Preliminary Evaluation of Derived Concentration Guideline Level for Surface Soil at Wolsong NPP Site Using RESRAD-ONSITE Code

Abstract

A preliminary derived concentration guideline level (DCGL) was calculated for the soil of the Wolsong nuclear power plant (NPP) (the first commercial pressurized heavy water reactor in South Korea) site using the RESRAD-ONSITE computational code. In total, fourteen selected radionuclides were analyzed after considering the preliminary evaluation information of radionuclides observed in the pressure tube specimen in the Wolsong Unit 1 heavy water reactor and previous NPP decommissioning cases. Furthermore, a geological structure model of the Wolsong NPP site was established according to the safety analysis report of the Wolsong NPP. In addition, the distribution coefficients (Kd) of various radionuclides were derived using the JAEA-SDB and the pH information of the groundwater around the Wolsong NPP site. The DCGL for surface soil of the Wolsong NPP site was derived via the application of criteria for the site release to facilitate unrestricted reuse. Moreover, preliminary dose evaluation and relevant analysis were performed according to the Wolsong NPP site resident scenario. The novelty of this study lies in the first calculation of the preliminary DCGL values for the case of the pressurized heavy water reactor (Wolsong NPP) site. It is expected that further reliable DCGL results might be achievable if more precise radionuclide information and site-specific parameters with respect to the Wolsong NPP site are secured and applied in the future.

1. Introduction

South Korea’s first commercial pressurized heavy water reactor (PHWR), Wolsong Unit 1, was permanently shut down in 2019 in the absence of any record of decommissioning heavy water reactors worldwide. Currently, South Korea is exerting efforts in preparation for nuclear power plant (NPP) decommissioning through the establishment of the world’s first heavy water reactor decommissioning institute.

Wolsong Unit 1 is a pressurized heavy-water reactor type (Canada Deuterium Uranium, CANDU) and, owing to the lack of research and results of decommissioning such NPPs, extensive prior research is necessary before the NPP decommissioning. The U.S. Multi-Agency Radiation Survey and Site Investigation Manual (MARSSIM) presents detailed standards for designing, planning, sampling, interpreting, and quality assuring for irradiation through consistent standard guidelines for site survey and instructs the inference of the derived concentration guideline level (DCGL) for site reuse [1]. In the U.S., 10 CFR Part 20, Subpart E prescribes that the exposure dose to the receptor considering all possible exposure pathways should not exceed 0.25 mSv·year⁻¹ based on the effective dose in case the business operator reuses the site in an unrestricted manner after decommissioning. In the case of South Korea, the Nuclear Safety and Security Commission Notification No. 2016-33 presents the standard of 0.1 mSv·year⁻¹.

To reuse the site after NPP decommissioning, the DCGL must be derived from the evaluation of the exposure dose for the site using an evaluation model [2] based on a computer code such as RESRAD. The RESRAD computational code, developed by the Argonne National Laboratory (ANL) in the U.S., is designed to evaluate the risk and exposure dose of residual radioactive materials and is used for the decommissioning of commercial NPPs in the U.S. under the approval of the U.S. Nuclear Regulatory Commission (NRC). The Rancho Seco [3] and Zion [4] NPPs in the U.S. are typical examples that applied the RESRAD computer code to NPP decommissioning. The Rancho Seco NPP was a pressurized water reactor (PWR) plant in California, USA which started commercial operation in 1975 and has now been fully decommissioned. To determine the list of potential radionuclides in this plant decommissioning case, a list of nuclides reflecting site characteristics was derived by referring to literature such as NUREG/CR-0130 [5], NUREG/CR-3474 [6], and NUREG/CR-4289 [7]. Furthermore, the DCGL criteria for surface soil, building, and materials were established via the application of the RESRAD computer code [3]. Meanwhile, the Zion NPP was another PWR type reactor that started commercial operation in 1973 and was permanently shut down in 1998 after 15 years. The information regarding the NUREG/CR-3474 [6], 4289 [7], WINCO-1191 [8], and site waste specimens needed to derive a list of site-specific nuclides and the DCGL criteria for site decommissioning were derived by using the RESRAD code [4].

The objective of this study was to obtain a preliminary DCGL for the soil of the Wolsong NPP site, which is a heavy water reactor, using the RESRAD-ONSITE computational code. The target radionuclides were selected by referring to previous studies on NPP decommissioning, and a geological structure model of the Wolsong NPP site was established. Further, the distribution coefficient reflecting the site characteristics was derived using the JAEA-SDB [9] and the pH information of groundwater around the Wolsong NPP site. The preliminary DCGL, satisfying the limit of the unrestricted reuse criteria, was derived based on the various site-specific input parameters as input data of the RESRAD-ONSITE code. Furthermore, the uncertainty of the input parameters for the result of exposure dose was quantified using the probabilistic analysis feature of the RESRAD-ONSITE. In addition, the effects of the variables used in the calculation of the RESRAD code on the result of exposure dose were analyzed by performing a sensitivity analysis.

2. Materials and Methods

2.1. RESRAD-ONSITE Computational Code

The RESRAD (RESidual RADioactivity) computer code [10] was developed in 1989 by the ANL in the U.S. and is used in various countries, institutions, companies, and research groups. It is used to evaluate the radiation exposure of humans, animals, and plants in the ecosystem by residual radioactive materials and the resulting potential exposure dose and risk. In particular, the RESRAD-ONSITE code is used to calculate the radiation exposure of receptors in contaminated sites. In the U.S., the RESRAD-ONSITE code is widely used to satisfy the site opening criteria considering both social/economic factors and technical validity by the Nuclear Regulatory Commission (NRC) and the Department of Energy (DOE) when decommissioning an NPP. The primary exposure pathways of the RESRAD-ONSITE include external radiation, inhalation, and ingestion. The exposure dose by external radiation is calculated by the dose to which the receptor was exposed at 1 m from the ground and via the application of correction factors such as area, thickness, shape, and shielding of the contaminated zone. Further, the exposure pathways by inhalation primarily include the inhalation of radon decay products and contaminated dust. Finally, the exposure pathways by ingestion include exposure of receptors to radioactive substances in food, water, and soil.

2.2. Target Radionuclides

The radionuclides to be analyzed were selected using the radionuclide selection method, applied to the decommissioning of the Rancho Seco [3] and Zion [4] NPPs in the U.S., for 37 radionuclides detected in the Wolsong Unit 1 pressure tube specimens from a study on the evaluation of heavy water reactor source terms [11].

