Loading...
HomeMy WebLinkAboutNC0004961_RBSS CAP Part I_Appx B_Final_20151116 Appendix B Background Well Analysis and Statistical Evaluation This page intentionally left blank 1 Background Monitoring Well Determinations Riverbend Steam Station RBSS is a former seven-unit coal-fired electricity generating facility with a capacity of 454 MW. Coal was delivered to the station by rail. The station began commercial operation in 1929 with operation of coal-fired Units 1-4. Coal-fired Units 5-7 began operations in 1952 through 1954. All of the coal-fired units were located in a single power plant. Units 1-3 were retired from service in the 1970s, and Units 4-7 were retired from service on April 1, 2013. From 1929 to 1957, coal ash residue from RBSS’s coal combustion process was deposited in an on-site cinder storage area. Following installation of precipitators and a wet sluicing system in 1957, coal ash residue was disposed of in the station’s ash basin located adjacent to the station and Mountain Island Lake. During operation, RBSS produced approximately 100,000 tons of ash per year. The RBSS site is generally forested along the Catawba River. The buildings and other structures associated with the power production facilities are located on the north side of Horseshoe Bend Beach Road, which extends from west to east and is generally located along a local topographic divide. The topography at the site generally slopes downward from this divide on the south to Mountain Island Lake on the north. In June and September 2015, groundwater elevations were collected from on-site National Pollution Discharge Elimination System (NPDES) compliance wells, voluntary wells, and newly installed Comprehensive Site Assessment (CSA) monitor wells. Groundwater elevations were measured in shallow, deep, and fractured bedrock flow layers (S, D, and BR wells respectively) and groundwater flow direction was estimated by contouring elevations in each flow layer. Groundwater at the site flows to the north, east, and west and discharges to the Catawba River. Groundwater in the southwest portion of the site under the ash storage area flows to the northwest under the cinder storage area to the Catawba River. Generalized groundwater contours for the shallow flow layer are shown in Figure B-1. The CSA Report submitted to NCDEQ in August 2015 identified the Primary and Secondary Cells of the ash basin, ash storage area, and cinder storage areas at the RBSS as potential source areas for groundwater contamination. Wells were chosen to represent background groundwater quality based on factors such as their horizontal distance from the source area or waste boundary, the relative topographic and groundwater elevation difference compared to the nearest ash basin surface or porewater, and the calculated groundwater flow direction. A summary table of information regarding the well and groundwater is provided on the following page. 2 Table B-1. Background Monitoring Well Information Well ID Distance and Direction from Source Area (Primary/Secondary Cells) Groundwater Flow Direction Ground Surface Elevation (feet) Elevation of Screened Interval (feet) Elevation of Groundwater (feet) Closest Pore Water Elevation to Well (feet) MW-7SR 380 feet SE NE 763 696-716 715 713 MW-7D 380 feet SE NE 763 660-665 716 713 MW-7BR 380 feet SE NE 763 561-566 711 713 BG-1S 1,600 feet ESE N 701 656-671 671 713 BG-1D 1,600 feet ESE N 701 502-507 674 713 BG-2S 1,450 feet E N 702.5 637-653 650 713 BG-2D 1,450 feet E NE 702.5 542-547 649 713 BG-2BR 1,450 feet E N 702.5 495-500 695 713 BG-3S 1,350 feet E N 672 635-651 645 713 BG-3D 1,350 feet E N 672 567-572 645 713 Notes: 1. Elevations presented in North American Vertical Datum (NAVD) 88. The methodology applied to determine if a well represents background water quality is explained in detail below for each monitoring well listed in the above table. The locations of the monitoring wells are depicted on Figure 1-4. MW-7D, MW-7SR, and MW-7BR Monitoring wells MW-7D and MW-7SR were installed as the compliance background wells for the NPDES groundwater monitoring program. The well group is located southeast of the ash basin primary cell and west of the ash storage area, located on a topographic high.  The ground surface elevation at this monitoring well group is approximately 763 feet. The ground surface elevation at the southern end of the primary cell of the ash basin is approximately 721 feet (ground surface at the wells is 42 feet higher than the edge of the primary cell of the ash basin).  The monitoring well group is located approximately 380 feet southeast of the southeast corner of the primary cell.  MW-7SR is screened from approximately 47 to 67 feet below ground surface in the shallow flow layer. MW-7SR well screen interval elevation is approximately 696 to 716 feet with a mean water level measurement (from 2010 to 2015) of 715 feet.  MW-7D is screened from approximately 98 to 103 feet below ground surface in the shallow flow layer. MW-7D screen interval elevation is approximately 660 to 665 feet with a mean water level measurement (from 2008 to 2015) of 716 feet.  MW-7BR is screened from approximately 196 feet to 201 feet below ground surface in the bedrock flow layer. MW-7BR well screen interval elevation is approximately 561 to 566 feet with a mean water level measurement from July 2015 at approximately 711.7 feet. 3  The water elevation in the primary cell of the ash basin pond (measured in July 2015) was 713 feet. The approximate maximum pond elevation is 724 feet for the primary cell of the ash basin and 714 feet for the secondary cell of the ash basin.  Based on extrapolation of groundwater flow data from on-site wells and evaluation of topographic data, groundwater in wells MW-7S/SR/BR originates from off-site properties south of the well group. These properties are used for residential and recreational purposes and are not expected to contribute site-specific COIs to groundwater at this location. Based on the horizontal distance from the waste boundary to the monitoring wells, the relative topographic and groundwater elevations measured elevation differences between the monitoring wells and the nearest ash basin surface water, and the determined groundwater flow direction, monitoring wells MW-7SR and MW-7D represent background water quality relative to the RBSS ash basin system. BG-1S and BG-1D Monitoring wells BG-1S and BG-1D were installed as background monitoring wells during the CSA assessment. The well pair is located east of the MW-7 well group in the southeast corner of the site.  The ground surface elevation at these monitoring wells pair is approximately 701 feet. The ground surface elevation at the southern end of the primary cell of the ash basin is approximately 721 feet (ground surface at the wells is approximately 20 feet lower than the primary cell of the ash basin). A creek divides the shallow flow layer between the ash basin and background wells BG-1S and BG-1D.  The monitoring wells are located approximately 1,600 feet east-southeast of the southeast corner of primary cell of the ash basin.  BG-1S is screened from approximately 30 feet to 45 feet below ground surface in the shallow flow layer. BG-1S well screen interval elevation is approximately 656 to 671 feet with a water level measurement from July 2015 at approximately 671 feet.  BG-1D is screened from approximately 194 feet to 199 feet below ground surface in the deep flow layer. BG-1D well screen interval elevation is approximately 502 to 507 feet with a water level measurement from July 2015 at approximately 674 feet.  The water elevation in the primary cell of the ash basin pond (measured in July 2015) was 713 feet. The approximate maximum pond elevation is 724 feet for the primary cell of the ash basin and 714 feet for the secondary cell of the ash basin. Mountain Island Lake is downgradient and has a normal maximum pool elevation of approximately 647 feet.  Based on extrapolation of groundwater flow data from on-site wells and evaluation of topographic data, groundwater in wells BG-1S and BG-1D originates from off-site properties southwest of the well pair. These properties are used for residential and recreational purposes and are not expected to contribute site-specific COIs to groundwater at this location. 4   Based on the horizontal distance from the waste boundary to the monitoring wells, the topographic divide that exists between the monitoring wells and the RBSS ash basins, and the determined groundwater flow direction (southeast toward the Catawba River), monitoring wells BG-1S and BG-1D represent background water quality relative to the RBSS ash basin system. BG-2S, BG-2D, and BG-2BR Monitoring well BG-2S, BG-2D, and BG-2BR were installed as background monitoring wells during the CSA assessment. This well group is located east of the ash basin and north of the BG-1 well group. The position of this well cluster is side gradient and possibly downgradient of the ash basin. Based on extrapolation of inferred groundwater contours, the groundwater in thiswell group may also be downgradient of the ash basin; however, the presence of a tributary between the basin and well cluster likely serves as a discharge for groundwater in the shallow flow layer. At the time of this report, the well cluster does not appear to be impacted by constituents from the ash basin; however, additional sampling and groundwater elevation measurements in 2015 will aid in determining the continued suitability of this well cluster for use as in determining background concentrations of COIs at the site. Data regarding this well cluster are present below.  The ground surface elevation at this monitoring well group is approximately 702.5 feet. The ground surface elevation at the eastern side of the ash basin waste boundary is approximately 720 feet (ground surface at the well group is approximately17.5 feet lower than the edge of the ash basin).  The monitoring well is located approximately 1,450 feet east of the east side of the ash basin waste boundary.  BG-2S is screened from approximately 49 feet to 65 feet below ground surface in the shallow flow layer. BG-2S well screen interval elevation is approximately 637 to 653 feet with a water level measurement from July 2015 at approximately 650.3 feet.  BG-2D is screened from approximately 155 feet to 160 feet below ground surface in the deep flow layer. BG-2D well screen interval elevation is approximately 542 to 547 feet with a water level measurement from July 2015 at approximately 649.5 feet.  BG-2BR is screened from approximately 203 feet to 208 feet below ground surface in the bedrock flow layer. BG-2BR well screen interval elevation is approximately 495 to 500 feet with a water level measurement from July 2015 at approximately 695.5 feet.  Groundwater flow direction determined by groundwater elevations gauged in the onsite monitoring wells during the July 2015 water level gauging event in the shallow flow layer at monitoring well BG-2S was generally from the south. Groundwater flow in the deep flow layer at monitoring well BG-1D was from the south/southwest. Groundwater flow in the bedrock flow layer at monitoring well BG-2BR was from the south. Based on the horizontal distance from the waste boundary to the monitoring wells, the relative topographic and groundwater elevations measured elevation differences between the 5 monitoring wells and the nearest ash basin surface water, and the determined groundwater flow direction, it is undetermined if monitoring wells BG-2S, BG-2D, and BG-2BR represent background water quality relative to the ash disposal areas at the site. however, as noted above, additional sampling and groundwater elevation measurements in 2015 will aid in determining the continued suitability of this well cluster for use as in determining background concentrations of COIs at the site BG-3S and BG-3D Monitoring wells BG-3S and BG-3D were installed as background monitoring wells during the CSA assessment. This well group is located east of the ash basin and north of the BG-2 well group, about 200 feet from the Catawba River. The position of this well cluster is side gradient and possibly downgradient of the ash basin and may not be suitable for use as a background well. Based on extrapolation of inferred groundwater contours, the groundwater within this well group may also be downgradient of the ash basin; however, the presence of a tributary between the basin and well cluster likely serves as a discharge for groundwater in the shallow flow layer. At the time of this report, the well cluster does not appear to be impacted by constituents from the ash basin; however, additional sampling and groundwater elevation measurements in 2015 will aid in determining the continued suitability of this well cluster for use as in determining background concentrations of COIs at the site. Data regarding this well cluster are present below.  The ground surface elevation at this monitoring well pair is approximately 672 feet. The ground surface elevation at the eastern side of the ash basin waste boundary is approximately 698 feet (ground surface at the wells is approximately 26 feet lower than the edge of the ash basin).  The monitoring well pair is located approximately 1,350 feet east of the eastern side of the ash basin waste boundary.  BG-3S is screened from approximately 21 feet to 37 feet below ground surface in the shallow flow layer. BG-3S well screen interval elevation is approximately 635 to 651 feet with a mean water level measurement from July 2015 at approximately 645 feet.  BG-3D is screened from approximately 100 feet to 105 feet below ground surface in the deep flow layer. BG-3D well screen interval elevation is approximately 567 to 572 feet with a mean water level measurement from July 2015 at approximately 644.9 feet.  Groundwater flow direction determined by groundwater elevations gauged in the onsite monitoring wells during the July 2015 water level gauging event in the shallow flow layer at monitoring wells BG-3S and BG-3D was generally to the north, toward Mountain Island Lake. Based on the horizontal distance from the waste boundary to the monitoring wells, the relative topographic and groundwater elevations measured elevation differences between the monitoring wells and the nearest ash basin surface water, and the determined groundwater flow direction, it was undetermined if monitoring wells BG-3S and BG-3D represent background water quality relative to the ash disposal areas at the site. However, as noted above, additional 6 sampling and groundwater elevation measurements in 2015 will aid in determining the continued suitability of this well cluster for use as in determining background concentrations of COIs at the site Turbidity and pH Observations in Background Wells Turbidity and pH were evaluated at each well to determine its suitability for use in monitoring. In some cases, turbidity was elevated above 10 NTU and pH was above 8.5. Analytical data from those wells has not been used to establish PPBCs at this time. Each well will be evaluated, re- developed as necessary, and new determinations made as to their continued suitability in future groundwater sampling events. MW-7SR/D/BR  Turbidity in MW-7SR have been less than 10 NTU for the 15 sampling events sampling events over the period of monitoring, with turbidity measurements less than 5 NTU during 11 of the 15 sampling events. The turbidity results indicate that turbidity should not be an issue with regard to utilizing the analytical results from this monitoring well. pH in MW-7SR have ranged from 5.0 to 5.5 SU over the period of monitoring. The relatively low pH readings indicate that the monitoring well is not being affected by grout/well construction issues.  Turbidity in MW-7D have been less than 10 NTU for the 19 sampling events sampling events over the period of monitoring, with turbidity measurements less than 5 NTU during 17 of the 19 sampling events. Turbidity results indicate that turbidity should not be an issue with regards to utilizing the analytical results from this monitoring well. pH in MW-7D have ranged from 5.5 to 5.9 SU over the period of monitoring. The relatively low pH readings indicate that the monitoring well is not being affected by grout/well construction issues.  