Rates of Nitrate and Perchlorate Removal in a 5-Year-Old Wood Particle Reactor Treating Agricultural Drainage

A 5-year-old wood particle reactor treating agricultural tile drainage in southern Ontario was monitored for its ongoing ability to treat both nitrate (NO₃^–) and perchlorate (ClO₄^–). Prior to sampling undertaken in the fifth year of operation, a highway safety flare containing ClO₄^– was immersed in the inlet pipe elevating influent ClO₄^– concentrations to up to 33.7 μg/L. ClO₄^– removal rates were inhibited in the presence of more than 1 to 2 mg/L NO₃^–-N, but increased rapidly to about 60 μg/L/d upon NO₃^– depletion. Nitrate removal rates, measured subsequently in the sixth and seventh years of operation, varied with temperature in the range of 2 to 16 mg N/L/d, but remained similar to rates measured in the second year. Additionally, no deterioration in the hydraulic conductivity (K) of the coarse core layer (0.5 3^– removal rates and can remain highly permeable over a number of years. The media can also provide high removal rates for other redox sensitive contaminants such as ClO₄^–. The ability to directly measure the reactor flow rate, in this case via an outlet pipe, greatly simplified the task of estimating hydraulic properties and reaction rates.     

High-permeability layers for remediation of ground water; go wide, not deep

A nitrate-reactive porous media layer comprising wood particles with very high hydraulic conductivity (K approximately 1 cm/s) was used to successfully treat nitrate in a shallow sand-and-gravel aquifer in southern Ontario. Nitrate concentrations of 1.3 to 14 mg/L as N in the aquifer were attenuated to <0.5 mg/L as N in the reactive layer. Borehole dilution testing indicated that ground water velocities in the reactive layer, although variable, averaged five times higher than in the surrounding aquifer, suggesting that the layer was capturing ground water flow from deeper in the aquifer. The use of high-K reactive media opens up the possibility of installing permeable reactive barriers as horizontal layers in the shallow water table zone that do not necessarily have to penetrate the full depth of a contaminant plume to be effective. Model simulations show that the depth of capture of a high-K layer increases as the layer width in the direction of flow increases. Shallower emplacement could decrease barrier costs at some sites.     

Metal Retention Experiments for the Design of Soil‐Mix Technology Permeable Reactive Barriers

Soil‐mix technology is effective for the construction of permeable reactive barriers (PRBs) for in situ groundwater treatment. The objective of this study was to perform initial experiments for the design of soil‐mix technology PRBs according to (i) sorption isotherm, (ii) reaction kinetics and (iii) mass balance of the contaminants. The four tested reactive systems were: (i) a granular zeolite (clinoptilolite–GZ), (ii) a granular organoclay (GO), (iii) a 1:1‐mixture GZ and model sandy clayey soil and (iv) a 1:1:1‐mixture of GZ, GO and model soil. The laboratory experiments consisted of batch tests (volume 900 mL and sorbent mass 18 g) with a multimetal solution of Pb, Cu, Zn, Cd and Ni. For the adsorption experiment, the initial concentrations ranged from 0.01 to 0.5 mM (2.5 to 30 mg/L). The maximum metal retention was measured in a batch test (300 mg/L for each metal, volume 900 mL, sorbent mass 90–4.5 g). The reactive material efficiency order was found to be GZ > GZ‐soil mix > GZ‐soil‐GO mix > GO. Langmuir isotherms modelled the adsorption, even in presence of a mixed cations solution. Adsorption was energetically favourable and spontaneous in all cases. Metals were removed according to the second order reaction kinetics; GZ and the 1:1‐mix were very similar. The maximum retention capacity was 0.1–0.2 mmol/g for Pb in the presence of clinoptilolite; for Cu, Zn, Cd and Ni, it was below 0.05 mmol/g for the four reactive systems. Mixing granular zeolite, organoclay and model soil increased the chemisorption. Providing that GZ is reactive enough for the specific conditions, GZ can be mixed to obtain the required sorption. Granular clinoptilolite addition to soil is recommended for PRBs for metal contaminated groundwater.     

