Evaluating Four Storm-Water Performance Metrics with a North Carolina Coastal Plain Storm-Water Wetland

Storm-water best management practices (BMPs) are typically assessed using the performance metric of pollutant concentration removal efficiencies. However, debate exists whether this is the most appropriate metric to use. In this study, a storm-water wetland constructed and monitored in the coastal plain of North Carolina is evaluated for water quality and hydrologic performance using four different metrics: concentration reduction, load reduction, comparison to nearby ambient water quality monitoring stations, and comparison to other wetlands studied in North Carolina. The River Bend storm-water wetland was constructed in spring 2007 and was monitored from June 2007 through May 2008. Twenty-four hydrologic and 11 water quality events were captured and evaluated. The wetland reduced peak flows and runoff volumes by 80 and 54%, respectively. Reductions were significant. Concentrations for the following pollutants increased: total kjeldahl nitrogen (TKN), NH4–N, total nitrogen (TN), and total suspended solids (TSS); inflow and outflow concentrations did not change for total phosphorus (TP), while only NO2–3–N and orthophosphorus (OP) concentrations were lower at the outlet. Using a load reduction metric, results were strikingly different, showing positive load reductions of 35, 41, 42, 36, 47, 61, and 49% for these respective pollutants: TKN, NO2–3–N, NH4–N, TN, TP, OP, and TSS. When comparing the effluent concentrations from the wetland to ambient water quality in the Trent River, all effluent nitrogen species concentration were either similar or lower. TP and TSS concentrations leaving the wetland were higher than ambient water quality data. Finally, by comparing pollutant concentrations among different North Carolina wetlands, it is apparent the River Bend wetland received relatively “clean” water and released water with pollutant concentrations comparable to all other studies examined. Major conclusions from this study include: (1) storm-water wetlands sited in sandier soils (such as those of the North Carolina coastal plain) should be considered a low impact development tool and (2) the selection of performance metric has a pronounced bearing on how a BMP’s performance is perceived. Sole reliance on a concentration reduction metric is discouraged.     

Establishing Storm-Water BMP Evaluation Metrics Based upon Ambient Water Quality Associated with Benthic Macroinvertebrate Populations

Storm-water experts agree that the currently used best management practice (BMP) percent removal methodology metric has many flaws, and some have suggested using a BMP effluent concentration metric. This case study examines establishing an effluent target concentration for BMPs that relates to the health of macroinvertebrates in the receiving water. In North Carolina, 193 ambient water quality monitoring stations were paired with benthic macroinvertebrate health ratings collected in very close proximity. Water quality for the sites ranged from excellent to poor and was divided into three distinct ecoregions: Mountain, Piedmont, and Coastal. Statistically significant relationships were found in one or more ecoregions for dissolved oxygen, fecal coliform, NH3, NO2?3?N, total Kjeldahl nitrogen, total nitrogen (TN), and total phosphorus (TP). BMPs can then be selected and designed to meet these target effluent concentrations. Based upon this research, a development, and therefore set of BMPs, in Piedmont North Carolina could be required to release TN and TP effluent concentrations of 0.99 mg/L and 0.11 mg/L, respectively. These concentrations are both associated with “good” benthos health. The new method was most effective in the Piedmont ecoregion, however with more data collection, the Mountain and Coastal ecoregions may also benefit.     

蚯蚓生物滤池处理农村分散生活污水效果研究

通过分析1-7月份蚯蚓生物滤池进出水水质情况,阐述蚯蚓生物滤池对农村分散生活污水处理效果.结果表明:蚯蚓生物滤池对BOD5、NH4+-N及CODcr有较好的去除效果,平均去除率分别为78.2％、77.8％与38.1％,且随着进水水质及温度的变化,其出水水质稳定,具有较强的抗冲击负荷能力;而对TN和TP的去除效果不明显,...     

Pollutant Removal and Peak Flow Mitigation by a Bioretention Cell in Urban Charlotte, N.C.