As a CANDU-type heavy water reactor, the Wolsong Unit 1 is composed of 380 nuclear fuel channels and each nuclear fuel channel is comprised of a pressure tube and calandria tube. The calandria tube acts as the pressure and thermal shielding boundary between the heavy water moderator and the coolant channel, whereas the pressure tube supports the nuclear fuel at the pressure boundary and nuclear reactor core to maintain the pressure so that the coolant does not boil. The pressure tube is made of Zr alloy containing 2.5% of Nb, with a total length, inner radius, outer radius, and density of 6.4 m, 5.2 cm, 5.6 cm, and 6.6 g·cm⁻³, respectively [11]. According to the Wolsong Unit 1 source term evaluation report, 37 radionuclides were derived from the preliminary evaluation of the radioactivity of the pressure tube using the ORIGEN2 code for 9 pressure tube specimens of the Wolsong Unit 1 [11]. When considering the License Termination Plan (LTP) of the Rancho Seco power plant, nuclides with a half-life of fewer than 2 years were excluded from the target radionuclides because their effect on the result was considered insignificant [3]. Thus, based on ICRP 107 [12], among the 37 radionuclides derived above from the pressure tube of the Wolsong Unit 1, 10 radionuclides were found to exhibit a half-life of fewer than 2 years (⁵⁴Mn, ⁹⁰Y, ¹⁰⁸Ag, ¹⁰⁹Cd, ^109mAg, ^125mTe, ^137mBa, ¹⁸²Ta, ¹⁹²Ir, and ^192mIr), while the remaining 27 radionuclides exhibited a half-life of 2 years or longer. Therefore, this study also excluded radionuclides with a half-life of fewer than 2 years from the evaluation by applying the sample case of the Rancho Seco LTP.

Further, a study related to the decommissioning of the Zion NPP in the U.S. reported a list of relative fractions of radionuclides based on the attenuation-corrected radioactivity concentration for the list of nuclides detected in 19 samples collected from the Zion NPP site [4]. Consequently, on analyzing the 27 target radionuclides while referring to the literature [4] related to the decommissioning of the Zion NPP and the result of the preliminary evaluation of the radioactivity of pressure tube specimens from the Wolsong Unit 1, 12 radionuclides (³H, ¹⁴C, ⁵⁵Fe, ⁶⁰Co, ⁶³Ni, ⁹⁰Sr, ⁹³Zr, ^93mNb, ⁹⁴Nb, ^108mAg, ^121mSn, and ¹²⁵Sb) were found to exhibit a half-life of 2 years or longer with a relative fraction of 0.01% or higher. The sum of their relative fractions was ca. 99.42%. In contrast, another 15 radionuclides yielded a relative fraction of less than 0.01% with the sum of their relative fractions being ca. 0.02%. Furthermore, the sum of relative fractions of the other 10 radionuclides with a half-life of fewer than 2 years was calculated to be ca. 0.56% [11]. In this study, the 25 radionuclides with a relative fraction of less than 0.01% were excluded from evaluation owing to the small sum of their fractions, which exerts a negligible effect on the result. Furthermore, in addition to the 12 radionuclides for analysis, the radionuclides ¹³⁴Cs and ¹³⁷Cs, which were detected in the previous study on the Kori Unit 1, Rancho Seco, and Zion NPPs, were included; thus, 14 radionuclides were finally selected as the target of analysis, and they are listed in

Table 1. Target radionuclides for the derivation of the preliminary DCGL.

Radionuclide	Relative Fraction of Radioactivity (%)	Decay Constant (s−1)
108mAg	5.1	5.26 × 10−11
14C	2.5	3.86 × 10−12
60Co	7.6	4.17 × 10−9
134Cs	7.9 × 10−5	1.06 × 10−8
137Cs	4.9 × 10−5	7.29 × 10−10
55Fe	8.1	8.03 × 10−9
3H	6.2 × 10−2	1.78 × 10−9
93mNb	1.5	1.36 × 10−9
94Nb	70.6	1.08 × 10−12
63Ni	1.2	2.20 × 10−10
125Sb	0.3	7.97 × 10−9
121mSn	1.2 × 10−2	5.01 × 10−10
90Sr	2.0 × 10−2	7.63 × 10−10
93Zr	2.5	1.44 × 10−14

[3,4,13].

2.3. Geological Structure Model

The Wolsong NPP site is located on the southeastern coast of South Korea and the geological structure of the site comprises sediments of the Cretaceous rocks strongly affected by volcanic activities, volcanic sediments, faults, or intrusions [14]. The rocks are mainly composed of argillite, quartz andesite, quartz diorite, granitic rocks, etc. Further, the thicknesses of the sedimentary layers distributed in the Wolsong NPP site have been reported to be in the range of 1–5 m (Wolsong unit 2), 6–15 m (Wolsong unit 3), and 6–12 m (Wolsong unit 4) [14]. In the present work, the thickness of the contaminated zone was assumed as 0.15 m by referring to the case of the Rancho Seco NPP, and the thickness of the unsaturated zone was set to be 1–15 m, considering the thickness of the sedimentary layer of Wolsong NPP site (see Figure 1). Furthermore, the geological features of the contaminated and unsaturated zones were assumed to be sand, soil, loam, etc., whereas that of the saturated zone was assumed to be granitic rocks, based on the typical geological features of South Korea [15].

2.4. Distribution Coefficient

The Japan Atomic Energy Agency (JAEA) provides the JAEA sorption database (JAEA-SDB) for utilization in the disposal safety evaluation modeling of radioactive waste [9]. The distribution coefficient (Kd) of radionuclide is a major variable used to predict the migration and retardation behavior of radionuclides in the groundwater system. The Kd values used in this study were selected from the JAEA-SDB considering the geological structure model of the Wolsong NPP site that was assumed above and the mean pH value of the groundwater around the Wolsong NPP site. The pH values of groundwater were obtained from the National Groundwater Information Center (https://www.gims.go.kr; accessed on 31 August 2021) which provides comprehensive groundwater information on South Korea. Consequently, the pH of the natural groundwater of Gyeongju City, which is near the Wolsong NPP site, 5.83–8.40, was used to derive the Kd values.

Table 2. Distribution coefficients of various radionuclides.