The turbidity measurement in MW-7BR was less than 10 NTU during the July 2015 sampling event. The pH in MW-7BR measured 12.76 SU from the July 2015 sampling event. The high pH readings indicate that the monitoring well may be affected by grout/well construction issues. BG-1S/D  The turbidity measurement in BG-1S was greater than 10 NTU during the July 2015 sampling event. Turbidity does appear to be an issue with monitoring well BG-1S. The pH in BG-1S measured 5.75 SU from the July 2015 sampling event. The relatively low pH readings indicate that the monitoring well is not being affected by grout contamination.  The turbidity measurement in BG-1D was greater than 10 NTU during the July 2015 sampling event. The pH in BG-1D measured 12.14 SU from the July 2015 sampling event. The high pH readings indicate that the monitoring well has been affected by grout contamination. 7 BG-2S/D/BR  The turbidity measurement in BG-2S was less than 10 NTU during the July 2015 sampling event. The pH in BG-2S was measured at 6.91 SU from the July 2015. The relatively neutral pH reading indicates that the monitoring well may not be affected by grout contamination.  The turbidity measurement in BG-2D was less than 10 NTU during the July 2015 sampling event. The pH in BG-2D was measured at 11.29 SU from the July 2015. The high pH reading indicates that the monitoring well has been affected by grout contamination.  The turbidity measurement in BG-2BR was less than 10 NTU during the July 2015 sampling event. The relatively neutral pH reading indicates that the monitoring well may not be affected by grout contamination. BG-3S/D  The turbidity measurement in BG-3S was less than 10 NTU during the July 2015 sampling event. The pH in BG-3S measured 6.2 SU in the July 2015 sampling event. The relatively neutral pH readings indicate that the monitoring well is not being affected by grout contamination.  The turbidity measurement in BG-3D was greater than 10 NTU during the July 2015 sampling event. Due to the elevated turbidity data, additional development and resampling will be needed to determine if this well is suitable for continued use. pH measured in BG-3D in the July 2015 sampling event was 7.1 SU. The relatively neutral pH readings indicate that the monitoring well is not being affected by grout contamination. In addition to looking at the pH and turbidity to assess the water quality, the data from the proposed background wells was compared to the regional background concentrations of constituents, where available, and the 2-10 Private Well data. These values are shown in Table 2-2 of the CAP Part 1 report. In general, the concentrations of COIs in the background wells are within the range of both the regional background and the 2-10 data, with the exception of the following:  TDS concentrations in the newly installed background wells are higher than the 2-10 data. No regional data are available.  Vanadium concentrations in the newly installed background wells are higher than the regional background concentrations and 2-10 data. The water quality in the proposed background wells appears to be similar in many respects to the regional published background concentrations for many constituents. Additional data being collected in 2015 (for a total of four sampling rounds in 2015) will help future decision-making and also a better understanding of groundwater flow direction on the eastern side of the RBSS site. 8 Based on horizontal and vertical differences from source areas to well locations, groundwater quality data, and regional background of site-specific COIs, monitoring wells MW-7SR, MW-7D, MW-7BR and BG-1S, BG-1D are considered to confidently reflect background water quality. BG-2S, BG-2D, BG-2BR, BG-3S, and BG-3D were also selected to represent background water quality at RBSS. At the time of this report, the well clusters do not appear to be impacted by constituents from the ash basin and the CSA sampling results were utilized to determine site specific PPBCs; however, additional sampling and groundwater elevation measurements in 2015 will aid in determining the continued suitability of these well clusters for use as in refining background concentrations of COIs at the site. During development of PPBC concentrations, the review of the analytical results and groundwater contour maps generated from the June/July sampling event suggest that certain background wells identified in the CSA may not truly represent a background condition. To obtain a representative sample of background groundwater additional background wells may be required. Soil Background Statistics for Riverbend Steps for determining background threshold values (BTV) for soils: Step 1: Collect an appropriate number of soil samples from the designated background or reference areas. Assume same population. Conduct data validation on analytical data to assess suitability of data for statistical analysis and decision making. Step 2: Determine the data distribution. Depending upon the data distribution, uses parametric or nonparametric methods to estimate BTVs. Step 3: Check for outliers in data set. Remove outliers if it can be justified. Step 4: Calculate BTVs  Upper percentiles  Upper prediction limits (UPLs)  Upper tolerance limits (UTLs)  Upper simultaneous limits (USLs) – New in ProUCL 5.0 Each BTV is described below (USEPA 2013):  Upper Percentile, x0.95: Based upon an established background data set, a 95th percentile represents that statistic such that 95% of the sampled data will be less than or equal to (≤) x0.95 . It is expected that an observation coming from the background population (or comparable to the background population) will be ≤ x0.95 with probability 0.95.  Upper Tolerance Limit (UTL): Based upon an established background data set, a UTL95-95 represents that statistic such that 95% observations (current and future) from the target population (background, comparable to background) will be less than or equal to the UTL95-95 with CC of 0.95. A UTL95-95 represents a 95% UCL of the 95th percentile of the data distribution (population). A UTL95-95 is designed to simultaneously provide coverage for 95% of all potential observations (current and future) from the background population (or comparable to background) with a CC of 0.95. A UTL95-95 can be used when many (unknown) current or future onsite observations need to be compared with a BTV. A parametric UTL95-95 takes the data variability into account.  Upper Prediction Limit (UPL): Based upon an established background data set, a 95% UPL (UPL95) represents that statistic such that an independently collected new/future observation from the target population (e.g., background, comparable to background) will be less than or equal to the UPL95 with confidence coefficient (CC) of 0.95. We are 95% sure that a single future value from the background population will be less than the UPL95 with CC= 0.95. A parametric UPL takes data variability into account.  Upper Simultaneous Limit (USL): Based upon an established background data set free of outliers and representing a single statistical population, a USL95 represents that statistic such that all observations from the “established” background data set are less than or equal to the USL95 with a CC of 0.95. A parametric USL takes the data variability into account. It is expected that all current or future observations coming from the background population (comparable to background population, unimpacted site locations) will be less than or equal to the USL95 with CC, 0.95. The use of a USL as a BTV estimate is suggested by the USEPA when a large number of onsite observations (current or future) need to be compared with a BTV. Approach:  Attachment A presents the Riverbend soil dataset. HDR completed a quality assurance and quality control (QA/QC) data validation assessment (presented in separate document) and has determined that the data meets project data quality objectives and is suitable for statistical analysis and for establishing BTVs. Table 1 list the constituents (17 samples and 30 constituents per sample)  Next, HDR conducted Dixon’s outlier test for each constituent using ProUCL Version 5.0 software (USEPA 2013). Statically significant outliers were identified for 3 out of 30 constituents at the 5% significant level. Outliers can inflate background concentration estimates (over estimate), where USEPA (2013) defines an outlier as Measurements (usually larger or smaller than the majority of the data values in a sample) that are not representative of the population from which they were drawn. The presence of outliers distorts most statistics if used in any calculations. However, an outlier should only be removed if there is justification for doing so (e.g., sample collected an area not representative of background conditions). Riverbend samples represent subsurface soils collected from drilling operations. HDR has determined that samples meet data quality objectives. As stated by the USEPA (2013), since the treatment and handling of outliers is a controversial and subjective topic, it is suggested that the outliers be treated on a site-specific basis using all existing knowledge about the site; and regional and site- specific background areas. Because soil samples were collected at depths greater than 5 feet below ground surface, and there is no evidence of anthropogenic impacts at depths greater than 5 feet below ground surface, the entire dataset was utilized for establishing BTV (assumes no outliers).  HDR used ProUCL Version 5.0, to calculate summary statistics, goodness of fit (population distribution), and the BTV upper limits (UTL, UPL, and USL). Results are summarized in Tables 1 and 2. The Kaplan Meier (KM) method was used for estimating statistics with censored data (data with non-detections). ProUCL printouts are presented in Appendix B. Table 1. Summary Statistics for Subsurface Soils Riverbend Variable n1 Detect Non- Detects KM Mean Detect Mean Detect Median KM SD Detect SD Detect Min Detect Max (mg/Kg) Aluminum 17 17 0 13,800 13,800 12,100 5,860 5,860 5,650 26,700 Antimony 17 0 17 NS1 NS NS NS NS     NS       NS    Arsenic 17 0 17     NS        NS        NS        NS        NS        NS        NS    Barium 17 17 0 164 164 161 79.5 79.5 20.4 282 Beryllium 17 14 3 0.55 0.61 0.63 0.33 0.35 0.23 1.4 Boron 17 2 15 12.5 15.8 15.8 1.96 5.37 12.0 19.6 Cadmium 17 0 17     NS        NS        NS        NS        NS        NS        NS    Calcium 17 11 6 741 1,090 1,150 716 697 96.0 2,480 Chloride 17 0 17     NS        NS        NS        NS        NS        NS        NS    Chromium 17 17 0 6.78 6.78 5.50 4.46 4.46 3.10 20.6 Cobalt 17 15 2 11.0 11.8 12.1 4.47 4.28 5.00 22.8 Copper 17 17 0 20.8 20.8 15.1 24.2 24.2 4.30 93.3 Iron 17 17 0 20,000 20,000 19,400 5,550 5,550 12,300 32,200 Lead 17 12 5 5.49 6.02 4.05 3.41 4.07 3.30 15.2 Magnesium 17 16 1 5,800 6,160 6,420 2,700 2,450 620 10,200 Manganese 17 17 0 691 691 689 347 347 29.1 1,440 Mercury 17 2 15 0.012 0.029 0.029 0.009 0.027 0.010 0.048 Molybdenum 17 0 17     NS        NS        NS        NS        NS        NS        NS    Nickel 17 17 0 5.06 5.06 4.90 2.07 2.07 1.30 8.50 Nitrate 17 0 17     NS        NS        NS        NS        NS        NS        NS    pH (field) 17 17 0 5.42 5.42 5.50 0.57 0.57 4.50 6.50 Potassium 17 16 1 4,870 5,160 5,810 2,330 2,150 216 8,700 Selenium 16 1 15 4.7 4.7 4.7 0      NS   4.7 4.7 Sodium 17 0 17     NS        NS        NS        NS        NS        NS        NS    Strontium 17 13 4 9.25 11.3 10.4 7.22 7.34 2.10 22.2 Sulfate 17 0 17     NS        NS        NS        NS        NS        NS        NS    Thallium 17 0 17     NS        NS        NS        NS        NS        NS        NS    TOC 17 1 16 448 448 448 0      NS   448 448 Vanadium 17 17 0 52.1 52.1 50.6 16.3 16.3 29.2 89.2 Zinc 17 16 1 35.0 36.8 36.7 14.7 13.7 6.60 59.8 1n = number of samples, KM = Kaplan Meier method (addresses data with non-detections, see USEPA 2013); NS = No statistical analysis run due to too few detections; SD = standard deviation; TOC = total organic carbon Table 2. Subsurface Soil Background Concentration Estimates Riverbend Constituent Distribution 95% Percentile 95% UTL 95% UPL 95% USL (mg/Kg) Aluminum Normal 23,500 28,400 24,400 28,400 Antimony NS1 NS NS NS NS  Arsenic Normal NS NS NS NS  Barium Normal 294 361 306 360 Beryllium Normal 1.10 1.38 1.15 1.38 Boron NS NS NS NS NS  Cadmium NS NS NS NS NS  Calcium Normal 1,960 2,580 2,070 2,570 Chloride NS NS NS NS NS  Chromium Log Normal 13.5 20.6 14.5 20.5 Cobalt Normal 18.4 22.1 19.1 22.1 Copper Log Normal 56.6 116 64.4 115 Iron Normal 29,100 22,800 30,000 33,700 Lead Log Normal 10.1 14.6 10.8 14.5 Magnesium Normal 10,200 12,500 10,700 12,500 Manganese Normal 1,260 1,550 1,310 1,550 Mercury NS NS NS NS NS  Molybdenum Normal NS NS NS NS  Nickel Normal 8.46 10.2 8.77 10.2 Nitrate NS NS NS NS NS  pH (field) NS NS NS NS NS  Potassium Normal 8,710 10,700 9,060 10,600 Selenium Normal NS NS NS NS  Sodium NS NS NS NS NS  Strontium Normal 21.1 27.2 22.2 27.1 Sulfate NS NS NS NS NS  Thallium NS NS NS NS NS  TOC NS NS NS NS NS  Vanadium Normal 78.9 92.7 81.4 92.5 Zinc Normal 59.1 71.4 61.3 71.3 1 NS = No statistical analysis run due to too few detections; UTL = upper tolerance limit; UPL = upper tolerance limit; USL = upper simultaneous limit References U.S. EPA. ProUCL Version 5.0.00 Technical Guide, EPA/600/R-07/041 Attachment A Riverbend Dataset Location ID Sample ID Sample Depth Aluminum D_Alumi num Antimony D_Antimo ny Arsenic D_Arsen ic Barium D_Bariu m Beryllium D_Berylliu m BG-1D BG-1D(14-15)14-15 11600 1 8.3 0 8.3 0 77.3 1 0.41 0 BG-1D BG-1D(24-25)24-25 14100 1 8.2 0 8.2 0 175.0 1 0.25 1 BG-1D BG-1D(30-31)30-31 13100 1 6.9 0 6.9 0 144.0 1 0.28 1 BG-1D BG-1D(34-35)34-35 10900 1 8.3 0 8.3 0 189.0 1 0.29 1 BG-1D BG-1D(44-45)44-45 9440 1 7.3 0 7.3 0 209.0 1 0.46 1 BG-1D BG-1D(5-6)5-6 12100 1 6.5 0 6.5 0 20.4 1 0.32 0 BG-2D BG-2D(48-49)48-49 10700 1 6.3 0 6.3 0 161.0 1 0.63 1 BG-2D BG-2D(53-54)53-54 11100 1 7.6 0 7.6 0 132.0 1 0.63 1 BG-2D BG-2D(60-62)60-62 15400 1 7.8 0 7.8 0 226.0 1 1.40 1 BG-2D BG-2D(74-75)74-75 11000 1 6.3 0 6.3 0 251.0 1 1.00 1 BG-3D BG-3D(18.5-20)18.5-20 26000 1 6.1 0 6.1 0 240.0 1 0.66 1 BG-3D BG-3D(23-24)23-24 26700 1 6.2 0 6.2 0 276.0 1 0.84 1 BG-3D BG-3D(27-28)27-28 21600 1 6.2 0 6.2 0 282.0 1 0.98 1 BG-3D BG-3D(34-35.5)34-35.5 15900 1 7.1 0 7.1 0 159.0 1 0.63 1 BG-3D BG-3D(73-75)73-75 7630 1 5.7 0 5.7 0 96.3 1 0.26 1 MW-7BR MW-7BR(43.5-45)43.5-45 5650 1 6.4 0 6.4 0 31.3 1 0.32 0 MW-7BR MW-7BR(53.5-55)53.5-55 12300 1 7.3 0 7.3 0 110.0 1 0.23 1 Location ID Sample ID BG-1D BG-1D(14-15) BG-1D BG-1D(24-25) BG-1D BG-1D(30-31) BG-1D BG-1D(34-35) BG-1D BG-1D(44-45) BG-1D BG-1D(5-6) BG-2D BG-2D(48-49) BG-2D BG-2D(53-54) BG-2D BG-2D(60-62) BG-2D BG-2D(74-75) BG-3D BG-3D(18.5-20) BG-3D BG-3D(23-24) BG-3D BG-3D(27-28) BG-3D BG-3D(34-35.5) BG-3D BG-3D(73-75) MW-7BR MW-7BR(43.5-45) MW-7BR MW-7BR(53.5-55) Boron D_Boron Cadmium D_Cadmium Calcium D_Calcium Chloride D_Chl oride Chromium D_Chromi um Cobalt D_Cobalt 20.7 0 0.99 0 207 0 427 0 5.5 1 5.0 1 20.5 0 0.98 0 205 0 394 0 5.4 1 22.8 1 17.2 0 0.83 0 96 1 330 0 3.8 1 12.1 1 20.8 0 1.00 0 208 0 402 0 3.1 1 13.9 1 18.3 0 0.88 0 244 1 353 0 4.6 1 9.9 1 16.2 0 0.78 0 162 0 321 0 9.0 1 6.5 0 15.8 0 0.76 0 524 1 314 0 6.0 1 14.6 1 19.0 0 0.91 0 580 1 365 0 4.5 1 8.5 1 19.6 0 0.94 0 1450 1 379 0 20.6 1 13.8 1 15.8 0 0.76 0 1020 1 296 0 5.5 1 9.2 1 15.4 0 0.74 0 1150 1 309 0 14.4 1 14.6 1 15.6 0 0.75 0 1420 1 318 0 6.1 1 11.0 1 15.5 0 0.75 0 1540 1 307 0 6.7 1 12.1 1 17.7 0 0.85 0 1520 1 346 0 5.7 1 8.0 1 14.3 0 0.69 0 2480 1 284 0 3.1 1 5.7 0 12.0 1 0.77 0 160 0 323 0 3.6 1 7.3 1 19.6 1 0.88 0 183 0 369 0 7.6 1 14.7 1 Location ID Sample ID BG-1D BG-1D(14-15) BG-1D BG-1D(24-25) BG-1D BG-1D(30-31) BG-1D BG-1D(34-35) BG-1D BG-1D(44-45) BG-1D BG-1D(5-6) BG-2D BG-2D(48-49) BG-2D BG-2D(53-54) BG-2D BG-2D(60-62) BG-2D BG-2D(74-75) BG-3D BG-3D(18.5-20) BG-3D BG-3D(23-24) BG-3D BG-3D(27-28) BG-3D BG-3D(34-35.5) BG-3D BG-3D(73-75) MW-7BR MW-7BR(43.5-45) MW-7BR MW-7BR(53.5-55) Copper D_Coppe r Iron D_Iron Lead D_Lead Magnesium D_Magnesi um Manganese D_Manganes e Mercur y D_Mercu ry 15.1 1 26800 1 4.9 1 2750 1 226.0 1 0.0095 1 25.2 1 28200 1 6.0 1 7340 1 1140.0 1 0.0130 0 16.1 1 20000 1 6.9 0 5980 1 518.0 1 0.0110 0 16.0 1 19300 1 6.3 1 5710 1 921.0 1 0.0130 0 10.3 1 13500 1 7.3 0 4350 1 817.0 1 0.0120 0 11.6 1 32200 1 13.7 1 162 0 29.1 1 0.0480 1 5.3 1 17100 1 15.2 1 6300 1 290.0 1 0.0100 0 4.3 1 16000 1 3.9 1 6530 1 