ML030080495

Phase I of the Large Area Open Land Survey for FSS

ML030080495
Person / Time
Site:	Saxton File:GPU Nuclear icon.png
Issue date:	09/30/2002
From:	Shonka Research Associates
To:	GPU Nuclear Corp, Office of Nuclear Reactor Regulation
References
Download: ML030080495 (98)

Text

Saxton Survey Project Final Report Phase I of the Large Area Open Land Survey for FSS Date of Final Report: September 2002 Conducted at Saxton Nuclear Experimental Corporation (SNEC) Facility For GPU Nuclear Corporation In November - December 2001 By Shonka Research Associates, Inc.

4939 Lower Roswell Road, Suite 106 Marietta, GA 30068 (770) 509-7606

Saxton Survey Project Final Report Table of Contents Discussion of Methods and Results

1. Introduction ............................................................................................................. 1

2. Methodology ........................................................................................................... 2 2.1 Establishment of Survey Areas ......................................................................... 2 2.1.1 Placement of Survey Area Blocks............................................................... 4 2.2 Survey Methods................................................................................................. 5 2.2.1 SMCM Scan Surveys .................................................................................. 5 2.2.2 Static In situ Measurements ........................................................................ 6 2.2.3 Calibration of the Spectrometers................................................................. 7 2.2.4 Data Handling and Analysis Techniques .................................................... 7

3. Survey Results......................................................................................................... 8 3.1 Background Data............................................................................................... 8 3.2 Scan Surveys ..................................................................................................... 9 3.2.1 SMCM Results Review............................................................................... 9 3.2.2 Data Quality Assessment .......................................................................... 13 3.2.3 SMCM Detection Limits........................................................................... 14 3.2.4 SMCM Quality Control............................................................................. 15 3.2.5 Performance Based Test Survey ............................................................... 15

4. Data ................................................................................................................... 15 4.1 SMCM Results ................................................................................................ 15 4.2 SAB Reports.................................................................................................... 15 4.2.1 A1001 ........................................................................................................ 16 4.2.2 B1001 ........................................................................................................ 19 4.2.3 B1002 ........................................................................................................ 22 4.2.4 B1003 ........................................................................................................ 25 4.2.5 B1004 ........................................................................................................ 28 4.2.6 B2001 ........................................................................................................ 31 4.2.7 B3001 ........................................................................................................ 34 4.2.8 B3002 ........................................................................................................ 37 4.2.9 B3003 ........................................................................................................ 40 4.2.10 C1001 ........................................................................................................ 43

5. References ............................................................................................................. 45 i

Saxton Survey Project Final Report Appendices:

Appendix A: Detector Quality Control Appendix B: Determination of Stripping Coefficients Appendix C: Determination of the Platform Factor Appendix D: Calibration Factor Methodology Appendix E: SMCM Scan Survey Characteristics Appendix F: Data Handling and Analysis Methods Appendix G: Performance-Based Test Survey Appendix H: Source Calibration Certificates Appendix I: Detector Quality Objectives ii

Saxton Survey Project Final Report

1. Introduction This document is an interim report for Phase I of the Radiological Site Survey of the Saxton Nuclear Experimental Corporation (SNEC) Facility located in Saxon, PA. This report may be embedded in an overall Final Status Survey report when Phase II is completed.

The Saxton Nuclear Experimental Corporation (SNEC) Facility was a deactivated, pressurized water reactor (PWR) that was originally licensed to operate at a power level of 23.5 megawatts (thermal). The facility, located near the town of Saxton, Pennsylvania, was built from 1960 to 1962. The SNEC Facility was operated from 1962 to 1972 primarily as a research and training reactor. The SNEC is owned by GPU Nuclear Corporation and the Saxton Nuclear Experimental Corporation.

The SNEC Facility is in the process of decontamination and decommissioning. General Public Utilities Service, Inc. (GPU Service) contracted with Shonka Research Associates, Inc. (SRA) to perform the Radiological Site Survey under Purchase Order 55001244, dated October 24, 2001.

SRA performed a radiological survey in late November and early December 2001 at the Saxton site. The survey was performed using sodium iodide NaI(Tl) scintillation spectrometers.

Approximately 7 hectares (15 acres) of area was surveyed with 100% coverage, with nearly 10,000 spectra acquired. The majority of the spectra (9500, comprising 3.8 hectares) were taken using the spectrometers in a scanning mode. 320 static in situ measurements were also made in areas that were heavily forested or otherwise difficult to scan (comprising 3.2 hectares).

In general nearly all measurements showed no evidence of contamination due to plant activities.

The average concentration site-wide of 137Cs was 0.3 +/- 0.15 pCi/g (1 standard deviation), a value similar to that measured in GPUs Background Soil Study and attributed to typical fallout levels from atmospheric testing of nuclear weapons. Nearly all of the data were within four standard deviations of the mean. This corresponds to less than 0.6 pCi/g of added activity in nearly 10,000 measurements. There were, however, three static measurements and one scanning measurement that appeared to be outliers in this data set.

This survey constituted the first phase of a two-phase effort to perform a Final Status Survey (FSS) for SNEC. During the first phase, an outer area around the site was surveyed along with an adjoining one-hectare (2.5 acre) area that was intended to be representative of background.

The second phase is anticipated to start later in CY 2002, and provide the surface contamination and the field gamma ray spectrometry measurements needed to complete the FSS. GPU will be responsible for sampling and analyzing soil samples that will be taken for the FSS.

The data from the spectrometers was analyzed using the industry standard methods for field gamma ray spectrometry detailed in the Procedures Manual of the Environmental Measurements Laboratory, HASL-300.

SHONKA RESEARCH ASSOCIATES, INC. 1

Saxton Survey Project Final Report

2. Methodology 2.1 Establishment of Survey Areas Survey Areas for Phase I were chosen by GPU staff. The intent was to choose areas of the site that were unlikely to be impacted by ongoing site D&D activities. The areas comprised 15 acres of Class 2 and bounding Class 3 areas surrounding the core of the Saxton Site, along with a background reference area. The Phase I survey area is shown superimposed on a site drawing in Figure 2-1. The survey areas are marked with a heavy black line. There are 4 major areas.

Figure 2 Defined Survey Areas.

The four areas were subdivided into ten Survey Area Blocks (SABs) that were similar in terrain and ground for reporting purposes. The SABs were designated as A1001 for the background reference area, located to the east of the site; B1001 to B1004 located to the south; B2001 located to the west; B3001 to B3003 located to the northeast; and a C1001 area that was a single strip of scanned data taken along the former site access road. The SABs are shown in Figure 2-2.

The C1001 road survey is contained in B1003 and B1004 and is not called out in Figure 2-2.

SHONKA RESEARCH ASSOCIATES, INC. 2

Saxton Survey Project Final Report Figure 2 Defined Survey Area Blocks (SABs)

The areas were surveyed on a scanning basis with the Subsurface Multispectral Contamination Monitor (SMCM). This system was used in a scanning mode and in a static mode. These modes are described in Section 2.2.

Figure 2-3 shows the SABs with the area scanned and subjected to static measurements shown.

About one half of the survey was performed with each method.

SHONKA RESEARCH ASSOCIATES, INC. 3

Saxton Survey Project Final Report Figure 2 Scanned and Static Coverage 2.1.1 Placement of Survey Area Blocks Table 2-1 shows the location of each SAB unit. The southwest (lower left) corner is noted in site coordinates along with the square meters of area surveyed.

SHONKA RESEARCH ASSOCIATES, INC. 4

Saxton Survey Project Final Report Table 2 SAB Location Table Area Area Static Scanned SAB X-SW Y-SW Square UTM X UTM Y Description Square Meters Meters

Background

A1001 111 AG 9220 1900 734893 4456511 Reference Area B1001 159 AC 1392 8200 734225 4456485 Spray Pond Area Eastside of Spray B1002 148 AC 6224 3200 734382 4456485 Pond Area Old Access Road B1003 135 AB 6960 1400 734553 4456474

- West Area Old Access Road B1004 122 AA 7392 0 734723 4456474

- East Area Substation -

B2001 145 BB 3440 2500 73421 4456617 Northwest Juniata River -

B3001 123 BC 840 6100 734710 4456622 West Area Juniata River -

B3002 115 BE 5412 2800 734815 4456633 East Area Northeast Dump B3003 104 BK 2076 5500 734946 4456665 Site C1001 135 AB 1680 0 734553 4456479 Old Access Road

• Coordinates represent points and not blocks.

2.2 Survey Methods 2.2.1 SMCM Scan Surveys Scan surveys were conducted using the Subsurface Multi-Spectral Contamination Monitor (SMCM) system developed by SRA. The SMCM, as configured for the Saxton survey, consisted of four sodium iodide (NaI(Tl)) gamma scintillators, of nominal size 5 inch by 2 inch right circular cylinder, packaged in heated enclosures for environmental stabilization. The scintillators were placed two meters apart and with the front face of the detector at one meter above the ground in a vertical orientation. The detector enclosures were also equipped with a thin cone shaped lead shield for Compton background suppression. The detectors and electronics were mounted to a Ford F250 Crew Cab 4-wheel drive vehicle. Figure 2-4 shows the SMCM as deployed at Saxton.

SHONKA RESEARCH ASSOCIATES, INC. 5

Saxton Survey Project Final Report Figure 2 SRAs SMCM platform as deployed at Saxton.

NaI(Tl) spectra were collected during each 10 second interval while the SMCM traveled 2 meters. This data was then assigned to the center of the 2-meter by 2-meter square under each detector. Each 100 square meter area has a total of 25 such measurements, with 250 seconds of counting time. The MCAs were operated using a field industrial personal computer (PC). The spectra were collected into 512 channels corresponding to a 0.3 to 3 MeV energy range. The energy range includes important photons from primordial nuclides series including potassium (K), uranium (U), and thorium (Th) (KUT) which are present in parts per million (ppm) levels in most soils.