Bioretention is a stormwater treatment practice that has gained popularity due to its aesthetics, potential to reduce flooding, and early documented improvements to stormwater quality. A bioretention cell in an urban setting was examined in Charlotte, N.C. from 2004 to 2006. Flow-weighted, composite water quality samples were collected for 23 events and analyzed for TKN, NH4-N, NO2-3-N, TP, TSS, BOD-5, Cu, Zn, Fe, and Pb. Grab samples were collected from 19 storms for fecal coliform and 14 events for Escherichia coli (E. coli). There were significant reductions (p<0.05) in the concentrations of TN, TKN, NH4-N, BOD-5, fecal coliform, E. Coli, TSS, Cu, Zn, and Pb. Iron concentrations significantly increased (p<0.05). NO2-3-N concentrations were essentially unchanged. Efficiency ratios for TN, TKN, NH4-N, TP, and TSS were 0.32, 0.44, 0.73, 0.31, and 0.60, respectively. Fecal coliform and E. coli efficiency ratios were 0.69 and 0.71, respectively. Efficiency ratios for Zn, Cu, and Pb were 0.77, 0.54, and 0.31, respectively. Concentrations of Fe increased by 330%. The peak outflow of the bioretention cell for 16 storms with less than 42?mm of rainfall was at least 96.5% less than the peak inflow, with a mean peak flow reduction being 99%. These results indicated that in an urban environment, bioretention systems can reduce concentrations of most target pollutants, including pathogenic bacteria indicator species. Additionally, bioretention can effectively reduce peak runoff from small to midsize storm events.     

Performance of Nitrogen-Removing Bioretention Systems for Control of Agricultural Runoff

This research evaluated nitrogen-removing bioretention systems for control of nutrients, organics, and solids in agricultural runoff. Pilot-scale experiments were conducted with bioretention systems incorporating aerobic nitrification and anoxic denitrification zones with sulfur or wood chips as denitrification substrates. Varying hydraulic loading rates (HLRs), influent concentrations, and wetting and drying periods were applied to the units during laboratory and two seasons of field tests with dairy farm runoff. Total N removal efficiencies greater than 88% were observed in both units with synthetic storm water. In first-season field tests, moderate removal efficiencies were observed for chemical oxygen demand (46%), suspended solids (69%), total phosphorous (TP) (66%), and total N (65%). During the second season, operational changes in the farm resulted in lower organic, solids, and nutrient loadings resulting in improved effluent quality, especially for suspended solids (81% removal) and total N (82% removal). The systems were not hydraulically overloaded even at 20 times the normal HLR.     

Indicator Bacteria Removal in Storm-Water Best Management Practices in Charlotte, North Carolina

Water quality degradation due to pathogen pollution is a major concern in the United States. Storm-water runoff is an important contributor to the transport of indicator bacteria from urbanized watersheds to nearby surface waters. With total maximum daily loads being established to reduce the export of indicator bacteria to surface waters, storm-water best management practices (BMPs) may be an important tool in treating indicator bacteria in runoff. However, the ability of these systems to remove indicator bacteria is not well established. A study in Charlotte, N.C., monitored nine storm-water BMPs (one wet pond, two storm-water wetlands, two dry detention basins, one bioretention area, and three proprietary devices) for fecal coliform and Escherichia coli (E. coli). A wet pond, two wetlands, a bioretention area, and a proprietary device all removed fecal coliform with an efficiency higher than 50%; however, only the wetlands and bioretention area had significantly different influent and effluent concentrations (p<0.05). For E. coli, only one of the wetlands and the bioretention area provided a concentration reduction greater than 50%, both of which had a significant difference in influent and effluent concentrations (p<0.05). Only one of the nine BMPs had a geometric mean effluent concentration of fecal coliform lower than the U.S. EPA target value, while four of the nine BMPs had geometric mean effluent concentrations lower than the U.S. EPA standard for E. coli. This study showed that some BMPs may be useful for treatment of indicator bacteria; however, other BMPs did not perform well. Because wet, nutrient-rich environments exist in many storm-water BMPs, there is a potential for indicator bacteria to persist in these systems.     