Hydrodynamic modelling for groundwater flow through permeable reactive barriers

Hydrodynamic modelling for analysis of groundwater flow through permeable reactive barriers (PRBs) is addressed in this paper. Permeable reactive barriers constitute an emerging technology for in situ remediation of groundwater contamination and have many advantages over the traditional ex situ treatment methods. The transport domains during groundwater flow through PRBs often may involve free‐flow or non‐porous sections. To model the fluid mobility efficiently in such situations, the free and porous flow zones (PRBs) must be studied in conjunction with each other. The present paper is devoted to the analysis of groundwater flow through combined free flow domains and PRBs. The free‐flow regime is modelled using the Navier–Stokes equations whereas the permeable barriers are simulated by either the Darcy or the Brinkman equation. In order to couple the governing equations of motions, well‐posed mathematical formulations of matching boundary conditions are prescribed at the interface between the free‐groundwater‐flow zones and the permeable barriers. Combination of the Navier–Stokes equations with the Brinkman equation is more straightforward owing to their analogous forms. However, the Navier–Stokes and Darcy equations are incompatible mathematically and cannot be linked directly. The problem is resolved in this paper by invoking validated hydrodynamical expressions for describing the flow behaviour at the interfaces between free‐flow and porous zones. Three schemes for the analyses of fluid flow in combined domains are applied to the case of groundwater flow through permeable reactive barriers and different model results are compared. Copyright © 2002 John Wiley & Sons, Ltd.     

Effects of gas generation and precipitates on performance of Fe0 PRBs

Long-term reactivity and permeability are critical factors in the performance of granular iron permeable reactive barriers (PRBs). Thus it is a topic of great practical importance, as well as scientific interest. In this study, four types of source solutions (distilled H2O, 10 mg/L TCE, 300 mg/L CaCO3, and 10 mg/L TCE + 300 mg/L CaCO3) were supplied to four columns containing a commercial granular iron material. In all four columns, gases accumulated to approximately 10% of the initial porosity and resulted in declines in permeability of approximately 50% to 80%. In the columns receiving CaCO3, carbonate precipitates accumulated to approximately 7% of the initial porosity, with no apparent decline in permeability. The data indicate that precipitates formed initially at the influent ends of the columns, reducing the reactivity of the iron in this region. As a consequence of the reduced reactivity, calcium and bicarbonate migrated further into the column, to precipitate in a region where the reactivity remained high. Thus precipitation occurred as a moving front through the columns. The results suggest improved methods for PRB design and rehabilitation, and also suggest improvements that are needed in the mathematical models developed for predicting long-term performance.     

Nitrate in the Intermediate Vadose Zone Beneath Irrigated Cropland

More than 1000 feet of fine-textured, unsaturated zone core beneath nitrogen-fertilized and irrigated farmland was collected, leached and analyzed for nitrate-nitrogen. Fertility plots treated with 200, 300, and 400 Ibs N/acre/yr accumulated significant quantities of nitrate-nitrogen in the vadose zone below the crop rooting zone. The average nitrate-nitrogen concentration approximately doubled with each 100 lbs N/acre/yr increment above the 100 lbs N/ acre/ yr treatment. Nitrate loading estimates for the plots treated with 400 lbs N/ acre/ yr indicate that over 1200 lbs N/ acre was in the vadose zone beneath the crop rooting zone. In 15 years, the nitrate moved vertically at least 60 feet through these fine-textured, unsaturated sediments. As much as 600 lbs N/acre have accumulated in the vadose zone under independent corn producers'fields.
Vadose zone sampling is effective in predicting future non-point nitrate-contaminated areas.     

