Journal of Environmental Radioactivity 143 (2015) 52e57
Contents lists available at ScienceDirect
Journal of Environmental Radioactivity journal homepage: www.elsevier.com/locate/jenvrad
Development of criteria used to establish a background environmental monitoring station Bradley G. Fritz*, J. Matthew Barnett, Sandra F. Snyder, Lynn E. Bisping, Jeremy P. Rishel Pacific Northwest National Laboratory, 902 Battelle Blvd., Richland, WA 99354, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 9 January 2015 Received in revised form 9 February 2015 Accepted 10 February 2015 Available online 2 March 2015
It is generally considered necessary to measure concentrations of contaminants-of-concern at a background location when conducting atmospheric environmental surveillance. This is because it is recognized that measurements of background concentrations can enhance interpretation of environmental monitoring data. Despite the recognized need for background measurements, there is little published guidance available that describes how to identify an appropriate atmospheric background monitoring location. This paper develops generic criteria that can guide the decision making process for identifying suitable locations for background atmospheric monitoring station. Detailed methods for evaluating some of these criteria are also provided and a case study for establishment of an atmospheric background surveillance station as part of an environmental surveillance program is described. While the case study focuses on monitoring for radionuclides, the approach is equally valid for any airborne constituent being monitored. The case study shows that implementation of the developed criteria can result in a good, defensible choice for a background atmospheric monitoring location. © 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Keywords: Background monitoring Environmental monitoring Airborne Radioactivity Air Sampling
1. Introduction Background monitoring data is generally considered a necessary component of atmospheric environmental surveillance (NCRP, 2010; IAEA, 2010; U.S. DOE, 1991; Keith, 1991; Kathren, 1984; Klement, 1982; WHO, 1968). This is because it is recognized that measurements of background concentrations can enhance interpretation of environmental monitoring data. For example, background concentrations provide a point of reference for other measurements on or near a site with emissions. If on-site samples were reported to have elevated concentrations, the initial assumption would be that the elevated concentrations resulted from on-site releases. However, results from samples collected at a background location could provide evidence for another explanation (e.g., regionally elevated concentrations). While many published works identify and stress the need for background atmospheric monitoring locations when establishing monitoring networks (i.e., IAEA, 2010; NCRP, 2010; Meinke and Essig, 1991), there is little published guidance provided about how to identify an appropriate background location. How far away
* Corresponding author. Tel.: þ1 509 371 7119. E-mail address: [email protected] (B.G. Fritz).
is far enough? How far is too far? These are questions not adequately addressed in available literature. This paper develops generic criteria that can guide the decision making process for identifying suitable locations for background atmospheric monitoring station. Additionally, some detailed methods for evaluating potential locations against the criteria are provided. Finally, a case study is presented that focuses on the establishment of an atmospheric background surveillance station for the measurement of radionuclides associated with an environmental surveillance program. 2. Background siting criteria Various definitions of background values and locations have been published. The NCRP (2010) defines background radiation as “the level of radiation from sources other than the source of interest”. Control samples are defined by Keith (1991) as being collected near the time and place where the analytes of interest may exist, and used to determine if concentrations measured on a site are truly different from background concentrations. The IAEA (2010) notes that “A reference sampler might be located in an area where the natural background levels are similar to those at the site, but where the influence of discharges from the facility is negligible”.