Radionuclides	Zones	Log of Kd (cm3·g−1)	Reference
Ag	Contaminated zone Unsaturated zone	2.07 ± 0.65	[9]
	Saturated zone	2.03 ± 1.09 a	[9,16]
C	Contaminated zone Unsaturated zone	0.54 ± 0.59	[9]
	Saturated zone	0.0 b	[16]
Co	Contaminated zone Unsaturated zone	1.87 ± 0.71	[9]
	Saturated zone	2.65 ± 0.24	[9]
Cs	Contaminated zone Unsaturated zone	2.52 ± 0.99	[9]
	Saturated zone	2.03 ± 1.09	[9]
Fe	Contaminated zone Unsaturated zone	1.55 ± 0.80	[9]
	Saturated zone	1.55 ± 0.80	[9]
H	Contaminated zone Unsaturated zone	non-absorbable	[9]
	Saturated zone	non-absorbable	[9]
Nb	Contaminated zone Unsaturated zone	2.93 ± 0.48	[9]
	Saturated zone	2.41 ± 0.57	[9]
Ni	Contaminated zone Unsaturated zone	2.82 ± 0.65	[9]
	Saturated zone	1.80 ± 0.72	[9]
Sb	Contaminated zone Unsaturated zone	0.87 ± 0.55	[9]
	Saturated zone	3.05 ± 0.37	[9]
Sn	Contaminated zone Unsaturated zone	3.52 ± 0.70	[9]
	Saturated zone	3.20 ± 0.94	[9]
Sr	Contaminated zone Unsaturated zone	1.18 ± 1.03	[9]
	Saturated zone	0.58 ± 0.94	[9]
Zr	Contaminated zone Unsaturated zone	2.78 ± 0.50	[9]
	Saturated zone	2.62 ± 1.22	[9]

^a: Kd value of Cs is applied based on the chemical analogy; ^b: deterministic value.

shows the Kd values for 14 target radionuclides derived from the JAEA-SDB. Since the JAEA-SDB does not provide the Kd value of Ag for granitic rock that satisfies the above pH condition, the present work replaced the Kd value of Ag with that of Cs for granitic rock by referring to the literature [16].

Further, there is a lack of data regarding Fe and C for granitic rock in the JAEA-SDB. Thus, it was assumed that the Kd values of Fe for unsaturated zone and granitic rock were identical to each other. Whereas for C, the Kd value suggested in the literature [16] was selectively applied; however, a deterministic single value was input, instead of a distribution function.

2.5. Input Parameters

The input parameters database of the RESRAD-ONSITE code used in the present work for the Wolsong NPP was established by referring to overseas NPP decommissioning cases [3,4], previous studies [17,18,19], and the site-specific variables of the Wolsong NPP site [10,14,20,21,22,23,24,25,26]. Consequently, based on the geological structure model of the Wolsong NPP site that was assumed above, the hydrological input parameters for each zone were taken from the NUREG/CR-6697 [17]. The area of the contaminated zone, which is a characteristic variable of the Wolsong NPP site, was applied based on a single unit of the Wolsong NPP, and the thickness of the contaminated zone was set to 0.15 m by referring to the case of the Rancho Seco NPP [3,14]. Furthermore, with respect to the density of the saturated zone, the averaged data for quartz andesite, quartz diorite, and andesite, reported as the main bedrocks in the Wolsong Final Safety Analysis Report (FSAR), were employed. The hydraulic conductivity was input by referring to the data of a previous study on field permeability tests in the Wolsong NPP site [14,20]. Moreover, for the water transport model, the non-dispersion model proposed by default in the RESRAD-ONSITE code for sites with a contaminated zone larger than 1000 m² was applied [10]. Furthermore, the input parameters related to diet in a contaminated site, assuming resident exposure scenario, were input by referring to the domestic annual consumption reports of South Korea [21,22,23,24,25]. In addition, regarding the precipitation and wind speed, the mean values of climate data measured in Gyeongju city from 2011 to 2020 were employed [26]. The detailed input parameters used in this study are listed in

Table 3. List of input parameters related to the characteristics of the Wolsong NPP site.

Parameter	Unit	Value
Area of contaminated zone	m2	49,518 [14]
Density of saturated zone	g·cm−3	2.6 [14]
Model for water transport parameters	-	non-dispersion [10]
Precipitation	m·year−1	1.1 [26]
Wind speed	m·s−1	2.5 [26]
Hydraulic conductivity in the saturated zone	m·year−1	15.8 [20]
Fruit, vegetable, and grain consumption	kg·year−1	286.4 [21]
Milk consumption	L·year−1	83.9 [22]
Meat and poultry consumption	kg·year−1	54.6 [23]
Fish consumption	kg·year−1	24.1 [24]
Other seafood consumption	kg·year−1	43.2 [24]
Drinking water intake	L·year−1	312.9 [25]

and

Table A1. Input parameter database for RESRAD-ONSITE code employed in the present work.

