GROUNDWATER MONITORING AT EARTHEN MANURE-STORAGE STRUCTURES IN IOWA
Modified from article in
Groundwater quality has been monitored since 1994 at three newly-constructed earthen manure storage (EMS) structures in Iowa. These are located in landform regions with differing surficial geologic materials: the Des Moines Lobe in north-central Iowa, the Iowan Surface in east-central Iowa, and within an area of thick loess in western Iowa. Three to seven shallow (<25 ft.) monitoring wells were installed around each structure. Water levels are measured and water samples are collected monthly for nitrate-N, ammonia-N, fecal coliform bacteria, and chloride analysis. Samples are analyzed quarterly for total organic carbon, sulfate, phosphate, and other parameters. No seepage has been detected at the western Iowa site. Samples of liquid manure from the Des Moines Lobe and Iowan Surface structures have also been analyzed. Seepage detected at the Des Moines Lobe and Iowan Surface sites is discussed below.
Des Moines Lobe Site
Surficial deposits at the Des Moines Lobe site consist of uniform, dense, loam-textured subglacial till, overlain by 12-15 feet of supraglacial till with low bulk density values and highly variable textures, which vary from loam, to sandy loam, to sand. An earthen basin, constructed from supraglacial till, is used to store wastes from a 4,500-head finishing operation. The basin began receiving wastes in January of 1994. Seven monitoring wells were installed around the basin. Grab samples of the waste liquid were collected three times. Chloride concentrations were relatively consistent, at 840-1,100 mg/L. Ammonia-N concentrations were more variable, ranging from 2,200-4,400 mg/L. The reason for this large difference is not known. Nitrate concentrations are negligible in the waste liquid. Phosphate-P concentrations were determined once, and were 200 mg/L.
Varying indications of seepage from the basin have been detected at all downgradient wells. Figure 1 shows pertinent monitoring results from Kan-5, which samples the groundwater immediately after it has passed beneath the basin. Indications of seepage include increases in the concentrations of chloride and total organic carbon (TOC), and a sharp decline in nitrate-N and sulfate concentrations. Chloride concentrations began increasing about three months after the basin began receiving wastes, reaching 400 mg/L after three years. A mass balance based on the highest chloride concentrations indicates the water sampled at Kan-5 consisted of a mixture of 55% groundwater and 45% basin seepage. Chloride concentrations have declined to 325 mg/L during the last 18 months of monitoring. This may reflect a decrease in the rate of seepage from the basin; some "sealing" (clogging of throat pores by particulate solids) of the basin may have occurred.
TOC concentrations at Kan-5 increased from less than 10 mg/L to over 200 mg/L during the summer of 1994, and have varied from 30 mg/L to over 1,000 mg/L since then. There appears to be a seasonal trend to the data, with higher concentrations occurring in the summer. The decrease in nitrate-N and sulfate concentrations is a result of denitrification and sulfate reduction, respectively, generated by the anaerobic, organic-carbon rich nature of the seepage from the basin. Fecal coliform bacteria were present in 21 of 34 monthly samples. Concentrations between 10 and 100 colonies/100 ml are most common, and are higher during summer, which suggests the bacteria are surviving longer, and are more likely to be transported to Kan-5, under warmer conditions.
While the chloride data from Kan-5 clearly show the basin is seeping, the major nutrients ammonia-N and phosphate (and presumably potassium) have not shown a similar response. Total phosphate-P concentrations have remained below the 0.5 mg/L detection limit. Ammonia-N concentrations have ranged from near zero to about 5 mg/L, but have often been below 1 mg/L. If ammonia-N and phosphate-P were being transported from the basin as effectively as the conservative ion chloride, water from Kan-5 would likely contain over 1,000 mg/L ammonia-N and 100 mg/L phosphate-P. This indicates these nutrients are largely being retained, via cation-exchange processes, on the clayey materials below the basin. A considerable build-up of nutrients may be occurring below the basin, depending upon the actual volume of seepage. For perspective, the maximum allowable seepage rate for earthen waste structures in Iowa is 1/16th inch/day. If the Des Moines Lobe basin were seeping at this limit, roughly 5,300 pounds of ammonia-N would be retained on the glacial materials beneath the ½-acre basin each year. While ammonia-N is largely retained below the basin, concentrations appear to be increasing over time. This suggests the adsorptive or cation-exchange capacity of the glacial materials beneath the structure is being exceeded, and that further increases in ammonia-N transport to Kan-5 will occur. Organic-nitrogen concentrations have occasionally been measured, and routinely exceed ammonia-N concentrations.
