[Home]
[Table of Contents] [Next Section]
B-20 Remedial Investigation Report for Former Open Burn/Open Detonation Area
Section 4 - Methodology
The remedial investigation of the B-20 area was conducted in accordance with 30 TAC 335 Subchapter S, the approved Partial Facility Closure Plan, and the Quality Assurance Project Plan (QAPP) appended to the closure plan (ES, 1994a), unless otherwise noted in the appropriate section below. Field tasks completed during this investigation include the following:
|
Prescreening of the entire site for UXO; |
|
|
Magnetometer survey at each of the fifteen ordnance disposal craters; |
|
|
Drilling and sampling of ten soil borings near selected craters and other areas on site; |
|
|
Collection of judgmental and systematic surface soil samples; |
|
|
Collection of groundwater, surface water, and sediment samples; |
|
|
Collection of background surface soil samples for use in statistical calculations of background metals concentrations in three soil types. |
Methods used to complete the field tasks are described below.
4.1 - Unexploded Ordnance Screening
Prior to drilling and sampling activities, the B-20 site was visually screened at the surface to identify potential UXO and clear pathways for the safe passage of field personnel and equipment. The prescreening was performed by a subcontracted team of qualified professionals from UXB International in Chantilly, Virginia. These UXO professionals, graduates of the U.S. Naval Explosive Ordnance Disposal (EOD) School in Indian Head, Massachusetts, are experienced in the identification and handling of UXO.
During screening activities, UXO personnel surveyed the entire B-20 site in 5-foot-wide traverses and visually identified potential UXO on the ground surface. Use of a magnetometer to identify buried UXO was not generally feasible due to the large amount of scrap metal at the entire site. Magnetometer surveys were only conducted at the fifteen craters, as described in Section 4.2.
Visual identification of UXO was based on the knowledge and experience of the UXO professionals. Some explosive compounds, such as TNT, can be visually identified based on their distinctive orange-red or white colors. In some cases, explosive residue was completely enclosed in the steel casing of an ordnance item. The UXO professionals took a conservative approach when these types of items were found and considered them to be UXO. The findings of the prescreening activities are described in Section 5.1.
A magnetometer survey was conducted by the UXO team at each of the fifteen craters identified during the preliminary investigation. These areas represented locations where buried UXO were most likely to be found.
A Foerster Ferex (Mk-26) Ordnance Locator was used at the craters to detect subsurface ferrous metal. This type of magnetometer is the most recent military-approved locator and is in use by U.S. military EOD teams. The Foerster Ferex locator is a hand-held unit which uses two fluxgate magnetometers aligned and mounted a fixed distance apart to detect changes in the earth's ambient magnetic field. These changes in the magnetic field can be due to ferrous metal or magnetic disturbances in soil. Both an audio and metered signal are provided to the operator. The metered signal indicates whether the disturbance is geodetic or metal related. The Foerster Ferex locator can detect large ferrous metal items to a depth of up to 19 feet, according to its manufacturer.
A less sensitive magnetometer, the Schonstedt GA-52B Locator, was used in cases where near-surface metal scrpas caused continuous high readings on the Foerster locator. The Schonstedt GA-52B is similar to the Foerster Ferex locator in that it is a passive dual fluxgate magnetometer which is used to detect subsurface ferrous metal items. However, the Schonstedt locator has only an audio signal, and according to manufacturer specifications, it detects large ferrous items to a maximum depth of 10 feet.
Factors generally affecting the performance of magnetic locators are magnetic properties of soil and rock (such as iron content), mass of a buried object, distance and depth to a buried object, and the amount of permanent magnetism of an object. The detection capabilities of the Foerster Ferex and Schodstedt locators are dependent upon the size of the item versus its depth. The Foerster Ferex locator will also detect disturbances caused by changes in soil conditions. However, according to the manufacturer, the unit's ability to detect metallic items is not affected by local soil conditions because it is designed to accept these soil conditions as nomal background readings on-site.
When magnetic properties of soil or rock are uniform, no local magnetic anomalies or fluctuations in readings are observed. When the magnetometer is passed over any object or native material having nonuniform magnetic properties, a response or fluctuation in reading is recorded. Nearby metal objects cause difficulty in evaluating the cause or source of the readings.
Prior to conducting the surveys, the UXO team removed as much scrap metal as possible at the crater sites to reduce surface interference. A minimum of six traverses were surveyed at each crater; three in a north-south direction and three in an east-west direction. To account for the size of the craters, the traverses were 30 to 50 feet in length. The results were used to identify appropriate drilling locations at selected craters.
Each drilling location was cleared with the magnetometer to a depth of 4 feet below ground surface (bgs). If soil was encountered at a drilling location to a depth of greater than 4 feet, the drilling rig was moved off the hole and a downhole magnetometer survey was performed. This survey cleared the borehole to a depth of 8 feet bgs. The drilling rig then proceeded to that depth. This process was repeated in 4-foot increments until competent limestone bedrock was encountered. Results of the surface and downhole magnetometer surveys are described in Section 5.2.
A total of ten soil borings were drilled and sampled, seven of which were located immediately adjacent to selected craters. These seven included two borings closely spaced, and one boring at each of craters 1, 3, 5, 8, and 9. Two other areas on site which were investigated for subsurface contamination were the rocket motor standpipe and the soil mounds. One soil boring was drilled and sampled next to the rocket standpipe, and two were drilled and sampled near the soil mounds. Soil boring locations are shown in Figure 4.1.