Nutrient Retention in Vegetated and Nonvegetated Bioretention Mesocosms

Thirty well-established 240L bioretention mesocosms were used to investigate retention of dissolved nutrients by bioretention systems. Ten mesocosms were comprised of 80?cm sandy loam, ten of 80?cm loamy sand, and ten of pea gravel with 20?cm of loamy sand. Half were vegetated with shrubs/grasses, while the other half had no vegetation (barren). In the first part of our study, the loam and sand mesocosms were dosed with synthetic storm water comprising 0.8?mg?L?1 total phosphorus (TP) and 4.8?mg?L?1 total nitrogen (TN). TP retention in the vegetated loam was 91% compared to 73% in the barren, and TN retention was 81% compared to 41% in the barren loam. TP retention was 86–88% in the sand treatments, while TN retention in the vegetated sand was 64%, compared to 30% in the barren. In the second part of our study, all 30 mesocosms were loaded weekly with 45?cm of tertiary effluent with high nutrient loads (22.3?m?year?1 hydraulic load at a flow-weighted average of 4.5?mg?L?1 TP and 4.8?mg?L?1 TN, or 1,012?kg?ha?1?year?1 TP and 1,073?kg?ha?1?year?1 TN). After 50 weeks of loading, cumulative TP retention was 92% in the vegetated loam, 67% in the sand, and 44% in the vegetated gravel. However, TP retention by barren media was 56% in the loam, 39% in the sand, and 14% in the gravel. Cumulative TN retention was 76% in the vegetated loam, 51% in the sand, and 40% in the vegetated gravel. In contrast, maximum TN removal by barren media was 18% in the loam. The increase in TP retention by vegetated systems substantially exceeds phosphorus uptake rates for plants, suggesting that other processes are involved. The increase in TN retention by vegetated systems also exceeds nitrogen uptake rates for plants, suggesting that denitrification is involved.     

Comparison of Steroid Hormone Concentrations in Domestic and Hospital Wastewater Treatment Plants

Influent and effluent samples originating from two wastewater treatment plants (WWTPs) (treating hospital wastewater and domestic wastewater, Belgium) have been analyzed in order to estimate their steroid hormone content. The natural estrogens estrone (E1), 17β-estradiol (E2), and the synthetic 17α-ethinylestradiol (EE2) together with other steroid hormones progesterone (P) and testosterone (T) metabolites were detected in these samples. The hormone concentrations in both the hospital and the domestic WWTP samples were not significantly different and ranged from <0.2?ng EE2/L to 114?ng EE2/L, from <0.2?ng E1/L to 58?ng E1/L and from <0.2?ng P/L to >100?ng P/L. E2 was detected once at a concentration of 17?ng/L. In the domestic WWTP which comprises a conventional activated sludge treatment in parallel with a membrane bioreactor, no differences in estrogen removal efficiency could be observed for both treatments. In comparison to chemical analysis data, the Yeast Estrogen Screen (YES) appears to underestimate the influent estrogen concentrations, probably due to influent toxicity for the YES. Effluent estrogen concentrations, on the other hand, were overestimated by the YES test, probably due to the presence of other estrogenic compounds in the effluent.     

Field Study of the Ability of Two Grassed Bioretention Cells to Reduce Storm-Water Runoff Pollution

Two grassed bioretention cells including internal storage zones (ISZs) were monitored for 16?months in central North Carolina. Each cell had a surface area of 106?m2 and fill media depths were 0.75 and 1.05?m for the north (North) and the south (South) cells, respectively. Asphalt parking lot inflow and outflows were analyzed for nitrogen and phosphorus forms and fecal coliform (FC). Outflow volumes and peak flows for individual storms were generally less than those of inflow. Overall, except for NO2,3–N, effluent nitrogen species event mean concentrations (EMCs) and loads were significantly (α = 0.05) lower than those of the inflow, and nitrogen species load reductions ranged from 47 to 88%. Apart from fall and winter, during which a longer hydraulic contact time seemed to be needed, the ISZs appeared to improve denitrification. Total phosphorus (TP) and OPO4-P EMCs were significantly lower than those of the inlet. Reductions were 58% (South) and 63% (North) for TP and 78% (North) and 74% (South) for OPO4–P. There was no significant difference in TP and OPO4–P loads between the inlet and the two outlets. Moreover, effluent concentrations for both phosphorus species were low, relative to other studies. The best nutrient EMC and load reductions occurred during the warm and humid seasons. When considering effluent concentrations in addition to removal rates, the grassed cells showed promising results for FC and nutrient pollution abatement when compared to conventionally vegetated bioretention (trees, shrubs, and mulch) previously studied in North Carolina.     