689.0 1 0.0120 0 20.3 1 20600 1 7.8 0 10200 1 1440.0 1 0.0130 0 5.5 1 14600 1 3.5 1 6910 1 975.0 1 0.0100 0 16.4 1 23800 1 4.2 1 8030 1 943.0 1 0.0100 0 10.4 1 18100 1 3.7 1 7660 1 626.0 1 0.0100 0 93.3 1 23500 1 3.3 1 9700 1 724.0 1 0.0098 0 12.6 1 19400 1 7.1 0 7530 1 542.0 1 0.0120 0 71.8 1 13500 1 5.7 0 4460 1 805.0 1 0.0093 0 4.5 1 12300 1 3.7 1 620 1 446.0 1 0.0100 0 15.4 1 21000 1 3.8 1 4440 1 615.0 1 0.0120 0 Location ID Sample ID BG-1D BG-1D(14-15) BG-1D BG-1D(24-25) BG-1D BG-1D(30-31) BG-1D BG-1D(34-35) BG-1D BG-1D(44-45) BG-1D BG-1D(5-6) BG-2D BG-2D(48-49) BG-2D BG-2D(53-54) BG-2D BG-2D(60-62) BG-2D BG-2D(74-75) BG-3D BG-3D(18.5-20) BG-3D BG-3D(23-24) BG-3D BG-3D(27-28) BG-3D BG-3D(34-35.5) BG-3D BG-3D(73-75) MW-7BR MW-7BR(43.5-45) MW-7BR MW-7BR(53.5-55) Molybdenum D_Molybden um Nickel D_Nickel Nitrate D_Nitrate Percent Moisture D_Percent Moisture pH (field) D_pH (field) 4.1 0 3.4 1 42.7 0 41.3 1 5.6 1 4.1 0 6.4 1 39.4 0 37.6 1 5.0 1 3.4 0 5.3 1 33.0 0 27.0 1 5.4 1 4.2 0 5.8 1 40.2 0 38.5 1 5.1 1 3.7 0 4.7 1 35.3 0 29.8 1 5.7 1 3.2 0 1.3 1 32.1 0 21.4 1 4.6 1 3.2 0 4.5 1 31.4 0 20.4 1 4.5 1 3.8 0 3.9 1 36.5 0 32.5 1 4.7 1 3.9 0 8.5 1 37.9 0 36.1 1 4.8 1 3.2 0 4.5 1 29.6 0 16.8 1 5.5 1 3.1 0 8.3 1 30.9 0 20.1 1 6.0 1 3.1 0 7.6 1 31.8 0 20.9 1 5.8 1 3.1 0 6.9 1 30.7 0 21.0 1 5.9 1 3.5 0 5.3 1 34.6 0 28.1 1 6.5 1 2.9 0 2.3 1 28.4 0 12.4 1 6.1 1 3.2 0 2.4 1 32.3 0 23.2 1 5.4 1 3.7 0 4.9 1 36.9 0 32.3 1 5.5 1 Location ID Sample ID BG-1D BG-1D(14-15) BG-1D BG-1D(24-25) BG-1D BG-1D(30-31) BG-1D BG-1D(34-35) BG-1D BG-1D(44-45) BG-1D BG-1D(5-6) BG-2D BG-2D(48-49) BG-2D BG-2D(53-54) BG-2D BG-2D(60-62) BG-2D BG-2D(74-75) BG-3D BG-3D(18.5-20) BG-3D BG-3D(23-24) BG-3D BG-3D(27-28) BG-3D BG-3D(34-35.5) BG-3D BG-3D(73-75) MW-7BR MW-7BR(43.5-45) MW-7BR MW-7BR(53.5-55) Potassium D_Potassiu m Selenium D_Seleniu m Sodium D_Sodiu m Strontium 2580 1 8.3 0 414 0 4.1 6500 1 8.2 0 410 0 4.1 5550 1 4.7 1 345 0 2.3 4740 1 8.3 0 417 0 3.1 3410 1 7.3 0 367 0 4.6 325 0 6.5 0 325 0 3.2 5380 1 6.3 0 316 0 7.2 6120 1 7.6 0 380 0 8.0 7020 1 7.8 0 392 0 16.7 6060 1 6.3 0 316 0 14.9 6450 1 6.1 0 307 0 10.4 6410 1 6.2 0 311 0 16.2 8700 1 6.2 0 311 0 22.2 7120 1 7.1 0 354 0 17.3 2760 1 5.7 0 286 0 22.2 216 1 321 0 3.2 3520 1 7.3 0 366 0 2.1 Location ID Sample ID BG-1D BG-1D(14-15) BG-1D BG-1D(24-25) BG-1D BG-1D(30-31) BG-1D BG-1D(34-35) BG-1D BG-1D(44-45) BG-1D BG-1D(5-6) BG-2D BG-2D(48-49) BG-2D BG-2D(53-54) BG-2D BG-2D(60-62) BG-2D BG-2D(74-75) BG-3D BG-3D(18.5-20) BG-3D BG-3D(23-24) BG-3D BG-3D(27-28) BG-3D BG-3D(34-35.5) BG-3D BG-3D(73-75) MW-7BR MW-7BR(43.5-45) MW-7BR MW-7BR(53.5-55) D_Strontium Sulfate D_Sulfate Thallium D_Thallium Total Organic Carbon D_Total Organic Carbon Vanadium D_Vanadiu m Zinc D_Zinc 0 427 0 8.3 0 1020 0 70.2 1 12.5 1 0 394 0 8.2 0 970 0 75.8 1 34.7 1 1 330 0 6.9 0 818 0 59.7 1 30.0 1 1 402 0 8.3 0 969 0 53.3 1 31.1 1 1 353 0 7.3 0 857 0 38.0 1 33.6 1 0 321 0 6.5 0 448 1 89.2 1 6.5 0 1 314 0 6.3 0 749 0 43.9 1 45.7 1 1 365 0 7.6 0 894 0 40.8 1 38.7 1 1 379 0 7.8 0 948 0 58.8 1 55.5 1 1 296 0 6.3 0 726 0 37.9 1 42.9 1 1 309 0 6.1 0 752 0 62.3 1 43.0 1 1 318 0 6.2 0 765 0 45.4 1 45.8 1 1 307 0 6.2 0 758 0 50.6 1 59.8 1 1 346 0 7.1 0 835 0 41.6 1 43.2 1 1 284 0 5.7 0 683 0 30.5 1 33.3 1 0 323 0 6.4 0 780 0 29.2 1 6.6 1 1 369 0 7.3 0 884 0 58.5 1 31.7 1 Attachment B ProUCL 5.0.00 Printout  Background Statistics for Data Sets with Non-Detects User Selected Options Date/Time of Computation 10/29/2015 16:31 From File Riverbend_ProUCL_data_Riverbend - REVISED_20151027121828.xls Full Precision OFF Confidence Coefficient 95% Coverage 95% Different or Future K Observation 1 Number of Bootstrap Operations 2000 Aluminum General Statistics Total Number of Observations 17 Number of Distinct Observations 17 Minimum 5650 First Quartile 10900 Second Largest 26000 Median 12100 Maximum 26700 Third Quartile 15400 Mean 13836 SD 5864 Coefficient of Variation 0.424 Skewness 1.2 Mean of logged Data 9.458 SD of logged Data 0.399 Critical Values for Background Threshold Values (BTVs) Tolerance Factor K (For UTL)2.486 d2max (for USL)2.475 Normal GOF Test Shapiro Wilk Test Statistic 0.863 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.892 Data Not Normal at 5% Significance Level Lilliefors Test Statistic 0.197 Lilliefors GOF Test 5% Lilliefors Critical Value 0.215 Data appear Normal at 5% Significance Level Data appear Approximate Normal at 5% Significance Level Background Statistics Assuming Normal Distribution 95% UTL with 95% Coverage 28415 90% Percentile (z)21352 95% UPL (t)24372 95% Percentile (z)23482 95% USL 28350 99% Percentile (z)27479 Gamma GOF Test A-D Test Statistic 0.592 Anderson-Darling Gamma GOF Test 5% A-D Critical Value 0.741 Detected data appear Gamma Distributed at 5% Significance Level K-S Test Statistic 0.155 Kolmogrov-Smirnoff Gamma GOF Test 5% K-S Critical Value 0.209 Detected data appear Gamma Distributed at 5% Significance Level Detected data appear Gamma Distributed at 5% Significance Level Gamma Statistics k hat (MLE)6.686 k star (bias corrected MLE)5.545 Theta hat (MLE)2069 Theta star (bias corrected MLE)2495 nu hat (MLE)227.3 nu star (bias corrected)188.5 MLE Mean (bias corrected)13836 MLE Sd (bias corrected)5876 Background Statistics Assuming Gamma Distribution 95% Wilson Hilferty (WH) Approx. Gamma UPL 25229 90% Percentile 21697 95% Hawkins Wixley (HW) Approx. Gamma UPL 25432 95% Percentile 24700 95% WH Approx. Gamma UTL with 95% Coverage 31340 99% Percentile 31016 95% HW Approx. Gamma UTL with 95% Coverage 31975 95% WH USL 31234 95% HW USL 31859 Lognormal GOF Test Shapiro Wilk Test Statistic 0.949 Shapiro Wilk Lognormal GOF Test 5% Shapiro Wilk Critical Value 0.892 Data appear Lognormal at 5% Significance Level Lilliefors Test Statistic 0.149 Lilliefors Lognormal GOF Test 5% Lilliefors Critical Value 0.215 Data appear Lognormal at 5% Significance Level Data appear Lognormal at 5% Significance Level Background Statistics assuming Lognormal Distribution 95% UTL with 95% Coverage 34591 90% Percentile (z)21381 95% UPL (t)26264 95% Percentile (z)24720 95% USL 34437 99% Percentile (z)32454 Nonparametric Distribution Free Background Statistics Data appear Approximate Normal at 5% Significance Level Nonparametric Upper Limits for Background Threshold Values Order of Statistic, r 17 95% UTL with 95% Coverage 26700 Approximate f 0.895 Confidence Coefficient (CC) achieved by UTL 0.582 95% Percentile Bootstrap UTL with 95% Coverage 26700 95% BCA Bootstrap UTL with 95% Coverage 26700 95% UPL 26700 90% Percentile 23360 90% Chebyshev UPL 31940 95% Percentile 26140 95% Chebyshev UPL 40140 99% Percentile 26588 95% USL 26700 Note: The use of USL to estimate a BTV is recommended only when the data set represents a background data set free of outliers and consists of observations collected from clean unimpacted locations. The use of USL tends to provide a balance between false positives and false negatives provided the data represents a background data set and when many onsite observations need to be compared with the BTV. Antimony General Statistics Total Number of Observations 17 Number of Missing Observations 0 Number of Distinct Observations 13 Number of Detects 0 Number of Non-Detects 17 Number of Distinct Detects 0 Number of Distinct Non-Detects 13 Minimum Detect N/A Minimum Non-Detect 5.7 Maximum Detect N/A Maximum Non-Detect 8.3 Variance Detected N/A Percent Non-Detects 100% Mean Detected N/A SD Detected N/A Mean of Detected Logged Data N/A SD of Detected Logged Data N/A Warning: All observations are Non-Detects (NDs), therefore all statistics and estimates should also be NDs! Specifically, sample mean, UCLs, UPLs, and other statistics are also NDs lying below the largest detection limit! The Project Team may decide to use alternative site specific values to estimate environmental parameters (e.g., EPC, BTV). The data set for variable Antimony was not processed! Arsenic General Statistics Total Number of Observations 17 Number of Missing Observations 0 Number of Distinct Observations 13 Number of Detects 0 Number of Non-Detects 17 Number of Distinct Detects 0 Number of Distinct Non-Detects 13 Minimum Detect N/A Minimum Non-Detect 5.7 Maximum Detect N/A Maximum Non-Detect 8.3 Variance Detected N/A Percent Non-Detects 100% Mean Detected N/A SD Detected N/A Mean of Detected Logged Data N/A SD of Detected Logged Data N/A Warning: All observations are Non-Detects (NDs), therefore all statistics and estimates should also be NDs! Specifically, sample mean, UCLs, UPLs, and other statistics are also NDs lying below the largest detection limit! The Project Team may decide to use alternative site specific values to estimate environmental parameters (e.g., EPC, BTV). The data set for variable Arsenic was not processed! Barium General Statistics Total Number of Observations 17 Number of Distinct Observations 17 Minimum 20.4 First Quartile 110 Second Largest 276 Median 161 Maximum 282 Third Quartile 226 Mean 163.5 SD 79.5 Coefficient of Variation 0.486 Skewness -0.255 Mean of logged Data 4.913 SD of logged Data 0.735 Critical Values for Background Threshold Values (BTVs) Tolerance Factor K (For UTL)2.486 d2max (for USL)2.475 Normal GOF Test Shapiro Wilk Test Statistic 0.967 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.892 Data appear Normal at 5% Significance Level Lilliefors Test Statistic 0.0783 Lilliefors GOF Test 5% Lilliefors Critical Value 0.215 Data appear Normal at 5% Significance Level Data appear Normal at 5% Significance Level Background Statistics Assuming Normal Distribution 95% UTL with 95% Coverage 361.1 90% Percentile (z)265.4 95% UPL (t)306.3 95% Percentile (z)294.3 95% USL 360.2 99% Percentile (z)348.4 Gamma GOF Test A-D Test Statistic 0.603 Anderson-Darling Gamma GOF Test 5% A-D Critical Value 0.746 Detected data appear Gamma Distributed at 5% Significance Level K-S Test Statistic 0.148 Kolmogrov-Smirnoff Gamma GOF Test 5% K-S Critical Value 0.211 Detected data appear Gamma Distributed at 5% Significance Level Detected data appear Gamma Distributed at 5% Significance Level Gamma Statistics k hat (MLE)2.872 k star (bias corrected MLE)2.405 Theta hat (MLE)56.92 Theta star (bias corrected MLE)67.99 nu hat (MLE)97.66 nu star (bias corrected)81.76 MLE Mean (bias corrected)163.5 MLE Sd (bias corrected)105.4 Background Statistics Assuming Gamma Distribution 95% Wilson Hilferty (WH) Approx. Gamma UPL 380.9 90% Percentile 304.7 95% Hawkins Wixley (HW) Approx. Gamma UPL 401.6 95% Percentile 366.3 95% WH Approx. Gamma UTL with 95% Coverage 513.4 99% Percentile 501.4 95% HW Approx. Gamma UTL with 95% Coverage 560.1 95% WH USL 511 95% HW USL 557.2 Lognormal GOF Test Shapiro Wilk Test Statistic 0.832 Shapiro Wilk Lognormal GOF Test 5% Shapiro Wilk Critical Value 0.892 Data Not Lognormal at 5% Significance Level Lilliefors Test Statistic 0.19 Lilliefors Lognormal GOF Test 5% Lilliefors Critical Value 0.215 Data appear Lognormal at 5% Significance Level Data appear Approximate Lognormal at 5% Significance Level Background Statistics assuming Lognormal Distribution 95% UTL with 95% Coverage 845.4 90% Percentile (z)348.8 95% UPL (t)509.3 95% Percentile (z)455.6 95% USL 838.5 99% Percentile (z)751.8 Nonparametric Distribution Free Background Statistics Data appear Normal at 5% Significance Level Nonparametric Upper Limits for Background Threshold Values Order of Statistic, r 17 95% UTL with 95% Coverage 282 Approximate f 0.895 Confidence Coefficient (CC) achieved by UTL 0.582 95% Percentile Bootstrap UTL with 95% Coverage 282 95% BCA Bootstrap UTL with 95% Coverage 282 95% UPL 282 90% Percentile 261 90% Chebyshev UPL 408.9 95% Percentile 277.2 95% Chebyshev UPL 520.1 99% Percentile 281 95% USL 282 Note: The use of USL to estimate a BTV is recommended only when the data set represents a background data set free of outliers and consists of observations collected from clean unimpacted locations. The use of USL tends to provide a balance between false positives and false negatives provided the data represents a background data set and when many onsite observations need to be compared with the BTV. Beryllium General Statistics Total Number of Observations 17 Number of Missing Observations 0 Number of Distinct Observations 14 Number of Detects 14 Number of Non-Detects 3 Number of Distinct Detects 12 Number of Distinct Non-Detects 2 Minimum Detect 0.23 Minimum Non-Detect 0.32 Maximum Detect 1.4 Maximum Non-Detect 0.41 Variance Detected 0.123 Percent Non-Detects 17.65% Mean Detected 0.61 SD Detected 0.351 Mean of Detected Logged Data -0.654 SD of Detected Logged Data 0.597 Critical Values for Background Threshold Values (BTVs) Tolerance Factor K (For UTL)2.486 d2max (for USL)2.475 Normal GOF Test on Detects Only Shapiro Wilk Test Statistic 0.899 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.874 Detected Data appear Normal at 5% Significance Level Lilliefors Test Statistic 0.176 Lilliefors GOF Test 5% Lilliefors Critical Value 0.237 Detected Data appear Normal at 5% Significance Level Detected Data appear Normal at 5% Significance Level Kaplan Meier (KM) Background Statistics Assuming Normal Distribution Mean 0.549 SD 0.334 95% UTL95% Coverage 1.38 95% KM UPL (t)1.149 90% KM Percentile (z)0.977 95% KM Percentile (z)1.099 99% KM Percentile (z)1.326 95% KM USL 1.376 DL/2 Substitution Background Statistics Assuming Normal Distribution Mean 0.533 SD 0.36 95% UTL95% Coverage 1.427 95% UPL (t)1.179 90% Percentile (z)0.994 95% Percentile (z)1.125 99% Percentile (z)1.37 95% USL 1.423 DL/2 is not a recommended method. DL/2 provided for comparisons and historical reasons Gamma GOF Tests on Detected Observations Only A-D Test Statistic 0.502 Anderson-Darling GOF Test 5% A-D Critical Value 0.742 Detected data appear Gamma Distributed at 5% Significance Level K-S Test Statistic 0.2 Kolmogrov-Smirnoff GOF 5% K-S Critical Value 0.23 Detected data appear Gamma Distributed at 5% Significance Level Detected data appear Gamma Distributed at 5% Significance Level Gamma Statistics on Detected Data Only k hat (MLE)3.296 k star (bias corrected MLE)2.638 Theta hat (MLE)0.185 Theta star (bias corrected MLE)0.231 nu hat (MLE)92.3 nu star (bias corrected)73.85 MLE Mean (bias corrected)0.61 MLE Sd (bias corrected)0.376 95% Percentile of Chisquare (2k)11.49 Gamma ROS Statistics using Imputed Non-Detects GROS may not be used when data set has > 50% NDs with many tied observations at multiple DLs GROS may not be used when kstar of detected data is small such as < 0.1 For such situations, GROS method tends to yield inflated values of UCLs and BTVs For gamma distributed detected data, BTVs and UCLs may be computed using gamma distribution on KM estimates Minimum 0.191 Mean 0.549 Maximum 1.4 Median 0.46 SD 0.345 CV 0.63 k hat (MLE)2.955 k star (bias corrected MLE)2.473 Theta hat (MLE)0.186 Theta star (bias corrected MLE)0.222 nu hat (MLE)100.5 nu star (bias corrected)84.07 MLE Mean (bias corrected)0.549 MLE Sd (bias corrected)0.349 95% Percentile of Chisquare (2k)10.99 90% Percentile 1.016 95% Percentile 1.219 99% Percentile 1.663 The following statistics are computed using Gamma ROS Statistics on Imputed Data Upper Limits using Wilson Hilferty (WH) and Hawkins Wixley (HW) Methods WH HW WH HW 95% Approx. Gamma UTL with 95% Coverage 1.703 1.776 95% Approx. Gamma UPL 1.264 1.287 95% Gamma USL 1.696 1.767 The following statistics are computed using gamma distribution and KM estimates Upper Limits using Wilson Hilferty (WH) and Hawkins Wixley (HW) Methods k hat (KM)2.692 nu hat (KM)91.52 WH HW WH HW 95% Approx. Gamma UTL with 95% Coverage 1.637 1.699 95% Approx. Gamma UPL 1.226 1.244 95% Gamma USL 1.63 1.691 Lognormal GOF Test on Detected Observations Only Shapiro Wilk Test Statistic 0.915 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.874 Detected Data appear Lognormal at 5% Significance Level Lilliefors Test Statistic 0.197 Lilliefors GOF Test 5% Lilliefors Critical Value 0.237 Detected Data appear Lognormal at 5% Significance Level Detected Data appear Lognormal at 5% Significance Level Background Lognormal ROS Statistics Assuming Lognormal Distribution Using Imputed Non-Detects Mean in Original Scale 0.553 Mean in Log Scale -0.76 SD in Original Scale 0.341 SD in Log Scale 0.591 95% UTL95% Coverage 2.033 95% BCA UTL95% Coverage 1.4 95% Bootstrap (%) UTL95% Coverage 1.4 95% UPL (t)1.352 90% Percentile (z)0.997 95% Percentile (z)1.236 99% Percentile (z)1.85 95% USL 2.019 Statistics