Impact of mineral fouling on hydraulic behavior of permeable reactive barriers

This paper describes reactive transport simulations conducted to assess the impact of mineral fouling on the hydraulic behavior of continuous-wall permeable reactive barriers (PRBs) employing granular zero-valent iron (ZVI) in carbonate-rich alluvial aquifers. The reactive transport model included a geochemical algorithm for simulating corrosion and mineral precipitation reactions that have been observed in ZVI PRBs. Results of simulations show that porosity and hydraulic conductivity of the ZVI decrease over time and that flows are redistributed throughout the PRB in response to fouling of the pore space. Under typical conditions, only subtle changes occur within the first 10 years (i.e., duration of the current field experience record with PRBs), and the most significant changes do not occur until the PRB has operated for at least 30 years. However, changes can occur sooner (or later) if the rate at which mineral-forming ions are delivered to the PRB is higher (or lower) than that expected under typical conditions (i.e., due to higher/lower flow rate or inflowing ground water that has higher/lower ionic strength). When the PRB is more permeable than the aquifer, the median Darcy flux in the PRB does not change appreciably over time because the aquifer controls the rate of flow through the PRB. However, seepage velocities in the PRB increase, and residence times decrease, due to porosity reductions caused by accumulation of minerals in the pore space. When fouling becomes extensive, bypassing and reductions in flow rate in the PRB occur.     

Injection-extraction treatment well pairs: an alternative to permeable reactive barriers

Two of the biggest drawbacks of using permeable reactive barriers (PRBs) to treat contaminated ground water are the high capital cost of installation, particularly when the contaminated ground water is deep below ground surface, and the uncertainty of whether or not PRBs remain effective for the long time scales (e.g., decades) needed for many contaminant plumes. The use of an injection-extraction treatment well pair (IETWP) for capture and treatment of contaminated ground water can circumvent these difficulties, while still providing many of the same advantages offered by PRBs. In this paper, the hydraulics of IETWPs and PRBs are compared, focusing primarily on the width of the captured plume. It is demonstrated that IETWPs act as hydraulic barriers in a manner similar to PRBs, and that IETWPs provide excellent plume capture. A mathematical expression is presented for the plume capture width of an IETWP oriented perpendicular to the ground water flow direction in a homogeneous aquifer. Also discussed are other practical considerations that might determine whether an IETWP is better suited than a PRB for a particular contaminated site; these considerations include operating and maintenance costs, and the conditions under which an IETWP system can be used for in situ remediation.     

Long-term performance of permeable reactive barriers using zero-valent iron: geochemical and microbiological effects

Geochemical and microbiological factors that control long-term performance of subsurface permeable reactive barriers were evaluated at the Elizabeth City, North Carolina, and the Denver Federal Center, Colorado, sites. These ground water treatment systems use zero-valent iron filings (Peerless Metal Powders Inc.) to intercept and remediate chlorinated hydrocarbon compounds at the Denver Federal Center (funnel-and-gate system) and overlapping plumes of hexavalent chromium and chlorinated hydrocarbons at Elizabeth City (continuous wall system). Zero-valent iron at both sites is a long-term sink for carbon, sulfur, calcium, silicon, nitrogen, and magnesium. After about four years of operation, the average rates of inorganic carbon (IC) and sulfur (S) accumulation are 0.09 and 0.02 kg/m2/year, respectively, at Elizabeth City where upgradient waters contain <400 mg/L of total dissolved solids (TDS). At the Denver Federal Center site, upgradient ground water contains 1000 to 1200 mg/L TDS and rates of IC and S accumulation are as high as 2.16 and 0.80 kg/m2/year, respectively. At both sites, consistent patterns of spatially variable mineral precipitation and microbial activity are observed. Mineral precipitates and microbial biomass accumulate the fastest near the upgradient aquifer-Fe0 interface. Maximum net reductions in porosity due to the accumulation of sulfur and inorganic carbon precipitates range from 0.032 at Elizabeth City to 0.062 at the Denver Federal Center (gate 2) after about four years. Although pore space has been lost due the accumulation of authigenic components, neither site shows evidence of pervasive pore clogging after four years of operation.     