No	Parameter	Value	Unit	Reference
Source
1	Radionuclide concentration	0.037	Bq·g−1	Initial concentration (=1 pCi·g−1)
Transport factors
2	Number of unsaturated zones	1	-	NUREG/CR-7267 [19]
3	Time since placement of material	0	year	NUREG/CR-7267 [19]
4	Groundwater concentration	0	pCi·L−1	NUREG/CR-7267 [19]
5	Leach rate	(Nuclide specific)	year−1	NUREG/CR-7267 [19]
6	Solubility limit	(Nuclide specific)	mol·L−1	NUREG/CR-7267 [19]
7	Use plant/soil ratio	Unchecked	-	NUREG/CR-7267 [19]
Calculation parameters
8	Times for calculations	1, 3, 10, 30, 100, 300, 1000	year	NUREG/CR-7267 [19]
Contaminated zone parameters
9	Area of contaminated zone	49,518	m2	Area of Wolsong NPP unit 1 [14]
10	Thickness of contaminated zone	0.15	m	Case of Rancho Seco NPP [3]
11	Length parallel to aquifer flow	100	m	NUREG/CR-7267 [19]
12	Does the initial contamination penetrate the water table	Unchecked	-	NUREG/CR-7267 [19]
13	Contaminated fraction below the water table	0	-	NUREG/CR-7267 [19]
Cover and contaminated zone hydrological data
14	Cover depth	0	m	NUREG/CR-7267 [19]
15	Cover erosion rate	0.001	m·year−1	NUREG/CR-7267 [19]
16	Density of contaminated zone	Normal distribution 1: 1.5635, 2: 0.2385	g·cm−3	NUREG/CR-6697 [17]
17	Contaminated zone total porosity	Normal distribution 1: 0.41, 2: 0.0899	-	NUREG/CR-6697 [17]
18	Contaminated zone field capacity	0.18	-	C. Yu [18]
19	Contaminated zone erosion rate	0.001	m·year−1	NUREG/CR-7267 [19]
20	Contaminated zone hydraulic conductivity	Bounded lognormal-n 1: 5.022, 2: 1.33, 3: 2.49, 4: 9250	m·year−1	NUREG/CR-6697 [17]
21	Contaminated zone b parameter	Bounded lognormal-n 1: 0.632, 2: 0.282, 3: 0.786, 4: 4.5	-	NUREG/CR-6697 [17]
22	Evapotranspiration coefficient	Uniform distribution min: 0.5, max: 0.75	-	NUREG/CR-6697 [17]
23	Wind speed	2.5	m·s−1	Average annual wind speed in Gyeongju city, 2011–2020 [26]
24	Precipitation	1.12	m·year−1	Average annual precipitation in Gyeongju city, 2011–2020 [26]
25	Irrigation	0.2	m·year−1	NUREG/CR-7267 [19]
26	Irrigation mode	Overhead	-	NUREG/CR-7267 [19]
27	Runoff coefficient	Uniform distribution min: 0.1, max: 0.8	-	NUREG/CR-6697 [17]
28	Watershed area for nearby stream or pond	1,000,000	m2	NUREG/CR-7267 [19]
29	Accuracy for water soil computation	0.001	-	NUREG/CR-7267 [19]
Saturated zone hydrological data
30	Density of saturated zone	2.56	g·cm−3	Case of Wolsong NPP unit 2 [14]
31	Saturated zone total porosity	0.45	-	C. Yu [18]
32	Saturated zone effective porosity	0.185	-	Average value of the effective porosity of volcanic tuff [18]
33	Saturated zone field capacity	0.2	-	NUREG/CR-7267 [19]
34	Saturated zone hydraulic conductivity	15.768	m·year−1	W. Sohn [20]
35	Saturated zone hydraulic gradient	Bounded lognormal-n 1: −5.11, 2: 1.77,3: 0.00007, 4: 0.5	-	NUREG/CR-6697 [17]
36	Saturated zone b parameter	5.3	-	NUREG/CR-7267 [19]
37	Water table drop rate	0.001	m·year−1	NUREG/CR-7267 [19]
38	Well pump intake depth (below the water table)	10	m	NUREG/CR-7267 [19]
39	Model for water transport parameters: non-dispersion or mass-balance	Non-dispersion	-	C. Yu [10]
40	Well pumping rate	250	m3·year−1	NUREG/CR-7267 [19]
Unsaturated zone parameters
41	Unsaturated zone thickness	Uniform distribution min: 1, max: 15	m	Case of Wolsong NPP [14]
42	Unsaturated zone density	Normal distribution 1: 1.5105, 2: 0.159	g·cm−3	NUREG/CR-6697 [17]
43	Unsaturated zone total porosity	Normal distribution 1: 0.43, 2: 0.06	-	NUREG/CR-6697 [17]
44	Unsaturated zone effective porosity	Normal distribution 1: 0.383, 2: 0.061	-	NUREG/CR-6697 [17]
45	Unsaturated zone field capacity	0.2	-	NUREG/CR-7267 [19]
46	Unsaturated zone, soil-specific b parameter	Bounded lognormal-n 1: −0.0253, 2: 0.216, 3: 0.501, 4: 1.9	-	NUREG/CR-6697 [17]
47	Unsaturated zone hydraulic conductivity	Bounded lognormal-n 1: 1.398, 2: 1.842, 3: 110, 4: 5870	m·year−1	NUREG/CR-6697 [17]
Occupancy, inhalation, and external gamma data
48	Inhalation rate	Triangular distribution 1: 4380, 2: 8400, 3: 13100	m3·year−1	NUREG/CR-6697 [17]
49	Mass loading for inhalation	Continuous linear distribution	g·m−3	NUREG/CR-6697 [17]
50	Exposure duration	30	year	NUREG/CR-7267 [19]
51	Inhalation shielding Factor	Uniform distribution min: 0.1, max: 0.95	-	NUREG/CR-6697 [17]
52	External gamma shielding factor	Bounded lognormal-n 1: −1.3, 2: 0.59, 3: 0.044, 4: 1	-	NUREG/CR-6697 [17]
53	Indoor time fraction	0.5	-	NUREG/CR-7267 [19]
54	Outdoor time fraction	0.25	-	NUREG/CR-7267 [19]
55	Shape of the contaminated zone (shape factor flag)	Circular	-	NUREG/CR-7267 [19]
Ingestion pathway, dietary data
56	Fruit, vegetable, and grain consumption	286.4	kg·year−1	The sum of fruit, vegetable, and grain consumption in South Korea in 2019 [21]
57	Leafy vegetable consumption	38.4	kg·year−1	Annual leafy vegetable consumption in South Korea in 2014 [34]
58	Milk consumption	83.9	L·year−1	Annual dairy product consumption in South Korea in 2020 [22]
59	Meat and poultry consumption	54.6	kg·year−1	Annual meat consumption in South Korea in 2019 [23]
60	Fish consumption	24.1	kg·year−1	Annual fish consumption in South Korea in 2019 [24]
61	Other seafood consumption	43.2	kg·year−1	Annual consumption of seaweed and shellfish in South Korea in 2019 [24]
62	Soil ingestion	Triangular distribution 1: 0, 2: 18.3, 3: 36.5	g·year−1	NUREG/CR-6697 [17]
63	Drinking water intake	312.87	L·year−1	Average annual water intake per male and female by average age group from 2013 to 2017 [25]
64	Drinking water contaminated fraction	1	-	NUREG/CR-7267 [19]
65	Household water contaminated fraction	1	-	NUREG/CR-7267 [19]
66	Livestock water contaminated fraction	1	-	NUREG/CR-7267 [19]
67	Irrigation water contaminated fraction	1	-	NUREG/CR-7267 [19]
68	Aquatic food contaminated fraction	0.5	-	NUREG/CR-7267 [19]
69	Plant food contaminated fraction	−1	-	NUREG/CR-7267 [19]
70	Meat contaminated fraction	−1	-	NUREG/CR-7267 [19]
71	Milk contaminated fraction	−1	-	NUREG/CR-7267 [19]
Ingestion pathway, non-dietary data
72	Livestock fodder intake for meat	68	kg·d−1	NUREG/CR-7267 [19]
73	Livestock fodder intake for milk	55	kg·d−1	NUREG/CR-7267 [19]
74	Livestock water intake for meat	50	L·d−1	NUREG/CR-7267 [19]
75	Livestock water intake for milk	160	L·d−1	NUREG/CR-7267 [19]
76	Livestock intake of soil	0.5	kg·d−1	NUREG/CR-7267 [19]
77	Mass loading for foliar deposition	0.0001	g·m−3	NUREG/CR-7267 [19]
78	Depth of soil mixing layer	Triangular distribution 1: 0, 2: 0.15, 3: 0.6	m	NUREG/CR-6697 [17]
79	Depth of roots	0.9	m	NUREG/CR-7267 [19]
80	Groundwater fractional usage for drinking water	1	-	NUREG/CR-7267 [19]
81	Groundwater fractional usage for household water	1	-	NUREG/CR-7267 [19]
82	Groundwater fractional usage for livestock water	1	-	NUREG/CR-7267 [19]
83	Groundwater fractional usage for irrigation water	1	-	NUREG/CR-7267 [19]
Plant factors
84	Wet-weight crop yields—non-leafy	0.7	kg·m−2	NUREG/CR-7267 [19]
85	Wet-weight crop yields—leafy	1.5	kg·m−2	NUREG/CR-7267 [19]
86	Wet-weight crop yields—fodder	1.1	kg·m−2	NUREG/CR-7267 [19]
87	Length of growing season—non-leafy	0.17	year	NUREG/CR-7267 [19]
88	Length of growing season—leafy	0.25	year	NUREG/CR-7267 [19]
89	Length of growing season—fodder	0.08	year	NUREG/CR-7267 [19]
90	Translocation factor—non-leafy	0.1	-	NUREG/CR-7267 [19]
91	Translocation factor—leafy and fodder	1	-	NUREG/CR-7267 [19]
92	Weathering removal constant	20	year−1	NUREG/CR-7267 [19]
93	Wet foliar interception fraction for non-leafy vegetable	0.25	-	NUREG/CR-7267 [19]
94	Wet foliar interception fraction for leafy vegetable	0.25	-	NUREG/CR-7267 [19]
95	Wet foliar interception fraction for fodder	0.25	-	NUREG/CR-7267 [19]
96	Dry foliar interception fraction for non-leafy vegetable	0.25	-	NUREG/CR-7267 [19]
97	Dry foliar interception fraction for leafy vegetable	0.25	-	NUREG/CR-7267 [19]
98	Dry foliar interception fraction for fodder	0.25	-	NUREG/CR-7267 [19]
Storage times before data use
99	Storage times for fruits, non-leafy vegetables, and grain	14	day	NUREG/CR-7267 [19]
100	Storage times for leafy vegetables	1	day	NUREG/CR-7267 [19]
101	Storage times for milk	1	day	NUREG/CR-7267 [19]
102	Storage times for meat	20	day	NUREG/CR-7267 [19]
103	Storage times for fish	7	day	NUREG/CR-7267 [19]
104	Storage times for crustacea and mollusks	7	day	NUREG/CR-7267 [19]
105	Storage times for well water	1	day	NUREG/CR-7267 [19]
106	Storage times for surface water	1	day	NUREG/CR-7267 [19]
107	Storage times for livestock fodder	45	day	NUREG/CR-7267 [19]
Carbon-14 data
108	C-12 concentration in local water	2.0 × 10−5	g·cm−3	NUREG/CR-7267 [19]
109	C-12 concentration in contaminated soil	0.03	g·g−1	NUREG/CR-7267 [19]
110	Fraction of vegetation carbon absorbed from soil	0.02	-	NUREG/CR-7267 [19]
111	Fraction of vegetation carbon absorbed from air	0.98	-	NUREG/CR-7267 [19]
112	Thickness of evasion layer of C-14 in soil	0.3	m	NUREG/CR-7267 [19]
113	C-14 evasion flux rate from soil	7.0 × 10−7	s−1	NUREG/CR-7267 [19]
115	C-12 evasion flux rate from soil	1.0 × 10−10	s−1	NUREG/CR-7267 [19]
116	Grain fraction in livestock feed-beef cattle	0.8	-	NUREG/CR-7267 [19]
117	Grain fraction in livestock feed-cow	0.2	-	NUREG/CR-7267 [19]