Pollutant Concentrations in Road Runoff: Southeast Queensland Case Study

This paper discusses the results of research into the pollutants in runoff from road pavement surfaces following natural rainfall events. Road runoff water quality was monitored at 21 sites centering around Brisbane, in southeast Queensland, Australia. The sites were selected according to traffic volumes, surrounding land use, pavement surface type, ease of access, and commercial vehicle percentage. Bridge sites were chosen for convenience of sample collection and minimized infrastructure modification. “First flush” grab samplers were permanently installed at each site to collect the first 20 L of runoff from one of the bridge drainage scuppers. The runoff samples were tested for a number of heavy metals, hydrocarbons, pesticides, and other physical characteristics. The observed results fall within the ranges of concentrations reported internationally and nationally but do not typically follow the “30,000 average annual daily traffic” results reported in the United States. Traffic volumes have not been found to be the best indicator of road runoff pollutant concentrations. Interevent duration has been found to be a statistically significant factor for pollutant concentrations. Sites incorporating exit lanes have recorded higher concentrations of acid-extractable copper and zinc, tending to support the hypothesis that brake pad and tire wear caused by rapid deceleration contributes to the concentrations of these metals in road runoff. Laser particle sizing has shown that a significant proportion of the sediment found in the runoff is <100 μm. However, these particulates do settle in water within 24 h, under laboratory conditions. This may be due to the presence of heavy metals.     

Evaluation and Optimization of Bioretention Media for Treatment of Urban Storm Water Runoff

Bioretention is a relatively new urban storm water best management practice. The objective of this study is to provide insight on media characteristics that control bioretention water management behavior. Eighteen bioretention columns and six existing bioretention facilities were evaluated employing synthetic runoff. In columns, the runoff infiltration rate through different media mixtures ranged from 0.28 to 8.15?cm/min at a fixed 15 cm head. For pollutant removals, the results showed excellent removal for oil/grease (>96%). Total lead removal (from 66 to >98%) decreased when the total suspended solids level in the effluent increased (removed from 29 to >96%). The removal efficiency of total phosphorus ranged widely (4–99%), apparently due to preferential flow patterns, and both nitrate and ammonium were moderate to poorly removed, with removals ranging from 1 to 43% and from 2 to 49%, respectively. Two more on-site experiments were conducted during a rainfall event to compare with laboratory investigation. For bioretention design, two media design profiles are proposed; >96%?TSS, >96%?O/G, >98%?lead, >70%?TP, >9%?nitrate, and >20%?ammonium removals are expected with these designs     

Effect of In-Lake Water Quality on Pollutant Removal in Two Ponds

An extensive field study examined pollutant removal in two regional wet detention ponds near High Point, N.C. Substantial differences in influent pollutant concentrations between the ponds caused significant differences in pond water quality and pollutant removal efficiency. In Davis Pond, influent fecal coliform and nutrient concentrations were high because of several large dairy farms in the watershed, resulting in hypereutrophic conditions as evidenced by high chlorophyll-a concentrations, high midday pH values and supersaturated midday oxygen concentrations. In Piedmont Pond, influent fecal coliform and nutrient concentrations were much lower, resulting in mesotrophic to slightly eutrophic conditions. Both ponds thermally stratified and developed an anaerobic hypolimnion. In Davis Pond, annual pollutant removal efficiencies for total suspended solids, volatile suspended solids, total organic carbon, total phosphorus, dissolved phosphorus, nitrate∕nitrite, total ammonia nitrogen, and total nitrogen were 56%, 32%, 15%, 41%, 54%, 16%, 2%, and 11%, respectively. In Piedmont Pond, annual pollutant removal efficiencies were 20%, 30%, 27%, 40%, 15%, 66%, ?64%, and 36%, respectively.     

Filtration Removals of Microorganisms and Particles

Filtration log removals (log?R’s) were determined by pilot plant seeding experiments for a selected array of organisms, i.e., protozoan cysts, algae, bacteria, and viruses, as well as turbidity/particles. Removals of these organisms and particles varied from one species to another within the range 1.4 ? log?R ? 5.1 for “reference,” i.e., specified, conditions. For other conditions, called here “excursion” conditions, log?R’s were found to be proportional to alum dose between zero and “optimum” the latter being defined as the smallest dosage that minimized filter effluent turbidity. Mono and dual media did not show performance differences, and conventional filtration showed slightly higher log?R’s than in-line filtration. The influent concentration did not show a “true” effect on log?R’s (but an “apparent” effect was seen when effluent values were consistently very low, making log?R’s proportional to influent concentration). log?R’s obtained in limited experiments at another site showed significant differences for two algae and two viruses when compared with log?R’s for reference conditions. The results illustrate the log?R variation that may be found among different groups of organisms/particles and indicate the influences of certain independent variables (e.g., influent concentration, dual media versus mono media, conventional filtration versus in-line, alum dose, zero alum dose, and filtration at a different site) on filtration removals. In addition, the importance of sampling repetitions was reinforced. The findings are significant to water treatment practice in that organism removals by filtration are likely to vary, depending on the organism/particle species as well as the filtration conditions. The implication is that a pilot plant study is advisable if confidence in log?R is desired for a given organism/particle.     