The system is operated as a spectrometer based in situ measurement system for both scanning and static modes. In static mode, a single five-minute (300 second) measurement was made on a ten-meter rectangular grid, and was assigned to 100 square meters. Comparable detection limits are attained for widely disperse contamination for either mode. The scan mode, when convenient, provides greater image resolution. The static mode is used in heavily forested areas where a large machine cannot be used.

2.2.2 Static In situ Measurements Fixed point static in situ measurements were taken to aid in the overall radiological characterization. A total of 320 in situ measurements were taken. SRAs static in situ system consisted of a NaI(Tl) detector and electronics enclosed in a free-in-air tripod geometry at a fixed height of one meter above the ground, measured from the detector face. The detectors were the same ones used in the SMCM system. Figure 2-5 provides a photograph of the system as deployed at Saxton.

SHONKA RESEARCH ASSOCIATES, INC. 6

Saxton Survey Project Final Report Each static measurement was five minutes in length.

2.2.3 Calibration of the Spectrometers The calibration method used was derived from procedures used for many years by the Department of Energys Environmental Measurements Laboratory, and is published in the long standing EML Manual for Field Procedures (EML, 1997).

Figure 2 Photograph of SRAs static in situ system in use at Saxton.

The energy and efficiency calibration of the scintillation NaI(Tl) spectrometers was accomplished by placing NIST traceable point sources (both 137Cs and 60Co) at a one meter distance from the front face of the 5-inch diameter by 2-inch long (5X2) crystal. The calibration certificates for the sources are provided in Appendix H. In this calibration geometry, the flux of photons is nearly parallel to the crystal axis. The calibration results are expressed in counts per second in the photopeak per disintegration per second from the source. For the in situ method, there are two other correction factors that are applied to the data: (1) the angular response of the detector, and (2) the flux arriving at the detector for the assumed source energy and geometry.

Appendix D has a detailed description of the calibration process.

For in situ measurements using spectrometers, the response of the detectors is calculated for a desired geometry. This calculation has two factors: one accounting for the anisotropic (angular) response of the detector, and the other accounting for the distribution of the source in the ground.

These factors were calculated using industry standard shielding codes.

The sensitivity of the method to detector height and scan speed is further discussed in Appendix E.

2.2.4 Data Handling and Analysis Techniques The data from the scanning measurements consisted of 10-second spectra from the nominal 5X2 NaI(Tl) detectors. The energy range from 0.3 to 3 MeV was spanned with 512 channels. The data was converted to count rate, and adjusted by a factor that included the detector efficiency (relative to detector #1) and the platform shielding. Following correction of the raw data by the platform-shielding factor, the spectra were analyzed and separated into components using the method of noise-adjusted single value decomposition (NASVD). (Hovgaard, 1997) This is a statistical process that computes the spectral shapes that are embedded in data. For a more detailed presentation of these data handling and analysis techniques, see Appendix F of this report.

SHONKA RESEARCH ASSOCIATES, INC. 7

Saxton Survey Project Final Report

3. Survey Results 3.1 Background Data This section describes the characteristics of background radiation levels at the Saxton Site.

The GPU Background Soil Study program was designed to obtain regional information about naturally occurring radioactive material and man-made contamination from atmospheric testing of nuclear weapons (GPU, 1999). In that program, locations approximately 10 miles from the site were selected in each of the 16 directional sectors. GPU staff conducted sampling and survey work during the week of July 13, 1999. Twenty soil samples were taken and split between SNEC and the GPUN Environmental Radioactivity Laboratory.

The average specific activity (in pCi/g) measured in the background soil samples for each of the typical nuclides is listed in Table 3-1. In addition, Table 3-1 lists the uncertainty (2 sigma) and the concentration in soil in ppm. The concentration was calculated using the specific activity of the elements present at equilibrium. The 226Ra value is indicative of the uranium concentration.

Table 3-1. Background Soil Study Results Nuclide Specific Act. Uncertainty Concentration (pCi/g, dry wt.) (2-sigma) ppm K-40 14 15.5 16,480 Cs-137 0.28 0.39 N/A Ra-226 1.8 1.1 2.7 Th-232 0.9 0.5 4.1 The lab analysis method measures the concentrations in soil that has been dried to remove all moisture. Typical soil moisture content (by weight) is 15% +/- 5%. Thus, in situ measurements at similar locations would be expected to average 85% of the values listed in the table due to the presence of water in the measured soil, with the exact value depending on the soil moisture at the location.

The 0.28 pCi/g average observed for 137Cs in the Background Soil Study project is consistent with reported fallout levels for Pennsylvania from atmospheric testing of nuclear weapons in the past. In a 5X2 NaI(Tl) detector used for the survey, this fallout level would produce about 2 cps in a window used for the characteristic gamma. Substantially higher values might indicate added activity. Substantially lower values can occur if the area has been disturbed, removing or covering the 137Cs bearing soils.

The regional variability of fallout levels of 137Cs (0.39 pCi/g, 2 sigma) could be interpreted in terms of a minimum detectable concentration (MDC) that a sampling program could attain, when sampling error is considered. That concentration is 0.9 pCi/g for a single sample, or 0.2 pCi/g for an average of 20 samples from an appropriate survey area.

SHONKA RESEARCH ASSOCIATES, INC. 8

Saxton Survey Project Final Report The reported naturally occurring radioactive materials (K,U, & T) listed in the Background Soil Study results are not completely applicable to the SNEC Site. The Saxton Facility was the site of a coal-fired plant that operated from 1923 to 1975. The treatment and disposal of ash (both fly and bottom ash) from that plant likely included local storage or disposal. While the concentration of the primordial radionuclides (K, U, & T) in coal varies, they are grossly similar to soil concentrations. Following use (combustion), the fifteen percent (weight) of impurities in coal, which include K, U, and T, become concentrated by factors of ten in the ash (Tadmore, 1986).

Qualitative observations of the surface soils made by the SRA survey team were made. The field team thought several Survey Area Blocks may be impacted by deposits of fly ash. Their observations included loose, unconsolidated soils, particle size similar to sawdust, black soil color in areas that were heavily impacted (where no vegetation could grow), and lack of organic matter or clay in the material. Surprisingly, the KUT concentrations were not elevated by a factor of 10 over other areas not impacted by fly ash. What is most striking in the data is the variability of the potassium across the site.

In order to assess the presence of added 137Cs at the Saxton Site, the contributions to the 137Cs window from the fallout levels of 137Cs, potassium, uranium, and thorium were assessed and subtracted from the gross counts in a 137Cs window. This is further described in Appendix B, Determination of Stripping Coefficients.

3.2 Scan Surveys 3.2.1 SMCM Results Review Nearly 80,000 gamma ray spectra were collected throughout the course of Phase I of the Large Area Open Land Survey for FSS. Each SAB was reviewed for the presence of 134Cs, 137Cs, and 60 Co using the methods described in Section 2.2 after electronic submittal from the field. No evidence of 134Cs and 60Co nuclides was found. B3000 series SABs may have 137Cs as discussed later. A complete listing of results from the scan survey can be found in Section 4.2 SAB Reports.

There is an extremely low level of fallout 137Cs present in a largely uniform fashion across the site (see Figure 3.1). The 137Cs is visible as a slight broadening of the 609 keV Bi-214 peak that comes from the decay of uranium in the ground and radon in the air. Because it is similar across the site, and below the detection limit of the method, it was not seen as a separate component.

SHONKA RESEARCH ASSOCIATES, INC. 9

Saxton Survey Project Final Report Figure 3-1 shows an image of 137Cs concentrations figure for the site.

Figure 3 Overlay of 137Cs results for all SABs for Saxton SHONKA RESEARCH ASSOCIATES, INC. 10

Saxton Survey Project Final Report The 40K levels across the Saxton site vary considerably. Results for all completed SABs as images overlaid on the same photograph are shown in Figure 3-2.

Figure 3 Overlay of 40K Results for all SABs for Saxton SHONKA RESEARCH ASSOCIATES, INC. 11

Saxton Survey Project Final Report The uranium and thorium levels at the site are shown in Figures 3-3 and 3-4 respectively. Care should be used interpreting the uranium data, as the contribution from radon has not been removed.

Figure 3 Overlay of Uranium data for all SABs for Saxton SHONKA RESEARCH ASSOCIATES, INC. 12

Saxton Survey Project Final Report Figure 3 Overlay of Thorium data for all SABs for Saxton 3.2.2 Data Quality Assessment The Phase I survey was performed as a characterization survey with the intent that the results be incorporated into a Final Status Survey if the results showed no action required. The Phase I area was a mixture of Class 2 and Class 3 areas. Largely, the Class 3 areas were a buffer area around the site, and the Class 2 areas were placed to completely encompass any potential for migration of site created radioactive material along some flow pathway from the inner core Class 1 center of the site that contained the reactor building itself (see Figure 2-2). The entire area was treated as Class 3. The 10,000 square meter background reference area (A1001) was also surveyed and the resulting data treated in the same manner as any survey area block. The investigation criteria are defined in Table 5-7 of the SNEC LTP (GPU, 2000).

The Data Quality Objectives (see Appendix I) are derived from the SNEC LTP, and assert that any Class 3 area would flag (for further investigation) any discrete measurement (static in situ count or soil sample), which was found to be greater than 10% of DCGLw. In addition, any scan measurement greater than DCGLw or the Scan MDC would also be flagged. The DCGLw is 8.5 SHONKA RESEARCH ASSOCIATES, INC. 13

Saxton Survey Project Final Report pCi/g 137Cs. The Scan MDC contractual requirement is 1.5 pCi/g. Thus, any static reading greater than 0.85 pCi/g 137Cs or any scan greater than 1.5 pCi/g would require further investigation.

The SMCM is a new measurement not described in MARSSIM. It is a scanning spectrometer.

