our understanding of peatland soil systems. In a comparative field experiment, concentrations of inorganic solutes and dissolved organic carbon (DOC) were measured in soil water extracted at a depth of 10 cm beneath the surface of deep peat by three techniques: zero-tension (z-t) lysimeters, PTFE suction samplers, and polysulfone suction samplers. The majority of solute concentrations were broadly similar, but mean concentrations of silicon, DOC, iron and aluminium in water extracted by z-t lysimeters and PTFE samplers were in ratios of 1:5; 1:2; 1:5 and 1:3 respectively. Mean conductivity and concentrations of chloride and hydrogen ion were significantly larger in the z-t lysimeter samples, which had sodium, potassium and magnesium to chloride ratios that were very similar to local rainfall. The z-t lysimeters appeared to sample macropores preferentially, while the suction samplers collected micropore water.     

Improvement of application method of liquid fertilizer with irrigation water for uniform topdressing and change in nitrogen concentration of flooded water after application

(pp. 811–816)
The application method of liquid fertilizer with irrigation water was tested in large scale paddy fields using a newly developed fertilizer supplier which can make a supplying rate of liquid fertilizer constant. Uniformity of fertilization, change in the nitrogen concentration of flooded water and absorption of applied nitrogen by paddy rice were examined. High uniformity of fertilization was achieved with this method as compared with a conventional broadcast application of granular fertilizer, or an inflow fertilization of granule-like fertilizer with irrigation water.
When urea was applied as liquid fertilizer, the fertilizer-N disappeared quickly from ponding water under shallow conditions of ponding water depth. In the case of the same depth of ponding water, the rate of reduction was large under conditions of large permeability of water.
The nitrogen utilization rates of urea and ammonium sulfate labeled with ¹⁵N were not related to the depth of ponding water at the time of fertilization. When ammonium sulfate was supplied the nitrogen utilization rate was around 50% regardless of fertilization conditions however, the urea-N utilization rate fell to 40% or less at a permeability of ponding water of 0.8 cm day⁻¹ or less, while it increased to 50% equivalent to ammonium sulfate, at 0.8 cm day⁻¹ or more permeability.     

Potential movement of methyl bromide through golf course greens

Abstract

Methyl bromide (MeBr) is routinely used for soil sterilization and stored food preservation. There is increasing concern for the continued use of MeBr due to the influence of methane and MeBr on the ozone layer in addition to its high mammalian toxicity. MeBr is commonly used for sterilizing golf course greens during construction and renovation. The high water solubility of MeBr and bromide ion (Br) and the high infiltration rate of most golf course greens allows for potential movement of these analytes into drainage channels and ultimately into surface contained water systems. The objective of this research was to develop a sensitive and accurate method for quantifying MeBr and Br in aqueous soil leachate and to determine the potential for MeBr and Br transport through golf course greens following fumigation. A method for utilizing a purge and trap system coupled to a gas Chromatograph equipped with an electron capture detector (P&T/GC‐ECD) was developed for measuring MeBr. This system along with the LACHAT system for Br analyses in the purged subsample were used to quantify MeBr and its metabolites in aqueous leachate obtained from lysimeters developed to simulate treatment of renovated greens using MeBr fumigation. MeBr was applied to the enclosed lysimeters at rates of 533 and 65 g m^‐2 in two experiments. At 2 days after treatment (DAT), tops of the lysimeters were removed and at 4 DAT simulated rainfall and irrigation events were initiated. The leachate was collected from the bottoms of the lysimeters for a 22 day period and analyzed for MeBr and Br transported through the soil mixture. The sum of transported MeBr and Br from the two experiments was proportional to the treatment rate. Considerably more Br was transported than MeBr. The highest concentration of MeBr and Br in the leachate from both experiments was 11.8 and 57.2 mg L^‐1, respectively. The total MeBr and Br transported through the golf course greens rooting medium, accounted for less than 1% of the analyte applied at both treatment rates.     

A Comparison of 18Oδ Composition of Water Extracted from Suction Lysimeters, Centrifugation, and Azeotropic Distillation