北太平洋上空臭氧总量长期变化趋势及其影响因素分析

本文基于1979—2014年臭氧总量的卫星遥感数据,利用多元线性回归模型对臭氧总量数据序列进行模拟计算,考察了北太平洋上空臭氧总量长期变化趋势及其影响因素的作用.结果表明,北太平洋地区大气臭氧总量长期变化呈现减少趋势,但是减少速率随季节和纬度带表现出差异性,在各纬度带臭氧峰值季节臭氧下降趋势最为显著.在0°—15°N地区臭氧高值出现在夏秋季节并在8月达到峰值,峰值月份臭氧年均下降率约为0.2DU/a;15°—30°N亚热带地区臭氧高值出现在春夏季并在5月达到峰值,峰值月份臭氧年均下降速率约为0.22DU/a;而在30°—45°N中纬度地区臭氧高值出现在冬春季并在2月达到峰值,峰值月份臭氧年均下降率0.75DU/a.在臭氧分布年平均态基础上,影响臭氧总量分布变化的因素主要有臭氧损耗物质(EESC)、太阳辐射周期(Solar)、准两年振荡(QBO)和厄尔尼诺-南方涛动(ENSO)等.其中,EESC导致臭氧损耗效应随着纬度升高而增大,在从低到高的三个纬度带损耗最大值分别为11DU、16DU和66DU;Solar增强导致臭氧增加,在三个纬度带的增加效应最大值分别为16DU、17DU和19DU;QBO@10hPa和QBO@30hPa对臭氧影响幅度基本在±10DU内波动,只有QBO@10hPa对30°—45°N区域的影响作用达到14DU,值得注意的是QBO影响作用随着纬度变化存在相位差异,在0°—15°N区域臭氧变化与QBO呈现相同相位,而在15°—30°N和30°—45°N区域臭氧变化与QBO呈现相反相位;ENSO对各个纬度带臭氧影响幅度也在±10DU内,ENSO影响作用在不同纬度带也存在相位差异,臭氧总量变化在0°—15°N、15°—30°N区域与ENSO相位相反,在30°—45°N区域与ENSO相位一致.     

Estimating Aquifer Sensitivity to Nitrate Contamination Using Geochemical Information

Methods for predicting aquifer sensitivity to contamination typically ignore geochemical factors that affect the occurrence of contaminants such as nitrate. Use of geochemical information offers a simple and accurate method for estimating aquifer sensitivity to nitrate contamination. We developed a classification method in which nitrate-sensitive aquifers have dissolved oxygen concentrations > 1.0 mg/L, Eh values >250 mV, and either reduced iron concentrations < 0.1 mg/L or total iron concentrations < 0.7 mg/L. We tested the method in four Minnesota aquifer systems having different geochemical and hydrologic conditions. A surficial sand aquifer in central Minnesota exhibited geochemical zonation, with a rapid shift from aerobic to anaerobic conditions 5 m below the water table. A fractured bedrock aquifer in east-central Minnesota remained aerobic to depths of 50 m, except in areas where anaerobic ground water discharged upward from an underlying aquifer. A bedrock aquifer in southeast Minnesota exhibited aerobic conditions when overlain by surficial deposits lacking shale, whereas anaerobic conditions occurred under deposits that contained shale. Surficial sand aquifers in northwest Minnesota contained high concentrations of sulfate and were anaerobic throughout their extent. Nitrate-nitrogen was detected at concentrations exceeding 1 mg/L in 135 of 149 samples classified as sensitive. Nitrate was not detected in any of the 109 samples classified as not sensitive. We observed differences between our estimates of sensitivity and existing sensitivity maps, which are based on methods that do not consider aquifer geochemistry. Because dissolved oxygen, reduced iron, and Eh are readily measured in the field, use of geochemistry provides a quick and accurate way of assessing aquifer sensitivity to nitrate contamination.     

Model parameters for simulating fate and transport of on-site wastewater nutrients

This paper presents a critical review of model-input parameters for transport of on-site wastewater treatment system (OWS) pollutants. Approximately 25% of the U.S. population relies on soil-based OWS for effective treatment and protection of public health and environmental quality. Mathematical models are useful tools for understanding and predicting the transport and fate of wastewater pollutants and for addressing water-budget issues related to wastewater reclamation from site to watershed scales. However, input parameters for models that simulate fate and transport of OWS pollutants are not readily obtained. The purpose of this analysis is to illustrate an objective, statistically supported method for choosing model-input parameters related to nitrogen (N) and phosphorus (P). Data were gathered from existing studies reported in the literature. Cumulative frequency distributions (CFDs) are provided for OWS effluent concentrations of N and P, nitrification and denitrification rates, and linear sorption isotherm constants for P. When CFDs are not presented, ranges and median values are provided. Median values for model-input parameters are as follows: total N concentration (44 mg/L), nitrate-N (0.2 mg/L), ammonium (60 mg/L), phosphate-P (9 mg/L), organic N (14 mg/L), zero-order nitrification rate (264 mg/L/d), first-order nitrification (2.9/d), first-order dentrification (0.025/d), maximum soil capacity for P uptake (237 mg/kg), linear sorption isotherm constant for P (15.1 L/kg), and OWS effluent flow rates (260 L/person/d).     