Drilling activities were performed by Jones Environmental Drilling, Inc. of San Antonio, Texas, using a Mobile B-61 drilling rig. The driller was licensed in the State of Texas. Hollow-stem augers were used to advance the boring through unconsolidated soil. This upper soil horizon was variable in thickness, ranging from 0.5 to 11.0 feet, depending upon location. Continuous, undisturbed soil samples were collected with a decontaminated 2-foot long, stainless-steel split-spoon. Downhole magnetometer surveys were conducted after every 4 feet of hole advancement while drilling in unconsolidated material.
Once competent limestone was reached (split-spoon refusal), drilling continued to total depth using air coring techniques. During air coring, continuous rosk (limestone) samples were collected in a decontaminated 5-foot-long, 3-inch-diameter core barrel. A compressor supplied filtered air for removal of cuttings from the borings. The total depths of individual soil borings ranged from 16.0 to 23.9 feet bgs.
During drilling, airborne dust was monitored with an air particulate meter (Miniram). Respiratory protection and goggles were worn when dust could not be physically avoided, and when the level of particulate matter in the breathing zone was above the action level.
The soil and rock core were described by a qualified geologist. The cost was described using the Unified Soil Classification System (USCS). Core samples were described according to rock type, lithology, color (using a Munsell color chart), fossil content, textural features (i.e., bedding), structural features (i.e., fractures, solution cavities), hardness, and moisture content. Lithologic descriptions were recorded on soil boring logs which can be found in Appendix E. These logs also contain sample recovery intervals, analytical sample intervals, a graphic log and comments concerning drilling and sampling.
Upon completion, all ten soil borings were grouted to ground surface with a cement/bentonite grout slurry. The slurry was a mixture of 8 gallons potable water, one 94 pound sack of Type I Portland cement, and 4 to 5 pounds of bentonite powder. The grout was placed into the open borehole through 1.5-inch polyvinyl chloride (PVC) tremie pipe. A brass monument was placed at the top of each grouted soil boring location for identification.
Subsurface soil samples were collected at B-20 to determine if soil contamination had occurred as a result of below grade detonation of explosives and to investigate teh possibility of shallow (perched) groundwater contamination in certain localized areas. Two subsurface soil or rock samples were collected from each soil boring at seven locations associated with craters. One subsurface soil or rock sample was collected at a depth associated with the historical pit bottom, estimated to be between 12 and 15 bgs. The other subsurface soil sample was collected at the total depth of drilling, approximately 20 feet bgs, to determine if potential contamination had migrated downwards.
Because the possible source of contamination was known to be aboveground or in the shallow subsurface at the mounds and the rocket motor standpipe, sample depths within the three borings at these locations were shallower than sample depths of borings drilled near craters. Collected samples are described in Table 4.1.
Rock core samples were broken into small pieces with a decontaminated hammer and chisel. The crushed sample was then transferred to the appropriate sample bottle. All sampling equipment was decontaminated as described in Section 4.9.
In addition to samples collected for chemical analysis, rock core was sampled for geotechnical characterization. Geotechnical samples were collected from three soil borings at depths adequate for geotechnical characterization of the unsaturated zone. Samples were collected into gallon-sized airtight plastic bags. These bags were labeled with the date, time of collection, location, and analysis to be performed. The geotechnical laboratory, IHS Geotechnical of San Antonio, Texas, picked up the samples at CSSA. The analyses are summarized in Table 4.2.
A thin perched water zone was encountered in one soil boring location (SB1) at approximately 10.5 feet bgs. The boring was left open for 24 hours to allow water to accumulate overnight. A grab water sample was collected from the open borehole SB1 using a decontaminated Teflon bailer and new nylon rope.
A total of 43 surface soil samples (or approximately 1.5 samples per acre over the site area) were collected during the remedial investigation at the B-20 site. These samples consisted of 21 judgmental samples and 22 random systematic samples. The samples, their locations, and the analytical methods are listed in Table 4.3. The locations of the surface soil samples are also shown in Figure 4.2.
All surface soil samples were collected at a depth of approximately 0.5 foot. The exposed surface soil was cleared away by a decontaminated shovel or trowel. The sample was then collected with a decontaminated trowel into a stainless steel bowl. Rocks and vegetative material were removed, and the soil was mixed to collect a homogeneous sample. The sample was transferred into appropriate sample jars, placed on ice, and prepared for shipment. All sampling equipment was decontaminated as described in Section 4.9.
The soil characteristics of each sample was described in the field logbook by a geologist. Descriptions included soil type, color (using a Munsell color chart), moisture content, and plasticity. Soil descriptions are provided in Table 4.3.
4.4.1 Judgmental Surface Soil Sampling
The locations of the judgmental surface soil samples were based on prior knowledge of site uses and observations made during sire prescreening. A total of 21 judgmental surface soil samples were collected at the site.
According to the Partial Facility Closure Plan (ES, 1994a), fourteen judgmental samples were to be collected at and around several of the craters. Surface soil within seven of the fifteen craters was sampled to determine if explosives demolition activities at the craters had caused soil contamination. Surface soil within 50 feet of each of these seven craters was also sampled to determine if explosive material had been "kicked out" during explosive demolition activities.
Based on site prescreening findings, seven additional judgmental surface soil samples were collected in areas having a high possibility of contamination. Three samples were collected at locations where raw explosives were found during site prescreening. Three samples were collected at small ammunition disposal areas at the site, including B-21. One sample was collected downgradient of the B-21 disposal area to determine if lead contamination had migrated toward a small branch of the ephemeral stream.
4.4.2 Systematic Surface Soil Sampling
The systematic random sampling consisted of surface soil sampling on a 250-foot grid which covered the entire site. The starting point for the layout of the grid was randomly chosen. Where the grid point and craters nearly intersected, only one sample was collected. A total of 22 systematic surface soil samples were collected.