using KM estimates on Logged Data and Assuming Lognormal Distribution KM Mean of Logged Data -0.775 95% KM UTL (Lognormal)95% Coverage 1.973 KM SD of Logged Data 0.585 95% KM UPL (Lognormal)1.318 95% KM Percentile Lognormal (z)1.206 95% KM USL (Lognormal)1.96 Background DL/2 Statistics Assuming Lognormal Distribution Mean in Original Scale 0.533 Mean in Log Scale -0.847 SD in Original Scale 0.36 SD in Log Scale 0.691 95% UTL95% Coverage 2.389 95% UPL (t)1.483 90% Percentile (z)1.039 95% Percentile (z)1.336 99% Percentile (z)2.139 95% USL 2.37 DL/2 is not a Recommended Method. DL/2 provided for comparisons and historical reasons. Nonparametric Distribution Free Background Statistics Data appear to follow a Discernible Distribution at 5% Significance Level Nonparametric Upper Limits for BTVs(no distinction made between detects and nondetects) Order of Statistic, r 17 95% UTL with95% Coverage 1.4 Approximate f 0.895 Confidence Coefficient (CC) achieved by UTL 0.582 95% UPL 1.4 95% USL 1.4 95% KM Chebyshev UPL 2.048 Note: The use of USL to estimate a BTV is recommended only when the data set represents a background data set free of outliers and consists of observations collected from clean unimpacted locations. The use of USL tends to provide a balance between false positives and false negatives provided the data represents a background data set and when many onsite observations need to be compared with the BTV. Boron General Statistics Total Number of Observations 17 Number of Missing Observations 0 Number of Distinct Observations 15 Number of Detects 2 Number of Non-Detects 15 Number of Distinct Detects 2 Number of Distinct Non-Detects 14 Minimum Detect 12 Minimum Non-Detect 14.3 Maximum Detect 19.6 Maximum Non-Detect 20.8 Variance Detected 28.88 Percent Non-Detects 88.24% Mean Detected 15.8 SD Detected 5.374 Mean of Detected Logged Data 2.73 SD of Detected Logged Data 0.347 Warning: Data set has only 2 Detected Values. This is not enough to compute meaningful or reliable statistics and estimates. Critical Values for Background Threshold Values (BTVs) Tolerance Factor K (For UTL)2.486 d2max (for USL)2.475 Normal GOF Test on Detects Only Not Enough Data to Perform GOF Test Kaplan Meier (KM) Background Statistics Assuming Normal Distribution Mean 12.54 SD 1.957 95% UTL95% Coverage 17.41 95% KM UPL (t)16.06 90% KM Percentile (z)15.05 95% KM Percentile (z)15.76 99% KM Percentile (z)17.1 95% KM USL 17.39 DL/2 Substitution Background Statistics Assuming Normal Distribution Mean 9.576 SD 2.889 95% UTL95% Coverage 16.76 95% UPL (t)14.77 90% Percentile (z)13.28 95% Percentile (z)14.33 99% Percentile (z)16.3 95% USL 16.72 DL/2 is not a recommended method. DL/2 provided for comparisons and historical reasons Gamma GOF Tests on Detected Observations Only Not Enough Data to Perform GOF Test Gamma Statistics on Detected Data Only k hat (MLE)16.95 k star (bias corrected MLE) N/A Theta hat (MLE)0.932 Theta star (bias corrected MLE) N/A nu hat (MLE)67.79 nu star (bias corrected) N/A MLE Mean (bias corrected) N/A MLE Sd (bias corrected) N/A 95% Percentile of Chisquare (2k) N/A The following statistics are computed using gamma distribution and KM estimates Upper Limits using Wilson Hilferty (WH) and Hawkins Wixley (HW) Methods k hat (KM)41.07 nu hat (KM)1396 WH HW WH HW 95% Approx. Gamma UTL with 95% Coverage 17.15 17.12 95% Approx. Gamma UPL 15.75 15.71 95% Gamma USL 17.13 17.09 Lognormal GOF Test on Detected Observations Only Not Enough Data to Perform GOF Test Background Lognormal ROS Statistics Assuming Lognormal Distribution Using Imputed Non-Detects Mean in Original Scale 12.5 Mean in Log Scale 2.518 SD in Original Scale 1.884 SD in Log Scale 0.124 95% UTL95% Coverage 16.87 95% BCA UTL95% Coverage 19.6 95% Bootstrap (%) UTL95% Coverage 19.6 95% UPL (t)15.49 90% Percentile (z)14.53 95% Percentile (z)15.2 99% Percentile (z)16.54 95% USL 16.85 Background DL/2 Statistics Assuming Lognormal Distribution Mean in Original Scale 9.576 Mean in Log Scale 2.228 SD in Original Scale 2.889 SD in Log Scale 0.238 95% UTL95% Coverage 16.78 95% UPL (t)14.24 90% Percentile (z)12.6 95% Percentile (z)13.74 99% Percentile (z)16.16 95% USL 16.74 DL/2 is not a Recommended Method. DL/2 provided for comparisons and historical reasons. Nonparametric Distribution Free Background Statistics Data do not follow a Discernible Distribution (0.05) Nonparametric Upper Limits for BTVs(no distinction made between detects and nondetects) Order of Statistic, r 17 95% UTL with95% Coverage 20.8 Approximate f 0.895 Confidence Coefficient (CC) achieved by UTL 0.582 95% UPL 20.8 95% USL 20.8 95% KM Chebyshev UPL 21.32 Note: The use of USL to estimate a BTV is recommended only when the data set represents a background data set free of outliers and consists of observations collected from clean unimpacted locations. The use of USL tends to provide a balance between false positives and false negatives provided the data represents a background data set and when many onsite observations need to be compared with the BTV. Cadmium General Statistics Total Number of Observations 17 Number of Missing Observations 0 Number of Distinct Observations 14 Number of Detects 0 Number of Non-Detects 17 Number of Distinct Detects 0 Number of Distinct Non-Detects 14 Minimum Detect N/A Minimum Non-Detect 0.69 Maximum Detect N/A Maximum Non-Detect 1 Variance Detected N/A Percent Non-Detects 100% Mean Detected N/A SD Detected N/A Mean of Detected Logged Data N/A SD of Detected Logged Data N/A Warning: All observations are Non-Detects (NDs), therefore all statistics and estimates should also be NDs! Specifically, sample mean, UCLs, UPLs, and other statistics are also NDs lying below the largest detection limit! The Project Team may decide to use alternative site specific values to estimate environmental parameters (e.g., EPC, BTV). The data set for variable Cadmium was not processed! Calcium General Statistics Total Number of Observations 17 Number of Missing Observations 0 Number of Distinct Observations 17 Number of Detects 11 Number of Non-Detects 6 Number of Distinct Detects 11 Number of Distinct Non-Detects 6 Minimum Detect 96 Minimum Non-Detect 160 Maximum Detect 2480 Maximum Non-Detect 208 Variance Detected 485060 Percent Non-Detects 35.29% Mean Detected 1093 SD Detected 696.5 Mean of Detected Logged Data 6.698 SD of Detected Logged Data 0.958 Critical Values for Background Threshold Values (BTVs) Tolerance Factor K (For UTL)2.486 d2max (for USL)2.475 Normal GOF Test on Detects Only Shapiro Wilk Test Statistic 0.946 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.85 Detected Data appear Normal at 5% Significance Level Lilliefors Test Statistic 0.17 Lilliefors GOF Test 5% Lilliefors Critical Value 0.267 Detected Data appear Normal at 5% Significance Level Detected Data appear Normal at 5% Significance Level Kaplan Meier (KM) Background Statistics Assuming Normal Distribution Mean 741.2 SD 715.8 95% UTL95% Coverage 2521 95% KM UPL (t)2027 90% KM Percentile (z)1659 95% KM Percentile (z)1919 99% KM Percentile (z)2406 95% KM USL 2513 DL/2 Substitution Background Statistics Assuming Normal Distribution Mean 740.4 SD 738.6 95% UTL95% Coverage 2577 95% UPL (t)2067 90% Percentile (z)1687 95% Percentile (z)1955 99% Percentile (z)2459 95% USL 2568 DL/2 is not a recommended method. DL/2 provided for comparisons and historical reasons Gamma GOF Tests on Detected Observations Only A-D Test Statistic 0.466 Anderson-Darling GOF Test 5% A-D Critical Value 0.74 Detected data appear Gamma Distributed at 5% Significance Level K-S Test Statistic 0.199 Kolmogrov-Smirnoff GOF 5% K-S Critical Value 0.259 Detected data appear Gamma Distributed at 5% Significance Level Detected data appear Gamma Distributed at 5% Significance Level Gamma Statistics on Detected Data Only k hat (MLE)1.824 k star (bias corrected MLE)1.387 Theta hat (MLE)599.3 Theta star (bias corrected MLE)788 nu hat (MLE)40.13 nu star (bias corrected)30.52 MLE Mean (bias corrected)1093 MLE Sd (bias corrected)928.1 95% Percentile of Chisquare (2k)7.42 Gamma ROS Statistics using Imputed Non-Detects GROS may not be used when data set has > 50% NDs with many tied observations at multiple DLs GROS may not be used when kstar of detected data is small such as < 0.1 For such situations, GROS method tends to yield inflated values of UCLs and BTVs For gamma distributed detected data, BTVs and UCLs may be computed using gamma distribution on KM estimates Minimum 70.31 Mean 732.1 Maximum 2480 Median 524 SD 746.3 CV 1.019 k hat (MLE)0.782 k star (bias corrected MLE)0.684 Theta hat (MLE)935.7 Theta star (bias corrected MLE)1071 nu hat (MLE)26.6 nu star (bias corrected)23.24 MLE Mean (bias corrected)732.1 MLE Sd (bias corrected)885.5 95% Percentile of Chisquare (2k)4.693 90% Percentile 1848 95% Percentile 2513 99% Percentile 4105 The following statistics are computed using Gamma ROS Statistics on Imputed Data Upper Limits using Wilson Hilferty (WH) and Hawkins Wixley (HW) Methods WH HW WH HW 95% Approx. Gamma UTL with 95% Coverage 4362 5081 95% Approx. Gamma UPL 2718 2943 95% Gamma USL 4331 5039 The following statistics are computed using gamma distribution and KM estimates Upper Limits using Wilson Hilferty (WH) and Hawkins Wixley (HW) Methods k hat (KM)1.072 nu hat (KM)36.45 WH HW WH HW 95% Approx. Gamma UTL with 95% Coverage 3960 4494 95% Approx. Gamma UPL 2529 2693 95% Gamma USL 3933 4459 Lognormal GOF Test on Detected Observations Only Shapiro Wilk Test Statistic 0.87 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.85 Detected Data appear Lognormal at 5% Significance Level Lilliefors Test Statistic 0.231 Lilliefors GOF Test 5% Lilliefors Critical Value 0.267 Detected Data appear Lognormal at 5% Significance Level Detected Data appear Lognormal at 5% Significance Level Background Lognormal ROS Statistics Assuming Lognormal Distribution Using Imputed Non-Detects Mean in Original Scale 757.8 Mean in Log Scale 6.086 SD in Original Scale 722.6 SD in Log Scale 1.142 95% UTL95% Coverage 7514 95% BCA UTL95% Coverage 2480 95% Bootstrap (%) UTL95% Coverage 2480 95% UPL (t)3420 90% Percentile (z)1900 95% Percentile (z)2876 99% Percentile (z)6262 95% USL 7419 Statistics using KM estimates on Logged Data and Assuming Lognormal Distribution KM Mean of Logged Data 5.945 95% KM UTL (Lognormal)95% Coverage 8690 KM SD of Logged Data 1.257 95% KM UPL (Lognormal)3653 95% KM Percentile Lognormal (z)3019 95% KM USL (Lognormal)8569 Background DL/2 Statistics Assuming Lognormal Distribution Mean in Original Scale 740.4 Mean in Log Scale 5.935 SD in Original Scale 738.6 SD in Log Scale 1.309 95% UTL95% Coverage 9797 95% UPL (t)3972 90% Percentile (z)2024 95% Percentile (z)3256 99% Percentile (z)7949 95% USL 9654 DL/2 is not a Recommended Method. DL/2 provided for comparisons and historical reasons. Nonparametric Distribution Free Background Statistics Data appear to follow a Discernible Distribution at 5% Significance Level Nonparametric Upper Limits for BTVs(no distinction made between detects and nondetects) Order of Statistic, r 17 95% UTL with95% Coverage 2480 Approximate f 0.895 Confidence Coefficient (CC) achieved by UTL 0.582 95% UPL 2480 95% USL 2480 95% KM Chebyshev UPL 3952 Note: The use of USL to estimate a BTV is recommended only when the data set represents a background data set free of outliers and consists of observations collected from clean unimpacted locations. The use of USL tends to provide a balance between false positives and false negatives provided the data represents a background data set and when many onsite observations need to be compared with the BTV. Chloride General Statistics Total Number of Observations 17 Number of Missing Observations 0 Number of Distinct Observations 17 Number of Detects 0 Number of Non-Detects 17 Number of Distinct Detects 0 Number of Distinct Non-Detects 17 Minimum Detect N/A Minimum Non-Detect 284 Maximum Detect N/A Maximum Non-Detect 427 Variance Detected N/A Percent Non-Detects 100% Mean Detected N/A SD Detected N/A Mean of Detected Logged Data N/A SD of Detected Logged Data N/A Warning: All observations are Non-Detects (NDs), therefore all statistics and estimates should also be NDs! Specifically, sample mean, UCLs, UPLs, and other statistics are also NDs lying below the largest detection limit! The Project Team may decide to use alternative site specific values to estimate environmental parameters (e.g., EPC, BTV). The data set for variable Chloride was not processed! Chromium General Statistics Total Number of Observations 17 Number of Distinct Observations 15 Minimum 3.1 First Quartile 4.5 Second Largest 14.4 Median 5.5 Maximum 20.6 Third Quartile 6.7 Mean 6.776 SD 4.459 Coefficient of Variation 0.658 Skewness 2.323 Mean of logged Data 1.774 SD of logged Data 0.502 Critical Values for Background Threshold Values (BTVs) Tolerance Factor K (For UTL)2.486 d2max (for USL)2.475 Normal GOF Test Shapiro Wilk Test Statistic 0.714 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.892 Data Not Normal at 5% Significance Level Lilliefors Test Statistic 0.272 Lilliefors GOF Test 5% Lilliefors Critical Value 0.215 Data Not Normal at 5% Significance Level Data Not Normal at 5% Significance Level Background Statistics Assuming Normal Distribution 95% UTL with 95% Coverage 17.86 90% Percentile (z)12.49 95% UPL (t)14.79 95% Percentile (z)14.11 95% USL 17.81 99% Percentile (z)17.15 Gamma GOF Test A-D Test Statistic 0.919 Anderson-Darling Gamma GOF Test 5% A-D Critical Value 0.743 Data Not Gamma Distributed at 5% Significance Level K-S Test Statistic 0.216 Kolmogrov-Smirnoff Gamma GOF Test 5% K-S Critical Value 0.21 Data Not Gamma Distributed at 5% Significance Level Data Not Gamma Distributed at 5% Significance Level Gamma Statistics k hat (MLE)3.755 k star (bias corrected MLE)3.132 Theta hat (MLE)1.805 Theta star (bias corrected MLE)2.164 nu hat (MLE)127.7 nu star (bias corrected)106.5 MLE Mean (bias corrected)6.776 MLE Sd (bias corrected)3.829 Background Statistics Assuming Gamma Distribution 95% Wilson Hilferty (WH) Approx. Gamma UPL 14.41 90% Percentile 11.91 95% Hawkins Wixley (HW) Approx. Gamma UPL 14.41 95% Percentile 14.05 95% WH Approx. Gamma UTL with 95% Coverage 18.9 99% Percentile 18.67 95% HW Approx. Gamma UTL with 95% Coverage 19.2 95% WH USL 18.83 95% HW USL 19.12 Lognormal GOF Test Shapiro Wilk Test Statistic 0.908 Shapiro Wilk Lognormal GOF Test 5% Shapiro Wilk Critical Value 0.892 Data appear Lognormal at 5% Significance Level Lilliefors Test Statistic 0.179 Lilliefors Lognormal GOF Test 5% Lilliefors Critical Value 0.215 Data appear Lognormal at 5% Significance Level Data appear Lognormal at 5% Significance Level Background Statistics assuming Lognormal Distribution 95% UTL with 95% Coverage 20.56 90% Percentile (z)11.23 95% UPL (t)14.54 95% Percentile (z)13.48 95% USL 20.45 99% Percentile (z)18.98 Nonparametric Distribution Free Background Statistics Data appear Lognormal at 5% Significance Level Nonparametric Upper Limits for Background Threshold Values Order of Statistic, r 17 95% UTL with 95% Coverage 20.6 Approximate f 0.895 Confidence Coefficient (CC) achieved by UTL 0.582 95% Percentile Bootstrap UTL with 95% Coverage 20.6 95% BCA Bootstrap UTL with 95% Coverage 20.6 95% UPL 20.6 90% Percentile 11.16 90% Chebyshev UPL 20.54 95% Percentile 15.64 95% Chebyshev UPL 26.78 99% Percentile 19.61 95% USL 20.6 Note: The use of USL to estimate a BTV is recommended only when the data set represents a background data set free of outliers and consists of observations collected from clean unimpacted locations. The use of USL tends to provide a balance between false positives and false negatives provided the data represents a background data set and when many onsite observations need to be compared with the BTV. Cobalt General Statistics Total Number of Observations 17 Number of Missing Observations 0 Number of Distinct Observations 15 Number of Detects 15 Number of Non-Detects 2 Number of Distinct Detects 13 Number of Distinct Non-Detects 2 Minimum Detect 5 Minimum Non-Detect 5.7 Maximum Detect 22.8 Maximum Non-Detect 6.5 Variance Detected 18.32 Percent Non-Detects 11.76% Mean Detected 11.83 SD Detected 4.28 Mean of Detected Logged Data 2.41 SD of Detected Logged Data 0.368 Critical Values for Background Threshold Values (BTVs) Tolerance Factor K (For UTL)2.486 d2max (for USL)2.475 Normal GOF Test on Detects Only Shapiro Wilk Test Statistic 0.929 