The use of woods-run chips in filter socks to control erosion and sedimentation during petroleum development in the Appalachian Basin

Filter socks frequently are used for erosion and sediment perimeter control during oil and gas development activities in the Appalachian Basin of the United States. Regulations specify the use of composted wood material for sock construction. This specification, as opposed to non-composted or fresh wood chips (woods-run), has created inefficiencies during well site construction. Rather than use fresh wood chips created during site construction, composted chips must be procured and used as filter sock media for erosion and sedimentation mitigation. If woods-run chips could be used as filter sock media instead of composted chips, there could be a significant reduction in energy/capital costs, truck traffic, and disposal costs. The primary objective of this research project was to compare the effectiveness of woods-run material versus traditionally composted wood chips in controlling sediment transport as well as other chemical and physical parameters in a laboratory setting. No significant differences in pH (5.96 versus 6.02) or conductivity (0.029 dS/m versus 0.035 dS/m) were found in sediment laden water filtered through woods-run versus composted chips, respectively. However, chip particle sizes were outside the allowable limits for composted sock media, and moisture content also was outside the specified limits for woods-run chips. Nitrate (NO₃) concentrations were significantly higher in woods-run, while phosphorus (P) and potassium (K) concentrations were greater in composted chips; however none of the N,P, or K concentrations were above the regulatory requirements. Finally, no difference in the filtering efficiency or time was found between woods-run and composted material. The laboratory results indicate that current regulations allowing the use of woods-run chips in all but the highest quality watersheds is justified.     

ANALYSIS OF SEISMICITY CHANGES PRIOR TO THE 2014 YUNNAN JINGGU MS6.6 EARTHQUAKE

According to the Region-Time-Length (RTL) algorithm,the analysis of seismicity changes prior to the 2014 Yunnan Jinggu M_S6.6 earthquake was conducted by using the earthquake catalogues about 6 and 15 years before this earthquake,respectively.When the studied period was nearly 6 years,an enhancement of seismic activity was detected around the epicenter since the beginning of 2013.The anomalies mainly distributed in the region of 22.5°~24.5°N and 99°~102°E.The range and degree of anomalies changed from small to large,and then to small chronologically.As the surface integral in respect to RTL,the physical parameter I_RTL,which could reflect the regional seismicity level,began to increase since August 2013,and then reduced after reaching the peak point.The time length from the peak point of I_RTL curve to the earthquake occurrence was 9 months.When the analyzed catalogue was nearly 15 years,the 2007 Ninger M_S6.4 occurred in the studied region.Seismicity quiescence was detected prior to the Ninger M_S6.4.Before the Jinggu M_S6.6,seismicity quiescence was detected firstly,and then enhanced activity was observed 1 year prior to the earthquake occurrence.The anomalies mainly distributed in the region of 22.5°~24.5°N and 99°~102°E.The time length from the peak point of I_RTL curve to the earthquake occurrence was 7 months.The above study showed that even the earthquakes location was near and the magnitude was close to each other,a big difference in seismic activity before the earthquakes may exist.Before the Jinggu M_S6.6,there was some difference in seismicity changes according to different beginning time of catalogues,but the distribution of anomalies and the time length from the peak point of I_RTL to the earthquake occurrence were uniform.So there was an important significance for exploring the relationship between the distribution of anomalies and the earthquake location,and the relationship between the time of the peak point of I_RTL and the earthquake occurrence time.     