In addition to samples collected for chemical analysis, six surface soil samples were collected for geotechnical characterization. At least one geotechnical sample was collected from each of the soil types located on the site. Samples were collected into gallon-sized airtight plastic bags labeled with the date, time of collection, location, and analyses to be performed.
surface water samples were collected from the ephemeral stream downgradient of B-20 and from the small onsite pond to determine if site waste disposal/thermal treatment activities had caused any surface water contamination. In addition, surface water in three craters (craters 8, 12, and 13) was sampled to determine if explosive demolition activities caused the water which occasionally accumulates in these craters to be contaminated.
Surface water samples were not collected from areas that did not contain contaminants above comparison criteria during the preliminary investigation (see Section 2.4). Therefore, no surface water sample was collected from the livestock pond. The ephemeral stream was not sampled on site because it was dry upstream of the small onsite pond. Figure 4.3 shows the locations of the samples. The samples are described in Table 4.4.
Surface water samples were collected using a decontaminated pond sampler or stainless steel bowl. Care was taken when collecting the sample to avoid collecting organic matter. The sample was then transferred into an appropriate sample bottle. All sampling equipment was decontaminated following the procedures described in Section 4.9.
Sediment samples were collected at the B-20 site and at the livestock pond to determine if sediments had been impacted by site waste disposal/thermal treatment activities. Samples were collected from three craters receiving some site drainage, the ephemeral stream crossing the site, the small pond located in the east portion of the site, and the livestock pond northeast of the site. Sediment sample locations are shown in Figure 4.3, and sample descriptions are provided in Table 4.4.
The three sediment samples collected at approximate 400-foot intervals along the ephemeral stream are considered to be random systematic samples. Judgmental sediment samples were collected from the livestock pond, the small pond, crater 8, crater 12, and crater 13.
Sediment samples were collected with either a decontaminated stainless-steel hand auger or shovel. The sample was then homogenized in a stainless-steel bowl and placed into the appropriate sample jar. Sediment samples for volatile organic analyses were collected first and were not homogenized. All sampling equipment was decontaminated according to the procedures described in Section 4.9.
A total of thirty-five background soil samples were collected for use in a statistical evaluation to establish background metals concentrations in three types of surface soil (Krum complex soils, Brackett-Tarrant association soils, and Crawford and Bexar stony soils) at the site. Locations representative of background conditions were chosen by the field team leader and a representative of CSSA familiar with present and past property uses. Areas of the site not known to be impacted by any waste management activities were chosen to represent background conditions in the statistical evaluation to determine background constituent concentrations. Soil type locations were based on the soil survey for Bexar County (USDA, 1966). Background sample locations are shown in Figure 4.4.
All background surface soil samples were collected at a depth of approximately 0.5 foot. The surface soil was cleared away with a decontaminated shovel or trowel. The sample was then collected with a decontaminated trowel into a stainless-steel bowl. The soil was mixed to collect a homogeneous sample and any rocks or vegetation were removed from the sample. The sample was then transferred into the appropriate sample jar and prepared for shipment. Normal sample handling, decontamination, and quality assurance procedures (described in Section 4.8, Section 4.9, and Section 4.13, respectively) were followed for the background sampling activities.
4.8 - Sample Handling Procedures
A sample numbering system was used to identify each environmental and blank sample collected during the field investigation. This numbering system was used for tracking to allow retrieval of information about a particular location and to ensure that each sample was uniquely numbered. A listing of sample numbers was maintained by the field team leader.
Each sample number consisted of a site identification code (B-20, or BKG for background), a location identification code and a consecutive sample number. The depths at which soil boring samples were collected follows the sample number in parentheses. The numbering system for duplicate samples began with the number 100. Equipment rinsate and trip blanks were also identified using an alphanumeric identification code.
The first two characters of the location identification number were one of the following:
|
SB = Soil boring sampling location |
|
|
SS = Surface soil sampling location |
|
|
SW = Surface water sampling location |
|
|
EB = Equipment rinsate blank |
|
|
TB = Trip blank |
4.8.2 Sample Handling Protocol
Sample handling procedures, as described in the Partial Facility Closure Plan (ES, 1994a), were followed to maintain sample integrity. All samples were collected into their appropriate glass and plastic bottles with Teflon-lined lids. All sample bottles used during this investigation were new bottle sent to CSSA by the analytical laboratory. Sample labels were affixed to each container to identify the collector's name, date and time of collection, project site name, sample number, analysis to be performed, and preservative added (if any).
Sample containers were placed on ice for storage and shipment. Individual sample bottles were wrapped in bubble pack and placed in sealed plastic bags to prevent breakage during shipment. The bags were placed into insulated shipping coolers with ice to maintain a proper temperature (4°C). A chain-of-custody (COC) record describing the contents of the cooler was placed in a sealed plastic bag and taped to the upper lid of the cooler.
Standard sample COC procedures were maintained. Samples were kept in a secured area when not in the immediate possession of the sampler. Custody seals were placed on the coolers to prevent tampering during shipment. The sealed sample coolers were shipped via overnight delivery.
4.9 - Decontamination Procedures
All drilling and sampling equipmen was decontaminated before use at the site. The drilling rig, augers, core barrel, and all downhole sampling equipment were decontaminated with a high-pressure steam cleaner. All steam cleaning activities were conducted in the southwestern corner of the site (Figure 4.1).
All sampling equipment, including the pond sampler, shovels, hand auger, split spoons, trowels, chisels, and Teflon bailer, were decontaminated prior to use with an Alconox soap scrub wash, potable water rinse, and distilled water rinse. Split sections of PVC pipe used to hold the core samples were also decontaminated in this manner. Decontaminated equipment that was not used immediately was allowed to air dry and then wrapped with aluminum foil for storage or transport.