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.881 Detected Data appear Normal at 5% Significance Level Lilliefors Test Statistic 0.185 Lilliefors GOF Test 5% Lilliefors Critical Value 0.229 Detected Data appear Normal at 5% Significance Level Detected Data appear Normal at 5% Significance Level Kaplan Meier (KM) Background Statistics Assuming Normal Distribution Mean 11.03 SD 4.465 95% UTL95% Coverage 22.13 95% KM UPL (t)19.05 90% KM Percentile (z)16.75 95% KM Percentile (z)18.37 99% KM Percentile (z)21.42 95% KM USL 22.08 DL/2 Substitution Background Statistics Assuming Normal Distribution Mean 10.8 SD 4.954 95% UTL95% Coverage 23.12 95% UPL (t)19.7 90% Percentile (z)17.15 95% Percentile (z)18.95 99% Percentile (z)22.33 95% USL 23.06 DL/2 is not a recommended method. DL/2 provided for comparisons and historical reasons Gamma GOF Tests on Detected Observations Only A-D Test Statistic 0.282 Anderson-Darling GOF Test 5% A-D Critical Value 0.738 Detected data appear Gamma Distributed at 5% Significance Level K-S Test Statistic 0.156 Kolmogrov-Smirnoff GOF 5% K-S Critical Value 0.222 Detected data appear Gamma Distributed at 5% Significance Level Detected data appear Gamma Distributed at 5% Significance Level Gamma Statistics on Detected Data Only k hat (MLE)8.335 k star (bias corrected MLE)6.713 Theta hat (MLE)1.42 Theta star (bias corrected MLE)1.763 nu hat (MLE)250.1 nu star (bias corrected)201.4 MLE Mean (bias corrected)11.83 MLE Sd (bias corrected)4.567 95% Percentile of Chisquare (2k)22.93 Gamma ROS Statistics using Imputed Non-Detects GROS may not be used when data set has > 50% NDs with many tied observations at multiple DLs GROS may not be used when kstar of detected data is small such as < 0.1 For such situations, GROS method tends to yield inflated values of UCLs and BTVs For gamma distributed detected data, BTVs and UCLs may be computed using gamma distribution on KM estimates Minimum 4.92 Mean 11.02 Maximum 22.8 Median 11 SD 4.615 CV 0.419 k hat (MLE)5.974 k star (bias corrected MLE)4.959 Theta hat (MLE)1.845 Theta star (bias corrected MLE)2.222 nu hat (MLE)203.1 nu star (bias corrected)168.6 MLE Mean (bias corrected)11.02 MLE Sd (bias corrected)4.949 95% Percentile of Chisquare (2k)18.19 90% Percentile 17.65 95% Percentile 20.22 99% Percentile 25.65 The following statistics are computed using Gamma ROS Statistics on Imputed Data Upper Limits using Wilson Hilferty (WH) and Hawkins Wixley (HW) Methods WH HW WH HW 95% Approx. Gamma UTL with 95% Coverage 25.98 26.76 95% Approx. Gamma UPL 20.71 21.01 95% Gamma USL 25.89 26.66 The following statistics are computed using gamma distribution and KM estimates Upper Limits using Wilson Hilferty (WH) and Hawkins Wixley (HW) Methods k hat (KM)6.102 nu hat (KM)207.5 WH HW WH HW 95% Approx. Gamma UTL with 95% Coverage 25.29 25.99 95% Approx. Gamma UPL 20.29 20.56 95% Gamma USL 25.21 25.9 Lognormal GOF Test on Detected Observations Only Shapiro Wilk Test Statistic 0.965 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.881 Detected Data appear Lognormal at 5% Significance Level Lilliefors Test Statistic 0.158 Lilliefors GOF Test 5% Lilliefors Critical Value 0.229 Detected Data appear Lognormal at 5% Significance Level Detected Data appear Lognormal at 5% Significance Level Background Lognormal ROS Statistics Assuming Lognormal Distribution Using Imputed Non-Detects Mean in Original Scale 11.1 Mean in Log Scale 2.328 SD in Original Scale 4.514 SD in Log Scale 0.415 95% UTL95% Coverage 28.75 95% BCA UTL95% Coverage 22.8 95% Bootstrap (%) UTL95% Coverage 22.8 95% UPL (t)21.6 90% Percentile (z)17.45 95% Percentile (z)20.29 99% Percentile (z)26.91 95% USL 28.62 Statistics using KM estimates on Logged Data and Assuming Lognormal Distribution KM Mean of Logged Data 2.316 95% KM UTL (Lognormal)95% Coverage 28.93 KM SD of Logged Data 0.422 95% KM UPL (Lognormal)21.63 95% KM Percentile Lognormal (z)20.28 95% KM USL (Lognormal)28.79 Background DL/2 Statistics Assuming Lognormal Distribution Mean in Original Scale 10.8 Mean in Log Scale 2.257 SD in Original Scale 4.954 SD in Log Scale 0.552 95% UTL95% Coverage 37.69 95% UPL (t)25.76 90% Percentile (z)19.39 95% Percentile (z)23.69 99% Percentile (z)34.51 95% USL 37.46 DL/2 is not a Recommended Method. DL/2 provided for comparisons and historical reasons. Nonparametric Distribution Free Background Statistics Data appear to follow a Discernible Distribution at 5% Significance Level Nonparametric Upper Limits for BTVs(no distinction made between detects and nondetects) Order of Statistic, r 17 95% UTL with95% Coverage 22.8 Approximate f 0.895 Confidence Coefficient (CC) achieved by UTL 0.582 95% UPL 22.8 95% USL 22.8 95% KM Chebyshev UPL 31.06 Note: The use of USL to estimate a BTV is recommended only when the data set represents a background data set free of outliers and consists of observations collected from clean unimpacted locations. The use of USL tends to provide a balance between false positives and false negatives provided the data represents a background data set and when many onsite observations need to be compared with the BTV. Copper General Statistics Total Number of Observations 17 Number of Distinct Observations 17 Minimum 4.3 First Quartile 10.3 Second Largest 71.8 Median 15.1 Maximum 93.3 Third Quartile 16.4 Mean 20.83 SD 24.22 Coefficient of Variation 1.163 Skewness 2.456 Mean of logged Data 2.638 SD of logged Data 0.85 Critical Values for Background Threshold Values (BTVs) Tolerance Factor K (For UTL)2.486 d2max (for USL)2.475 Normal GOF Test Shapiro Wilk Test Statistic 0.619 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.892 Data Not Normal at 5% Significance Level Lilliefors Test Statistic 0.337 Lilliefors GOF Test 5% Lilliefors Critical Value 0.215 Data Not Normal at 5% Significance Level Data Not Normal at 5% Significance Level Background Statistics Assuming Normal Distribution 95% UTL with 95% Coverage 81.04 90% Percentile (z)51.87 95% UPL (t)64.34 95% Percentile (z)60.67 95% USL 80.77 99% Percentile (z)77.17 Gamma GOF Test A-D Test Statistic 1.15 Anderson-Darling Gamma GOF Test 5% A-D Critical Value 0.758 Data Not Gamma Distributed at 5% Significance Level K-S Test Statistic 0.257 Kolmogrov-Smirnoff Gamma GOF Test 5% K-S Critical Value 0.213 Data Not Gamma Distributed at 5% Significance Level Data Not Gamma Distributed at 5% Significance Level Gamma Statistics k hat (MLE)1.397 k star (bias corrected MLE)1.19 Theta hat (MLE)14.91 Theta star (bias corrected MLE)17.51 nu hat (MLE)47.5 nu star (bias corrected)40.45 MLE Mean (bias corrected)20.83 MLE Sd (bias corrected)19.1 Background Statistics Assuming Gamma Distribution 95% Wilson Hilferty (WH) Approx. Gamma UPL 60.66 90% Percentile 45.95 95% Hawkins Wixley (HW) Approx. Gamma UPL 60.93 95% Percentile 58.72 95% WH Approx. Gamma UTL with 95% Coverage 89.6 99% Percentile 88 95% HW Approx. Gamma UTL with 95% Coverage 93.23 95% WH USL 89.06 95% HW USL 92.62 Lognormal GOF Test Shapiro Wilk Test Statistic 0.909 Shapiro Wilk Lognormal GOF Test 5% Shapiro Wilk Critical Value 0.892 Data appear Lognormal at 5% Significance Level Lilliefors Test Statistic 0.19 Lilliefors Lognormal GOF Test 5% Lilliefors Critical Value 0.215 Data appear Lognormal at 5% Significance Level Data appear Lognormal at 5% Significance Level Background Statistics assuming Lognormal Distribution 95% UTL with 95% Coverage 115.6 90% Percentile (z)41.55 95% UPL (t)64.36 95% Percentile (z)56.58 95% USL 114.5 99% Percentile (z)101 Nonparametric Distribution Free Background Statistics Data appear Lognormal at 5% Significance Level Nonparametric Upper Limits for Background Threshold Values Order of Statistic, r 17 95% UTL with 95% Coverage 93.3 Approximate f 0.895 Confidence Coefficient (CC) achieved by UTL 0.582 95% Percentile Bootstrap UTL with 95% Coverage 93.3 95% BCA Bootstrap UTL with 95% Coverage 93.3 95% UPL 93.3 90% Percentile 43.84 90% Chebyshev UPL 95.6 95% Percentile 76.1 95% Chebyshev UPL 129.5 99% Percentile 89.86 95% USL 93.3 Note: The use of USL to estimate a BTV is recommended only when the data set represents a background data set free of outliers and consists of observations collected from clean unimpacted locations. The use of USL tends to provide a balance between false positives and false negatives provided the data represents a background data set and when many onsite observations need to be compared with the BTV. Iron General Statistics Total Number of Observations 17 Number of Distinct Observations 16 Minimum 12300 First Quartile 16000 Second Largest 28200 Median 19400 Maximum 32200 Third Quartile 23500 Mean 19994 SD 5546 Coefficient of Variation 0.277 Skewness 0.635 Mean of logged Data 9.868 SD of logged Data 0.274 Critical Values for Background Threshold Values (BTVs) Tolerance Factor K (For UTL)2.486 d2max (for USL)2.475 Normal GOF Test Shapiro Wilk Test Statistic 0.956 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.892 Data appear Normal at 5% Significance Level Lilliefors Test Statistic 0.134 Lilliefors GOF Test 5% Lilliefors Critical Value 0.215 Data appear Normal at 5% Significance Level Data appear Normal at 5% Significance Level Background Statistics Assuming Normal Distribution 95% UTL with 95% Coverage 33782 90% Percentile (z)27102 95% UPL (t)29958 95% Percentile (z)29117 95% USL 33720 99% Percentile (z)32897 Gamma GOF Test A-D Test Statistic 0.184 Anderson-Darling Gamma GOF Test 5% A-D Critical Value 0.738 Detected data appear Gamma Distributed at 5% Significance Level K-S Test Statistic 0.0975 Kolmogrov-Smirnoff Gamma GOF Test 5% K-S Critical Value 0.209 Detected data appear Gamma Distributed at 5% Significance Level Detected data appear Gamma Distributed at 5% Significance Level Gamma Statistics k hat (MLE)14.29 k star (bias corrected MLE)11.81 Theta hat (MLE)1399 Theta star (bias corrected MLE)1693 nu hat (MLE)486 nu star (bias corrected)401.6 MLE Mean (bias corrected)19994 MLE Sd (bias corrected)5818 Background Statistics Assuming Gamma Distribution 95% Wilson Hilferty (WH) Approx. Gamma UPL 30846 90% Percentile 27718 95% Hawkins Wixley (HW) Approx. Gamma UPL 31003 95% Percentile 30425 95% WH Approx. Gamma UTL with 95% Coverage 36144 99% Percentile 35950 95% HW Approx. Gamma UTL with 95% Coverage 36569 95% WH USL 36054 95% HW USL 36473 Lognormal GOF Test Shapiro Wilk Test Statistic 0.977 Shapiro Wilk Lognormal GOF Test 5% Shapiro Wilk Critical Value 0.892 Data appear Lognormal at 5% Significance Level Lilliefors Test Statistic 0.0883 Lilliefors Lognormal GOF Test 5% Lilliefors Critical Value 0.215 Data appear Lognormal at 5% Significance Level Data appear Lognormal at 5% Significance Level Background Statistics assuming Lognormal Distribution 95% UTL with 95% Coverage 38117 90% Percentile (z)27410 95% UPL (t)31560 95% Percentile (z)30276 95% USL 38000 99% Percentile (z)36486 Nonparametric Distribution Free Background Statistics Data appear Normal at 5% Significance Level Nonparametric Upper Limits for Background Threshold Values Order of Statistic, r 17 95% UTL with 95% Coverage 32200 Approximate f 0.895 Confidence Coefficient (CC) achieved by UTL 0.582 95% Percentile Bootstrap UTL with 95% Coverage 32200 95% BCA Bootstrap UTL with 95% Coverage 32200 95% UPL 32200 90% Percentile 27360 90% Chebyshev UPL 37116 95% Percentile 29000 95% Chebyshev UPL 44871 99% Percentile 31560 95% USL 32200 Note: The use of USL to estimate a BTV is recommended only when the data set represents a background data set free of outliers and consists of observations collected from clean unimpacted locations. The use of USL tends to provide a balance between false positives and false negatives provided the data represents a background data set and when many onsite observations need to be compared with the BTV. Lead General Statistics Total Number of Observations 17 Number of Missing Observations 0 Number of Distinct Observations 16 Number of Detects 12 Number of Non-Detects 5 Number of Distinct Detects 11 Number of Distinct Non-Detects 5 Minimum Detect 3.3 Minimum Non-Detect 5.7 Maximum Detect 15.2 Maximum Non-Detect 7.8 Variance Detected 16.53 Percent Non-Detects 29.41% Mean Detected 6.017 SD Detected 4.066 Mean of Detected Logged Data 1.646 SD of Detected Logged Data 0.52 Critical Values for Background Threshold Values (BTVs) Tolerance Factor K (For UTL)2.486 d2max (for USL)2.475 Normal GOF Test on Detects Only Shapiro Wilk Test Statistic 0.664 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.859 Data Not Normal at 5% Significance Level Lilliefors Test Statistic 0.306 Lilliefors GOF Test 5% Lilliefors Critical Value 0.256 Data Not Normal at 5% Significance Level Data Not Normal at 5% Significance Level Kaplan Meier (KM) Background Statistics Assuming Normal Distribution Mean 5.484 SD 3.409 95% UTL95% Coverage 13.96 95% KM UPL (t)11.61 90% KM Percentile (z)9.853 95% KM Percentile (z)11.09 99% KM Percentile (z)13.41 95% KM USL 13.92 DL/2 Substitution Background Statistics Assuming Normal Distribution Mean 5.271 SD 3.581 95% UTL95% Coverage 14.17 95% UPL (t)11.7 90% Percentile (z)9.86 95% Percentile (z)11.16 99% Percentile (z)13.6 95% USL 14.13 DL/2 is not a recommended method. DL/2 provided for comparisons and historical reasons Gamma GOF Tests on Detected Observations Only A-D Test Statistic 1.417 Anderson-Darling GOF Test 5% A-D Critical Value 0.737 Data Not Gamma Distributed at 5% Significance Level K-S Test Statistic 0.258 Kolmogrov-Smirnoff GOF 5% K-S Critical Value 0.247 Data Not Gamma Distributed at 5% Significance Level Data Not Gamma Distributed at 5% Significance Level Gamma Statistics on Detected Data Only k hat (MLE)3.529 k star (bias corrected MLE)2.702 Theta hat (MLE)1.705 Theta star (bias corrected MLE)2.226 nu hat (MLE)84.7 nu star (bias corrected)64.86 MLE Mean (bias corrected)6.017 MLE Sd (bias corrected)3.66 95% Percentile of Chisquare (2k)11.69 Gamma ROS Statistics using Imputed Non-Detects GROS may not be used when data set has > 50% NDs with many tied observations at multiple DLs GROS may not be used when kstar of detected data is small such as < 0.1 For such situations, GROS method tends to yield inflated values of UCLs and BTVs For gamma distributed detected data, BTVs and UCLs may be computed using gamma distribution on KM estimates Minimum 3.3 Mean 5.533 Maximum 15.2 Median 4.503 SD 3.462 CV 0.626 k hat (MLE)4.495 k star (bias corrected MLE)3.741 Theta hat (MLE)1.231 Theta star (bias corrected MLE)1.479 nu hat (MLE)152.8 nu star (bias corrected)127.2 MLE Mean (bias corrected)5.533 MLE Sd (bias corrected)2.861 95% Percentile of Chisquare (2k)14.76 90% Percentile 9.368 95% Percentile 10.92 99% Percentile 14.24 The following statistics are computed using Gamma ROS Statistics on Imputed Data Upper Limits using Wilson Hilferty (WH) and Hawkins Wixley (HW) Methods WH HW WH HW 95% Approx. Gamma UTL with 95% Coverage 14.37 14.43 95% Approx. Gamma UPL 11.15 11.08 95% Gamma USL 14.31 14.37 The following statistics are computed using gamma distribution and KM estimates Upper Limits using Wilson Hilferty (WH) and Hawkins Wixley (HW) Methods k hat (KM)2.588 nu hat (KM)88 WH HW WH HW 95% Approx. Gamma UTL with 95% Coverage 14.22 14.29 95% Approx. Gamma UPL 11.02 10.96 95% Gamma USL 14.16 14.23 Lognormal GOF Test on Detected Observations Only Shapiro Wilk Test Statistic 0.775 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.859 Data Not Lognormal at 5% Significance Level Lilliefors Test Statistic 0.241 Lilliefors GOF Test 5% Lilliefors Critical Value 0.256 Detected Data appear Lognormal at 5% Significance Level Detected Data appear Approximate Lognormal at 5% Significance Level Background Lognormal ROS Statistics Assuming Lognormal Distribution Using Imputed Non-Detects Mean in Original Scale 5.554 Mean in Log Scale 1.601 SD in Original Scale 3.452 SD in Log Scale 0.438 95% UTL95% Coverage 14.71 95% BCA UTL95% Coverage 15.2 95% Bootstrap (%) UTL95% Coverage 15.2 95% UPL (t)10.88 90% Percentile (z)8.684 95% Percentile (z)10.18 99% Percentile (z)13.72 95% USL 14.64 Statistics using KM estimates on Logged Data and Assuming Lognormal Distribution KM Mean of Logged Data 1.579 95% KM UTL (Lognormal)95% Coverage 14.62 KM SD of Logged Data 0.444 95% KM UPL (Lognormal)10.76 95% KM Percentile Lognormal (z)10.06 95% KM USL (Lognormal)14.54 Background DL/2 Statistics Assuming Lognormal Distribution Mean in Original Scale 5.271 Mean in Log Scale 1.527 SD in Original Scale 3.581 SD in Log Scale 0.475 