The farthest downgradient well at the Des Moines Lobe site is located about 150 feet away. Water-quality trends are similar to those found at Kan-5, but seepage-related changes are less significant. A mass-balance based on chloride suggests the groundwater passing 150 feet from the basin contains 15% seepage.
Iowan Surface Site
The geology of the Iowan Surface site is characterized by fractured Pre-Illinois glacial till, mantled by colluvium (slopewash materials) and a thin deposit of windblown-silt (loess). Wastes from a 130-sow farrowing and 680-hog finishing operation are stored in a two-cell lagoon, constructed primarily in uniform loam textured, fractured till. Samples of waste were collected twice, in 1995 and 1996, from the north lagoon cell, with higher concentrations observed in the later sampling. Ammonia-N concentrations varied from 270-310 mg/L, and chloride ranged from 160-360 mg/L. One sample was analyzed for total phosphate-P and showed 70 mg/L. Note that concentrations are considerably lower than those measured for the liquid waste at the Des Moines Lobe basin. This reflects the more dilute nature of the wastes in a lagoon, relative to a basin.
Seepage has been detected at all downgradient wells at the IS site. Figure 2 shows the results of the monitoring at well K-4, which is drilled through the downgradient berm of the north lagoon cell. Chloride concentrations increased from 50 to over 350 mg/L, indicating waste liquid was seeping from the basin to the water table. A mass-balance based on chloride concentrations suggests that the groundwater flowing out from the lagoon was comprised of 80% seepage during the last half of the monitoring period. As chloride concentrations have not declined, the lagoon has not "sealed." Nitrate-N concentrations were below 0.5 mg/L for all samples other than the initial collection in October 1993, when 1 mg/L nitrate-N was present. Sulfate concentrations declined across the period, from 25 to 5 mg/L. The relatively low initial concentrations, and their continued decline, is suggestive of denitrification and sulfate reduction, as was noted for the downgradient wells at the Des Moines Lobe site.
TOC concentrations were less than 10 mg/L for the first samples that were collected in 1994. Samples collected during the remainder of the period ranged from 50 to 400 mg/L. The increase likely reflects the input of TOC-rich seepage to the water table. Fecal coliform bacteria were detected in only three monthly samples, at concentrations of 10 to 20 mg/L. Fecal coliforms do not appear to be routinely transported from the north lagoon cell. Ammonia-N concentrations remained below 1 mg/L and appeared to be following a general declining trend through the spring of 1996. Since the summer of 1996, however, ammonia-N concentrations have increased rather sharply, exceeding 5 mg/L in the fall of 1997. The liquid waste in the north cell contains 300 mg/L ammonia-N and 70 mg/L phosphate-P, and chloride analyses indicate that the groundwater at K-4 is roughly 80% seepage. This suggests that if ammonia-N and phosphate-P were being transported from the basin and to Kan-4 as readily as chloride, samples from K-4 would have contained 200 mg/L ammonia-N and 50 mg/L phosphate-P during the latter half of the monitoring period. Ammonia-N and phosphate-P appeared to be effectively adsorbed and retained by the sediments beneath the lagoon cell. However, ammonia-N concentrations are beginning to rise, suggesting the exchange capacity of the sediments below the basin has been utilized. Organic-nitrogen concentrations have occasionally been measured, and routinely exceed ammonia-N concentrations. Ammonia- plus organic-N in September 1997 exceeded 20 mg/L.
The farthest downgradient well at the Iowan Surface site is located 150 feet away. Water-quality trends are similar to those found at K-4, but seepage-related changes are less significant. A mass-balance based on chloride suggests the groundwater passing 150 feet from the basin contains 40% seepage.
Seepage to the water table is occurring at two of the three monitored structures. Indications of this seepage are generally similar at all affected monitoring wells, and include rising concentrations of chloride and organic carbon, and declining concentrations of sulfate and nitrate. Organic-N concentrations appear to be increasing. Concentrations of nutrients such as ammonia-N and phosphate-P have not significantly changed from background conditions until the latter part of the monitoring period. Adsorption of these nutrients has largely removed them from the seepage. Movement of fecal coliforms from the structures does not seem significant. The structures where seepage is readily identified have not sealed during three years of operation, although some slowing of seepage may be occurring at one site, as indicated by declining chloride concentrations.
The results discussed above are from only three of the over six hundred earthen manure-storage structures that have been constructed in Iowa in the last seven years. They are from structures that are relatively new, having been in use four years or less. They are also from "moderately-sized" operations that have been constructed on predominantly fine-grained, clayey materials. How typical the results of this investigation are is unknown. In particular, monitoring of earthen structures built on -- and of -- coarser materials, structures that have been in service longer, or significantly larger structures, may yield quite different results.