Horizontal coordinates of all sampling locations at the B-20 site were surveyed by a licensed surveyor. Horizontal and vertical coordinates of site features (such as craters, ponds, and soil mounds) were also surveyed. The boundary of the site was surveyed for future deed certification purposes. All surveying was conducted by UXB International of Chantilly, Virginia.
Horizontal control was referenced to existing CSSA datum. All points required to control the survey were occupied as stations within a closed and adjusted traverse using procedures to meet or exceed third-order accuracy standards. The surveyors established temporary horizontal control referenced to the North American Datum of 1983 [NAD83(86)] and to Texas South Central (4204) State Plane Grid System, U.S. Foot Definition (1 inch = 0.08333 foot). Vertical control was referenced to the National Vertical Geodetic Datum (NVGD) of 1985. Controlling points were part of a closed and adjusted level loop.
4.11 - Management of Investigation-Derived Waste
Management of Investigation-Derived Wastes During Site Inspection (EPA, 1991a) and TNRCC comments to the B-20 partial facility closure plan (ES, 1994a) were used as a guidance for waste disposal methods during this project. Wastes resulting from this investigation consisted of soil cuttings, decontamination fluids, and personal protective equipment (PPE).
Soil cuttings and rock core from each drilling location were placed into plastic garbage bags, labeled with the soil boring number, and placed into a 55-gallon drum. The drum was closed tightly and labeled with information including site name, date, and contents. The drum was left on site for proper disposal. Disposal options will be determined after all investigative actions and corrective measures have been completed.
All decontamination rinsate was disposed of on the ground surface in one designated area of the site. This included decontamination rinsate from all steam cleaning activities and from the decontamination of sampling equipment (see Section 4.9).
Personal protective equipment, which consisted solely of nitrile gloves, were disposed of in general waste dumpsters at CSSA.
Analytical techniques followed procedures described in Test Methods for Evaluating Solid Waste, U.S. Environmental Protection Agency, SW-846, November 1986 (EPA, 1986). All data were analyzed and reported using level 3 quality control and reporting requirements. Level 3 data can be used for site characterization, risk assessment, and remedial alternatives screening. Laboratory analytical methods are summarized on Table 4.5. Practical quantitation limits (PQL) for each method are listed in Table 4.6. All analyses, except method 8330, were performed by Terra Laboratories in League City, Texas. Method 8330 analyses were performed by Inchcape Laboratory in Richardson, Texas.
A summary of the number of surface soil, subsurface soil, and sediment samples collected per analytical method is shown on Table 4.7. Background samples and QA/QC samples (such as trip blanks, duplicates, and matrix spike/matrix spike duplicates) are included in the total sample count per analysis. A summary of the number of surface water and groundwater samples collected per analytical method is shown on Table 4.8.
In general, target compounds during this investigation were explosives compounds and metals. In addition, sediment samples were analyzed for volatile organic compounds (VOCs) and semivolatile organic compounds (SVOCs). Surface water samples were also analyzed for biological oxygen demand (BOD). This section describes the sampling program in detail.
4.12.1.1 Judgmental Surface Soil Samples
A total of fourteen judgmental surface soil samples were collected from within or nearby craters. Eight of the fourteen samples were analyzed for nitroaromatics and nitramines by EPA SW-846 method 8330. All of the judgmental samples collected within and nearby craters were analyzed for arsenic by method 7062; barium, cadmium, and chromium by method 6010; lead by method 6010 or 7420; and mercury by method 7471. Since lead levels in many samples, particularly in surface soil samples, exceeded the practical quantitation limit of both the more sensitive method 7420 and the more general method 6010, method 6010 was often used to avoid the necessity of multiple sample dilutions.
In addition, three judgmental surface soil samples were collected at small ammunition disposal areas at the site. One additional sample was collected downgradient of one of the piles. All four samples were analyzed for the metals listed above. Three judgmental samples were collected at locations where raw explosives were found during site prescreening. Two of these samples were analyzed for nitroaromatics and nitramines. The remaining sample was inadvertently analyzed for metals only.
4.12.1.2 Systematic Surface Soil Samples
A total of twenty-two systematic surface soil samples were collected. All of the samples were analyzed for arsenic by EPA SW-846 method 7062; barium, cadmium, and chromium by method 6010; lead by method 6010 or 7420; and mercury by method 7471. In addition, thirteen of the samples were analyzed for nitroaromatics and nitramines by method 8330.
4.12.1.3 Subsurface Soil Samples
A total of twenty subsurface soil or rock samples were collected during the drilling operations. All subsurface soil or rock samples were analyzed for nitroaromatics and nitramines by method 8330; arsenic by method 7062; barium, cadmium, and chromium by method 6010; lead by method 7420 (with the exception of one sample which was analyzed for lead by method 6010); and mercury by method 7471.
Eight sediment samples were collected during this investigation. All of the samples were analyzed for semivolatile organic compounds by EPA SW-846 method 8270; arsenic by method 7062; barium, cadmium, and chromium by method 6010; and mercury by method 7471. Six of the sediment samples were analyzed for lead by method 6010, and the remaining two were analyzed by method 7420. Six of the samples were analyzed for VOCs by method 8240, and seven were analyzed for nitroaromatics and nitramines by method 8330.
The groundwater sample collected from SB1 was analyzed for nitroaromatics and nitramines by method 8330; arsenic by method 7062; barium, cadmium, and chromium by method 6010; lead by method 7421; and mercury by method 7471.
4.12.1.6 Background Surface Soil Samples
A total of thirty-five background surface soil samples were collected as part of this investigation. The samples were analyzed for arsenic by method 7062; barium, cadmium, and chromium by method 6010; lead by method 6010 or 7420; and mercury by method 7471.