The data from each ten-second measurement is significantly improved as compared to a scan performed with a gross gamma detector or a detector with a single channel analyzer. As such, the technology can be thought of as a fusion between static in situ spectrometer systems and gross gamma scans. The SMCM technology provides a scanning in situ measurement. The detection limit during the scan is comparable to the static in situ measurement results.

To fulfill the Data Quality Objectives, no difference between static and scanning in situ measurements has been made for reporting purposes. That is, rather than assess the data using a scan MDC of 1.5 pCi/g, all data was assessed with the 0.85 pCi/g investigation level.

There were two static in situ measurements (out of 320) made that exceeded 0.85 pCi/g, and one static measurement that was 0.83. All three measurements were found in the 3000 area of the site, in SAB 3001, 3002, and 3003. In addition, there was a single 10-second scan measurement indicating a value greater than 1.5 pCi/g in SAB B3003. There is a 20-meter line of scan measurements in B3003 located near the 1.5 pCi/g data point that are at the 0.85 pCi/g Investigation Level. Without these exceptions, all other measurements exhibit an average of about 0 +/- 0.13 pCi/g. Nearly all other measurements are less than 0.6 pCi/g net activity over fallout with few exceptions.

The B3000 area is the area to the northwest corner of the site, and appears to be a former landfill that has not been actively used for some time. Over the years, trees have covered the area, typically reaching up to 10 inches in diameter, with isolated trees that are larger. As can be seen in the site image for 137Cs (see Figure 3-1), there appears to be more 137Cs present in that area than in the rest of the site. In most cases, the material is below the investigation level from the SNEC LTP. This assumes, however, that the material is present as uniform contamination The scanning anomaly, consisting of a single measurement above 1.5 pCi/g, could be a statistical anomaly, or could be due to the presence of a surface or buried radioactive source of small diameter. Additional measurements, including soil sampling, could help resolve the source of the added counts from 137Cs.

The data was reported by subtracting out the low level of 137Cs background present from the fallout from atmospheric testing of nuclear weapons. The sites Background Soil Study Program was used (0.28 +/- 0.4 pCi/g 137Cs) as an estimate of typical fallout levels in this area (GPU, 1999). Thus, these areas indicate observations above fallout levels.

3.2.3 SMCM Detection Limits Minimum detectable concentrations (MDCs, in terms of pCi/g for uniformly-distributed material) were computed a posteriori for each SAB. The MDC was determined by taking the SHONKA RESEARCH ASSOCIATES, INC. 14

Saxton Survey Project Final Report standard deviation of the measurements in each SAB and multiplying it by 4.65 to arrive at a 95%-confidence level. The MDC data for the 10 SABs are provided in Section 4.2 of this report.

3.2.4 SMCM Quality Control A discussion of the SMCM Quality Control (QC) procedures and results may be found in Appendix A, Detector Quality Control.

3.2.5 Performance Based Test Survey A sealed source was placed in B1004 at depths of 0, 15, and 30 centimeters. This qualitative test is discussed in Appendix G.

4. Data 4.1 SMCM Results Section 4.2 contains the reports for each of the Survey Area Blocks (SABs). Each SAB report has two data tables and four images. The Header Block has the Survey Location Code, date of survey and other basic information. A second data block shows the area surveyed in square meters, the mean 137Cs value seen for the area and a detection limit, which is calculated as 4.65 times the standard deviation of the measurements. The data is presented for the scanning survey and the static measurements separately.

Four figures are presented. The first is a figure that shows where the SAB is in relation to the site. The second shows an image file of the 137Cs data. The third shows the area covered by static measurements and scanning measurements on the SAB. Finally, a cumulative frequency distribution (CFD) is shown for all static and scanning data taken together. The data is shown in pCi/g of 137Cs, assuming an exponential distribution in soil.

4.2 SAB Reports The MDC is calculated for the actual uncertainty of all measurements, both scan and static in situ data, for an SAB. When the MDC is shown as N/A (for Cs-137 and SABs B3001, B3002, and B3003), the data is not shown because the spatial variability of the data from the SABs is large, resulting in large uncertainties in the average for the SAB.

SHONKA RESEARCH ASSOCIATES, INC. 15

Saxton Survey Project Final Report 4.2.1 A1001 Survey Report Survey Location Code A1001C Southwest Corner AG-111 Survey Equipment SMCM Survey Date December 10th Surveyor(s) D. DeBord Criteria DCGL-W 8.5 pCi/g Measurement Area Cs-137 Mean Cs-137 MDC Type (#) Surveyed Activity in in pCi/g m^2 pCi/g Scanning 9220 0.053 0.664 Static (19) 1900 0.024 0.533 Figure 1: The Location of the SAB relative to the site SHONKA RESEARCH ASSOCIATES, INC. 16

Saxton Survey Project Final Report Figure 2: Color map of Cs-137 in the SAB Figure 3: The Location of Scanned and Static areas SHONKA RESEARCH ASSOCIATES, INC. 17

Saxton Survey Project Final Report Figure 4: Cs-137 CFD for Entire SAB SHONKA RESEARCH ASSOCIATES, INC. 18

Saxton Survey Project Final Report 4.2.2 B1001 Survey Report Survey Location Code B1001A Southwest Corner AC-159 Survey Equipment SMCM Survey Date December 4th Surveyor(s) D. DeBord Criteria DCGL-W 8.5 pCi/g Measurement Area Cs-137 Mean Cs-137 MDC Type (#) Surveyed Activity in in pCi/g m^2 pCi/g Scanning 1392 -0.059 0.730 Static (82) 8200 -0.010 0.619 Figure 1: The Location of the SAB relative to the site SHONKA RESEARCH ASSOCIATES, INC. 19

Saxton Survey Project Final Report Figure 2: Color map of Cs-137 in the SAB Figure 3: The Location of Scanned and Static areas SHONKA RESEARCH ASSOCIATES, INC. 20

Saxton Survey Project Final Report Figure 4: Cs-137 CFD for Entire SAB SHONKA RESEARCH ASSOCIATES, INC. 21

Saxton Survey Project Final Report 4.2.3 B1002 Survey Report Survey Location Code B1002A Southwest Corner AC-148 Survey Equipment SMCM Survey Date December 2nd Surveyor(s) D. DeBord Criteria DCGL-W 8.5 pCi/g Measurement Area Cs-137 Mean Cs-137 MDC Type (#) Surveyed Activity in in pCi/g m^2 pCi/g Scanning 6224 0.033 0.807 Static (32) 3200 -0.142 0.320 Figure 1: The Location of the SAB relative to the site SHONKA RESEARCH ASSOCIATES, INC. 22

Saxton Survey Project Final Report Figure 2: Color map of Cs-137 in the SAB Figure 3: The Location of Scanned and Static areas SHONKA RESEARCH ASSOCIATES, INC. 23

Saxton Survey Project Final Report Figure 4: Cs-137 CFD for Entire SAB SHONKA RESEARCH ASSOCIATES, INC. 24

Saxton Survey Project Final Report 4.2.4 B1003 Survey Report Survey Location Code B1003A Southwest Corner AB-135 Survey Equipment SMCM Survey Date December 1st Surveyor(s) D. DeBord Criteria DCGL-W 8.5 pCi/g Measurement Area Cs-137 Mean Cs-137 MDC Type (#) Surveyed Activity in in pCi/g m^2 pCi/g Scanning 6960 0.032 0.752 Static (14) 1400 0.029 0.415 Figure 1: The location of the SAB relative to the site SHONKA RESEARCH ASSOCIATES, INC. 25

Saxton Survey Project Final Report Figure 2: Color map of Cs-137 in the SAB Figure 3: The Location of Scanned and Static Areas SHONKA RESEARCH ASSOCIATES, INC. 26

Saxton Survey Project Final Report Figure 4: Cs-137 CFD for Entire SAB SHONKA RESEARCH ASSOCIATES, INC. 27

Saxton Survey Project Final Report 4.2.5 B1004 Survey Report Survey Location Code B1004A Southwest Corner AA-122 Survey Equipment SMCM Survey Date December 3rd Surveyor(s) D. DeBord Criteria DCGL-W 8.5 pCi/g Measurement Area Cs-137 Mean Cs-137 MDC Type (#) Surveyed Activity in in pCi/g m^2 pCi/g Scanning 7392 0.012 0.526 Static (0) 0 0 0 Figure 1: The Location of the SAB relative to the site SHONKA RESEARCH ASSOCIATES, INC. 28

Saxton Survey Project Final Report Figure 2: Color map of Cs-137 in the SAB Figure 3: The Location of Scanned and Static areas SHONKA RESEARCH ASSOCIATES, INC. 29

Saxton Survey Project Final Report Figure 4: CFD for Scanned Areas SHONKA RESEARCH ASSOCIATES, INC. 30

Saxton Survey Project Final Report 4.2.6 B2001 Survey Report Survey Location Code B2001A Southwest Corner BB-145 Survey Equipment SMCM Survey Date December 2nd Surveyor(s) D. DeBord Criteria DCGL-W 8.5 pCi/g Measurement Area Cs-137 Mean Cs-137 MDC Type (#) Surveyed Activity in in pCi/g m^2 pCi/g Scanning 3440 0.290 0.730 Static (25) 2500 0.087 0.816 Figure 1: The Location of the SAB relative to the site SHONKA RESEARCH ASSOCIATES, INC. 31

Saxton Survey Project Final Report Figure 2: Color map of Cs-137 in the SAB Figure 3: The Location of Scanned and Static areas SHONKA RESEARCH ASSOCIATES, INC. 32

Saxton Survey Project Final Report Figure 4: Cs-137 CFD for Entire SAB SHONKA RESEARCH ASSOCIATES, INC. 33