95% UTL95% Coverage 14.99 95% UPL (t)10.8 90% Percentile (z)8.462 95% Percentile (z)10.05 99% Percentile (z)13.89 95% USL 14.91 DL/2 is not a Recommended Method. DL/2 provided for comparisons and historical reasons. Nonparametric Distribution Free Background Statistics Data appear to follow a Discernible Distribution at 5% Significance Level Nonparametric Upper Limits for BTVs(no distinction made between detects and nondetects) Order of Statistic, r 17 95% UTL with95% Coverage 15.2 Approximate f 0.895 Confidence Coefficient (CC) achieved by UTL 0.582 95% UPL 15.2 95% USL 15.2 95% KM Chebyshev UPL 20.77 Note: The use of USL to estimate a BTV is recommended only when the data set represents a background data set free of outliers and consists of observations collected from clean unimpacted locations. The use of USL tends to provide a balance between false positives and false negatives provided the data represents a background data set and when many onsite observations need to be compared with the BTV. Magnesium General Statistics Total Number of Observations 17 Number of Missing Observations 0 Number of Distinct Observations 17 Number of Detects 16 Number of Non-Detects 1 Number of Distinct Detects 16 Number of Distinct Non-Detects 1 Minimum Detect 620 Minimum Non-Detect 162 Maximum Detect 10200 Maximum Non-Detect 162 Variance Detected 6017823 Percent Non-Detects 5.88% Mean Detected 6157 SD Detected 2453 Mean of Detected Logged Data 8.589 SD of Detected Logged Data 0.665 Critical Values for Background Threshold Values (BTVs) Tolerance Factor K (For UTL)2.486 d2max (for USL)2.475 Normal GOF Test on Detects Only Shapiro Wilk Test Statistic 0.968 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.887 Detected Data appear Normal at 5% Significance Level Lilliefors Test Statistic 0.115 Lilliefors GOF Test 5% Lilliefors Critical Value 0.222 Detected Data appear Normal at 5% Significance Level Detected Data appear Normal at 5% Significance Level Kaplan Meier (KM) Background Statistics Assuming Normal Distribution Mean 5804 SD 2702 95% UTL95% Coverage 12521 95% KM UPL (t)10658 90% KM Percentile (z)9267 95% KM Percentile (z)10248 99% KM Percentile (z)12089 95% KM USL 12491 DL/2 Substitution Background Statistics Assuming Normal Distribution Mean 5799 SD 2795 95% UTL95% Coverage 12748 95% UPL (t)10821 90% Percentile (z)9382 95% Percentile (z)10397 99% Percentile (z)12302 95% USL 12717 DL/2 is not a recommended method. DL/2 provided for comparisons and historical reasons Gamma GOF Tests on Detected Observations Only A-D Test Statistic 0.877 Anderson-Darling GOF Test 5% A-D Critical Value 0.742 Data Not Gamma Distributed at 5% Significance Level K-S Test Statistic 0.198 Kolmogrov-Smirnoff GOF 5% K-S Critical Value 0.216 Detected data appear Gamma Distributed at 5% Significance Level Detected data follow Appr. Gamma Distribution at 5% Significance Level Gamma Statistics on Detected Data Only k hat (MLE)3.82 k star (bias corrected MLE)3.145 Theta hat (MLE)1612 Theta star (bias corrected MLE)1957 nu hat (MLE)122.2 nu star (bias corrected)100.7 MLE Mean (bias corrected)6157 MLE Sd (bias corrected)3472 95% Percentile of Chisquare (2k)13.02 Gamma ROS Statistics using Imputed Non-Detects GROS may not be used when data set has > 50% NDs with many tied observations at multiple DLs GROS may not be used when kstar of detected data is small such as < 0.1 For such situations, GROS method tends to yield inflated values of UCLs and BTVs For gamma distributed detected data, BTVs and UCLs may be computed using gamma distribution on KM estimates Minimum 620 Mean 5916 Maximum 10200 Median 6300 SD 2574 CV 0.435 k hat (MLE)3.43 k star (bias corrected MLE)2.864 Theta hat (MLE)1725 Theta star (bias corrected MLE)2066 nu hat (MLE)116.6 nu star (bias corrected)97.36 MLE Mean (bias corrected)5916 MLE Sd (bias corrected)3496 95% Percentile of Chisquare (2k)12.18 90% Percentile 10604 95% Percentile 12584 99% Percentile 16888 The following statistics are computed using Gamma ROS Statistics on Imputed Data Upper Limits using Wilson Hilferty (WH) and Hawkins Wixley (HW) Methods WH HW WH HW 95% Approx. Gamma UTL with 95% Coverage 17172 18667 95% Approx. Gamma UPL 13001 13693 95% Gamma USL 17098 18577 The following statistics are computed using gamma distribution and KM estimates Upper Limits using Wilson Hilferty (WH) and Hawkins Wixley (HW) Methods k hat (KM)4.615 nu hat (KM)156.9 WH HW WH HW 95% Approx. Gamma UTL with 95% Coverage 20809 24197 95% Approx. Gamma UPL 14938 16488 95% Gamma USL 20703 24054 Lognormal GOF Test on Detected Observations Only Shapiro Wilk Test Statistic 0.731 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.887 Data Not Lognormal at 5% Significance Level Lilliefors Test Statistic 0.251 Lilliefors GOF Test 5% Lilliefors Critical Value 0.222 Data Not Lognormal at 5% Significance Level Data Not Lognormal at 5% Significance Level Background Lognormal ROS Statistics Assuming Lognormal Distribution Using Imputed Non-Detects Mean in Original Scale 5878 Mean in Log Scale 8.51 SD in Original Scale 2639 SD in Log Scale 0.72 95% UTL95% Coverage 29756 95% BCA UTL95% Coverage 10200 95% Bootstrap (%) UTL95% Coverage 10200 95% UPL (t)18110 90% Percentile (z)12498 95% Percentile (z)16236 99% Percentile (z)26524 95% USL 29517 Background DL/2 Statistics Assuming Lognormal Distribution Mean in Original Scale 5799 Mean in Log Scale 8.342 SD in Original Scale 2795 SD in Log Scale 1.204 95% UTL95% Coverage 83667 95% UPL (t)36483 90% Percentile (z)19628 95% Percentile (z)30396 99% Percentile (z)69038 95% USL 82547 DL/2 is not a Recommended Method. DL/2 provided for comparisons and historical reasons. Nonparametric Distribution Free Background Statistics Data appear to follow a Discernible Distribution at 5% Significance Level Nonparametric Upper Limits for BTVs(no distinction made between detects and nondetects) Order of Statistic, r 17 95% UTL with95% Coverage 10200 Approximate f 0.895 Confidence Coefficient (CC) achieved by UTL 0.582 95% UPL 10200 95% USL 10200 95% KM Chebyshev UPL 17922 Note: The use of USL to estimate a BTV is recommended only when the data set represents a background data set free of outliers and consists of observations collected from clean unimpacted locations. The use of USL tends to provide a balance between false positives and false negatives provided the data represents a background data set and when many onsite observations need to be compared with the BTV. Manganese General Statistics Total Number of Observations 17 Number of Distinct Observations 17 Minimum 29.1 First Quartile 518 Second Largest 1140 Median 689 Maximum 1440 Third Quartile 921 Mean 690.9 SD 347 Coefficient of Variation 0.502 Skewness 0.147 Mean of logged Data 6.315 SD of logged Data 0.89 Critical Values for Background Threshold Values (BTVs) Tolerance Factor K (For UTL)2.486 d2max (for USL)2.475 Normal GOF Test Shapiro Wilk Test Statistic 0.991 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.892 Data appear Normal at 5% Significance Level Lilliefors Test Statistic 0.0889 Lilliefors GOF Test 5% Lilliefors Critical Value 0.215 Data appear Normal at 5% Significance Level Data appear Normal at 5% Significance Level Background Statistics Assuming Normal Distribution 95% UTL with 95% Coverage 1554 90% Percentile (z)1136 95% UPL (t)1314 95% Percentile (z)1262 95% USL 1550 99% Percentile (z)1498 Gamma GOF Test A-D Test Statistic 0.701 Anderson-Darling Gamma GOF Test 5% A-D Critical Value 0.748 Detected data appear Gamma Distributed at 5% Significance Level K-S Test Statistic 0.184 Kolmogrov-Smirnoff Gamma GOF Test 5% K-S Critical Value 0.211 Detected data appear Gamma Distributed at 5% Significance Level Detected data appear Gamma Distributed at 5% Significance Level Gamma Statistics k hat (MLE)2.393 k star (bias corrected MLE)2.01 Theta hat (MLE)288.7 Theta star (bias corrected MLE)343.7 nu hat (MLE)81.38 nu star (bias corrected)68.35 MLE Mean (bias corrected)690.9 MLE Sd (bias corrected)487.3 Background Statistics Assuming Gamma Distribution 95% Wilson Hilferty (WH) Approx. Gamma UPL 1694 90% Percentile 1342 95% Hawkins Wixley (HW) Approx. Gamma UPL 1828 95% Percentile 1636 95% WH Approx. Gamma UTL with 95% Coverage 2319 99% Percentile 2288 95% HW Approx. Gamma UTL with 95% Coverage 2613 95% WH USL 2308 95% HW USL 2599 Lognormal GOF Test Shapiro Wilk Test Statistic 0.749 Shapiro Wilk Lognormal GOF Test 5% Shapiro Wilk Critical Value 0.892 Data Not Lognormal at 5% Significance Level Lilliefors Test Statistic 0.236 Lilliefors Lognormal GOF Test 5% Lilliefors Critical Value 0.215 Data Not Lognormal at 5% Significance Level Data Not Lognormal at 5% Significance Level Background Statistics assuming Lognormal Distribution 95% UTL with 95% Coverage 5057 90% Percentile (z)1730 95% UPL (t)2737 95% Percentile (z)2391 95% USL 5006 99% Percentile (z)4387 Nonparametric Distribution Free Background Statistics Data appear Normal at 5% Significance Level Nonparametric Upper Limits for Background Threshold Values Order of Statistic, r 17 95% UTL with 95% Coverage 1440 Approximate f 0.895 Confidence Coefficient (CC) achieved by UTL 0.582 95% Percentile Bootstrap UTL with 95% Coverage 1440 95% BCA Bootstrap UTL with 95% Coverage 1440 95% UPL 1440 90% Percentile 1041 90% Chebyshev UPL 1762 95% Percentile 1200 95% Chebyshev UPL 2247 99% Percentile 1392 95% USL 1440 Note: The use of USL to estimate a BTV is recommended only when the data set represents a background data set free of outliers and consists of observations collected from clean unimpacted locations. The use of USL tends to provide a balance between false positives and false negatives provided the data represents a background data set and when many onsite observations need to be compared with the BTV. Mercury General Statistics Total Number of Observations 17 Number of Missing Observations 0 Number of Distinct Observations 8 Number of Detects 2 Number of Non-Detects 15 Number of Distinct Detects 2 Number of Distinct Non-Detects 6 Minimum Detect 0.0095 Minimum Non-Detect 0.0093 Maximum Detect 0.048 Maximum Non-Detect 0.013 Variance Detected 7.41E-04 Percent Non-Detects 88.24% Mean Detected 0.0288 SD Detected 0.0272 Mean of Detected Logged Data -3.847 SD of Detected Logged Data 1.145 Warning: Data set has only 2 Detected Values. This is not enough to compute meaningful or reliable statistics and estimates. Critical Values for Background Threshold Values (BTVs) Tolerance Factor K (For UTL)2.486 d2max (for USL)2.475 Normal GOF Test on Detects Only Not Enough Data to Perform GOF Test Kaplan Meier (KM) Background Statistics Assuming Normal Distribution Mean 0.0117 SD 0.00908 95% UTL95% Coverage 0.0343 95% KM UPL (t)0.028 90% KM Percentile (z)0.0233 95% KM Percentile (z)0.0266 99% KM Percentile (z)0.0328 95% KM USL 0.0341 DL/2 Substitution Background Statistics Assuming Normal Distribution Mean 0.0083 SD 0.0103 95% UTL95% Coverage 0.0339 95% UPL (t)0.0268 90% Percentile (z)0.0215 95% Percentile (z)0.0252 99% Percentile (z)0.0322 95% USL 0.0338 DL/2 is not a recommended method. DL/2 provided for comparisons and historical reasons Gamma GOF Tests on Detected Observations Only Not Enough Data to Perform GOF Test Gamma Statistics on Detected Data Only k hat (MLE)1.83 k star (bias corrected MLE) N/A Theta hat (MLE)0.0157 Theta star (bias corrected MLE) N/A nu hat (MLE)7.321 nu star (bias corrected) N/A MLE Mean (bias corrected) N/A MLE Sd (bias corrected) N/A 95% Percentile of Chisquare (2k) N/A The following statistics are computed using gamma distribution and KM estimates Upper Limits using Wilson Hilferty (WH) and Hawkins Wixley (HW) Methods k hat (KM)1.651 nu hat (KM)56.13 WH HW WH HW 95% Approx. Gamma UTL with 95% Coverage 0.0295 0.0289 95% Approx. Gamma UPL 0.023 0.0224 95% Gamma USL 0.0294 0.0288 Lognormal GOF Test on Detected Observations Only Not Enough Data to Perform GOF Test Background Lognormal ROS Statistics Assuming Lognormal Distribution Using Imputed Non-Detects Mean in Original Scale 0.008 Mean in Log Scale -5.272 SD in Original Scale 0.0109 SD in Log Scale 0.879 95% UTL95% Coverage 0.0456 95% BCA UTL95% Coverage 0.048 95% Bootstrap (%) UTL95% Coverage 0.048 95% UPL (t)0.0249 90% Percentile (z)0.0158 95% Percentile (z)0.0218 99% Percentile (z)0.0397 95% USL 0.0452 Background DL/2 Statistics Assuming Lognormal Distribution Mean in Original Scale 0.0083 Mean in Log Scale -5.038 SD in Original Scale 0.0103 SD in Log Scale 0.544 95% UTL95% Coverage 0.0251 95% UPL (t)0.0172 90% Percentile (z)0.013 95% Percentile (z)0.0159 99% Percentile (z)0.023 95% USL 0.0249 DL/2 is not a Recommended Method. DL/2 provided for comparisons and historical reasons. Nonparametric Distribution Free Background Statistics Data do not follow a Discernible Distribution (0.05) Nonparametric Upper Limits for BTVs(no distinction made between detects and nondetects) Order of Statistic, r 17 95% UTL with95% Coverage 0.048 Approximate f 0.895 Confidence Coefficient (CC) achieved by UTL 0.582 95% UPL 0.048 95% USL 0.048 95% KM Chebyshev UPL 0.0524 Note: The use of USL to estimate a BTV is recommended only when the data set represents a background data set free of outliers and consists of observations collected from clean unimpacted locations. The use of USL tends to provide a balance between false positives and false negatives provided the data represents a background data set and when many onsite observations need to be compared with the BTV. Molybdenum General Statistics Total Number of Observations 17 Number of Missing Observations 0 Number of Distinct Observations 10 Number of Detects 0 Number of Non-Detects 17 Number of Distinct Detects 0 Number of Distinct Non-Detects 10 Minimum Detect N/A Minimum Non-Detect 2.9 Maximum Detect N/A Maximum Non-Detect 4.2 Variance Detected N/A Percent Non-Detects 100% Mean Detected N/A SD Detected N/A Mean of Detected Logged Data N/A SD of Detected Logged Data N/A Warning: All observations are Non-Detects (NDs), therefore all statistics and estimates should also be NDs! Specifically, sample mean, UCLs, UPLs, and other statistics are also NDs lying below the largest detection limit! The Project Team may decide to use alternative site specific values to estimate environmental parameters (e.g., EPC, BTV). The data set for variable Molybdenum was not processed! Nickel General Statistics Total Number of Observations 17 Number of Distinct Observations 15 Minimum 1.3 First Quartile 3.9 Second Largest 8.3 Median 4.9 Maximum 8.5 Third Quartile 6.4 Mean 5.059 SD 2.067 Coefficient of Variation 0.409 Skewness 0.00703 Mean of logged Data 1.522 SD of logged Data 0.498 Critical Values for Background Threshold Values (BTVs) Tolerance Factor K (For UTL)2.486 d2max (for USL)2.475 Normal GOF Test Shapiro Wilk Test Statistic 0.975 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.892 Data appear Normal at 5% Significance Level Lilliefors Test Statistic 0.101 Lilliefors GOF Test 5% Lilliefors Critical Value 0.215 Data appear Normal at 5% Significance Level Data appear Normal at 5% Significance Level Background Statistics Assuming Normal Distribution 95% UTL with 95% Coverage 10.2 90% Percentile (z)7.707 95% UPL (t)8.772 95% Percentile (z)8.458 95% USL 10.17 99% Percentile (z)9.867 Gamma GOF Test A-D Test Statistic 0.316 Anderson-Darling Gamma GOF Test 5% A-D Critical Value 0.741 Detected data appear Gamma Distributed at 5% Significance Level K-S Test Statistic 0.161 Kolmogrov-Smirnoff Gamma GOF Test 5% K-S Critical Value 0.21 Detected data appear Gamma Distributed at 5% Significance Level Detected data appear Gamma Distributed at 5% Significance Level Gamma Statistics k hat (MLE)5.181 k star (bias corrected MLE)4.306 Theta hat (MLE)0.976 Theta star (bias corrected MLE)1.175 nu hat (MLE)176.1 nu star (bias corrected)146.4 MLE Mean (bias corrected)5.059 MLE Sd (bias corrected)2.438 Background Statistics Assuming Gamma Distribution 95% Wilson Hilferty (WH) Approx. Gamma UPL 9.881 90% Percentile 8.326 95% Hawkins Wixley (HW) Approx. Gamma UPL 10.13 95% Percentile 9.619 95% WH Approx. Gamma UTL with 95% Coverage 12.56 99% Percentile 12.37 95% HW Approx. Gamma UTL with 95% Coverage 13.13 95% WH USL 12.51 95% HW USL 13.07 Lognormal GOF Test Shapiro Wilk Test Statistic 0.915 Shapiro Wilk Lognormal GOF Test 5% Shapiro Wilk Critical Value 0.892 Data appear Lognormal at 5% Significance Level Lilliefors Test Statistic 0.192 Lilliefors Lognormal GOF Test 5% Lilliefors Critical