in Appendix A. Table A1 presents various input parameters used in the RESRAD-ONSITE code taken from various research works. The data obtained from NUREG/CR-7267 [19] were alternatively used in case of the absence of site-specific parameters available for the Wolsong NPP site.

2.6. Probabilistic Analysis

The RESRAD-ONSITE code provides probabilistic analysis features for calculating the results from the input parameters that contain uncertainty. Following various parameters being input as distribution functions, the correlations of the results derived from them can be quantified. Subsequently, the assigned values for specific parameters can be derived and reflected in the result.

In this study, the non-linear monotonic relationship between the input parameters and the exposure dose results was estimated by referring to the LTP case of the Rancho Seco NPP. In addition, the partial rank correlation coefficient (PRCC), which provides a unique contribution to the exposure dose result, was used. If the absolute value of the PRCC exceeded 0.25, the corresponding input parameter was treated as a sensitive parameter. The uncertainty of the input parameter can be quantified by replacing the deterministic value corresponding to the 75% quartile if the PRCC value is positive, and the 25% quartile if it is negative, with the input parameter. The list of sensitive parameters for which the absolute value of PRCC exceeded 0.25 in this study is listed in

Table A2. Partial rank correlation coefficient (PRCC) and assigned value for each radionuclide derived in this study.

Radionuclides	Parameters	PRCC	Assigned Value
108mAg	External gamma shielding factor	0.97	3.98 × 10−1
	Kd of 108mAg in contaminated zone	0.74	1.23 × 101
	Density of contaminated zone	0.71	1.72 × 100
	Runoff coefficient	0.44	6.24 × 10−1
	Evapotranspiration coefficient	0.36	6.87 × 10−1
14C	Density of contaminated zone	1	1.72 × 100
	Kd of 14C in contaminated zone	0.81	2.54 × 100
	Runoff coefficient	0.59	6.24 × 10−1
	Evapotranspiration coefficient	0.55	6.87 × 10−1
60Co	External gamma shielding factor	0.96	3.98 × 10−1
	Kd of 60Co in contaminated zone	0.79	1.04 × 101
	Density of contaminated zone	0.77	1.72 × 100
	Runoff coefficient	0.44	6.24 × 10−1
	Evapotranspiration coefficient	0.31	6.87 × 10−1
134Cs	External gamma shielding factor	0.96	3.98 × 10−1
	Kd of 134Cs in contaminated zone	0.73	2.42 × 101
	Density of contaminated zone	0.62	1.72 × 100
	Runoff coefficient	0.26	6.24 × 10−1
137Cs	External gamma shielding factor	0.95	3.98 × 10−1
	Kd of 137Cs in contaminated zone	0.73	2.42 × 101
	Density of contaminated zone	0.59	1.72 × 100
	Runoff coefficient	0.27	6.24 × 10−1
55Fe	Depth of soil mixing layer	−0.94	1.50 × 10−1
	Kd of 55Fe in contaminated zone	0.58	8.03 × 100
3H	Density of contaminated zone	0.65	1.72 × 100
	Saturated zone hydraulic gradient	0.56	1.95 × 10−2
	Evapotranspiration coefficient	−0.38	5.62 × 10−1
	Thickness of unsaturated zone 1	−0.32	4.49 × 100
	Runoff coefficient	0.26	6.24 × 10−1
93mNb	External gamma shielding factor	0.93	3.98 × 10−1
	Kd of 93mNb in contaminated zone	0.74	2.58 × 101
	Soil ingestion	0.62	2.36 × 101
	Depth of soil mixing layer	−0.62	1.50 × 10−1
	Runoff coefficient	0.6	6.25 × 10−1
	Evapotranspiration coefficient	0.45	6.87 × 10−1
	Density of contaminated zone	0.39	1.72 × 100
94Nb	External gamma shielding factor	0.99	3.98 × 10−1
	Density of contaminated zone	0.87	1.72 × 100
	Kd of 94Nb in contaminated zone	0.54	2.58 × 101
	Runoff coefficient	0.35	6.25 × 10−1
63Ni	Depth of soil mixing layer	−0.94	1.50 × 10−1
	Kd of 63Ni in contaminated zone	0.63	2.58 × 101
	Runoff coefficient	0.26	6.24 × 10−1
	Density of contaminated zone	0.25	1.72 × 100
125Sb	External gamma shielding factor	0.94	3.98 × 10−1
	Kd of 125Sb in contaminated zone	0.84	3.46 × 100
	Density of contaminated zone	0.64	1.72 × 100
	Runoff coefficient	0.63	6.24 × 10−1
	Evapotranspiration coefficient	0.54	6.87 × 10−1
121mSn	Depth of soil mixing layer	−0.89	1.50 × 10−1
	External gamma shielding factor	0.8	3.98 × 10−1
90Sr	Kd of 90Sr in contaminated zone	0.97	6.45 × 100
	Runoff coefficient	0.73	6.24 × 10−1
	Evapotranspiration coefficient	0.57	6.87 × 10−1
	Density of contaminated zone	0.51	1.72 × 100
93Zr	Kd of 93Zr in saturated zone	−0.65	5.98 × 100
	Saturated zone hydraulic gradient	0.64	1.95 × 10−2
	Density of contaminated zone	0.5	1.72 × 100
	Runoff coefficient	0.41	6.24 × 10−1
	Thickness of unsaturated zone 1	−0.39	4.49 × 100
	Kd of 93Zr in unsaturated zone 1	−0.33	1.15 × 101
	Evapotranspiration coefficient	0.33	6.87 × 10−1