Saxton Survey Project Final Report 4.2.7 B3001 Survey Report Survey Location Code B3001A Southwest Corner BC-123 Survey Equipment SMCM Survey Date December 2nd Surveyor(s) D. DeBord Criteria DCGL-W 8.5 pCi/g Measurement Area Cs-137 Mean Cs-137 MDC Type (#) Surveyed Activity in in pCi/g m^2 pCi/g Scanning 840 0.157 N/A Static (61) 6100 0.052 0.655 Figure 1: The location of the SAB relative to the site SHONKA RESEARCH ASSOCIATES, INC. 34

Saxton Survey Project Final Report Figure 2: Color map of Cs-137 in the SAB Figure 3: The location of Scanned and Static Areas SHONKA RESEARCH ASSOCIATES, INC. 35

Saxton Survey Project Final Report Figure 4: CFD for Scanned Areas SHONKA RESEARCH ASSOCIATES, INC. 36

Saxton Survey Project Final Report 4.2.8 B3002 Survey Report Survey Location Code B3002A Southwest Corner BE-115 Survey Equipment SMCM Survey Date December 3rd Surveyor(s) D. DeBord Criteria DCGL-W 8.5 pCi/g Measurement Area Cs-137 Mean Cs-137 MDC Type (#) Surveyed Activity in in pCi/g m^2 pCi/g Scanning 5412 -0.039 0.712 Static (28) 2800 0.020 N/A Figure 1: The Location of the SAB relative to the site SHONKA RESEARCH ASSOCIATES, INC. 37

Saxton Survey Project Final Report Figure 2: Color map of Cs-137 in the SAB Figure 3: The Location of Scanned and Static areas SHONKA RESEARCH ASSOCIATES, INC. 38

Saxton Survey Project Final Report Figure 4: CFD for Scanned Areas SHONKA RESEARCH ASSOCIATES, INC. 39

Saxton Survey Project Final Report 4.2.9 B3003 Survey Report Survey Location Code B3003A Southwest Corner BK-104 Survey Equipment SMCM Survey Date December 4th Surveyor(s) D. DeBord Criteria DCGL-W 4 pCi/g Measurement Area Cs-137 Mean Cs-137 MDC Type (#) Surveyed Activity in in pCi/g m^2 pCi/g Scanning 2076 0.030 N/A Static (55) 5500 0.012 N/A Figure 1: The Location of the SAB relative to the site SHONKA RESEARCH ASSOCIATES, INC. 40

Saxton Survey Project Final Report Figure 2: Color map of Cs-137 in the SAB Figure 3: The Location of Scanned and Static areas SHONKA RESEARCH ASSOCIATES, INC. 41

Saxton Survey Project Final Report Figure 4: Cs-137 CFD for Entire SAB SHONKA RESEARCH ASSOCIATES, INC. 42

Saxton Survey Project Final Report 4.2.10 C1001 Survey Report Survey Location Code C1001A Southwest Corner AB-135 Survey Equipment SMCM Survey Date December 1st Surveyor(s) D. DeBord Criteria DCGL-W 8.5 pCi/g Measurement Area Cs-137 Mean Cs-137 MDC Type (#) Surveyed Activity in in pCi/g m^2 pCi/g Scanning 1680 -0.137 0.357 Static (0) 0 0 0 Figure 1: Color map of Cs-137 in the SAB SHONKA RESEARCH ASSOCIATES, INC. 43

Saxton Survey Project Final Report Figure 2: Cs-137 CFD for Entire SAB SHONKA RESEARCH ASSOCIATES, INC. 44

Saxton Survey Project Final Report

5. References EML, 1997 The Procedures Manual of the Environmental Measurements Laboratory, Volume I, 28th Edition, February, 1997.

GPU, 2000 GPU, SNEC Facility License Termination Plan, Rev 0, February, 2000.

GPU, 1999 SNEC Soil Background Study 1999, Table 2-21, pg. 2-49. Contained in SNEC Facility License Termination Plan, Rev 0. GPU. February 2000.

Hovgaard, 1997 Hovgaard, Jens. A New Processing Technique for Airborne Gamma-ray Spectrometer Data (Noise Adjusted Singular Value Decomposition).

Presented at the American Nuclear Society Sixth Topical Meeting on Emergency Preparedness and Response, San Francisco, California, April 22-25, 1997.

Tadmore, 1986 Tadmore, J., 1986, Radioactivity from coal-fired power plants: A review:

Journal of Environmental Radioactivity, v. 4, p. 177-204.

SHONKA RESEARCH ASSOCIATES, INC. 45

Saxton Survey Project Final Report, Appendix A Detector Quality Control Appendix A Detector Quality Control

Saxton Survey Project Final Report, Appendix A Detector Quality Control Quality Control SMCM and in situ quality control (QC) measurements were performed at least once every 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />. The QC routine consisted of a MCA gain check and adjustment (if necessary) and integration of peak counts in a common region of interest (ROI) from check sources to ensure consistent equipment performance. Sources were placed in identical fixtures on the sides of each 5x2 NaI(Tl) enclosure. 60Co and 137Cs sources were alternately used throughout the survey for QC purposes. The QC spectra were logged in the same fashion as normal survey data, using the SMCM software to log survey strips of data to the field computer with unique QC filenames.

The QC results for 60Co for each detector are shown in Figures 1-1A through 1-2A; 137Cs QCs for each detector are shown in Figures 1-3A through 1-4A. QC data was investigated for trending and that no two consecutive QC measurements of similar isotopes were outside of two sigma.

Page A-1

Saxton Survey Project Final Report, Appendix A Detector Quality Control Detector 1 Co-60 Performance Based Test Mean Co-60 Detector 1

+ 2 Sigma

- 2 Sigma 30 25 20 Gross Cps 15 10 5

0 2 7 12 17 22 27 32 PBC #

Performance Based Test Detector 2 Co-60 Detector 2 Mean

+ 2 Sigma

- 2 Sigma 35 30 25 Gross Cps 20 15 10 5

0 2 7 12 17 22 27 32 PBC #

Figure 1-1A - Co-60 QC control charts for SMCM detectors 1 and 2.

Page A-2

Saxton Survey Project Final Report, Appendix A Detector Quality Control Detector 3 Performance Based Test Mean Co-60 Detector 3

+ 2 Sigma

- 2 Sigma 30 25 20 Gross Cps 15 10 5

0 2 4 6 8 10 12 14 PBC #

Detector 4 Performance Based Test Mean Co-60 Detector 4

+ 2 Sigma

- 2 Sigma 30 25 20 Gross Cps 15 10 5

0 2 4 6 8 10 12 14 PBC #

Figure 1-2A- Co-60 QC control charts for SMCM detectors 3 and 4.

Page A-3

Saxton Survey Project Final Report, Appendix A Detector Quality Control Detector 1 Cs-137 Performance Based Test Mean Cs-137 Detector 1

+ 2 Sigma

- 2 Sigma 120 110 100 Gross Cps 90 80 70 60 50 1 6 11 16 21 26 31 PBC #

Detector 1 Performance Based Test Mean Cs-137 Detector 2

+ 2 Sigma

- 2 Sigma 180 170 160 Gross Cps 150 140 130 120 110 100 1 6 11 16 21 26 31 PBC #

Figure 1-3A - Cs-137 QC control charts for SMCM detectors 1 and 2.

Page A-4

Saxton Survey Project Final Report, Appendix A Detector Quality Control Detector 1 Performance Based Test Mean Cs-137 Detector 3

+ 2 Sigma

- 2 Sigma 100 90 80 Gross Cps 70 60 50 40 30 1 3 5 7 9 11 13 15 17 PBC #

Detector 1 Performance Based Test Mean Cs-137 Detector 4

+ 2 Sigma

- 2 Sigma 100 90 80 Gross Cps 70 60 50 40 30 1 3 5 7 9 11 13 15 17 PBC #

Figure 1-4A - Cs-137 QC control charts for SMCM detectors 3 and 4.

Page A-5

Saxton Survey Project Final Report - Appendix B Determination of Stripping Coefficients Appendix B SRA Technical Note 01-001 Determination of Stripping Coefficients

Saxton Survey Project Final Report - Appendix B Determination of Stripping Coefficients SRA Tech Note no.01-001 Rev. 2 Dated 01/25/02 Author: R.E. Burmeister Coefficients for 40K The SRA Subsurface Multi-spectral Contamination Monitor (SMCM) made use of the potassium (40K) window in its data analysis. It was recognized that there would be spectral interferences from other primordial nuclides. This technical note describes the method used to account for the interferences and details the determination of certain parameters used in the calculations.

Spectral interferences among K, U and Th radiation occur due to the combined effects of full-energy-peak overlaps and gamma ray scattering in the source, in the transport path from source to detector, and as a result of partial absorption processes in the detector. There are standard methods to treat these interferences, and the traditional IAEA recommended method was applied. In this method, only the relative contributions from the Th source into the U and K windows, denoted as and , respectively, and the uranium series contribution to K, denoted as , were taken into account (IAEA 1979). The contributions are displayed symbolically in the following equations:

Thc = Th - Thb Uc = U - Ub - Thc Kc = K - Kb - Thc - Uc where Kb , Ub , Thb are background count rates; K , U , Th are uncorrected count rates; Kc , Uc , Thc are corrected count rates.

The stripping coefficients or stripping ratios are defined as the ratio of the number of counts due to a nuclide in other windows to the number of counts in the window for the nuclide. The ratios are usually determined from calibration pads, which are large concrete pads that are doped with uranium, thorium, and potassium sources.

Stripping ratios for a few cylindrical detectors are available (Grasty, 1997). The 5x2 detectors used by the SMCM were not part of the available lists. It was therefore necessary to estimate the ratios for the 5x2s from the lists of ratios for other detectors. For each of the detectors in the list in Table 1-1B, the source-detector geometry was the same, and the detectors were all cylinders. The quantity that changed among the detectors was the volume, or equivalently the mass, since the detectors had the same density. For a given photon energy, the mass-attenuation coefficient is the same for the detectors, but the varying volumes imply varying amounts of event collection. Thus, the stripping coefficients should be functions of the photon interaction collection volume of the detectors.