Value 0.215 Data appear Lognormal at 5% Significance Level Data appear Lognormal at 5% Significance Level Background Statistics assuming Lognormal Distribution 95% UTL with 95% Coverage 15.8 90% Percentile (z)8.67 95% UPL (t)11.2 95% Percentile (z)10.39 95% USL 15.71 99% Percentile (z)14.59 Nonparametric Distribution Free Background Statistics Data appear Normal at 5% Significance Level Nonparametric Upper Limits for Background Threshold Values Order of Statistic, r 17 95% UTL with 95% Coverage 8.5 Approximate f 0.895 Confidence Coefficient (CC) achieved by UTL 0.582 95% Percentile Bootstrap UTL with 95% Coverage 8.5 95% BCA Bootstrap UTL with 95% Coverage 8.5 95% UPL 8.5 90% Percentile 7.88 90% Chebyshev UPL 11.44 95% Percentile 8.34 95% Chebyshev UPL 14.33 99% Percentile 8.468 95% USL 8.5 Note: The use of USL to estimate a BTV is recommended only when the data set represents a background data set free of outliers and consists of observations collected from clean unimpacted locations. The use of USL tends to provide a balance between false positives and false negatives provided the data represents a background data set and when many onsite observations need to be compared with the BTV. Nitrate General Statistics Total Number of Observations 17 Number of Missing Observations 0 Number of Distinct Observations 17 Number of Detects 0 Number of Non-Detects 17 Number of Distinct Detects 0 Number of Distinct Non-Detects 17 Minimum Detect N/A Minimum Non-Detect 28.4 Maximum Detect N/A Maximum Non-Detect 42.7 Variance Detected N/A Percent Non-Detects 100% Mean Detected N/A SD Detected N/A Mean of Detected Logged Data N/A SD of Detected Logged Data N/A Warning: All observations are Non-Detects (NDs), therefore all statistics and estimates should also be NDs! Specifically, sample mean, UCLs, UPLs, and other statistics are also NDs lying below the largest detection limit! The Project Team may decide to use alternative site specific values to estimate environmental parameters (e.g., EPC, BTV). The data set for variable Nitrate was not processed! Percent Moisture General Statistics Total Number of Observations 17 Number of Distinct Observations 17 Minimum 12.4 First Quartile 20.9 Second Largest 38.5 Median 27 Maximum 41.3 Third Quartile 32.5 Mean 27.02 SD 8.391 Coefficient of Variation 0.311 Skewness 0.142 Mean of logged Data 3.248 SD of logged Data 0.33 Critical Values for Background Threshold Values (BTVs) Tolerance Factor K (For UTL)2.486 d2max (for USL)2.475 Normal GOF Test Shapiro Wilk Test Statistic 0.957 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.892 Data appear Normal at 5% Significance Level Lilliefors Test Statistic 0.16 Lilliefors GOF Test 5% Lilliefors Critical Value 0.215 Data appear Normal at 5% Significance Level Data appear Normal at 5% Significance Level Background Statistics Assuming Normal Distribution 95% UTL with 95% Coverage 47.88 90% Percentile (z)37.78 95% UPL (t)42.1 95% Percentile (z)40.83 95% USL 47.79 99% Percentile (z)46.55 Gamma GOF Test A-D Test Statistic 0.331 Anderson-Darling Gamma GOF Test 5% A-D Critical Value 0.739 Detected data appear Gamma Distributed at 5% Significance Level K-S Test Statistic 0.144 Kolmogrov-Smirnoff Gamma GOF Test 5% K-S Critical Value 0.209 Detected data appear Gamma Distributed at 5% Significance Level Detected data appear Gamma Distributed at 5% Significance Level Gamma Statistics k hat (MLE)10.41 k star (bias corrected MLE)8.611 Theta hat (MLE)2.596 Theta star (bias corrected MLE)3.138 nu hat (MLE)353.9 nu star (bias corrected)292.8 MLE Mean (bias corrected)27.02 MLE Sd (bias corrected)9.209 Background Statistics Assuming Gamma Distribution 95% Wilson Hilferty (WH) Approx. Gamma UPL 44.49 90% Percentile 39.29 95% Hawkins Wixley (HW) Approx. Gamma UPL 44.93 95% Percentile 43.74 95% WH Approx. Gamma UTL with 95% Coverage 53.32 99% Percentile 52.91 95% HW Approx. Gamma UTL with 95% Coverage 54.37 95% WH USL 53.17 95% HW USL 54.21 Lognormal GOF Test Shapiro Wilk Test Statistic 0.952 Shapiro Wilk Lognormal GOF Test 5% Shapiro Wilk Critical Value 0.892 Data appear Lognormal at 5% Significance Level Lilliefors Test Statistic 0.124 Lilliefors Lognormal GOF Test 5% Lilliefors Critical Value 0.215 Data appear Lognormal at 5% Significance Level Data appear Lognormal at 5% Significance Level Background Statistics assuming Lognormal Distribution 95% UTL with 95% Coverage 58.44 90% Percentile (z)39.28 95% UPL (t)46.55 95% Percentile (z)44.28 95% USL 58.23 99% Percentile (z)55.44 Nonparametric Distribution Free Background Statistics Data appear Normal at 5% Significance Level Nonparametric Upper Limits for Background Threshold Values Order of Statistic, r 17 95% UTL with 95% Coverage 41.3 Approximate f 0.895 Confidence Coefficient (CC) achieved by UTL 0.582 95% Percentile Bootstrap UTL with 95% Coverage 41.3 95% BCA Bootstrap UTL with 95% Coverage 41.3 95% UPL 41.3 90% Percentile 37.96 90% Chebyshev UPL 52.93 95% Percentile 39.06 95% Chebyshev UPL 64.66 99% Percentile 40.85 95% USL 41.3 Note: The use of USL to estimate a BTV is recommended only when the data set represents a background data set free of outliers and consists of observations collected from clean unimpacted locations. The use of USL tends to provide a balance between false positives and false negatives provided the data represents a background data set and when many onsite observations need to be compared with the BTV. pH (field) General Statistics Total Number of Observations 17 Number of Distinct Observations 15 Minimum 4.5 First Quartile 5 Second Largest 6.1 Median 5.5 Maximum 6.5 Third Quartile 5.8 Mean 5.418 SD 0.568 Coefficient of Variation 0.105 Skewness -0.0187 Mean of logged Data 1.684 SD of logged Data 0.106 Critical Values for Background Threshold Values (BTVs) Tolerance Factor K (For UTL)2.486 d2max (for USL)2.475 Normal GOF Test Shapiro Wilk Test Statistic 0.971 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.892 Data appear Normal at 5% Significance Level Lilliefors Test Statistic 0.135 Lilliefors GOF Test 5% Lilliefors Critical Value 0.215 Data appear Normal at 5% Significance Level Data appear Normal at 5% Significance Level Background Statistics Assuming Normal Distribution 95% UTL with 95% Coverage 6.83 90% Percentile (z)6.146 95% UPL (t)6.438 95% Percentile (z)6.352 95% USL 6.824 99% Percentile (z)6.739 Gamma GOF Test A-D Test Statistic 0.259 Anderson-Darling Gamma GOF Test 5% A-D Critical Value 0.736 Detected data appear Gamma Distributed at 5% Significance Level K-S Test Statistic 0.148 Kolmogrov-Smirnoff Gamma GOF Test 5% K-S Critical Value 0.208 Detected data appear Gamma Distributed at 5% Significance Level Detected data appear Gamma Distributed at 5% Significance Level Gamma Statistics k hat (MLE)95.6 k star (bias corrected MLE)78.77 Theta hat (MLE)0.0567 Theta star (bias corrected MLE)0.0688 nu hat (MLE)3250 nu star (bias corrected)2678 MLE Mean (bias corrected)5.418 MLE Sd (bias corrected)0.61 Background Statistics Assuming Gamma Distribution 95% Wilson Hilferty (WH) Approx. Gamma UPL 6.489 90% Percentile 6.213 95% Hawkins Wixley (HW) Approx. Gamma UPL 6.496 95% Percentile 6.459 95% WH Approx. Gamma UTL with 95% Coverage 6.943 99% Percentile 6.938 95% HW Approx. Gamma UTL with 95% Coverage 6.959 95% WH USL 6.936 95% HW USL 6.952 Lognormal GOF Test Shapiro Wilk Test Statistic 0.966 Shapiro Wilk Lognormal GOF Test 5% Shapiro Wilk Critical Value 0.892 Data appear Lognormal at 5% Significance Level Lilliefors Test Statistic 0.155 Lilliefors Lognormal GOF Test 5% Lilliefors Critical Value 0.215 Data appear Lognormal at 5% Significance Level Data appear Lognormal at 5% Significance Level Background Statistics assuming Lognormal Distribution 95% UTL with 95% Coverage 7.012 90% Percentile (z)6.173 95% UPL (t)6.519 95% Percentile (z)6.415 95% USL 7.004 99% Percentile (z)6.895 Nonparametric Distribution Free Background Statistics Data appear Normal at 5% Significance Level Nonparametric Upper Limits for Background Threshold Values Order of Statistic, r 17 95% UTL with 95% Coverage 6.5 Approximate f 0.895 Confidence Coefficient (CC) achieved by UTL 0.582 95% Percentile Bootstrap UTL with 95% Coverage 6.5 95% BCA Bootstrap UTL with 95% Coverage 6.5 95% UPL 6.5 90% Percentile 6.04 90% Chebyshev UPL 7.172 95% Percentile 6.18 95% Chebyshev UPL 7.966 99% Percentile 6.436 95% USL 6.5 Note: The use of USL to estimate a BTV is recommended only when the data set represents a background data set free of outliers and consists of observations collected from clean unimpacted locations. The use of USL tends to provide a balance between false positives and false negatives provided the data represents a background data set and when many onsite observations need to be compared with the BTV. Potassium General Statistics Total Number of Observations 17 Number of Missing Observations 0 Number of Distinct Observations 17 Number of Detects 16 Number of Non-Detects 1 Number of Distinct Detects 16 Number of Distinct Non-Detects 1 Minimum Detect 216 Minimum Non-Detect 325 Maximum Detect 8700 Maximum Non-Detect 325 Variance Detected 4638340 Percent Non-Detects 5.88% Mean Detected 5159 SD Detected 2154 Mean of Detected Logged Data 8.355 SD of Detected Logged Data 0.87 Critical Values for Background Threshold Values (BTVs) Tolerance Factor K (For UTL)2.486 d2max (for USL)2.475 Normal GOF Test on Detects Only Shapiro Wilk Test Statistic 0.943 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.887 Detected Data appear Normal at 5% Significance Level Lilliefors Test Statistic 0.166 Lilliefors GOF Test 5% Lilliefors Critical Value 0.222 Detected Data appear Normal at 5% Significance Level Detected Data appear Normal at 5% Significance Level Kaplan Meier (KM) Background Statistics Assuming Normal Distribution Mean 4868 SD 2333 95% UTL95% Coverage 10669 95% KM UPL (t)9060 90% KM Percentile (z)7858 95% KM Percentile (z)8706 99% KM Percentile (z)10296 95% KM USL 10643 DL/2 Substitution Background Statistics Assuming Normal Distribution Mean 4865 SD 2412 95% UTL95% Coverage 10860 95% UPL (t)9197 90% Percentile (z)7955 95% Percentile (z)8832 99% Percentile (z)10475 95% USL 10833 DL/2 is not a recommended method. DL/2 provided for comparisons and historical reasons Gamma GOF Tests on Detected Observations Only A-D Test Statistic 1.3 Anderson-Darling GOF Test 5% A-D Critical Value 0.746 Data Not Gamma Distributed at 5% Significance Level K-S Test Statistic 0.232 Kolmogrov-Smirnoff GOF 5% K-S Critical Value 0.217 Data Not Gamma Distributed at 5% Significance Level Data Not Gamma Distributed at 5% Significance Level Gamma Statistics on Detected Data Only k hat (MLE)2.736 k star (bias corrected MLE)2.265 Theta hat (MLE)1886 Theta star (bias corrected MLE)2278 nu hat (MLE)87.55 nu star (bias corrected)72.46 MLE Mean (bias corrected)5159 MLE Sd (bias corrected)3428 95% Percentile of Chisquare (2k)10.33 Gamma ROS Statistics using Imputed Non-Detects GROS may not be used when data set has > 50% NDs with many tied observations at multiple DLs GROS may not be used when kstar of detected data is small such as < 0.1 For such situations, GROS method tends to yield inflated values of UCLs and BTVs For gamma distributed detected data, BTVs and UCLs may be computed using gamma distribution on KM estimates Minimum 216 Mean 4987 Maximum 8700 Median 5550 SD 2202 CV 0.442 k hat (MLE)2.685 k star (bias corrected MLE)2.251 Theta hat (MLE)1857 Theta star (bias corrected MLE)2216 nu hat (MLE)91.3 nu star (bias corrected)76.52 MLE Mean (bias corrected)4987 MLE Sd (bias corrected)3324 95% Percentile of Chisquare (2k)10.29 90% Percentile 9435 95% Percentile 11400 99% Percentile 15727 The following statistics are computed using Gamma ROS Statistics on Imputed Data Upper Limits using Wilson Hilferty (WH) and Hawkins Wixley (HW) Methods WH HW WH HW 95% Approx. Gamma UTL with 95% Coverage 15866 17887 95% Approx. Gamma UPL 11749 12703 95% Gamma USL 15792 17792 The following statistics are computed using gamma distribution and KM estimates Upper Limits using Wilson Hilferty (WH) and Hawkins Wixley (HW) Methods k hat (KM)4.352 nu hat (KM)148 WH HW WH HW 95% Approx. Gamma UTL with 95% Coverage 18233 21346 95% Approx. Gamma UPL 12945 14347 95% Gamma USL 18137 21215 Lognormal GOF Test on Detected Observations Only Shapiro Wilk Test Statistic 0.638 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.887 Data Not Lognormal at 5% Significance Level Lilliefors Test Statistic 0.237 Lilliefors GOF Test 5% Lilliefors Critical Value 0.222 Data Not Lognormal at 5% Significance Level Data Not Lognormal at 5% Significance Level Background Lognormal ROS Statistics Assuming Lognormal Distribution Using Imputed Non-Detects Mean in Original Scale 4919 Mean in Log Scale 8.274 SD in Original Scale 2308 SD in Log Scale 0.905 95% UTL95% Coverage 37214 95% BCA UTL95% Coverage 8700 95% Bootstrap (%) UTL95% Coverage 8700 95% UPL (t)19937 90% Percentile (z)12509 95% Percentile (z)17380 99% Percentile (z)32207 95% USL 36839 Background DL/2 Statistics Assuming Lognormal Distribution Mean in Original Scale 4865 Mean in Log Scale 8.163 SD in Original Scale 2412 SD in Log Scale 1.156 95% UTL95% Coverage 62073 95% UPL (t)27976 90% Percentile (z)15428 95% Percentile (z)23478 99% Percentile (z)51613 95% USL 61275 DL/2 is not a Recommended Method. DL/2 provided for comparisons and historical reasons. Nonparametric Distribution Free Background Statistics Data appear to follow a Discernible Distribution at 5% Significance Level Nonparametric Upper Limits for BTVs(no distinction made between detects and nondetects) Order of Statistic, r 17 95% UTL with95% Coverage 8700 Approximate f 0.895 Confidence Coefficient (CC) achieved by UTL 0.582 95% UPL 8700 95% USL 8700 95% KM Chebyshev UPL 15334 Note: The use of USL to estimate a BTV is recommended only when the data set represents a background data set free of outliers and consists of observations collected from clean unimpacted locations. The use of USL tends to provide a balance between false positives and false negatives provided the data represents a background data set and when many onsite observations need to be compared with the BTV. Selenium General Statistics Total Number of Observations 16 Number of Missing Observations 1 Number of Distinct Observations 12 Number of Detects 1 Number of Non-Detects 15 Number of Distinct Detects 1 Number of Distinct Non-Detects 11 Minimum Detect 4.7 Minimum Non-Detect 5.7 Maximum Detect 4.7 Maximum Non-Detect 8.3 Variance Detected N/A Percent Non-Detects 93.75% Mean Detected 4.7 SD Detected N/A Mean of Detected Logged Data 1.548 SD of Detected Logged Data N/A Warning: Only one distinct data value was detected! ProUCL (or any other software) should not be used on such a data set! It is suggested to use alternative site specific values determined by the Project Team to estimate environmental parameters (e.g., EPC, BTV). The data set for variable Selenium was not processed! Sodium General Statistics Total Number of Observations 17 Number of Missing Observations 0 Number of Distinct Observations 15 Number of Detects 0 Number of Non-Detects 17 Number of Distinct Detects 0 Number of Distinct Non-Detects 15 Minimum Detect N/A Minimum Non-Detect 286 Maximum Detect N/A Maximum Non-Detect 417 Variance Detected N/A Percent Non-Detects 100% Mean Detected N/A SD Detected N/A Mean of Detected Logged Data N/A SD of Detected Logged Data N/A Warning: All observations are Non-Detects (NDs), therefore all statistics and estimates should also be NDs! Specifically, sample mean, UCLs, UPLs, and other statistics are also NDs lying below the largest detection limit! The Project Team may decide to use alternative site specific values to estimate environmental parameters (e.g., EPC, BTV). The data set for variable Sodium was not processed! Strontium General Statistics Total Number of Observations 17 Number of Missing Observations 0 Number of Distinct Observations 14 Number of Detects 13 Number of Non-Detects 4 Number of Distinct Detects 12 Number of Distinct Non-Detects 2 Minimum Detect 2.1 Minimum Non-Detect 3.2 Maximum Detect 22.2 Maximum Non-Detect 4.1 Variance Detected 53.84 Percent Non-Detects 23.53% Mean Detected 11.32 SD Detected 7.337 Mean of Detected Logged Data 2.152 SD of Detected Logged Data 0.85 Critical Values for Background Threshold Values (BTVs) Tolerance Factor K (For UTL)2.486 d2max (for USL)2.475 Normal GOF Test on Detects Only Shapiro Wilk