in Appendix A. Consequently, the deterministic value for each radionuclide was derived and the preliminary DCGL was calculated.

3. Results

NUREG/CR-1757 recommends the use of the peak of the mean dose to derive the DCGL value. According to the previous studies [2,13,27], the DCGL formula can be expressed as follows:

$$\mathrm{DCGL}=\frac{\mathrm{Regulatory}\mathrm{Dose}\mathrm{Limit}}{\mathrm{Peak}\mathrm{of}\mathrm{Mean}\mathrm{Dose}}$$

(1)

Table 4. DCGL results for various radionuclides were calculated with two different regulatory dose limits.

	Regulatory Dose Limit	DCGL (Bq·g−1)
Radionuclides		0.25 mSv·year−1	0.1 mSv·year−1
108mAg		2.56 × 10−1	1.02 × 10−1
14C		2.07 × 100	8.27 × 10−1
60Co		1.73 × 10−1	6.90 × 10−2
134Cs		2.93 × 10−1	1.17 × 10−1
137Cs		6.65 × 10−1	2.66 × 10−1
55Fe		1.47 × 103	5.87 × 102
3H		1.88 × 102	7.51 × 101
93mNb		5.63× 103	2.25 × 103
94Nb		2.62 × 10−1	1.05 × 10−1
63Ni		5.03 × 102	2.01 × 102
125Sb		1.21 × 100	4.86 × 10−1
121mSn		6.01 × 102	2.41 × 102
90Sr		8.42 × 10−1	3.37 × 10−1
93Zr		6.75 × 101	2.70 × 101

shows the preliminary DCGL results derived with the assigned value for each radionuclide through probabilistic analysis (see Table A2). As presented in Table A2, the highest PRCC values were mostly due to the external gamma shielding factor, density of contaminated zone, runoff coefficient, distribution coefficients, evapotranspiration coefficient, etc. The PRCC and assigned values will be further discussed in the following chapter. The regulatory dose limit applied in this study was 0.25 mSv·year⁻¹ according to the 10 CFR Part 20, subpart E in the U.S., and 0.1 mSv·year⁻¹ according to the Nuclear Safety and Security Commission Notification No. 2016-33 in South Korea.

On deriving the preliminary DCGLs for 14 target radionuclides, the radionuclide ^93mNb yielded the largest DCGL value, followed by ⁵⁵Fe, ^121mSn, and ⁶³Ni. Those large DCGL values of several radionuclides were seemingly caused by their large Kd values (e.g., log Kd_{saturated zone} = 2.93 ± 0.48 for Nb, 3.52 ± 0.70 for Sn, 2.82 ± 0.65 for Ni, respectively) inducing remarkable retardation of migration along with relatively short half-life (e.g., t_1/2 = 2.7 years for ⁵⁵Fe and 16 years for ^93mNb).

The result of the preliminary DCGL of the Wolsong NPP site soil derived in this study was compared with other studies [13,28,29,30,31] as presented in

Table 5. Comparison of DCGL results (Bq·g⁻¹) for various radionuclides derived with a regulatory dose limit of 0.25 mSv·year⁻¹.

Cases	CY a [28]	Yankee Rowe [29]	Zion [30]	Kori-1 [13]	IAEA [31]	Wolsong (p.w.)
Methods	RESRAD v6.2	RESRAD v6.2	RESRAD v7.0	RESRAD v7.2	IAEA Standard	RESRAD v7.2
108mAg	2.60 × 10−1	2.70 × 10−1	-	-	-	2.56 × 10−1
14C	2.10 × 10−1	2.04 × 10−1	-	8.24 × 100	2.50 × 101	2.07 × 100
60Co	1.40 × 10−1	1.48 × 10−1	1.74 × 10−1	2.10 × 10−1	2.50 × 100	1.73 × 10−1
134Cs	1.70 × 10−1	1.85 × 10−1	2.78 × 10−1	3.60 × 10−1	2.50 × 100	2.93 × 10−1
137Cs	2.90 × 10−1	3.18 × 10−1	5.81 × 10−1	8.20 × 10−1	2.50 × 100	6.65 × 10−1
55Fe	1.01 × 103	1.07 × 103	-	-	2.50 × 104	1.47 × 103
3H	1.52 × 101	1.37 × 101	-	-	2.50 × 103	1.88 × 102
93mNb	-	-	-	-	2.50 × 102	5.63 × 103
94Nb	2.60 × 10−1	2.74 × 10−1	-	-	2.50 × 100	2.62 × 10−1
63Ni	2.68 × 101	3.00 × 101	1.48 × 102	5.47 × 102	2.50 × 103	5.03 × 102
125Sb	-	1.18 × 100	-	-	2.50 × 100	1.21 × 100
121mSn	-	-	-	-	-	6.01 × 102
90Sr	6.00 × 10−2	6.29 × 10−2	5.29 × 10−1	1.75 × 100	2.50 × 101	8.42 × 10−1
93Zr	-	-	-	-	2.50 × 102	6.75 × 101

^a: Connecticut Yankee.