Page B-1

Saxton Survey Project Final Report - Appendix B Determination of Stripping Coefficients The data for the stripping coefficients for the standard detector sizes in Table 1-1B were plotted against detector volume. The plot is shown in Figure 1-1B. The values for , , were read off of the plot for the volume of the 5x2 detectors for a detector volume of 644 cc. The values are given in Table 1-1B in bold.

Detector 3x3 0.71 0.88 1.0 5x5 0.43 0.62 0.95 9x4 0.39 0.52 0.90 5x2 0.63 0.81 0.99 Table 1-1B - Stripping Coefficients for some Standard Detectors (Grasty1997) and the fitted SRA 5x2 (in bold)

Stripping Coefficients for Primordials 1

0.9 0.8 Alpha 0.7 Beta 0.6 Gamma 0.5 0.4 0.3 0 1000 2000 3000 4000 5000 Detector Volume (cc)

Figure 1-1B - Plot of Stripping Coefficients in Table 1 with smooth line interpolation Page B-2

Saxton Survey Project Final Report - Appendix B Determination of Stripping Coefficients Coefficients for 137Cs Like for 40K, the SMCM made use of the 137Cs window, and also likewise, it was recognized that there would be spectral interferences from the primordial nuclides. The contributions are displayed symbolically in the following equation:

137 Csgross = aKnet + bUnet + cThnet where Knet , Unet , Thnet are the count rates given by the IAEA stripping equations above; 137 Csgross is the gross count rate in the 137Cs region-of-interest; a, b, c are coefficients that account for the respective contributions of the primordial 137 nuclides to the Cs window.

The coefficients were determined from a least-squares solution for approximately 2700 sets of 137

( Csgross , Knet , Unet , Thnet) values from the SMCM 4 NaI detector array and for approximately 5000 sets for the 2 detector array. The source check surveys and the special salted survey area were not included in the sets of data. The values are given Tables 1-2B and 1-3B. Prior to calculating the solution, the average fallout of 137Cs was subtracted from the 137Csgross values. The fallout value was 2.03 counts-per-second.

Table 1-2B. Coefficients for Primordial Contributions to Cs-137 Window for 4 Detector array Detector no. a b c 1 0.18 3.17 3.95 2 0.25 3.12 3.69 3 0.23 2.84 3.66 4 0.20 3.33 3.52 Table 1-3B - Coefficients for Primordial Contributions to Cs-137 Window for 2 Detector array Detector no. a b c 1 0.35 2.96 3.68 2 0.31 3.09 3.45 Page B-3

Saxton Survey Project Final Report - Appendix B Determination of Stripping Coefficients (Grasty 1997) Grasty, Bob. Standardization of Airborne Gamma-Ray Surveys.

Presentation at HPS 42nd Annual Meeting, Summer 1997.

(IAEA 1979) International Atomic Energy Agency. Gamma Ray Surveys in Uranium Exploration. Technical Report Series 186. International Atomic Energy Agency, Vienna.

Page B-4

Saxton Survey Project Final Report - Appendix C Determination of the Platform Factor Appendix C SRA Technical Note 01-004 Determination of the Platform Factor

Saxton Survey Project Final Report - Appendix C Determination of the Platform Factor SRA Tech Note no.01-004 Dated 01/08/01 Author: R.E. Burmeister When the SRA Subsurface Multi-spectral Contamination Monitor (SMCM) is deployed with multiple detectors on a survey platform, the possibility exists that the platform will obstruct the field of view of some of the detectors. This obstruction of the field of view results in different responses from the detectors. When surveying a field area for background, the detectors should have as similar a response as possible. This technical note describes the determination of shielding factors that account for the obstructed field of view and put the detectors on the same response basis.

The method begins with the collection of a series of region-of-interest (ROI) integrations. In general, there are more than one regions-of-interest. Each detector collects spectral data that can be integrated to determine ROI values. One of the detectors with an unobstructed view is then chosen to be the basis for comparing the responses of the detectors. The ratio of ROIs of the other detectors to the chosen detector, one ratio per ROI, are numbers that report how similarly the other detectors responded compared to the basis detector, and are also numbers that can be used to scale the responses of the other detectors so that shielding effects from obstructed views are removed.

The ratios are determined for a number of background surveys until the statistics yield a standard deviation of the mean of 1-5% for each ROI, and then the mean ratio for each ROI-detector combination is recorded.

For example, assume four detectors and three regions-of-interest and five initial surveys. Also assume that detector 1 was chosen as the basis detector. Thus, the calculation of the ratio of detectors 2, 3, and 4 to detector 1 for each ROI would yield a table of ratios like that shown in Table 1-1C, filled with hypothetical data.

Table 1-1C - Hypothetical Table of Ratios for Three Regions-of-Interest and Four Detectors ROI 1 ROI 2 ROI 3 survey 2 3 4 2 3 4 2 3 4 1 0.961 0.953 1.031 0.982 0.966 1.007 0.978 0.935 1.074 2 0.950 0.936 1.024 0.948 0.971 1.013 0.968 0.948 1.048 3 0.955 0.950 1.048 0.952 0.973 1.027 1.004 0.968 1.072 4 0.978 0.945 1.046 0.925 1.038 1.020 1.007 0.934 1.059 5 0.967 0.956 1.025 0.916 0.919 0.974 1.026 1.000 1.088 Mean 0.962 0.948 1.035 0.944 0.973 1.008 0.997 0.957 1.068 Page C-1

Saxton Survey Project Final Report - Appendix C Determination of the Platform Factor The mean reported in Table 1-1C is the mean of the corresponding five ratios; this is the mean ratio for each ROI-detector combination. For ROI 1, the data imply that for the five surveys, detector 2 was on average 96% of detector 1. Similarly, detectors 3 and 4 were 95% and 104% of detector 1.

There are similar numbers for the other regions-of-interest. To place the detectors on the same basis as detector 1, future responses would need to be divided by these ratios.

Page C-2

Saxton Survey Project Final Report - Appendix D Calibration Factor Methodology Appendix D Calibration Factor Methodology

Saxton Survey Project Final Report - Appendix D Calibration Factor Methodology

1. Objective This document develops calibration factors for the five-inch by two-inch sodium iodide (5-by-2 NaI) detector system employed at the Saxton site for in situ counting and SMCM surveys of designated areas from November 28 through December 10, 2001. Factors are developed for the common power plant nuclides, 60Co and 137Cs, for a plane geometry and for uniform distribution in soil.

2. Summary The detectors in both the in situ system (Detectors 1 & 2) and the SMCM array ( 2 or 4 detectors) were at a fixed height of one meter above the ground, measured from the detector face. Each detector was a 5-inch by 2-inch NaI(Tl) Bicron model 5M2/5, coupled to an EG&G ORTEC model 296 ScintiPak solid-state tube base. The Scintipack is equipped with integral high voltage power supplies and preamplifier stage in each unit. An ORTEC MicroAce MCA card provided the amplifier stage and the analog-to-digital conversion. Table 2-1D and 2-2D give the factors for uniform and infinite plane distributions for the 4 detectors, respectively.

Page D-1

Saxton Survey Project Final Report - Appendix D Calibration Factor Methodology Table 2-1D - Calibration factors for in situ and SMCM counting of infinite plane source (/ = ) at a height of one meter.

Calibration N0 / Factor Intensity (cps//cm2- /A (cps per pCi Nuclide Detector Energy ( s-1 per-Bq) sec) Nf/N0 (g cm-2) per cm^2)

Cs-137 1 0.662 .8512 5.538 1 6.1456 2.02 Cs-137 2 0.662 .8512 5.537 1 6.1456 2.02 Cs-137 3 0.662 .8512 5.809 1 6.1456 2.12 Cs-137 4 0.662 .8512 5.404 1 6.1456 1.97 Co-60 1 1.173 1 3.947 1 8 1.99 Co-60 2 1.173 1 3.474 1 8 1.75 Co-60 3 1.173 1 3.853 1 8 1.94 Co-60 4 1.173 1 3.735 1 8 1.89 Co-60 1 1.3325 1 2.867 1 8 1.54 Co-60 2 1.3325 1 2.780 1 8 1.49 Co-60 3 1.3325 1 2.832 1 8 1.52 Co-60 4 1.3325 1 2.586 1 8 1.39

• Unscattered flux per unit source strength in soil for a planar source is dimensionless.

Table 2-2D - Calibration factors for SMCM and in situ counting of uniformly distributed source (/ = 0) at a height of one meter.

Calibration N0 / Factor Intensity (cps//cm2- /A (cps per pCi Nuclide Detector Energy ( s-1 per-Bq) sec) Nf/N0 (g cm-2) per gram)

Cs-137 1 0.662 .8512 5.538 1 6.1456 7.38 Cs-137 2 0.662 .8512 5.537 1 6.1456 7.38 Cs-137 3 0.662 .8512 5.809 1 6.1456 7.75 Cs-137 4 0.662 .8512 5.404 1 6.1456 7.21 Co-60 1 1.173 1 3.947 1 8 8.79 Page D-2

Saxton Survey Project Final Report - Appendix D Calibration Factor Methodology Co-60 2 1.173 1 3.474 1 8 7.74 Co-60 3 1.173 1 3.853 1 8 8.58 Co-60 4 1.173 1 3.735 1 8 8.32 Co-60 1 1.3325 1 2.867 1 8 6.91 Co-60 2 1.3325 1 2.780 1 8 6.70 Co-60 3 1.3325 1 2.832 1 8 6.82 Co-60 4 1.3325 1 2.586 1 8 6.23 Page D-3

Saxton Survey Project Final Report - Appendix D Calibration Factor Methodology

3. Technical Approach The parameter of interest in in situ gamma ray spectrometry is the ratio between the count rate in the full-energy peak of interest and the corresponding quantity of the nuclide of interest in the soil below (either per unit mass or per unit area). This parameter is known as the calibration factor. The general expression employed in establishing calibration factors for in situ gamma ray spectrometry is Nf Nf N o

= çç ÷÷çç ÷÷ç ÷ Equation 1 A No ø ø A ø where each ratio in the above expression is defined as follows (Helfer and Miller 1988):

Nf is the calibration factor desired. The dimensions are count rate (cpm, typically) per unit A inventory (Bq m-2) or per unit concentration (Bq g-1) of the nuclide of interest in the soil.

