The sediment sample was collected from the vicinity of the Nine Mile Point Reactor located in the town of Scriba, approximately five miles northeast of Oswego, New York (Figure 2). This is a sediment sample collected from the shoreline of Lake Ontario boat launch collected right where the water meets land. Distance from the reactor is ~1.44 miles E of Unit 2 with a heading of 82 deg. Depth is skimmed off the top at approximately 2 cm and sifted in the field to remove large rocks. The sample was kept in a cooler and brought to the lab where the sample was kept at 105˚C overnight to dry and gain constant weight. The sample was then weighed and transferred into 800 ML Marinelli beaker up to the top and sealed with Phenoseal Vinyl Adhesive Caulk (PHENOSEAL, Baltimore, MD 21224) and black electric tape (S-17841; Uline, Pleasant Prairie, WI

53158) to avoid the escape of ²²²Rn gas from the container. The sediment sample was kept for four weeks to reach ²²⁶Ra in equilibrium with ²²²Rn Progeny (²¹⁴Pb and ²¹⁴Bi) before counting. The sample was counted for 60,000 s on all HPGe detectors. The background was counted for 240,000 s. The characteristics of HPGe detectors are given in Table 1.

In this study, we have utilized five p-type high purity Germanium co-axial detectors (HPGe) having relative efficiencies varying from 80% to 140% by Mirion Technologies (Canberra) Inc. (Meriden, CT, USA). The detectors (except GE11) are situated inside a 15-cm thick wall steel room made of pre-World War II steel (Dixie Manufacturing Co., Baltimore, MD, USA), which is located under a 47-story building providing 33 meters of water equivalent (mwe) shielding from cosmic rays in the vertical direction. GE03, GE08, and GE09 detectors have general low-background lead shields and are equipped with custom active muon shields made of plastic scintillators positioned on the top, which reduce cosmic muon background by a factor of 2. The 140% detector (GE12) has a custom 3-layer ultra-low background lead shield of a 17 cm total thickness, which is surrounded by plastic scintillators from five sides for muon rejection ( Khan et al., 2014 ). This spectrometer has an integrated background rate of 2.4 counts per minute (cpm) in the gamma energy range between 50 - 2700 keV, corresponding to a background rate of 15 counts∙ks⁻¹∙kg⁻¹ Ge.

To quantify radionuclides by their gamma rays, calibration of Ge gamma spectrometers is performed which involves determination of detector peak efficiency as a function of gamma energy and geometry. In addition, the density and coincidence-summing (cascade-summing) corrections to the peak efficiency are applied, the latter requiring either measured or calculated total efficiency. One can categorize the variety of the HPGe spectrometer calibration techniques into four major categories: empirical, semi-empirical, computational Monte Carlo (MC), and efficiency transfer. We used a semi-empirical method that is based on a direct measurement of the efficiency at a given geometry as a function of gamma energy for a set of radionuclides in a mild acidic solution (often called calibration in water). The measured water peak efficiencies, ε_p, are corrected for the coincidence-summing (as in Equation (5) below). Then, a smooth efficiency curve is often fitted to the water calibration using algorithms ln(ε_p) against polynomials of (lnE)ⁿ or (ln(c/E))ⁿ, where E is gamma energy, n is the polynomial order, and c is a constant. There are also other algorithms that were not used here ( Gilmore, 2011 ). When analyzing activity in real samples, the corrections are applied to the efficiency curve, to convert the peak efficiency for water at a particular gamma energy to the peak efficiency of the matrix and a particular radionuclide at the same energy as follows:

ε p ( matrix ; radionuclide ) = ε p ( water ; energy ) C d C s (5)

where C_d is the density correction factor and C_s is the coincidence-summing correction factor.