. The authors note that the regulatory dose limit was unified to the U.S. criterion of 0.25 mSv·year⁻¹ to establish the same criterion for the comparison among various studies.

The comparisons indicated relatively similar results for most of the radionuclides. In particular, the results of major radionuclides, such as ¹⁴C, ⁶⁰Co, ¹³⁴Cs, ¹³⁷Cs, ⁶³Ni, and ⁹⁰Sr, were relatively consistent with the values derived from Kori-1 [13]. Furthermore, there were no data in the other decommissioning cases for comparing the DCGL values for ^93mNb and ⁹³Zr. However, the radionuclide ^93mNb showed a significantly higher DCGL value compared to the IAEA standard. As explained above, this is considered to be the effect of the considerably high Kd value of ^93mNb, hindering the spread of the contamination, together with a relatively short half-life. However, although the Kd values of ^93mNb and ⁹³Zr are similar to each other, the DCGL of ⁹³Zr obtained in the present work indicated a relatively decreased tendency in comparison to the IAEA standard. It might be caused by a considerably longer half-life of ⁹³Zr (t_1/2 = 1.53 × 10⁶ years) compared with that of ^93mNb (t_1/2 = 16 years). A longer half-life means a larger number of radionuclides required to provide the same radioactivity. Due to the limited retention capacity of contaminated, unsaturated, and saturated zones, the DCGL of ⁹³Zr cannot be as high as ^93mNb. Nevertheless, it is expected that a more reliable DCGL value of various radionuclides could be derived by using the Kd value directly obtained from the practical adsorption test for Wolsong NPP site soil in the future.

4. Discussion

The sensitivity parameters and assigned values were calculated with PRCC based on various site-specific input parameters listed in Table A1 in Appendix A. For preliminary calculation of the timing for site release for unrestricted reuse after NPP decommissioning, the total exposure dose was derived by assuming the concentration in the contaminated soil of various radionuclides to 0.037 Bq·g⁻¹ (1 pCi·g⁻¹), the same initial condition employed in the previous works [13,32], and is presented in Figure 2. Consequently, the timing when the regulatory limit of 0.1 mSv·year⁻¹ (allowing unrestricted reuse of the site after decommissioning the NPP in South Korea according to Article 4 of the Nuclear Safety and Security Commission Notification No. 2016-33) is satisfied was derived to be 4.07 years after the NPP decommissioning and a final investigation of the site condition.

In addition, this study evaluated the effect of a change of input parameters on the final exposure dose result by using the sensitivity analysis feature of the RESRAD-ONSITE code. To that end, the sensitivity parameters for the 14 radionuclides were identified by applying the PRCC. Consequently, eight representative sensitivity parameters were derived as listed in

Table 6. Input parameters derived from the sensitivity analysis.

Parameters	PRCC	Quartile	Assigned Value
External gamma shielding factor	0.98	75%	3.98 × 10−1
Density of contaminated zone (g·cm−1)	0.86	75%	1.72 × 100
Runoff coefficient	0.63	75%	6.24 × 10−1
Kd of 60Co in contaminated zone (cm3·g−1)	0.51	75%	1.04 × 101
Evapotranspiration coefficient	0.46	75%	6.87 × 10−1
Kd of 108mAg in contaminated zone (cm3·g−1)	0.39	75%	1.23 × 101
Kd of 134Cs in contaminated zone (cm3·g−1)	0.38	75%	2.41 × 101
Kd of 90Sr in contaminated zone (cm3·g−1)	0.27	75%	6.52 × 100

. The sensitivity parameter that yielded the highest PRCC was the external gamma shielding factor, followed by the density of the contaminated zone, runoff coefficient, distribution coefficient (Kd), and evapotranspiration coefficient. Moreover, all the eight sensitivity parameters had positive correlations, and the deterministic values corresponding to the 75% quartile were derived accordingly.

Furthermore, an analysis to evaluate the relative effect of changes in individual sensitivity parameters on the unrestricted site reuse criteria was also performed. To that end, the change of the site reuse criteria according to the change of input parameters was evaluated while increasing and decreasing the values of each sensitivity parameter by a factor of two. However, the input values for runoff coefficient and evapotranspiration coefficient among the sensitivity parameters were limited to the range of 0 and 1. Thus, the effect on site reuse criteria was evaluated while increasing or decreasing the runoff coefficient by 60% and the evapotranspiration coefficient by 45%.

Table 7. Year for site release for the unrestricted reuse of Wolsong NPP site calculated based on the sensitivity analysis for various input parameters.

Parameters	Multiplication Factor		Year for Site Release (Years)
External gamma shielding factor	upper	×2	7.14
	mid	×1	4.07
	lower	×0.5	- a
Density of contaminated zone	upper	×2	11.19
	mid	×1	4.07
	lower	×0.5	- a
Runoff coefficient	upper	×1.6	3.44
	mid	×1	4.07
	lower	×0.63	3.64
Kd of 60Co in contaminated zone	upper	×2	4.56
	mid	×1	4.07
	lower	×0.5	1.96
Evapotranspiration coefficient	upper	×1.45	7.99
	mid	×1	4.07
	lower	×0.69	3.44
Kd of 108mAg in contaminated zone	upper	×2	4.56
	mid	×1	4.07
	lower	×0.5	3.44
Kd of 134Cs in contaminated zone	upper	×2	4.07
	mid	×1	4.07
	lower	×0.5	4.07
Kd of 90Sr in contaminated zone	upper	×2	4.31
	mid	×1	4.07
	lower	×0.5	3.85

^a: Total exposure dose is less than 0.1 mSv·year⁻¹ at t = 0 year.

summarizes the analysis results for all the eight sensitivity parameters and the relevant changes in the criteria for site release for unrestricted reuse.

The result of sensitivity analysis showed that the parameters that have a major impact on the year for site release to facilitate unrestricted reuse were the density of contaminated zone, evapotranspiration coefficient, and external gamma shielding factor. As shown in Figure 3a, if the input for the density of the contaminated zone through sensitivity analysis is doubled, the year for site release for unrestricted reuse was increased to 11.2 years, which is 2.8 times longer than the standard value of 4.07 years, because of the distinct increase in the total exposure dose. This tendency might be caused by the largest impact of external radiation among various exposure pathways on the total exposure dose, together with the increased number of radionuclides contained in the soil due to the increased density of the contaminated zone (see Figure 3b).