Nf is the angular correction factor for the detector at the energy of interest and for a given No source distribution.

No is the detector peak response, which is the peak count rate per unit uncollided flux from a plane-parallel source of the photon energy of interest.

is the total uncollided flux (for the photon energy of interest) arriving at the detector per A unit inventory or concentration in the soil.

To establish a calibration factor for a nuclide of interest, the three ratios in Equation 1 are determined for a given detector configuration and source distribution. Note that the calculated calibration factor is not strongly influenced by the estimate of the source distribution, so inaccuracies in this estimation should not adversely affect the calculated factor. This is due to the fact that the ratio between the uncollided flux at the detector and the source inventory or concentration does not vary significantly with source distribution (EML 1990).

The sections that follow describe the three factors from Equation 1 and how these are established.

Values are given for each factor for the 5-by-2 NaI detectors. These factors are then used to calculate calibration factors for the detectors for two different deposition geometries (i.e. uniform and planar).

Page D-4

Saxton Survey Project Final Report - Appendix D Calibration Factor Methodology

4. Description of Individual Factors 4.1 Angular Correction Factor (Nf/N0)

The angular correction factor Nf/N0 depends on both the detector and the source geometry. It corrects for the fact that the cylindrical detector does not respond isotropically to sources incident from different angles with respect to its vertical axis (when oriented normal to the ground). Values of Nf/N0 for intrinsic germanium (Ge) detectors have been published by Helfer and Miller (1988) as a function of energy and detector length-to-diameter ratio (L/D) for planar and uniform sources.

(Uniform means constant source concentration with depth, and refers to the distribution of the primordial species within the soil. Planar means a surface source with uniform dispersion, i.e., an infinite plane source.) However, no such data was identified for NaI detectors. Hence, for the purpose of this document, the angular correction factor for the 5 by 2 NaI detector was assumed to be unity. This assumption is supported by the data given by Helfer and Miller that show, for an intrinsic Ge detector with a length-to-diameter ratio (L/D) of unity, the angular correction factor deviates from unity by no more than 4% for energies up to 2.5 MeV for either the uniform or planar source geometries.

4.2 Peak Count Rate per Unit Uncollided Flux (N0/)

The values of N0/ for a given detector can be obtained by counting point sources of the nuclides of interest at a distance of at least one meter from the face of the detector (Helfer and Miller 1988). At this distance, a point source provides a good approximation of a plane-parallel field. N0/ is computed by dividing the observed peak count rate by the flux at the detector. The flux is computed by dividing the source photon emission rate by 4r2 and correcting (if necessary) for attenuation by air and the source holder. The distance r to be used depends on the photon energy.

For energies > 1 MeV, r should be taken as the distance to the center of the crystal; and for energies

< 100 keV, the distance should be to the detector face (EML 1990). For energies in between these values, the distance to be used can be computed using the mean free path in sodium iodide for the photon energy of interest. However, one should be careful when doing so to be sure that the mean free path is not greater than the location of the midpoint of the crystal. If this does occur, then r should be chosen to correspond with the detector midpoint. The preferred dimensions of N0/ are cpm per photon cm-2 second-1.

Values of N0/ were established for each detector by making measurements at a distance of one meter using NIST traceable 137Cs and 60Co point sources, serial numbers 619-38-1 and 578-32-17, respectively. The original activities were 9.301 µCi for the 137Cs standard (on 7/1/98) and 0.8029

µCi for the 60Co standard (on 7/1/98). The calibration certificates for these standards are included in Appendix H of this report. The 137Cs source was counted for 5 minutes; the 60Co and backgrounds were counted for 10 minutes. Net count rates for each region of interest Page D-5

Saxton Survey Project Final Report - Appendix D Calibration Factor Methodology were established by subtracting the background counts from the gross counts for each energy region.

Table 4-1D shows the calculation of the N0/ values using the measured net count rates from the tables above.

Table 4-1D - Calculation of N0/ values for detector 1.

137-Cs Assay activity (µCi) for the 137Cs standard = 9.301 Assay date for the 137Cs standard = 7/1/98 137 Cs half-life (days) = 11019.593 Measurement date = 11-28-01 137 Cs activity (µCi) on measurement date = 8.60 Distance from source to center of detector (cm) = 102.54 137 Cs gamma-ray intensity = 0.851 Gamma flux at detector center (in vacuum) (cm2-sec) = 2.067 N0 /. (cps/cm2-sec) = 55.38 60-Co 1173 keV Assay activity (µCi) for the 60Co standard = 0.803 Assay date for the 60Co standard = 7/1/98 60Co half-life (days) = 1923.915 Measurement date = 11-28-01 60 Co activity (µCi) on measurement date = 0.513 Distance from source to center of detector (cm) = 102.54 60 Co gamma-ray intensity = 1.0 Gamma flux at detector center (in vacuum) (cm2-sec) = 0.151 N0 /. (cps/cm2-sec) = 39.47 60-Co 1333 keV Assay activity (µCi) for the 60Co standard = 0.803 Assay date for the 60Co standard = 7/1/98 60Co half-life (days) = 1923.915 Measurement date = 11-28-01 60 Co activity (µCi) on measurement date = 0.513 Distance from source to center of detector (cm) = 102.54 60 Co gamma-ray intensity = 1.0 Gamma flux at detector center (in vacuum) (cm2-sec) = 0.151 N0 /. (cps/cm2-sec) = 28.67 4.3 Total Uncollided Flux per Unit Source Inventory or Concentration (/A)

The factor /A is not detector dependent, but is a function of soil composition and density, air attenuation and the distribution of the source in the soil. Values of for sources in soil having strengths of either 1 photon gram-1 second-1 (for a uniform source) or 1 photon cm-2 second-1 (for planar sources), are given by Helfer and Miller (1988). This data is reproduced as Table 4-2D.

Note this data is flux (at the detector) per unit photon emission in soil. Hence, the data have dimensions of g cm-2 for the uniform source and are dimensionless for the distributed (deposited) sources. To get the quantity desired (/A), these values must be multiplied by the intensity for the Page D-6

Saxton Survey Project Final Report - Appendix D Calibration Factor Methodology photon of interest. Doing so gives /A in terms of cm-2 s-1 per g-1 s-1 (g cm-2) for the uniform source and cm-2 s-1 per cm-2 s-1 (dimensionless) for the planar sources.

The parameter / is a measure of the source depth profile, where the profile is assumed to be exponential. is the inverse of the relaxation length and is the soil density. Thus, / equals zero for the uniform (primordial) source geometry (where the source profile is a constant) and equals infinity for an infinite plane. Deeply distributed sources have values of / that are less than 0.1, where a range 0.1 / 0.5 would be characteristic of fallout from historical weapons testing that has not penetrated far into the soil (Helfer and Miller 1988). In the case of aged fallout in the U.S., Helfer and Miller give ranges for / of 0.03 to 0.2 for open fields in areas having moist climates and 0.2 to 1.0 for semi-arid regions. The / value of 6.25 in Table 4-2D corresponds to fresh deposition with little penetration into the soil.

It should be noted that while the parameter / is favored domestically for characterizing source profiles in soil, the ICRU uses a different parameter for the same purpose (ICRU 1994). They chose to define a parameter that is essentially the inverse of / called the relaxation mass per unit area, . has the dimensions of g cm-2. The most notable difference between and / is that is defined in such a manner that the soil density profile can be accounted for, whereas / can only be defined for constant soil density. Beyond this, the only important thing to remember is that is the inverse of /, and so = 0 corresponds to the infinite plane (where / = ) and =

corresponds to the uniform source (where / = 0).

Page D-7

Saxton Survey Project Final Report - Appendix D Calibration Factor Methodology Table 4-2D - values per unit source strength* as a function of energy and source distribution published by Helfer and Miller (1988)

• For the uniform profile (/ = 0), the source strength is one gamma per second per gram for soil at all depths. For the exponential profiles (deposited nuclides), the source strength is one gamma per cm2 per second. Thus, the data for the uniform source have dimensions of g cm-2 and those for the exponential sources are dimensionless.

Page D-8

Saxton Survey Project Final Report - Appendix D Calibration Factor Methodology

5. Results 5.1 Calibration Factors for Depositions of Common Power Plant Nuclides Calibration factors for the NaI detector system for the common power plant nuclides 137Cs, and 60Co have been estimated as follows. Values for the peak count rate per unit uncollided flux (also known as the response factor) for each system were established using the peak count rates measured using the NIST traceable 137Cs and 60Co sources. The calibration factor for common power plant nuclides was inferred from this data through linear interpolation. Values for the angular correction factors were assumed to be unity, as discussed in section 4.1. Values for the uncollided flux per unit source concentration or inventory were taken from Table 4-2D above. Using these values, calibration factors for in situ counting were established for two deposition geometries for the detector at a distance of one meter above the ground. The two source geometries were uniform profile (/ = 0),

and an infinite plane (/ = ).

The values used in the calculation of the calibration factors for deposited nuclides for the detector system are given in Table 5-1D. These values were established from linear interpolation of the data in Table 4-2D where necessary.