The NIST traceable mixed-gamma radionuclide standards containing ²⁴¹Am, ¹⁰⁹Cd, ⁵⁷Co, ¹³⁹Ce, ²⁰³Hg, ¹¹³Sn, ⁸⁵Sr, ¹³⁷Cs, ⁵⁴Mn, ⁸⁸Y, ⁶⁵Zn, ⁶⁰Co were provided by EZ Analytics, Atlanta, GA, USA. The radionuclides and their standardized gamma energies are listed elsewhere ( Semkow et al., 2015 ). Gamma-ray spectra were acquired using Apex Gamma software of Mirion Technologies (Canberra) Inc. (Meriden, CT, USA) however, the spectra were analyzed using Genie 2000 software for the sediment sample as well as for efficiency calibration.

Calculations of radionuclide activities need to account for emission probabilities (Table 2), full energy peak efficiency, geometry-dependent true coincidence summing, density correction factor, and background characteristics of the detector. The gamma activity using the HPGe detector can be measured by the

following equation ( Joel et al., 2017 ):

A ( Bq ⋅ kg − 1 ) = N s t s − N B t B M s ∗ E ∗ I γ ∗ C d ∗ C s (6)

where A denotes the activity concentration of a radionuclide, N_s/t_s, and N_B/t_B are the count rates of the radionuclide and background, M_s is the mass of the sample, E is full energy peak efficiency, and I_γ denotes an emission probability of the specific γ-ray, C_d is the density correction factor and C_s is the coincidence-summing correction factor. Density and coincidence summing corrections were calculated with the Gespecor program ( Sima et al., 2001 ). Table 2 also lists gamma energies used for the activity of the radionuclides and emission probabilities for key members of the ²³⁸U, ²³²Th, and ²³⁵U series. For radionuclides with multiple energy lines, a weighted mean of each gamma line’s measured activities was taken. The other main natural source, ⁴⁰K, emits 1461 keV photons with a probability of 10.66% ± 0.16%.

The error σ_A for a radionuclide activity is calculated using:

( σ A A ) 2 = ( σ s S ) 2 ± ( σ E E ) 2 ± ( σ I γ I γ ) 2 ± ( σ C d C d ) 2 ± ( σ C s C s ) 2 (7)

The uncertainty (σ_A) of activity calculation is a built-in function of the Genie2000 software. Our reported total uncertainty is expressed at the 95% confidence level, i.e., 1.96σ.

Minimum detectable activity (MDA) is defined by ( Gilmore, 2011 ) as being the smallest quantity of activity that can be detected in a sample with a predefined probability of 95% (approximately k = 2), considering of 5% probability of making type I errors (false positive) and making type II errors (false negative). MDA of a particular radionuclide is calculated by Genie 2000 software (Mirion Tech) which used the Currie (1968) formulation as given below:

MDA = k 2 ± 2 k B + B N 2 m M s ∗ E ∗ I γ ∗ t ∗ C d ∗ C s (8)

where B is the background, E is the detection efficiency, I_γ is the emission probability of the specific γ-ray, t is the measurement time in sec, N is the number of channels of the region of interest (ROI) and m is the number of channels to the left and to the right of the ROI. k is a confidence level equal to 1.645. The sediment sample was counted for 60,000 s at each detector and the background was counted for 240,000 s. Table 3 gives the MDA of radionuclides for specific energy in the sediment sample as well as in the distilled water sample (method blank) measured at each detector for 60,000 s. The verification of the radionuclides’ activity and the capability in terms of precision and accuracy of our gamma spectrometry measurements were established by using a Certified Reference Material called Irish sea sediment (IAEA-385) under the same conditions.