Test Statistic 0.91 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.866 Detected Data appear Normal at 5% Significance Level Lilliefors Test Statistic 0.149 Lilliefors GOF Test 5% Lilliefors Critical Value 0.246 Detected Data appear Normal at 5% Significance Level Detected Data appear Normal at 5% Significance Level Kaplan Meier (KM) Background Statistics Assuming Normal Distribution Mean 9.247 SD 7.215 95% UTL95% Coverage 27.18 95% KM UPL (t)22.21 90% KM Percentile (z)18.49 95% KM Percentile (z)21.11 99% KM Percentile (z)26.03 95% KM USL 27.1 DL/2 Substitution Background Statistics Assuming Normal Distribution Mean 9.088 SD 7.592 95% UTL95% Coverage 27.96 95% UPL (t)22.73 90% Percentile (z)18.82 95% Percentile (z)21.58 99% Percentile (z)26.75 95% USL 27.88 DL/2 is not a recommended method. DL/2 provided for comparisons and historical reasons Gamma GOF Tests on Detected Observations Only A-D Test Statistic 0.486 Anderson-Darling GOF Test 5% A-D Critical Value 0.744 Detected data appear Gamma Distributed at 5% Significance Level K-S Test Statistic 0.2 Kolmogrov-Smirnoff GOF 5% K-S Critical Value 0.239 Detected data appear Gamma Distributed at 5% Significance Level Detected data appear Gamma Distributed at 5% Significance Level Gamma Statistics on Detected Data Only k hat (MLE)1.972 k star (bias corrected MLE)1.568 Theta hat (MLE)5.742 Theta star (bias corrected MLE)7.221 nu hat (MLE)51.27 nu star (bias corrected)40.77 MLE Mean (bias corrected)11.32 MLE Sd (bias corrected)9.042 95% Percentile of Chisquare (2k)8.049 Gamma ROS Statistics using Imputed Non-Detects GROS may not be used when data set has > 50% NDs with many tied observations at multiple DLs GROS may not be used when kstar of detected data is small such as < 0.1 For such situations, GROS method tends to yield inflated values of UCLs and BTVs For gamma distributed detected data, BTVs and UCLs may be computed using gamma distribution on KM estimates Minimum 1.701 Mean 9.325 Maximum 22.2 Median 7.2 SD 7.381 CV 0.792 k hat (MLE)1.525 k star (bias corrected MLE)1.295 Theta hat (MLE)6.114 Theta star (bias corrected MLE)7.2 nu hat (MLE)51.85 nu star (bias corrected)44.04 MLE Mean (bias corrected)9.325 MLE Sd (bias corrected)8.194 95% Percentile of Chisquare (2k)7.092 90% Percentile 20.14 95% Percentile 25.53 99% Percentile 37.81 The following statistics are computed using Gamma ROS Statistics on Imputed Data Upper Limits using Wilson Hilferty (WH) and Hawkins Wixley (HW) Methods WH HW WH HW 95% Approx. Gamma UTL with 95% Coverage 39.51 43.13 95% Approx. Gamma UPL 27.05 28.27 95% Gamma USL 39.28 42.85 The following statistics are computed using gamma distribution and KM estimates Upper Limits using Wilson Hilferty (WH) and Hawkins Wixley (HW) Methods k hat (KM)1.643 nu hat (KM)55.85 WH HW WH HW 95% Approx. Gamma UTL with 95% Coverage 37.88 41.02 95% Approx. Gamma UPL 26.08 27.08 95% Gamma USL 37.67 40.76 Lognormal GOF Test on Detected Observations Only Shapiro Wilk Test Statistic 0.889 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.866 Detected Data appear Lognormal at 5% Significance Level Lilliefors Test Statistic 0.202 Lilliefors GOF Test 5% Lilliefors Critical Value 0.246 Detected Data appear Lognormal at 5% Significance Level Detected Data appear Lognormal at 5% Significance Level Background Lognormal ROS Statistics Assuming Lognormal Distribution Using Imputed Non-Detects Mean in Original Scale 9.336 Mean in Log Scale 1.888 SD in Original Scale 7.357 SD in Log Scale 0.894 95% UTL95% Coverage 60.91 95% BCA UTL95% Coverage 22.2 95% Bootstrap (%) UTL95% Coverage 22.2 95% UPL (t)32.89 90% Percentile (z)20.76 95% Percentile (z)28.72 99% Percentile (z)52.81 95% USL 60.3 Statistics using KM estimates on Logged Data and Assuming Lognormal Distribution KM Mean of Logged Data 1.858 95% KM UTL (Lognormal)95% Coverage 59.03 KM SD of Logged Data 0.893 95% KM UPL (Lognormal)31.89 95% KM Percentile Lognormal (z)27.85 95% KM USL (Lognormal)58.44 Background DL/2 Statistics Assuming Lognormal Distribution Mean in Original Scale 9.088 Mean in Log Scale 1.786 SD in Original Scale 7.592 SD in Log Scale 1.005 95% UTL95% Coverage 72.5 95% UPL (t)36.26 90% Percentile (z)21.61 95% Percentile (z)31.14 99% Percentile (z)61.75 95% USL 71.69 DL/2 is not a Recommended Method. DL/2 provided for comparisons and historical reasons. Nonparametric Distribution Free Background Statistics Data appear to follow a Discernible Distribution at 5% Significance Level Nonparametric Upper Limits for BTVs(no distinction made between detects and nondetects) Order of Statistic, r 17 95% UTL with95% Coverage 22.2 Approximate f 0.895 Confidence Coefficient (CC) achieved by UTL 0.582 95% UPL 22.2 95% USL 22.2 95% KM Chebyshev UPL 41.61 Note: The use of USL to estimate a BTV is recommended only when the data set represents a background data set free of outliers and consists of observations collected from clean unimpacted locations. The use of USL tends to provide a balance between false positives and false negatives provided the data represents a background data set and when many onsite observations need to be compared with the BTV. Sulfate General Statistics Total Number of Observations 17 Number of Missing Observations 0 Number of Distinct Observations 17 Number of Detects 0 Number of Non-Detects 17 Number of Distinct Detects 0 Number of Distinct Non-Detects 17 Minimum Detect N/A Minimum Non-Detect 284 Maximum Detect N/A Maximum Non-Detect 427 Variance Detected N/A Percent Non-Detects 100% Mean Detected N/A SD Detected N/A Mean of Detected Logged Data N/A SD of Detected Logged Data N/A Warning: All observations are Non-Detects (NDs), therefore all statistics and estimates should also be NDs! Specifically, sample mean, UCLs, UPLs, and other statistics are also NDs lying below the largest detection limit! The Project Team may decide to use alternative site specific values to estimate environmental parameters (e.g., EPC, BTV). The data set for variable Sulfate was not processed! Thallium General Statistics Total Number of Observations 17 Number of Missing Observations 0 Number of Distinct Observations 13 Number of Detects 0 Number of Non-Detects 17 Number of Distinct Detects 0 Number of Distinct Non-Detects 13 Minimum Detect N/A Minimum Non-Detect 5.7 Maximum Detect N/A Maximum Non-Detect 8.3 Variance Detected N/A Percent Non-Detects 100% Mean Detected N/A SD Detected N/A Mean of Detected Logged Data N/A SD of Detected Logged Data N/A Warning: All observations are Non-Detects (NDs), therefore all statistics and estimates should also be NDs! Specifically, sample mean, UCLs, UPLs, and other statistics are also NDs lying below the largest detection limit! The Project Team may decide to use alternative site specific values to estimate environmental parameters (e.g., EPC, BTV). The data set for variable Thallium was not processed! Total Organic Carbon General Statistics Total Number of Observations 17 Number of Missing Observations 0 Number of Distinct Observations 17 Number of Detects 1 Number of Non-Detects 16 Number of Distinct Detects 1 Number of Distinct Non-Detects 16 Minimum Detect 448 Minimum Non-Detect 683 Maximum Detect 448 Maximum Non-Detect 1020 Variance Detected N/A Percent Non-Detects 94.12% Mean Detected 448 SD Detected N/A Mean of Detected Logged Data 6.105 SD of Detected Logged Data N/A Warning: Only one distinct data value was detected! ProUCL (or any other software) should not be used on such a data set! It is suggested to use alternative site specific values determined by the Project Team to estimate environmental parameters (e.g., EPC, BTV). The data set for variable Total Organic Carbon was not processed! Vanadium General Statistics Total Number of Observations 17 Number of Distinct Observations 17 Minimum 29.2 First Quartile 40.8 Second Largest 75.8 Median 50.6 Maximum 89.2 Third Quartile 59.7 Mean 52.1 SD 16.31 Coefficient of Variation 0.313 Skewness 0.671 Mean of logged Data 3.908 SD of logged Data 0.311 Critical Values for Background Threshold Values (BTVs) Tolerance Factor K (For UTL)2.486 d2max (for USL)2.475 Normal GOF Test Shapiro Wilk Test Statistic 0.956 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.892 Data appear Normal at 5% Significance Level Lilliefors Test Statistic 0.13 Lilliefors GOF Test 5% Lilliefors Critical Value 0.215 Data appear Normal at 5% Significance Level Data appear Normal at 5% Significance Level Background Statistics Assuming Normal Distribution 95% UTL with 95% Coverage 92.66 90% Percentile (z)73.01 95% UPL (t)81.41 95% Percentile (z)78.93 95% USL 92.47 99% Percentile (z)90.05 Gamma GOF Test A-D Test Statistic 0.184 Anderson-Darling Gamma GOF Test 5% A-D Critical Value 0.739 Detected data appear Gamma Distributed at 5% Significance Level K-S Test Statistic 0.107 Kolmogrov-Smirnoff Gamma GOF Test 5% K-S Critical Value 0.209 Detected data appear Gamma Distributed at 5% Significance Level Detected data appear Gamma Distributed at 5% Significance Level Gamma Statistics k hat (MLE)11.2 k star (bias corrected MLE)9.26 Theta hat (MLE)4.653 Theta star (bias corrected MLE)5.627 nu hat (MLE)380.7 nu star (bias corrected)314.8 MLE Mean (bias corrected)52.1 MLE Sd (bias corrected)17.12 Background Statistics Assuming Gamma Distribution 95% Wilson Hilferty (WH) Approx. Gamma UPL 84.42 90% Percentile 74.89 95% Hawkins Wixley (HW) Approx. Gamma UPL 84.98 95% Percentile 83.08 95% WH Approx. Gamma UTL with 95% Coverage 100.6 99% Percentile 99.95 95% HW Approx. Gamma UTL with 95% Coverage 102.1 95% WH USL 100.4 95% HW USL 101.8 Lognormal GOF Test Shapiro Wilk Test Statistic 0.98 Shapiro Wilk Lognormal GOF Test 5% Shapiro Wilk Critical Value 0.892 Data appear Lognormal at 5% Significance Level Lilliefors Test Statistic 0.11 Lilliefors Lognormal GOF Test 5% Lilliefors Critical Value 0.215 Data appear Lognormal at 5% Significance Level Data appear Lognormal at 5% Significance Level Background Statistics assuming Lognormal Distribution 95% UTL with 95% Coverage 107.8 90% Percentile (z)74.15 95% UPL (t)87.02 95% Percentile (z)83.01 95% USL 107.4 99% Percentile (z)102.6 Nonparametric Distribution Free Background Statistics Data appear Normal at 5% Significance Level Nonparametric Upper Limits for Background Threshold Values Order of Statistic, r 17 95% UTL with 95% Coverage 89.2 Approximate f 0.895 Confidence Coefficient (CC) achieved by UTL 0.582 95% Percentile Bootstrap UTL with 95% Coverage 89.2 95% BCA Bootstrap UTL with 95% Coverage 89.2 95% UPL 89.2 90% Percentile 72.44 90% Chebyshev UPL 102.5 95% Percentile 78.48 95% Chebyshev UPL 125.3 99% Percentile 87.06 95% USL 89.2 Note: The use of USL to estimate a BTV is recommended only when the data set represents a background data set free of outliers and consists of observations collected from clean unimpacted locations. The use of USL tends to provide a balance between false positives and false negatives provided the data represents a background data set and when many onsite observations need to be compared with the BTV. Zinc General Statistics Total Number of Observations 17 Number of Missing Observations 0 Number of Distinct Observations 17 Number of Detects 16 Number of Non-Detects 1 Number of Distinct Detects 16 Number of Distinct Non-Detects 1 Minimum Detect 6.6 Minimum Non-Detect 6.5 Maximum Detect 59.8 Maximum Non-Detect 6.5 Variance Detected 186.2 Percent Non-Detects 5.88% Mean Detected 36.76 SD Detected 13.65 Mean of Detected Logged Data 3.498 SD of Detected Logged Data 0.555 Critical Values for Background Threshold Values (BTVs) Tolerance Factor K (For UTL)2.486 d2max (for USL)2.475 Normal GOF Test on Detects Only Shapiro Wilk Test Statistic 0.936 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.887 Detected Data appear Normal at 5% Significance Level Lilliefors Test Statistic 0.185 Lilliefors GOF Test 5% Lilliefors Critical Value 0.222 Detected Data appear Normal at 5% Significance Level Detected Data appear Normal at 5% Significance Level Kaplan Meier (KM) Background Statistics Assuming Normal Distribution Mean 34.98 SD 14.66 95% UTL95% Coverage 71.43 95% KM UPL (t)61.32 90% KM Percentile (z)53.77 95% KM Percentile (z)59.09 99% KM Percentile (z)69.09 95% KM USL 71.26 DL/2 Substitution Background Statistics Assuming Normal Distribution Mean 34.79 SD 15.51 95% UTL95% Coverage 73.35 95% UPL (t)62.65 90% Percentile (z)54.66 95% Percentile (z)60.3 99% Percentile (z)70.87 95% USL 73.17 DL/2 is not a recommended method. DL/2 provided for comparisons and historical reasons Gamma GOF Tests on Detected Observations Only A-D Test Statistic 1.091 Anderson-Darling GOF Test 5% A-D Critical Value 0.741 Data Not Gamma Distributed at 5% Significance Level K-S Test Statistic 0.265 Kolmogrov-Smirnoff GOF 5% K-S Critical Value 0.216 Data Not Gamma Distributed at 5% Significance Level Data Not Gamma Distributed at 5% Significance Level Gamma Statistics on Detected Data Only k hat (MLE)4.869 k star (bias corrected MLE)3.998 Theta hat (MLE)7.549 Theta star (bias corrected MLE)9.194 nu hat (MLE)155.8 nu star (bias corrected)127.9 MLE Mean (bias corrected)36.76 MLE Sd (bias corrected)18.38 95% Percentile of Chisquare (2k)15.5 Gamma ROS Statistics using Imputed Non-Detects GROS may not be used when data set has > 50% NDs with many tied observations at multiple DLs GROS may not be used when kstar of detected data is small such as < 0.1 For such situations, GROS method tends to yield inflated values of UCLs and BTVs For gamma distributed detected data, BTVs and UCLs may be computed using gamma distribution on KM estimates Minimum 6.6 Mean 35.38 Maximum 59.8 Median 34.7 SD 14.37 CV 0.406 k hat (MLE)4.287 k star (bias corrected MLE)3.57 Theta hat (MLE)8.254 Theta star (bias corrected MLE)9.913 nu hat (MLE)145.8 nu star (bias corrected)121.4 MLE Mean (bias corrected)35.38 MLE Sd (bias corrected)18.73 95% Percentile of Chisquare (2k)14.27 90% Percentile 60.49 95% Percentile 70.72 99% Percentile 92.7 The following statistics are computed using Gamma ROS Statistics on Imputed Data Upper Limits using Wilson Hilferty (WH) and Hawkins Wixley (HW) Methods WH HW WH HW 95% Approx. Gamma UTL with 95% Coverage 94.28 100.5 95% Approx. Gamma UPL 72.91 75.77 95% Gamma USL 93.9 100 The following statistics are computed using gamma distribution and KM estimates Upper Limits using Wilson Hilferty (WH) and Hawkins Wixley (HW) Methods k hat (KM)5.69 nu hat (KM)193.5 WH HW WH HW 95% Approx. Gamma UTL with 95% Coverage 98.91 106.7 95% Approx. Gamma UPL 75.28 78.85 95% Gamma USL 98.5 106.2 Lognormal GOF Test on Detected Observations Only Shapiro Wilk Test Statistic 0.759 Shapiro Wilk GOF Test 5% Shapiro Wilk Critical Value 0.887 Data Not Lognormal at 5% Significance Level Lilliefors Test Statistic 0.306 Lilliefors GOF Test 5% Lilliefors Critical Value 0.222 Data Not Lognormal at 5% Significance Level Data Not Lognormal at 5% Significance Level Background Lognormal ROS Statistics Assuming Lognormal Distribution Using Imputed Non-Detects Mean in Original Scale 35.21 Mean in Log Scale 3.431 SD in Original Scale 14.67 SD in Log Scale 0.605 95% UTL95% Coverage 139.1 95% BCA UTL95% Coverage 59.8 95% Bootstrap (%) UTL95% Coverage 59.8 95% UPL (t)91.65 90% Percentile (z)67.11 95% Percentile (z)83.61 99% Percentile (z)126.3 95% USL 138.2 Background DL/2 Statistics Assuming Lognormal Distribution Mean in Original Scale 34.79 Mean in Log Scale 3.362 SD in Original Scale 15.51 SD in Log Scale 0.778 95% UTL95% Coverage 199.5 95% UPL (t)116.7 90% Percentile (z)78.17 95% Percentile (z)103.7 99% Percentile (z)176.2 95% USL 197.8 DL/2 is not a Recommended Method. DL/2 provided for comparisons and historical reasons. Nonparametric Distribution Free Background Statistics Data appear to follow a Discernible Distribution at 5% Significance Level Nonparametric Upper Limits for BTVs(no distinction made between detects and nondetects) Order of Statistic, r 17 95% UTL with95% Coverage 59.8 Approximate f 0.895 Confidence Coefficient (CC) achieved by UTL 0.582 95% UPL 59.8 95% USL 59.8 95% KM Chebyshev UPL 100.7 Note: The use of USL to estimate a BTV is recommended only when the data set represents a background data set free of outliers and consists of observations collected from clean unimpacted locations. The use of USL tends to provide a balance between false positives and false negatives provided the data represents a background data set and when many onsite observations need to be compared with the BTV.