Five surface water samples were collected. All five samples were analyzed for arsenic by EPA SW-846 method 7062; barium, cadmium, and chromium by method 6010; lead by method 7421; and mercury by method 7471. In addition, the three samples collected within craters 8, 12, and 13 were also analyzed for nitroaromatics and nitramines by method 8330. Four of the samples were analyzed for BOD by method 405.1.
A total of nine samples were collected for geotechnical analysis (see Table 4.2). Core samples collected from three soil borings were tested for porosity by American Society of Testing and Materials (ASTM) method D4404 and permeability by ASTM method D4525. In addition, six geotechnical samples were collected along with the systematic random surface sampling. These were tested for grain size by ASTM method D422, moisture content by method D2216, and pH by EPA method 150.1. Cation exchange capacity (CEC) analyses were conducted in accordance with standard Soil Science Society of America (SSSA) methods (SSSA, 1982). All geotechnical analyses were performed by IHS Geotechnical in San Antonio, Texas.
4.13 - Quality Assurance/Quality Control
To check field and laboratory QA/QC procedures, trip blanks, rinsate blanks, field duplicate samples, and matrix spike/matrix spike duplicates (MS/MSDs) were sent to the laboratory.
A trip blank was sent to the laboratory whenever samples to be analyzed for VOCs were included in the shipment. A trip blank is a sample bottle filled by the laboratory with analyte-free laboratory reagent-grade water, transported to the site, handled like a sample but not opened, and returned to the laboratory for analysis. The purpose of the trip blank is to determine if any VOCs are introduced into samples during shipment.
Equipment rinsate blanks were collected at a frequency of one per twenty samples per matrix (surface soil, subsurface soil, sediment, surface water, and groundwater). Equipment blanks were collected by pouring analyte-free deionized water into the sampling equipment before transferring the water into the sample bottle. These samples were collected to determine if decontamination procedures were sufficient.
Field duplicate samples were collected to determine the accuracy and precision of he laboratory analysis. These samples were collected at a frequency of one per ten samples per matrix. Wehn it was time for a duplicate sample, sufficient volume for two samples was collected. The sample was then split between two bottle sets. The duplicate sample identifications (IDs) were disguised to prevent laboratory bias.
MS/MSD samples were collected at a frequency of one per twenty samples per matrix to determine the effect of matrix interference on the analytical results. Collection of MS/MSDs was similar to collection of duplicate samples.
All analytical reports were validated by a chemist to ensure that the data met EPA level 3 reporting and methodology requirements. The procedures used during this validation are described below. Results of the data validation are included in the data validation report (Appendix F).
All analytical reports submitted by the laboratory were checked to determine if reporting requirements specified in the QAPP appended to the B-20 partial facility closure plan (ES, 1994a) were met. If any deficiencies or problems were found, the laboratory was contacted and the problem was resolved, either through verbal or written communication.
Holding times were established by comparing the sampling dates on the COC forms with dates of analysis on the raw sample data. If holding times listed in Table 4.3 and Table 4.4 of the QAPP were exceeded, all appropriate results greater than the detection limit were qualified as estimated (flagged "J"), and all appropriate results less than the detection limit were qualified as estimated nondetected (flagged "UJ").
4.14.3 Instrument Performance Check
Data for each performance check were compared with criteria for instrument performance checks listed in the QAPP. If mass assignment was in error, associated data was qualified as unusable (flagged "R"). If ion abundance criteria are not met, professional judgment can be applied to determine to what extent the data may be used. These data validation measures were not necessary during this investigation.
Data presented for each instrument calibration were compared to calibration criteria listed in EPA Laboratory Data Validation Functional Guidelines for Evaluating Inorganic Analyses (EPA, 1988) and the EPA National Functional Guidelines for Organic Data Review (EPA, 1991b). If gas chromatograph/mass spectrometer (GC/MS) methods did not meet continuing calibration relative response factor (RRF) criteria (>0.005), results greater than the detection limit were qualified as unusable (flagged "R"). If GC/MS methods did not meet percent difference (%D) criteria (>25.0) between the initial calibration RRF and the continuing calibration RRF, results that were greater than the detection limit were qualified as estimated (flagged "J") and results less than the detection limit were qualified as nondetected estimated (flagged "UJ").
If the minimum number of standards were not used for each metal method calibration, the associated sample results were qualified as unusable (flagged "R"). If the initial calibration correlation coefficient was less than 0.995 for any metal, the associated sample results greater than the detection limit were qualified as estimated (flagged "J") and results less than the detection limit were qualified as nondetected estimated (flagged "UJ").
The submitted data were reviewedto determine that all required method blanks were analyzed. The results of all associated blanks were reviewed to evaluate the presence of analytes in the blanks. If the appropriate method blanks were not analyzed at the required frequency, professional judgment was applied to determine the extent to which the data may be qualified. If a blank was found to contain an analyte, no action was taken if the analyte was not detected in associated samples. If the blank contained an analyte and the result in associated samples was less than five times (ten times for common laboratory organic contaminants such as acetone and methylene chloride), the concentration in the blank qualified results as nondetected (flagged "U"). If the results in the associated samples were greater than five times the blank result, no action was taken.
Submitted GC/MS data were reviewed to determine if surrogate spiking criteria specified in the QAPP were met. If one or more surrogate in a method 8240 analysis, or two or more surrogates in either the base/neutral or acid fraction of method 8270 did not meet surrogate spike recoveries (%R), the following procedures were reviewed to determine appropriate actions:
If the surrogate %R was greater than its upper acceptance limit, analytes in the sample (fraction) with results greater than the detection limit were qualified as estimated (flagged "J").