A Full-Scale Porous Reactive Wall for Prevention of Acid Mine Drainage

The generation and release of acidic drainage containing high concentrations of dissolved metals from decommissioned mine wastes is an environmental problem of international scale. A potential solution to many acid drainage problems is the installation of permeable reactive walls into aquifers affected by drainage water derived from mine waste materials. A permeable reactive wall installed into an aquifer impacted by low-quality mine drainage waters was installed in August 1995 at the Nickel Rim mine site near Sudbury, Ontario. The reactive mixture, containing organic matter, was designed to promote bacterially mediated sulfate reduction and subsequent metal sulfide precipitation. The reactive wall is installed to an average depth of 12 feet (3.6 m) and is 49 feet (15 m) long perpendicular to ground water flow. The wall thickness (flow path length) is 13 feet (4 m). Initial results, collected nine months after installation, indicate that sulfate reduction and metal sulfide precipitation is occurring. Comparing water entering the wall to treated water exiting the wall, sulfate concentrations decrease from 2400 to 4600 mg/L to 200 to 3600 mg/L; Fe concentrations decrease from 250 to 1300 mg/L to 1.0 to 40 mg/L; pH increases from 5.8 to 7.0; and alkalinity (as CaCO₃) increases from 0 to 50 mg/L to 600 to 2000 mg/L. The reactive wall has effectively removed the capacity of the ground water to generate acidity on discharge to the surface. Calculations based on comparison to previously run laboratory column experiments indicate that the reactive wall has potential to remain effective for at least 15 years.     

Analytical Solutions for Flow Fields near Drain-and-Gate Reactive Barriers

Permeable reactive barriers (PRBs) are a popular technology for passive contaminant remediation in aquifers through installation of reactive materials in the pathway of a plume. Of fundamental importance are the degree of remediation inside the reactor (residence time) and the portion of groundwater intercepted by a PRB (capture width). Based on a two-dimensional conformal mapping approach (previously used in related work), the latter is studied in the present work for drain-and-gate (DG) PRBs, which may possess a collector and a distributor drain (“full” configuration) or a collector drain only (“simple” configuration). Inherent assumptions are a homogeneous unbounded aquifer with a uniform far field, in which highly permeable drains establish constant head boundaries. Solutions for aquifer flow fields in terms of the complex potential are derived, illustrated, and analyzed for doubly symmetric DG configurations and arbitrary reactor hydraulic resistance as well as ambient groundwater flow direction. A series of practitioner-friendly charts for capture width is given to assist in PRB design and optimization without requiring complex mathematics. DG PRBs are identified as more susceptible to flow divergence around the reactor than configurations using impermeable side structures (e.g., funnel-and-gate), and deployment of impermeable walls on drains is seen to mitigate this problem under certain circumstances.     

Effect of iron type on kinetics and carbon isotopic enrichment of chlorinated ethylenes during abiotic reduction on Fe(0)

Four samples of two commercially available iron brands used as substrate for iron permeable reactive barriers (PRBs) were tested for suitability for remediation of perchloroethylene (PCE), trichloroethylene (TCE), cis-dichloroethylene (cDCE) and vinyl chloride (VC). Kinetic studies indicate that rates of reaction are enhanced for cDCE and VC on Connelly iron (2.8 x 10(-4) to 6.9 x 10(-4) L/m2/hr and 2.0 x 10(-4) to 9.0 x 10(-4) L/m2/hr, for cDCE and VC, respectively) vs. Peerless iron (3.1 x 10(-5) to 4.6 x 10(-5) L/m2/hr and 2.4 x 10(-5) to 4.1 x 10(-5) L/m2/hr, for cDCE and VC, respectively). Carbon isotopic analyses of the residual chlorinated ethylene (CE) during degradation indicate significant fractionation occurs during reductive dechlorination, with, for example, up to 70% enrichment in carbon isotopic values observed when VC is more than 99% degraded. Comparison of fractionation factors (epsilon) indicates significant differences in carbon isotopic fractionation for different iron types and for different CEs. For the lower CEs (cDCE and VC) in particular, both slower reaction rates and larger fractionation are observed for degradation on Peerless vs. Connelly iron. This is the first study to establish a correlation between the rate of abiotic degradation on Fe(0) and the extent of isotopic fractionation, and the first to confirm consistent differences in these two parameters as a function of iron type. The possibility that these differences in kinetics and carbon isotopic fractionation for cDCE and VC are related to differences in branching ratios between competing hydrogenolysis and beta-elimination reactions during reductive dechlorination on the iron surfaces is discussed.     