Table 5-1D values used in the calculation of the calibration factors for deposited nuclides Uncollided Flux per Unit Source (/A)

Nuclide Energy (MeV) 0-Uniform (g cm-2) -plane Cs-137 0.66165 6.1456 2.054 Co-60 1.17322 8.1472 2.189 Co-60 1.33249 8.7504 2.224 To compute the calibration factors for in situ soil counting for deposited nuclides, the values given in Table 4-1D and Table 5-1D were substituted into Equation 1. The values from Table 5-1D are converted to /A by multiplying by the appropriate photon intensity. The calculated factors are then multiplied by 1 x 10-4 m2 per cm2 to get the desired dimensions. The factors are given in terms of cpm per µCi per square meter. The results are shown in Tables 2-1D and 2-2D.

6. References (EML 1990) Krey, P. W.; Beck, H. L. EML Procedures Manual, 27th edition, Vol. 1, HASL-300, Environmental Measurements Laboratory, New York, NY, November, 1990.

Page D-9

Saxton Survey Project Final Report - Appendix D Calibration Factor Methodology (Helfer and Miller 1988) Helfer, I. K.; Miller, K. M. Calibration Factors for Field Ge Detectors Health Physics, Vol. 55, No. 1 (July), pp. 15-29, 1988.

(ICRU 1994) Gamma-Ray Spectrometry in the Environment, ICRU Report 53, International Commission on Radiation Units and Measurements, Bethesda, MD, December 1, 1994.

Page D-10

Saxton Survey Project Final Report - Appendix E SMCM Scan Survey Characteristics Appendix E SMCM Scan Survey Characteristics

Saxton Survey Project Final Report - Appendix E SMCM Scan Survey Characteristics

1. Vehicle Speed Considerations The SMCM software can trigger recording a spectrum of data for either a fixed time or for a fixed distance. For this survey, the data was triggered with a time based trigger set for ten-second intervals. The speed of the vehicle affects the meters that are traveled in ten seconds. At the desired pixel size of two meters, this corresponds to 0.2 meters per second (0.4 mph). In order to permit an operator to control a vehicle at that low speed, a specialized operator interface was developed. The user interface provided for the operator gave speed, time and distance traveled along a strip. In addition, indications were provided for speed required to complete the strip at the desired average speed. A screen capture of the user interface is shown in Figure 1-1E.

Figure1-1E- Screen capture of SMCM software user interface.

Page E-1

Saxton Survey Project Final Report - Appendix E SMCM Scan Survey Characteristics

2. Detector Height Considerations The detectors in the large array were subject to varying heights above the terrain. The height of the detectors could vary because of terrain variations, or because of roll or pitch motions on the vehicle.

The height dependence on pitch (vehicle oriented up or down from horizontal) was minimized by placing the array as close as possible to the front axle of the vehicle, which minimized the dependence of detector height on pitch. The dependence on roll (vehicle rotated left or right from horizontal) was greater. The inner detectors were one meter from the centerline of the vehicle and were less subject to height variation as a function of roll angle. The outer detectors were 3 meters from the centerline of the vehicle, and the front face of the detector was one meter above the terrain.

They were in an enclosure, with the front of the enclosure at 92 centimeters above the ground.

Thus, the outer detectors would contact the ground if the roll angle exceeded 17 degrees. To avoid damage and lost time, the operator prevented the detectors from contacting the ground by maneuvering the vehicle. However, at times, one of the outer two detectors came close (front face at 10 centimeters) to the ground, while the detector on the opposite side was lifted to a height of less than two meters.

The flux from a uniform source is constant as a function of detector height. The technical basis documents from the DOE Environmental Measurements Lab on in situ gamma spectroscopy state that the response function is nearly constant from one half meter to two meters (EML 1990). Thus, the impact on the survey from height variation with roll angle on large area sources is small and can be neglected.

The impact on small area sources can be estimated from the nearly inverse distance squared relationship of the detector to a point source. The detector near the ground has a substantial improvement in detection limit for small sources, perhaps as much as 100 fold. The detector that lifts away from the ground suffers a degradation of detection limit of a factor of 3.6. This was estimated to affect less than 1% of all of the measurements. In addition, as a result of the typical survey pattern (back and forth), the adjoining strip of data collected would reverse the travel over the ditch or obstacle and adjoining pixels in the next strip would be affected in the opposite manner as any affected pixel.

Page E-2

Saxton Survey Project Final Report - Appendix F Data Handling and Analysis Methods Appendix F Data Handling and Analysis Methods

Saxton Survey Project Final Report - Appendix F Data Handling and Analysis Methods The data from the SMCM for a given SAB consisted of 10-second spectra from the nominal 5X2 NaI(Tl) detectors. The energy range from 0 to 3 MeV was spanned with 512 channels. The data was converted to count rate, and adjusted by a factor that included the detector efficiency (relative to detector 1) and the platform shielding. This platform-shielding factor was applied to place the detectors on an equivalent response basis. This practice simplified the development of images from the data, and avoided banding in the image due to slight differences in response. It is mathematically equivalent to corrections applied to medical (gamma) cameras. The inner detectors were shielded from the ground by the frame, body and engine of the Ford F-250 4x4 truck platform.

As a result, the count rate for naturally occurring radioactive potassium were slightly reduced. The tripod measurements were not subject to vehicle shielding; therefore no correction was needed.

The factor was determined in an iterative fashion. Raw data was examined in each of the energy windows in the gamma ray spectra that were studied, and the count rates for the two detectors mounted near the Ford were found to be systematically lower, on average, than the outer two detectors. The reduction was less than 10% when uncorrected. This difference provides an image that is banded in appearance, since all of the pixels along a row of data are affected.

Following correction of the raw data by the platform-shielding factor, the spectra were analyzed and separated into components using the method of noise-adjusted single value decomposition (NASVD). (Hovgaard 1997). This is a statistical process that computes the spectral shapes that are embedded data. Following decomposition, the spectra are re-assembled without adding the noise back in. This process reduces the noise by a factor of 4 or more, and improves the signal to noise by more than a factor of two. The same gain in signal to noise would require counting four times as long. This means each 10 second spectra was equivalent to 40 seconds of counting (if the NASVD method was not used). The larger detectors (5x2) are about 2.5 times more efficient than the (3x3) detectors often used for in situ measurements.

Another significant advantage of NASVD is that the components that are present are presented as separated spectra. If any nuclides of concern were present at detectable levels, the spectral shape from these nuclides would be observed in the components. Following the NASVD, the components were examined for any sign of 137Cs.

Page F-1

Saxton Survey Project Final Report - Appendix G Point Source Test Survey Appendix G Point Source Test Survey

Saxton Survey Project Final Report - Appendix G Point Source Test Survey As a test of the system, a high-intensity 137Cs source was placed in SAB A1004. Three 70-meter strips of data were taken with the source placed on the surface, and buried at 6 and 12 inches. The SRA and GPU staff observed the computer display during passage of the source, and qualitatively noted that the surface source was easily seen for much of the 70-meter strip on all detectors, and was highly collimated and localized to a single detector when buried at 30 centimeters. The data was processed in the same manner as all reported data from this survey. The data is designated as T0, T6 and T12 for 0, 6 inches and 12 inches source depth. Figures 1-1G, 1-2G, and 1-3G show the 70-meter strip of data.

Figure 1-1G - Test Strip with Source at 0 Centimeter Depth Figure 1-2G - Test Strip with Source at 15 Centimeter Depth Page G-1

Saxton Survey Project Final Report - Appendix G Point Source Test Survey Figure 1-3G - Test Strip with Source at 30 centimeter Depth The images above show the data from all 4 detectors for the one strip. There are several noteworthy points. When the image of the source placed on the surface (noted as T0) is reviewed, a black projection can be seen extending upward into the region where the source was placed. This region of reduced counts is caused by the inherent shielding of the truck that carried the detectors. We call this effect the platform shielding. The magnitude of the platform shielding depends on the source geometry. For a point source shown in the three images, the two inner detectors in the array are strongly impacted and have reduced counts once the source passes under the array and under the truck. That creates the dark region. For a source that is uniformly mixed into soil, such as natural potassium, uranium, and thorium, the vehicle shielding, on average, reduces the count rate but does not create the artifact that is seen with a point source. The factor is described in Appendix C.

Because the source was very strong, other nuclides show increases in count rates for both the 0 and 15 centimeters of depth. This effect, called coincidence counting, occurs in spectrometers at high count rates when two or more photons interact at the same time. When the source is placed at 30 centimeters of depth, the earth acts as a collimator, since the slant range to a detector not immediately above the detector is much greater than 30 centimeters. Thus, the T12 image shows only a few pixels impacted by the buried source. The SAB B3003 report shows a single pixel which has an implied concentration of 1.5 pCi/g 137Cs if the activity was uniformly distributed over a large area. That data might be a statistical outlier. Because the B3003 image shows that the activity is not distributed, a smaller area with added activity that was a factor of 20 times higher, perhaps 30 pCi/g or more (the area factor) might be present. Alternatively, the image is consistent with a buried small diameter source since the appearance is similar to the T12 image above. If the source was buried at a similar depth as the T12 experiment (30 centimeters), the source activity would be approximately 10 times lower than the source used for the images shown above.

Page G-2

Saxton Survey Project Final Report - Appendix H Source Calibration Certificates Appendix H Source Calibration Certificates

Saxton Survey Project Final Report - Appendix H Source Calibration Certificates Page H-1

Saxton Survey Project Final Report - Appendix H Source Calibration Certificates Page H-2

Saxton Survey Project Final Report - Appendix I Detector Quality Objectives Appendix I Detector Quality Objectives

Saxton Survey Project Final Report - Appendix I Detector Quality Objectives

1. Introduction Data Quality Objectives (DQOs) are established to specify the objectives of a data collection effort, to define the most appropriate data to collect, and to specify tolerable limits on decision errors.