If the surrogate %R was greater than 10% but less than the lower acceptance limit, analytes in the sample (fraction) with results greater than the detection limit were qualified as estimated (flagged "J"), and analytes in the sample (fraction) with results less than the detection limit were qualified as estimated nondetects (flagged "UJ").
If any one surrogate %R was less than 10%, analytes in the sample (fraction) with results greater than the detection limit were qualified as estimated (flagged "J"), and analytes in the sample (fraction) with results less than the detection limit were qualified as unusable (flagged "R").
4.14.7 Matrix Spike Sample Recovery
MS/MSD %R results were inspected and compared to advisory limits. No action is taken on MS/MSD data alone for the samples other than for the sample which was spiked. However, using professional judgment, the MS/MSD results may be used in conjunction with other quality control criteria in the determination for some qualification of the data. Samples for which an MS/MSD was analyzed and had one or more MS analytes fail to meet criteria were qualified as estimated (flagged "J" or "UJ"). If the MS analyte had %R less than 10% (30% for metals), all analytes in the sample with results less than the detection limit were qualified as unusable (flagged "R").
4.14.8 Laboratory Control Samples
The GC/MS laboratory control sample (LCS) results were analyzed to determine if they fell within the advisory limits. All metals analyses (except mercury) and GC/MS water analyses require that an LCS be analyzed. Action was taken on LCS data for GC/MS using professional judgment based on the number of compounds exceeding criteria and the magnitude of exceedance. Action was taken for metals if the %R limits of 80 to 120% were exceeded. If %R was between 50 and79% or greater than 120%, associated sample results for that analyte were qualified as estimated (flagged "J"). If a metals LCS had %R between 50 and 79%, associated sample results less than the detection limit were qualified as estimated non-detects (flagged "UJ"). If a metals LCS had %R of less than 50%, all associated samples were qualified as unusable (flagged "R").
Internal standards were checked for meeting the following criteria: 1) internal standards used for each GC/MS analyses must have areas within 50 to 200% of the area of the internal standard in the continuing calibration, and 2) the internal standard must also elute within 30 seconds of the time that is eluted during the continuing calibration. If an internal standard did not meet the area criteria, all associated analyte results in the sample that were greater than the detection limit were qualified as estimated (flagged "J"). Analytes with results less than the detection limit that were associated with an internal standard that had %R less than 50 are qualified as estimated nondetects (flagged "UJ"). If extremely low area counts were reported, or other factors indicated a severe loss of sensitivity, analytes with results less than the detection limit were qualified as unusable (flagged "R"). If an internal standard did not meet retention time criteria, professional judgment was used in qualifying the results.
4.14.10 Duplicate Sample Recovery
Atomic adsorption (AA) metals analysis reports were checked to determine if samples were injected twice and if the relative percent difference (RPD) between the two samples was less than 20%. If these duplicate injection criteria were not met, the analyte result for the sample was qualified as estimated (flagged "J" or "UJ").
4.14.11 AA Post-Digestion Spike Recovery
The %R for post-digestion spikes (PDS) of AA metals must be within 85 to 115%. If PDS recovery did not meet criteria, the foolowing action was taken:
If PDS %R was less than 40%, the analyte results greater than the detection limit were qualified as estimated "flagged "J").
If PDS %R was greater than 10% and less than 40%, the analyte results less than the detection limit were qualified as estimated nondetects (flagged "UJ").
If PDS %R was less than 10%, analyte results less than detection limit were qualified as unusable (flagged "R").
If PDS %R were not within 85 to 115% and the sample absorbency was less than 50% of the PDS absorbency, the analyte results were qualified as estimted (flagged "J" or "UJ").
If PDS %R is not within 85 to 115% and the sample absorbency is greater than 50% of the PDS absorbency, the method of standard additions (MSA) was employed.
a. If MSA was required but not doem, results were qualified as estimated (flagged "J").
b. If incorrect MSA spiking levels were used, results were qualified as estimated (flagged "J").
c. If the MSA correlation coefficient was less than 0.995, results were qualified as estimated (flagged "J").
4.14.12 Field Duplicate Samples
Field duplicate samples were checked to determine if they were analyzed at the required frequency. Results were inspected to determine if %R results were within advisory limits. Using professional judgment, the field duplicate results may have been used in conjunction with other quality control criteria in the determination for some qualification of the data.
4.15 - Statistical Evaluation of Background Metals Concentrations
Analytical results for background surface soil samples collected during two investigations at CSSA were combined to calculate the background metals concentrations of three soil types. In accordance with the TNRCC risk reduction rules (30 TAC 335, Subchapter S), statistically determined background levels for a facility can be used as cleanup levels for SWMUs at that facility. Since all of the background samples were collected at CSSA, the statistically determined background levels can be used as cleanup levels for all SWMUs at CSSA on a soil-specific basis. Background soil samples were collected during the following investigations:
|
F-14 Closure Investigation, February 1994 (ES, 1994c). Soil samples BKGR-SS1 through BKGR-SS10 were collected from background locations shown on Figure 4.4. These background samples consisted of seven different soil types: Brackett soils, Brackett-Tarrant association, Crawford and Bexar stony soils, Krum complex, Lewisville silty clay, Tarrant association, and Trinity and Frio soils. Samples were collected between 0.5 and 1.0 foot below ground surface and analyzed for metals using method SW-6010. |
|
|
B-20 Investigation, December 1994. Soil samples BKGR-SS11 through BKGR-SS35 were collected from background locations shown on Figure 4.4. These background samples consisted of three soil types: Brackett-Tarrant association, Crawford and Bexar stony soils, and Krum complex. These three soil types are found at the B-20 site. Samples were collected between 0.5 and 1.0 foot below ground surface. Samples were analyzed for arsenic (SW-7062); barium, cadmium, and chromium (SW-6010); lead (SW-7420 or SW-6010); and mercury (SW-7471). |
Background metals concentrations in the Glen Rose Formation were also statistically calculated. Background Glen Rose limestone samples were collected during the F-14 closure investigation (ES, 1994c). These Glen Rose limestone samples, which were collected at various depths ranging from 4.5 to 20 feet bgs, were analyzed for metals by EPA method SW-6010.