²³²Th decays to stable ²⁰⁸Pb by a series of alpha and beta decay modes. The decay of the daughters is always accompanied by the emission of gamma radiation and this mode of decay provides an opportunity to apply low-level gamma spectrometry for the determination of ²²⁸Ra and its progeny. ²³²Th is less soluble than ²³⁸U and found mostly in secular equilibrium with its progeny i.e., ²²⁸Ra, ²¹²Pb, ²¹²Bi, and ²⁰⁸Tl in rocks, soil, and sediments. However, it should be noted that in the case of ²⁰⁸Tl, only 35.94% of ²¹²Bi decay leads to this nuclide while 100% of ²²⁸Ac or ²¹²Pb activity is in secular equilibrium with ²³²Th. The results of the ²²⁸Ra, ²¹²Pb, ²¹²Bi, and ²⁰⁸Tl are presented in Table 4. ²²⁸Ra activity in the sediment sample is measured by the weighted mean of the activities of the ²²⁸Ac peaks of 270.24 keV, 338.32 keV, 911.20 keV, 964.77 keV, 968.97 keV, and 1588.2 keV. ²²⁸Ac (T_1/2 = 6.15 h) is the immediate daughter of ²²⁸Ra (T_1/2 = 5.75 ± 0.03 y) and expected in secular equilibrium with ²²⁸Ra in the sediment sample. ²¹²Pb activity was measured by the gamma peaks of 238.63 keV and 300.09 keV; ²¹²Bi activity was determined by using gamma energy of 727.33 keV and ²⁰⁸Tl activity was calculated using the weighted mean activities of gamma peaks 583.19 keV, 860.56 keV and 2614.51 keV. The gamma activity of the radionuclides in the sediment sample measured by available HPGe detectors is agreed very well within uncertainty. The average activities of the radionuclides ²²⁸Ra, ²¹²Pb, ²¹²Bi and ²⁰⁸Tl in the sediment sample are found to be 22.7 ± 0.27 Bq∙kg⁻¹, 22.5 ± 0.23 Bq∙kg⁻¹, 21.8 ± 0.32 Bq∙kg⁻¹and 21.7 ± 0.33 Bq∙kg⁻¹, respectively. The ²³²Th series radionuclides activity agrees with each other which shows that the ²³²Th is in secular equilibrium with its radioactive progeny.

²³⁵U emits several gamma lines between 70 and 300 keV however the main gamma line at 185.71 keV (emission probability 57.2%) is interfered with the ²²⁶Ra gamma line at 186.21 keV. The best gamma line is at 205.31 keV which is sufficiently well separated from other likely peaks (e.g., 204 keV peak from ²²⁸Ac) to use for determination although its low intensity (4.7%) precludes its use in many samples. Two other lines of interest are those at 143.76 keV and 163.33 keV with intensities of 10.5% and 4.7%, respectively, however, the former is interfered with ²³⁰Th peak at (143.87 keV; 0.048%), and by the 144.2 keV emission of the daughter ²²³Ra. The interference at 163.1 keV (0.154%) of the short-lived ²³¹Th is assumed to be in equilibrium with ²³⁵U. We used the weighted mean of the activities of gamma lines at 143.76 keV, 163.33 keV, and 205.31 keV and the results are presented in Table 4.

The measured ²³⁵U activity in the sediment sample shows (Figure 9) good precision in all HPGe detectors. The ²³⁵U activity varied from 1.12 ± 0.4 Bq∙kg⁻¹ to 1.39 ± 0.4 Bq∙kg⁻¹ with a mean value of 1.22 ± 0.18. ²³⁵U activity measured by GE11 and GE12 detectors show 1.26 ± 0.4 Bq∙kg⁻¹ and 1.39 ± 0.4 Bq∙kg⁻¹, respectively. Our results suggest that ²³⁵U activity in the environmental sample can be

measured precisely and accurately by low background gamma spectrometry. The activity ratio of the radionuclides ²³⁵U to ²³⁸U is constant in all environmental samples, except where contaminated by processed uranium or taken from certain specific, well-known locations. The ²³⁵U/²³⁸U activity ratio is 0.046 ± 0.007 ( Dowdall et al., 2004 ). The mean activity ratio of the ²³⁵U to ²³⁸U in the sediment sample is found to be 0.055 ± 0.007. The high ²³⁵U/²³⁸U ratio either may be uncertainty in ²³⁵U measurements or washed-out of ²³⁸U from the surficial sediment.