Nitrate transport in the unsaturated zone: a case study of the riverbank filtration system Karany,Czech Republic

Nitrate transport in the unsaturated zone of a riverbank filtration (RBF) system in Karany, Czech Republic, was studied. Previous study of the system estimated RBF recharge as 60% riverbank filtrate and 40% local groundwater contaminated by nitrates. Nitrate concentrations observed in RBF recently cannot be explained by simple groundwater contamination and a new conception of groundwater recharge is suggested. A two‐component model based on water ¹⁸O data modelled recharge of local groundwater. One component of groundwater recharge is rainfall and irrigation water moving through the unsaturated zone of the Quaternary sediments in piston flow. The second component is groundwater from the Cretaceous deposits with a free water table. Both the components of groundwater recharge have different nitrate concentrations, and resulting contamination of groundwater depends on the participation of water from Quaternary and Cretaceous deposits. Nitrates' origins and their mixing in the subsurface were traced by ¹⁵N data. Nitrate transport from the unsaturated zone is important and time variable source of groundwater contamination. Copyright © 2011 John Wiley & Sons, Ltd.     

Wood export prediction at the watershed scale

Wood export from a watershed is a function of peak annual discharge, but one hydrologic relationship alone does not fully explain observed variability. Consideration of physical processes that influence the amount of wood available for transport is needed. However, wood recruitment, storage, mobilization, breakage, and transport rates and processes remain difficult to quantify. A theoretical wood transport equation focused on variations in discharge was the motivation for investigation into watershed‐specific wood export rates. Herein, multiplicative coefficients categorized by water year type are developed, paired with the equation, and validated to provide a new method for prediction of wood export at the watershed scale. The coefficients are defined as representing a broad suite of watershed processes that encompass spatio‐temporally variable scales. Two complementary datasets from the 1097 km² mountainous North Yuba River, California watershed were used. Wood surveys above New Bullards Bar Reservoir yielded a wood availability estimate of 250 000–300 000 m³ along the channel network. Annual wood export into the reservoir was field‐surveyed in 2010, 2012 and 2013, and estimated in seven years via remotely sensed images over the 30 year study period of water years 1985–2014. Empirical, watershed‐scale wood export rates ranged from 0.3–5.6%. Comparison of predicted quantities using the new DVWP (discharge variations modified by watershed processes) wood export equation to observed wood export quantities resulted in an aggregate error rate of ±10%. When individual wood export quantities were compared, predicted to observed varied by 0.5–3.0 times. Total wood export of 59 000–71 000 m³ was estimated over the 30 year period, yielding a rate of 1.8 to 2.2 m³/year/km². Wood export predictive capabilities at the watershed scale may help water resource and regulatory agencies plan for wood transfers to augment downstream ecosystems. Copyright © 2017 John Wiley & Sons, Ltd.     

Reactive barriers: hydraulic performance and design enhancements

The remediation of contaminated ground water is a multibillion-dollar global industry. Permeable reactive barriers (PRBs) are one of the leading technologies being developed in the search for alternatives to the pump-and-treat method. Improving the hydraulic performance of these PRBs is an important part of maximizing their potential to the industry. Optimization of the hydraulic performance of a PRB can be defined in terms of finding the balance between capture, residence time, and PRB longevity that produces a minimum-cost acceptable design. Three-dimensional particle tracking was used to estimate capture zone and residence time distributions. Volumetric flow analysis was used for estimation of flow distribution across a PRB and in the identification of flow regimes that may affect the permeability or reactivity of portions of the PRB over time. Capture zone measurements extended below the base of partially penetrating PRBs and were measured upgradient from the portion of aquifer influenced by PRB emplacement. Hydraulic performance analysis of standard PRB designs confirmed previously presented research that identified the potential for significant variation in residence time and capture zone. These variations can result in the need to oversize the PRB to ensure that downgradient contaminant concentrations do not exceed imposed standards. The most useful PRB design enhancements for controlling residence time and capture variation were found to be customized downgradient gate faces, velocity equalization walls, deeper emplacement of the funnel than the gate, and careful manipulation of the hydraulic conductivity ratio between the gate and the aquifer.