The representativeness of soil pore water extracted by suction lysimeters in ground-water monitoring studies is a problem that often confounds interpretation of measured data. Current soil water sampling techniques cannot delineate from which soil volume a pore water sample is extracted, neither macroscopic, microscopic, or preferential flowpath. This research was undertaken to compare δ¹⁸O and Br^? values of extracted suction lysimeters samples from intact soil cores with samples obtained by the direct extraction methods of centrifugation and azeotropic distillation. Also, the study was concerned with determining what portion of soil pore water is sampled by each method and explaining differences in concentrations of the extracted water from each method to allow a determination of the accuracy and viability of the three methods of extraction. Intact soil cores (30 cm diameter by 40 cm height) were extracted from two different sites. Site 1 was rapid infiltration basin number 50, near Altamonte Springs in Seminole County, Florida. Site 2 was the Missouri Management System Evaluation Area (MSEA) near Centralia in Boone County, Missouri. Isotopically (¹⁸Oδ) labeled water and bromide concentrations within water samples taken by suction lysimeters was compared with samples obtained by methods of centrifugation and azeotropic distillation. The ¹⁸Oδ water was analyzed by mass spectrometry while bromide concentration, applied in the form of KBr was measured using standard IC procedures. Water collected by centrifugation and azeotropic distillation data were about 0.25‰ more negative than that collected by suction lysimeter values from a sandy soil and about 2–7‰ more negative from a well structured soil. Results indicate that the majority of soil water in well-structured soil is strongly bound to soil grain surfaces and is not easily sampled by suction lysimeters. In cases where a sufficient volume of water has passed through the soil profile and displaced previous pore water, suction lysimeters will collect a representative sample of soil pore water from the sampled depth interval. It is suggested that for stable isotope studies monitoring precipitation and soil water, suction lysimeter be installed at shallow depths (10 cm). Samples should also be coordinated with precipitation events. The data also suggest that each extraction method samples a separate component of soil-pore water. Centrifugation can be used with success, particularly for efficient sampling of large areas. Azeotropic distillation is more appropriate when strict qualitative and quantitative data on sorption desorption, and various types of kinetic studies may be needed.     

Effects of fertilizer placement on solute leaching under ridge tillage and no tillage

Elevated nitrate concentrations in ground water can be a problem in agricultural areas, especially where soils are sandy. Tillage operations, such as ridge tillage (RT) and no tillage (NT) can reduce runoff and erosion but leaching of soluble nutrients could adversely impact groundwater. In a 2-year study, Br was used to trace the effects of fertilizer placement on solute movement under corn (Zea mays L.) in RT and NT systems on a Monmouth fine sandy loam (Typic Hapludult) in Maryland. Treatments included 120 kg ha⁻¹ of Br⁻ or NO₃⁻-N applied in a narrow band near the ridge top (RT-RA) or in the furrow (RT-FA) with ridge tillage, or in the inter-row with NT. Two-dimensional arrays of tensiometers and suction lysimeters were used to follow the movement of water and solutes during and after the corn-growing season. Tillage and fertilizer placement did not significantly affect N uptake when averaged across years. A pronounced argillic horizon beginning at 60 cm depth caused lateral movement of Br. It appears that Br leaching in RT-RA increased slightly due to the crop canopy funneling rain towards the ridge top. Therefore, when fertilizer is applied near the row, rain occurring after full corn canopy may cause greater solute leaching in RT-RA compared to other treatments. Rain during the beginning of the growing season or after harvest caused less leaching in RT-RA. Corn yield could be maximized and N leaching minimized by applying fertilizer to the upper portion of the ridge in RT.     

Laboratory studies of effects of lime on soluble Al, Ca, Mg and Mn in a pelostagnogley soil

The cation composition of solutions and leachates from small-diameter laboratory soil columns was examined over a 23-week period after the addition of lime (0, 3 and 6 t ha^–1) and/or nitrogen fertilizer (0 or 200 kg N ha^–1) to an acid soil (pH 4.2). Water was applied at regular intervals to the surface of the columns and 17 leachate samples collected. Initially, the pH of the leachate was high (6.6) in all treatments (including those without lime) but fell rapidly to approach a steady value of 3.8. Large losses of calcium occurred from all columns; the total equivalent amounts of lime lost ranged from 0.88 (no addition) to 2.38 (with added lime) t CaCO₃ ha^–1 High concentrations of aluminium (181–325 μM) were present in leachates from all treatments; the addition of 200 kg N ha^–1 increased the leaching of Al by 94%; addition of lime also increased the amounts of Al leached (by 52%).
The pH of the soil solution (separated by centrifugation) was influenced by treatment, especially in the top 0–40 mm of the column. Aluminium concentration was related to pH, but the form of the relationship differed amongst the treatments.     

Scaling the saturated hydraulic conductivity of an alfisol

Alfisols exhibit a high degree of spatial variability in their physical properties. As a result, it is difficult to use information on physical parameters measured at one location to model larger-scale hydrologic processes. In this study, the saturated hydraulic conductivity, K_S , of an Alfisol was determined on 109 undisturbed monoliths using the falling-head permeameter method. The model developed by Arya & Paris (1981) was used to calculate the pore volume from sand and clay fractions. Scaling factors were calculated from the measured K_s , sand pore-volume, clay pore-volume, clay content and effective porosity, using the similar media concept. Prediction of K_s of gravelly Alfisol using clay pore-volume is confounded by high gravel content which, when discounted, improves the prediction remarkably. The scaled mean saturated hydraulic conductivity K* for all horizons of the Alfisol was approximately l.0x 10⁻⁵ms⁻¹.