The background concentrations were calculated using information presented in two U.S. Environmental Protection Agency documents"
|
"Statistical Analysis of Ground-Water Monitoring Data at RCRA Facilities," Interim Final Guidance, February 1989 (EPA, 1989a). |
|
|
"Statistical Analysis of Ground-Water Monitoring Data at RCRA Facilities," Draft Addendum to Interim Final Guidance, July 1992 (*EPA, 1992a). |
Although the Partial Facility Closure Plan indicated that the student-t test would be used for determining background metals concentrations, this test proved to be impractical for this application. Use of the student-t test is suggested in the risk reduction rules, but is only suitable for populations with normal distribution. Preliminary use of the student-t test consistently resulted in "cleanup" values below at least three of the ten actual background values. Clearly, this statistical method is not acceptable for this situation since it suggests that background metals concentrations require cleanup, or that the background samples were collected in areas of waste management activities. The background sample locations were based on records of all SWMUs at CSSA (ES, 1993b) and were carefully reviewed by Parsons ES and CSSA personnel in the field prior to collection of any background samples. Therefore, after evaluation of sample locations and a preliminary student-t test analysis, it was concluded that the background sample locations were acceptable for calculation of background metals levels but that the student-t test was not appropriate for the number of samples collected.
The background concentrations were therefore calculated with a Tolerance Interval test, using a one-sded tolerance interval to estimate the upper bound on a large fraction of the concentration distribution. Use of the Tolerance Interval test for this purpose was recently approved by the TNRCC in a similar study of background metals levels at a nearby U.S. Air Force facility (Kelly AFB, 1994). In addition, the UTL is referenced in the EPA documents listed above as an approved method to compare background monitoring data to compliance wells (EPA, 1989a) and 1992b). For background soil data, the UTL predicts the upper range of background concentrations from a relatively small data set.
The UTL is designed for use on data that consist mainly of positive detections. Since background data sets typically contain many non-detects, several tests and procedures must be conducted on those sets of data. Non-detect data must be evaluated and manipulated in one of three ways depending on the percentage of non-detects within the sample population. After screening for non-detects, the data must be tested for normality. The UTL assumes a normal or lognormal distribution. Specific procedures used to evaluate background levels are described below. Figure 4.5 is a flowchart of the procedures followed to determine the UTL.
4.15.3 Procedures for Non-Detects
If an analyte was present at a concentration that is less than the detection limit of the method, the analytical result was reported as not detected with a detection limit rather than as a concentration. All data with "U" or "UJ" qualifiers were considered to be non-detect. the procedures for addressing below detection limit values depended on the percentage of non-detect values in the data set. There were three possibilities:
For less than 15% non-detect results, the non-detect values were replaced with a value equal to half the PQL. The distribution of the data and the appropriate UTL (normal or lognormal, based on the distribution) were then determined.
For between 15% and 50% non-detect results, Cohen's or Aitchinson's adjustment was made to the sample mean and the standard deviation to continue with a parametric UTL.
For greater than 50% non-detect results, a non-parametric UTL was used. A non-parametric UTL is not based on a normal or lognormal distribution. The largest value observed in the data set was used as the non-parametric tolerance limit. Detection limits were included (for non-detects values) in determining the largest observed value unless the analysis was performed on a laboratory dilution.
Probability plots, probability plot correlation coefficients, and the Shapiro-Wilk test were used to determine if the data were normally or lognormally distributed. Probability plots were also used to visually screen the data for statistical outliers. Some outliers were eliminated from the statistical evaluation for background concentrations. Elimination of outliers is clearly noted on the probability plots.
Probability plots show the normal cumulative probability plotted against the sample concentrations. If the points plot approximately on a straight line, the underlying distribution is approximately normal (or lognormal). Probability plots are also useful for identifying irregularities within the data and indicating the presence outlier values that do not follow the basic pattern of the data (EPA, 1992b). Because computer software was used to evaluate the data, the y-axis of the plot was generated using normal quantile values, instead of the normal cumulative probability. The y-coordinate was computed using the following formula:
yi = Ø-1 (i/ n+1),
where Ø-1 denotes the inverse of the cumulative normal (or lognormal) distribution, n represents the size of the data set, and i represents the rank position of the ith ordered concentration (EPA, 1992b).
The Shapiro-Wilk Test is based on the premise that if a set of dta are normally (or lognormally) distributed, the ordered values should be highly correlated with corresponding quantiles taken from a norm (or lognormal) distribution (Shapiro and Wilk, 1965). The Shapiro-Wilk test can be used for data sets with 50 samples or less. The Shapiro-Wilk Test statistic (W) will tend to be large when a probability plot of the data indicates a nearly straight line (normal distribution). When the data show significant bends or curves, W is small. The test statistic W was calculated using the following formula:
W = [b/(S*SQRT(n-1))]-2
where b = S xj* aj for values from (n-1) to k
The value x(j) represents the jth smallest ordered value in the sample, coefficient aj depends on the sample size n, S is the standard deviation, n is the number of samples, and k is the greatest integer less than or equal to n/2.