Furthermore, the groundwater contamination in the NPP site was mainly caused by the leaching of radionuclides from the water infiltration. This is simulated in the RESRAD-ONSITE code as follows [10]:

$${\mathrm{L}}_{\mathrm{i}}=\frac{\mathrm{R}\left(\mathrm{t}\right)}{\mathsf{\rho }·\mathrm{A}·\mathrm{T}\left(\mathrm{t}\right)·{\mathrm{S}}_{\mathrm{i}}\left(\mathrm{t}\right)}$$

(2)

where

L_i = leach rate for radionuclide i (year⁻¹)

R(t) = source strength at time t (pCi·year⁻¹)

ρ = bulk density of the contaminated zone (kg·m⁻³)

A = area of the contaminated zone (m²)

T(t) = thickness of the contaminated zone at time t (m)

S_i(t) = average concentration of the i-th principal radionuclide in the contaminated zone available for leaching at time t (Bq·kg⁻¹)

According to Equation (2), the leach rate (L_i) has an inverse relationship with the density of the contaminated zone (ρ). In other words, with an increase in the density of the contaminated zone, the leaching of radionuclides to groundwater decreases relatively. Thus, the radionuclides stay longer in the contaminated zone, which increases the direct external exposure to receptors, and consequently, the total exposure dose increases as well.

Further, the relationship between the evapotranspiration coefficient and the infiltration rate is expressed in the RESRAD-ONSITE code as follows [10]:

$$\mathrm{I}=\left(1-{\mathrm{C}}_{\mathrm{e}}\right)[\left(1-{\mathrm{C}}_{\mathrm{r}}\right)\mathrm{Pr}+\mathrm{Irr}$$

(3)

where

I = infiltration rate (m·year⁻¹)

C_e = evapotranspiration coefficient (dimensionless)

C_r = runoff coefficient (dimensionless)

Pr = precipitation rate (m·year⁻¹)

Irr = irrigation rate (m·year⁻¹)

Here, if the evapotranspiration coefficient increases, the infiltration rate of the infiltration water decreases relatively. Thus, similar to the case of the density of contaminated zone, the transport of radionuclides is relatively delayed. Consequently, there is an increase in the external exposure to receptors which increases the total exposure dose. The results of this study showed that when the evapotranspiration coefficient was increased by 1.45 times, the year for site release for unrestricted reuse increased to approximately 8 years from the reference case of 4.07 years because of a significant increase in the total exposure dose.

The external gamma shielding factor is also considered to be a major factor that significantly affects the evaluation of the receptors’ radiation exposure to radionuclides remaining in the contaminated zone. Among the parameters related to external radiation, which is a major exposure path, the external gamma shielding factor is used to derive the environmental transport factor (ETF), which is expressed as follows:

$$\mathrm{ETF}\left(\mathrm{t}\right)=\mathrm{FO}\times \mathrm{FS}\left(\mathrm{t}\right)\times \mathrm{FA}\left(\mathrm{t}\right)\times \mathrm{FCD}\left(\mathrm{t}\right)$$

(4)

$$\mathrm{FO}={\mathrm{f}}_{\mathrm{otd}}+\left({\mathrm{f}}_{\mathrm{ind}}\times {\mathrm{F}}_{\mathrm{sh}}\right)$$

(5)

where

ETF(t) = environment transport factor (dimensionless)

FO = occupancy and shielding factor (dimensionless)

FS(t) = shape factor (dimensionless)

FA(t) = nuclide-specific area factor (dimensionless)

FCD(t) = depth-and-cover factor (dimensionless)

f_otd = fraction of a year spent outdoors, on site (dimensionless)

f_ind = fraction of a year spent indoors, on site (dimensionless)

F_sh = indoor shielding factor for external gamma (dimensionless)

Thus, according to Equations (4) and (5), the environmental transport factor (ETF) increases together with the external gamma shielding factor. Consequently, the exposure dose that receptors receive via the external radiation pathway also increases. Therefore, it can be evaluated as a major factor that increases the year for site release for unrestricted reuse.

5. Conclusions

The preliminary DCGLs for the soil of the Wolsong NPP site, which is a pressurized heavy water reactor, were derived in this study according to the methodology suggested by the MARSSIM. The decommissioning cases of overseas NPPs such as Rancho Seco and Zion NPPs were considered, and various literature was referred to as well for the evaluation. The target radionuclides for calculating the DCGL were selected based on the data of radionuclides identified in the pressure tube specimens of Wolsong Unit 1 that have been reported in the literature. The input parameters were selected considering the site-specific variables of the Wolsong NPP site and employed in the RESRAD-ONSITE code. In addition, input data such as distribution coefficients were derived and applied to simulate the migration and retardation behavior of radionuclides using the JAEA-SDB [9]. Consequently, the DCGLs for the Wolsong NPP site soil with respect to the resident scenario were preliminarily derived by using the input parameter database built in the present work. A comparison of the results with that of previous studies showed that they were relatively consistent with each other.

To preliminarily derive the year for site release for unrestricted reuse, the inventory of 14 target radionuclides was set to 0.037 Bq·g⁻¹ (1 pCi·g⁻¹) in the contaminated zone, and their total exposure dose was calculated accordingly. The result showed that the exposure dose of 0.1 mSv·year⁻¹ for site reuse was satisfied at 4.07 years after the decommissioning of the Wolsong NPP and the final investigation of the site condition. Furthermore, as a result of sensitivity analysis to identify input parameters that affect the calculation of the year for site release for unrestricted reuse, the density of contaminated zone, evapotranspiration coefficient, and external gamma shielding factor were found to be major sensitivity parameters. These parameters were evaluated to have a significant impact on the external radiation pathway by radionuclide numbers remaining in the contaminated zone.

As the first study that evaluated the DCGL for the pressurized heavy water reactor, Wolsong NPP, the input parameter database and calculation model developed in the present work are expected to be useful as a preliminary study that can be referenced when decommissioning heavy water reactors in South Korea in the future. Furthermore, it is foreseen that the economic and efficient site release standards for pressurized heavy water reactors can be established through the derivation of further reliable exposure dose results if the uncertainties in radionuclide inventory, hydrogeological parameters, and other input parameters are reduced through practical experiments on soil samples in the NPP site, along with additional field surveys in the future. For example, previous research work has confirmed the considerable relationship between the distribution coefficient and exposure dose results in the safety assessment model of nuclear waste repositories [33]. In the former investigation, a relatively increased distribution coefficient provided a remarkably delayed release rate of radionuclides owing to the enhanced retention capacity of geologic barriers and presented a relatively delayed peak time and decreased peak value of exposure dose. Since the distribution coefficient is an element-specific and conditional parameter, depending on various hydro-geochemical conditions, securing more precise distribution coefficient data for various radionuclides through a practical field test is expected to provide a significant contribution to further reliable model calculation.