⁴⁰K is a primordial natural radionuclide and is present in all rocks, sediment, and soil however ¹³⁷Cs is a manmade radioactive isotope released in the atmosphere by nuclear testing during the 1950s and 1960s and nuclear accidents like Chernobyl in 1986 and Fukushima in 2011. The activity of ¹³⁷Cs was calculated from the 661.6 keV peak and ⁴⁰K from the 1461.0 keV peak by HPGe detectors. The ⁴⁰K activity in the sediment sample is presented in Table 4. The ⁴⁰K activity varies from 549 ± 11 Bq∙kg⁻¹ to 557 ± 11 Bq∙kg⁻¹ with a mean of 554 ± 5 Bq∙kg⁻¹. ⁴⁰K activity in the sediment sample was measured very precisely with our gamma spectrometers. ¹³⁷Cs activity in the sediment sample varies from 0.84 ± 0.06 to 0.95 ± 0.05 Bq∙kg⁻¹ among the five HPGe detectors. ¹³⁷Cs value in the sediment sample is very low and measured very precisely by all detectors presented in Table 4. The ¹³⁷Cs values are in good agreement with uncertainty.

Certified reference materials (CRMs) and Standard Reference Materials (SRMs) are important for method development and validation. They can indicate the need to improve or change existing methods and/or the need for further training. It is well known that the use of Proficiency Tests (PTs) and CRMs is among the most important quality control tools to produce reliable and valid measurement results and to fulfill international standards requirements regarding method validation, quality control, and metrological traceability. IAEA-385 Irish Sea sediment ( Pham et al., 2008 ) was used for verification of our gamma spectrometry results for radionuclides in the sediment sample. 55 g of IAEA-385 sample was measured on GE08, GE09 and GE12 detector. The sample was counted for 60,000 s on each detector. The results of the radionuclide activities are presented in Table 5. The same gamma energy lines were chosen to calculate the activities of individual radionuclides as discussed in section 3.4. Good agreement was found with the certificate values and the measured values by our low background gamma spectrometry measurement of radionuclides with some exceptions. Th-234 (63.3 keV) activity by GE08 shows 16.8 ± 13.4 Bq∙kg⁻¹ which is lower than the certificate value with large uncertainty may be due to the low intensity (I_γ = 3.7% ± 0.4%) of gamma-ray (Figure 10), however, the activity of ²³⁸U measured by GE09 and GE12 show good agreement with the certificate value. The precision of the measured activity of ²³⁵U in GE08, GE09, and GE12 show

*Not detected.

2.0 ± 1.8 Bq∙kg⁻¹, 1.85 ± 1.3 Bq∙kg^−1, and 2.03 ± 0.6 Bq∙kg⁻¹ which agrees with the certificate value of 1.36 ± 0.11 Bq∙kg⁻¹ (95%) confidence interval is (1.24 – 1.51 Bq∙kg⁻¹) as shown in Figure 11. The certificate value of ²³⁵U activity is based on the 16 measurements either by Alpha Spectrometry or ICPMS. Our measured activity values of 15.3 ± 0.5 Bq∙kg⁻¹, 15.1 ± 0.5 Bq∙kg⁻¹_, and 15.8 Bq∙kg⁻¹ on GE08, GE09, and GE12 HPGe detectors for ¹³⁷Cs do not agree with the certificate value of 19.8 ± 0.6 Bq∙kg⁻¹ presented in Table 5, however, our ¹³⁷Cs activity results show good agreement and precision among three detectors. To validate our ¹³⁷Cs results further, we used a PT soil sample from IAEA-CU-2006-03 World-Wide Open Proficiency Test on the Determination of Gamma Emitting Radionuclides ( Shakhashiro et al., 2007 ). Our laboratory participated in this PT exercise in 2006. Soil sample from the 2006 PT batch was analyzed for this study on GE08 and GE12 for 60000 s and the results agree with the PT value (Figure 12). The results of the Certified Reference Material IAEA-385 sediment and 2006 IAEA PT soil sample show the accuracy and precision of NCL’s low background gamma measurement capability for measuring radionuclides in environmental samples, particularly for ²³⁸U and ²³⁵U.