Normality (or lognormality) of the data was rejected if the Shapiro-Wilk statistic was lower than the critical values (Shapiro and Wilk, 1965). In cases where both the normal and lognormal distributions tested positive, the distribution was selected based on the value of the test statistic, W. Because a higher value of W indicates a more normal distribution of the data examined (whether original data or log-transformed data), the test with higher value is considered to indicate the distribution which is a closer fit to the data.
4.15.4.3 Probability Plot Correlation Coefficient
The probability plot correlation coefficient measures the linearity of data in a probability plot, thereby indicating how normal (or lognormal) the data is. We used the probability plot correlation coefficient in cases where it was not visually evident which probability plot was most linear.
These correlations involved only the points actually plotted (i.e. detected concentrations). Since the correlation coefficient is a measure of the linearity of the points on a plot, the correlation coefficient will be high when the plotted points fall along a straight line and low when there are significant bends and curves in the probability plot. The formulas below (Filliben, 1975 and EPA, 1992b) were used in calculating probability plot correlation coefficients.
r = Corr (X, M)
r = (S (Xi-Xavg) * (Mi - Mavg))/(SQRT (S (Xi-Xavg)2 * S (Mi - Mavg)2))
where i ranges from 1 to n.
Mi = f-1 (mi) = ith Normal order statistic median
where f-1 = inverse of standard Normal cumulative distribution,
or f-1 = f(z,0,1)-1 = SQRT (2p)*e(z*z)/2
mi = 1-(0.5)1/n for i = 1
mi = (i-0.3175)/(n+0.365) for 1<i<n
mi = (0.5)1/n for i=n
Mavg = average of Mi
Xavg = average of Xi
X(i) represents the ith smallest ordered concentration value, Mi is the median of the ith order statistic from a standard normal or lognormal distribution, and X and M represent the average values of X(i) and M(i). The plot with the higher correlation coefficient (r) represented the most linear trend and indicated which method was used to adjust the data.
4.15.5 Adjustment of Sample Mean and Standard Deviation
Both Cohen's Adjustment and Aitchinson's Adjustment can be used to adjust the sample mean and sample standard deviation to account for data below the detection limit. To determine if Cohen's or Aitchinson's adjustment was more appropriate for a particular set of data during this statistical evaluation, two separate probability plots were constructed.
4.15.5.1 Censored Probability Plot
A censored probability plot was constructed to test Cohen's underlying assumption that nondetects have been "censored" at their detection limit. To construct the censored probability plots, the combined set of detects and nondetects was ordered, and normal quantiles were computed for the data set as in a regular probability plot. However, only the detected values and their associated normal quantiles were actually plotted. If the shape of the censored data probability plot was more linear than the detects-only probability plot, then Cohen's assumption was considered to be acceptable, and Cohen's adjustment was made to estimate the sample mean and standard deviation.
4.15.5.2 Detects Only Probability Plot
To test the assumptions of the Aitchinson method, a detects-only probability plot was constructed. The assumptions underlying Aitchinson's adjustment are that non-detects represent zero concentrations and that detects and nondetects follow separate probability distributions. Only detected measurements were used to construct the detects-only probability plots. Nondetects were completely ignored. Normal quantiles were computed only for the ordered detected values. The same number of points and concentration values were plotted on both the detects-only and censored probability plots; however, different normal quantiles were associated with each detected concentration. If the detects-only probability plot was more linear than the censored data probability plot, then the underlying assumptions of Aitchinson's method were considered to be reasonable.
To determine the adjusted sample mean and standard deviation, a number of calculations were made. First the mean (Xd) and standard deviation (Sd) of the data above the detection limit were calculated. Then, two parameters, h and g, were calculated using the following equations:
h = (n-m)/n
g = Sd2/(Xd-DL)2
where DL is the average detection limit. Based on the values of h and g, the value of the parameter l was determined (EPA, 1989a).
The adjusted sample mean (Xa) and standard deviation (Sa) were determined using the equations:
Xa = Xd - l(Xd-DL)
Sa = SQRT(Sd2 +l(Xd-DL)2)
The adjusted sample mean and standard deviation were then used to determine the upper tolerance limit.
4.15.5.4 Aitchinson's Adjustment
Aitchinson's adjustment was calculated on either the normal or lognormal data, depending on the distribution. The adjusted mean was calculated using the equation:
Xa = (1-d/n)Xd
where d is the number of detects, n is the number of samples, and Xd is the sample mean of the detected values. The adjusted standard deviation was calculated using the equation
Sa2 = ((n-(d+1)*Sd2)/(n-1) + (d*(n-d)*Xd2)/(n*(n-1))
The adjusted mean and adjusted standard deviation were then used to determine the upper tolerance limit.
A tolerance interval was constructed from the background soil data to establish a basis for determining if there is statistically significant evidence of contamination present in B-20 soil samples. The Tolerance Interval test consists of defining an interval, based on background data, which is expected to contain (with a statistical level of confidence) a given percentage of the population. A tolerance interval establishes a range that is constructed to contain a specified proportion (P%) of the population with a specified confidence coefficient, Y. The proportion of the population included, P, is referred to as the coverage. The probability with which the tolerance interval includes the proportion P% of the population is referred to as the tolerance coefficient.
A coverage of 95% and a tolerance coefficient of 95% were used in the evaluation of all the background soil data. Therefore, there is a confidence level of 95% that the tolerance limit will contain 95% of the distribution of observations from background data.
The UTL was calculated using the following equation:
UTL = X + KS
where X is the mean of the data, K is the one-sided normal tolerance factor (EPA, 1989a), and S is the standard deviation. The tolerance intervals were constructed assuming that the data or the log-transformed data are normally distributed. The appropriate distribution indicated by the Shapiro-Wilk Tests were used for the calculation.