Comparison of Pollutant Removal Efficiency for Two Residential Storm Water Basins

Detention basins with a low-flow concrete channel or a vegetated channel are two types of storm water collection basins examined in this study to assess effectiveness in water quality improvement. Influent and effluent data collected from four storm events include flow, petroleum hydrocarbons, nutrients, total suspended solids, three major ions, and indicator organisms. The calculation of influent and effluent mass loading for each basin determines the removal efficiency, which is used to rank the more effective basin for water quality improvement. As expected, the detention basin with a low-flow concrete channel was found to be ineffective for improving the water quality of storm water. The vegetated detention basin was also found essentially ineffective for water quality improvement for all four storms, with low influent mass loading and flushing of stored water the most probable reasons for this result.     

Storm-Water Filter Media Pollutant Retention under Aerobic versus Anaerobic Conditions

Storm-water runoff entering filters is usually aerobic and therefore the removal processes in the filter normally occur under oxidizing and aerobic conditions. However, storm-water filters differ from water and wastewater treatment filters because there are quiescent times when no influent enters the filter and the pore water stagnates. During this stagnation period, anaerobic conditions on a macro- or microscale could develop. This note presents the results of experiments conducted to determine if four potential filter media (sand, activated carbon, peat moss, and compost) could retain previously trapped pollutants when anaerobic conditions develop during interevent periods. The results indicated that permanent retention of heavy metals may occur even in an anaerobic environment (for the media and metals investigated). However, retention of some nutrients may not occur under these conditions, particularly for the organic media. This is an area of concern when the design of filters and bioretention devices includes an internal water storage zone where, between events, anaerobic conditions for nitrate removal are encouraged.     

Water Quality Improvement through Reductions of Pollutant Loads Using Bioretention

As an increasingly adopted storm-water best management practice (BMP) to remedy hydrology and water quality impairment from urban development, bioretention facilities need rigorous investigation to quantify performance benefits and to allow design improvements. This study examines water quality improvements [total arsenic, total cadmium, chloride, total chromium, total and dissolved copper, E. coli, fecal coliform, lead, mercury, nitrogen species, oil and grease, phosphorus, total organic carbon (TOC), total suspended solids, and total zinc] via monitoring for a 15-month period at two bioretention cells in Maryland. Both bioretention cells effectively removed suspended solids, lead, and zinc from runoff through concentration reduction. Runoff volume reduction promotes pollutant mass removal and links BMP water quality benefits with hydrologic performance. From a load perspective (kg/ha?year), all but TOC at one cell showed pollutant reduction. Bioretention effluents exhibited good water quality for all significant pollutants except for nitrate, copper, and phosphorus in one cell, the latter two of which may be attributed to media organic matter dissolution. Copper dissolved/particulate analyses showed that significant changes in copper speciation behavior result from transport through the bioretention media.     

Cell Immobilized FOG-Trap System for Fat, Oil, and Grease Removal from Restaurant Wastewater

Cell immobilized lipase-producing bacteria on three different matrices were incorporated in a fat-, oil-, and grease (FOG) trap system for restaurant wastewater treatment. During a 16-day laboratory-scale experiment for the treatment of synthetic FOG wastewater containing soybean oil, no significant difference (two-tailed t test at 95% confidence interval) in the FOG removal between two systems was observed at FOG influent ≤ 1,000?mg/L. However, the typical trap showed lower FOG removal efficiency than the matrix-based system when the influent FOG concentration was increased to ≥ 5,000?mg/L. In addition, the matrix-based trap system was able to sustain a stable high FOG removal, with <100?mg/L effluent, even at 10,000 mg/L influent FOG. Based on FOG heights measured and mass balance calculations, 97.4 and 99.5% of the total FOG load for 16 days were removed in a typical trap and matrix-based system, respectively. About 93.6% of the removal in the matrix-based was accounted to biodegradation. The 30-day full-scale operations demonstrated a distinguishably better performance in the matrix-based system (92.7±9.06% of 1,044.8±537.27?mg FOG/L) than in the typical trap system (74.6±27.13% of 463.4±296.87?mg FOG/L) for the treatment of barbeque restaurant wastewater. Similarly, matrix-based system revealed higher chemical oxygen demand removal (85.9±11.99%) than the typical trap system (60.4±31.26%). Characterizations of the influent, emulsified, adsorbed and effluent FOG indicated that straight saturated fatty acids constituted the cause of clogging problems in the FOG-trap and piping system.     