Decision error limits in turn provide the basis for establishing the quantity and quality of data needed to support the decision. The DQO process is a series of seven steps, as depicted in Figure 1-1I below.

Figure 1-1I - DQO process Page I-1

Saxton Survey Project Final Report - Appendix I Detector Quality Objectives A brief description of each step of the DQO process is provided below.

• Step 1 - State the Problem - Concisely describe the problem to be studied.

• Step 2 - Identify the Decision - Identify the questions the study will attempt to answer, and what actions may result.

• Step 3 - Identify Inputs to the Decision - Identify the information that needs to be obtained and the measurements that need to be taken.

• Step 4 - Define the Study Boundaries - Specify the time periods and spatial areas to which decisions will apply (to facilitate data interpretation).

• Step 5 - Develop a Decision Rule - Define the statistical parameter(s) of interest, specify the action level(s), and describe the logical basis for choosing among alternative actions.

• Step 6 - Specify Tolerable Limits on Decision Errors - Define the decision makers tolerable decision error rates based on the consequences of making an incorrect decision.

• Step 7 - Optimize the Design - Evaluate information from the previous steps and develop a resource-effective design that meets all DQOs.

Following are discussions for the first six steps of the DQO process described above as they pertain to Phase I of the site survey for the Saxton Nuclear Experimental Corporation (SNEC) site. The survey design is as defined by the Contract for the survey services and its associated references.

Step One: State the Problem The purpose of the site survey is to provide a thorough, quantitative assessment of radionuclide concentrations in the soil at the SNEC facility in Saxton, PA. Phase I of the site survey is to cover 100% of accessible areas where remediation activities have been completed or are not anticipated.

The purpose of the survey is to characterize radionuclide concentrations in soil to identify any areas where remediation may be required or for which it may be necessary to revise the Final Status Survey Plan (e.g., reclassify an area).

Soil radionuclide concentrations will be estimated via gamma-ray spectrometry, with the spectrometry equipment utilized in either a scanning or a static mode, as necessary.

Step Two: Identify the Decision The Phase I site survey should identify any areas within the Phase I survey area where concentrations of plant-related radioactivity in soil:

• exceed the applicable derived concentration guideline level (DCGL) for unrestricted use,

• exceed levels that would be considered ALARA (as defined by the site License Termination Plan), or

• represent levels that would indicate the need for revision of the License Termination Plan (e.g., reclassify an area).

Page I-2

Saxton Survey Project Final Report - Appendix I Detector Quality Objectives The results from the Phase I survey may be used as input to the final status survey process, to guide remediation activities, or as input to the evaluation of remediation alternatives. Possible actions that may result from analysis of the data from the Phase I site survey are summarized in Table 5-8 of the SNEC License Termination Plan (LTP). This is reproduced below in Table 1-1I.

Table 1-1I - Analysis of data from Phase I site survey Step Three: Identify Inputs to the Decision The following inputs will be used to identify areas where concentrations of plant-related radionuclides in soil indicate the need for investigation or action.

• Quantitative assessment of the magnitude and spatial variability of background in a suitable background reference area. The background reference area will consist of 10,000 m2 selected within a non-impacted area.

• A spatially representative, quantitative assessment of the concentration in soil of primordial radionuclides (uranium series, thorium series, and 40K) and any plant-related radioactivity identified within the Phase I study area.

Page I-3

Saxton Survey Project Final Report - Appendix I Detector Quality Objectives

• Derived concentration guideline levels (DCGLs) that relate radionuclide concentrations in soil to annual dose for plant-related radioisotopes.

• Area factors to adjust DCGL values so assessments can be made for small areas of elevated activity.

• Detection limits for instruments and analysis techniques relative to the applicable DCGLs.

• SNEC Action/Investigation levels as defined in Table 5-7 of the LTP.

DCGL values will be as defined in the SNEC LTP or as otherwise prescribed by the Purchaser. The minimum detectable concentration for measurements performed for the Phase I site survey will be no greater than 1.5 pCi/g 137Cs. This value applies to both static and scan measurements.

Step Four: Define the Boundaries of the Study The spatial boundaries of the Phase I site survey are 100% of the accessible area within that defined in Items 13(c)(1) and 13(c)(2) of the Contract. The Phase I area may be subdivided for reporting purposes into smaller areas. If so, logical subdivisions of the overall study area will be defined at the time of data evaluation.

The temporal boundary of the study, for purposes of dose assessment for a residential occupant, is 1,000 years.

Step Five: Develop a Decision Rule Decisions regarding the need for investigation or other actions will be made based on the median concentration in soil for plant-related radioactivity (if found) relative to the applicable DCGLs.

Given that the data acquired for the Phase I site survey will be nuclide-specific and will cover 100%

of the accessible area, the median concentration of any plant-related activity identified can be established by inspection. The non-parametric statistical tests described in the MARSSIM will not be required.

The investigation and action levels to be employed for the Phase I site survey as are defined in Table 5-7 of the SNEC LTP. This is reproduced below in Table 1-2J.

The limiting investigation level is that for Class 3 areas. Specifically, any measurement that exceeds the required MDC of 1.5 pCi/g for 137Cs will be flagged for investigation. The 1.5 pCi/g value applies to an area of 10,000 m2, which is the basis for the DCGLW value for 137Cs from the SNEC LTP. If an area of interest is smaller than 10,000 m2, then the 1.5 pCi/g value should be multiplied by an appropriate area factor.

Page I-4

Saxton Survey Project Final Report - Appendix I Detector Quality Objectives Table 1-2I- Summary of SNEC Investigation/Action Levels All of the investigation/action levels in the above table assume background values have been subtracted (when appropriate).

Page I-5

Saxton Survey Project Final Report - Appendix I Detector Quality Objectives Step Six: Specify Limits on Decision Errors Decision errors refer to making false decisions by either rejecting a null hypothesis when it is true (a Type I error) or accepting a null hypothesis when it is false (a Type II error). With respect to the Phase I SNEC site survey, the null hypothesis being tested is that the survey area of interest contains residual activity in excess of the applicable investigation or action levels. Thus, a Type I error refers to concluding that an area does not require investigation or action when in fact it does.

The probability of making a Type I error is referred to as alpha (). Likewise, a Type II error refers to concluding that an area requires investigation or action when in fact it does not. The probability of making a Type II error is denoted beta ().

Under the current regulatory models, an value that is too large equates to greater risk to the public in that there is a greater chance of releasing a survey area that does not meet the unrestricted use criteria. Selecting a value that is too large can result in excessive costs in that survey areas that meet the release criterion could be subjected to superfluous investigations or other actions.

The Type I decision error rate for the Phase I SNEC site survey will be limited to a maximum of 5%

as per the Contract. For Type II errors, an initial limit of 5% will be used unless it is found that this value can be relaxed without compromising the Type I error limit. The effect of the Type II error rate on the ability to meet the required Type I error rate will depend on the difference between the applicable investigation level and background for a given survey area. The benefit of increasing the allowable Type II error rate is that this reduces the probability of falsely concluding a survey area requires investigation when in fact it does not. This benefit is proportional to the total number of survey areas to be characterized.

In the case where the null hypothesis is that an area contains residual activity that is in excess of the applicable investigation or action levels, the Type I error rate depends on the variability of additional radioactivity above background (should any exist) and is controlled by requiring that a net result exceed the critical level by some number of standard deviations. The Type II error rate depends on the variability of background and is controlled by requiring that the net count rate exceed some multiple of the measurement standard deviation. The critical level is computed as LC = z1 2 where LC = critical level; z1 = the 1- percentile of the cumulative normal distribution function; and

= standard deviation of the measurements.

Note that this differs from the traditional critical level expression used in asserting instrument detection limits. The difference is that the null hypothesis being tested is the survey area contains activity in excess of the investigation level. In counting statistics, the null being tested is that the Page I-6

Saxton Survey Project Final Report - Appendix I Detector Quality Objectives sample being measured does not contain activity in excess of background. Thus, one would use the 1- percentile in place of the 1- percentile in the above expression.

In the case where the survey protocol is such that 100% of an area is scanned using a technique sensitive enough to detect concentrations sufficiently less than the applicable investigation level, compliance with the Type I error rate is demonstrated by using the survey data to directly compute the critical value for the data set and then subtracting this value from the applicable investigation level. The Type I decision error limit was met if the following inequality is shown to be true (Investigation Level) LC z1 where z1 is the 1- percentile of the cumulative normal distribution function and is the measurement standard deviation. The measurement standard deviation has been used as an approximation of the standard deviation of the distribution of the null hypothesis (where mean =

median = investigation level).

For and values of 0.05, the value of the cumulative normal distribution function for 1- or 1-is 1.645. Values of Z1- or Z1- for various decision error limits ( or ) are tabulated below in Table 1-3I.

or Z1- or Z1- or Z1- or Z1-0.01 2.3263 0.14 1.0803 0.02 2.0537 0.15 1.0364 0.03 1.8808 0.16 0.9945 0.04 1.7507 0.17 0.9542 0.05 1.6449 0.18 0.9154 0.06 1.5548 0.19 0.8779 0.07 1.4758 0.20 0.8416 0.08 1.4051 0.21 0.8064 0.09 1.3408 0.22 0.7722 0.10 1.2816 0.23 0.7388 0.11 1.2265 0.24 0.7063 0.12 1.1750 0.25 0.6745 0.13 1.1264 0.30 0.5244 Table 1-3I - Values of the cumulative normal distribution function for various decision error limits In cases where the MARSSIM statistical tests may be used to evaluate survey data against a given release criterion, compliance with decision error limits is a matter of ensuring that at least some minimum number of measurements are collected for both the survey area of interest and its reference area. The required minimum number of measurements depends on the statistical test to be used, the relative shift (as defined in the MARSSIM), and the applicable Type I and Type II decision error limits. The minimum number of measurements required will be computed as prescribed in the MARSSIM for the statistical test to be employed.

Page I-7