Simulating Activated Sludge System by Simple-to-Advanced Models

Models ranging through simple, intermediate, and International Water Association complex activated sludge models (ASMs) were evaluated to compare their ability to describe biomass growth and substrate removal in an activated sludge system. A membrane-activated sludge bench-scale system was used to treat a complex synthetic wastewater over a wide range of operating conditions, ranging from 1 to 15 days solids retention time and 4 to 12 h hydraulic retention time. Total suspended solids, volatile suspended solids (VSSs), and total and soluble chemical oxygen demands (CODs) were monitored in the influent, the reactor, and the effluent. A variety of substrate removal formulations were used with the simple and intermediate models. Although all models provide excellent prediction of biomass growth, the intermediate model was best. Prediction of substrate removal was good with models that incorporated a nonbiodegradable component in the influent. ASM3 was the best model for predicting effluent soluble COD, but overall, the intermediate model was judged best for prediction of mixed liquor VSS and effluent soluble COD.     

Comparison of Continuous and Sequencing Batch Operated Biofilters for Treatment of Gas-Phase Methyl Ethyl Ketone

Although biofiltration has been used successfully to remove and biodegrade a wide variety of gas-phase organic contaminants generated by industrial facilities and environmental remediation efforts, the ability of conventional biofilters to maintain high removal efficiency during short-term, unsteady-state, elevated loading conditions is limited. A promising alternative for improving biofilter performance during transient elevated loading conditions while minimizing the disadvantages of conventional treatment technologies is utilization of adsorption packing media and implementation of sequencing batch operating strategies. In the studies described herein, a continuous-flow biofilter (CFB) and a sequencing batch biofilter (SBB) were operated for more than 300 days to treat a methyl ethyl ketone (MEK) contaminated gas stream. The packing medium for both biofilters consisted of activated carbon coated polyurethane foam cubes. Both biofilters exhibited stable long-term performance with greater than 99% removal of the influent 106?ppmv MEK concentration during “normal” loading conditions. To assess performance during transient loading conditions, on a regular basis the influent MEK concentration of each biofilter was temporarily increased by a factor of 10 to 1,060?ppmv MEK for a duration of either 45 or 60 min. Results are presented which demonstrate how the operational flexibility of SBB systems can be utilized to minimize or eliminate contaminant emissions from biofilters during unsteady-state loading conditions. The SBB was able to remove more than 99% of the influent MEK at a transient loading rate of 380?g?m?3?h?1 and 83% of the influent MEK at a transient loading rate of 760?g?m?3?h?1. In comparison, the CFB exhibited lower MEK removal efficiency.     

Evaluating Bioretention Hydrology and Nutrient Removal at Three Field Sites in North Carolina

Three bioretention field sites in North Carolina were examined for pollutant removal abilities and hydrologic performance. The cells varied by fill media type or drainage configuration. The field studies confirmed high annual total nitrogen mass removal rates at two conventionally drained bioretention cells (40% reduction each). Nitrate-nitrogen mass removal rates varied between 75 and 13%, and calculated annual mass removal of zinc, copper, and lead from one Greensboro cell were 98, 99, and 81%, respectively. All high mass removal rates were due to a substantial decrease in outflow volume. The ratio of volume of water leaving the bioretention cell versus that which entered the cell varied from 0.07 (summer) to 0.54 (winter). There was a significant (p<0.05) change in the ratio of outflow volume to inflow volume when comparing warm seasons to winter. Cells using a fill soil media with a lower phosphorus index (P-index), Chapel Hill cell C1 and Greensboro cell G1, had much higher phosphorus removal than Greensboro cell G2, which used a high P-index fill media. Fill media selection is critical for total phosphorus removal, as fill media with a low P-index and relatively high CEC appear to remove phosphorus much more readily.