http://dx.doi.org/10.1016/j.jenvrad.2015.02.010 0265-931X/© 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
B.G. Fritz et al. / Journal of Environmental Radioactivity 143 (2015) 52e57
While these definitions provide some context for background locations, none of them are ideal or specific to background levels of contaminants in ambient air. Using these definitions as a guide, the following definition of an ideal background air monitoring location is proposed: “An ideal background air monitoring location is a point where the measured concentrations of analytes of interest are equal to the concentrations that would be measured at the site if operational emissions did not occur.” Based on this definition of an ideal background monitoring location, a list of general criteria, and approaches for evaluating potential sites against those criteria, were developed. These are generic requirements that could be applied anywhere to assist in establishing an environmental surveillance background air monitoring station, and are presented in order of importance. A. Air concentration of each constituent of concern measured at a background location should be relatively uninfluenced by facility emissions. The acceptable level of influence a facility emission has on the concentration measured at a background location will be different for different programs. However, the increase in concentration at the background location caused by facility emissions should be less than the total acceptable error associated with the measurement. 1. Atmospheric modeling can be used to estimate the dilution of emissions, and the corresponding impact to background concentrations, at varying distances away from the source. 2. The estimated change in measured background concentrations caused by influence of facility emissions can be evaluated relative to the program's stated acceptable error. B. The air sampled at a background location should be typical of the air sampled at or near the facility (except for those constituents of concern [COCs] emitted from the facility). That is to say, analytes other than the COCs should have similar concentrations at the background location and the facility. 1. Qualitative assessment of the source facility and potential background locations are sufficient to meet this criterion. Background monitoring locations should be in an area with comparable land use and cover, similar anthropogenic emissions, etc. C. Typical weather conditions (e.g., inversions, dust storms, precipitation, prevailing wind patterns) at the facility should also occur at the background station (Glantz, 1990). 1. Knowledge of current and historic local weather patterns can be sufficient to qualitatively assess the representativeness of the background location with respect to weather. For examples, wind roses and precipitation maps could be useful. D. The background location should be established at a reasonable distance away from the emission source (i.e., not too close or too far away). A reasonable distance is a function of the size of the emission source and magnitude of emission, but generally should be as close as possible while still meeting the other requirements. The reasonable distance should also consider a worst-case scenario with wind blowing directly from the source to the background location. 1. Gaussian plume dispersion modeling under worst case dispersion conditions is sufficient for determining the minimum distance for a background location. 2. Project resources should be considered in determining the maximum acceptable distance (e.g. cost of driving to station for sample collection). E. Terrain should be a secondary consideration in this evaluation, considered after the initial modeling effort (for models that do
53
not consider terrain in the dispersion calculation). This consideration is related to Criterion D. 1. Atmospheric modeling with terrain effects or an evaluation of wind patterns and topographic maps can be used to qualitatively assess the representativeness of the background location. F. All necessary infrastructure must be available (i.e., power, pavement, communications) 1. Once a general area is identified as meeting the large scale requirements (Criteria AeE), potential specific locations within that area can be identified. G. The sampling location must meet general siting requirements for an air sampling location (e.g., minimal obstructions, no nearby sources, minimal impact to environment, adequate security and safety provisions, accessible by staff). 1. Potential sampling locations should be evaluated against siting requirements. If projects do not have established siting criteria, refer to published meteorological tower siting requirements for guidance (i.e. U.S. EPA, 2000; U.S. NRC, 2007). 2. Consider if there are any unique siting requirements specific to the sampling equipment used. Some optional considerations include: H. Co-located sampling by other agencies can be useful to provide backup data in the event of equipment failure, and for QA purposes. 1. Local regulatory agencies should be able to provide a list of other active and relevant monitoring programs in the area. I. Historic data from previous/other sampling program(s) can be useful for comparison and QA purposes. 1. A literature review should provide information about historic projects in the area. 2.1. Source to background dilution factor Atmospheric dispersion models are used to estimate the dilution factor at varying distances from the source (Criterion A). The dilution factor is used to identify the distance away from the COC source at which concentrations would be diluted enough to be negligible. One consideration is the impact that overestimation of the true background can have on other measurements made on or near the site. If the background monitoring station is ‘too close’ to the site, then site emissions will be collected by the background monitor, and the reported background will be higher than the true background (as defined above). Therefore, it is necessary to locate the background station sufficiently far from the site such that the systematic error in the measured background created by collection of site effluent at the background location is less than the total acceptable error. For example, consider a program where the required accuracy of the reported concentration is ±20%. If the estimated random errors in the sample volume and analytical measurements are ±10 and 15% respectively, then the total combined error (calculated as the root mean square of the individual error terms for random errors) is 18% (Equation (1)). An additional 2% systematic error could then be contributed by collection of site emissions at the background location and still result in the total combined error being 20% (Equation (2)). Therefore, if a potential background location has an annual average concentration 1/50th of the concentration estimated at the site boundary (or less), then that location might be considered acceptable for use as a background location because the small amounts of effluent collected at the background station will be indistinguishable from the random sampling error. For programs with lower tolerance for error, a lower dilution factor may be necessary
54
B.G. Fritz et al. / Journal of Environmental Radioactivity 143 (2015) 52e57
error ¼
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 0:152 þ 0:102 ¼ 0:18
(1)
error ¼
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 0:152 þ 0:102 þ 0:02 ¼ 0:20
(2)
(3)
consists of air sampling stations at several on-site locations. In 2013, it was determined that a background air monitoring station should be established as part of the PNNL air monitoring program. Prior to identifying the need for a background station, potential options for acquiring background concentration data for constituents of concern were considered (Fritz et al., 2014). A summary of the advantages and disadvantages of each option was developed, and costs associated with each option were estimated. After considering the various options for obtaining background data, it was decided that installation and operation of a background air monitoring station was necessary. It was the only option that could provide data of the quality and pedigree necessary to fully meet the needs of the project. While a formal DQO process could be used to establish a suitable background monitoring location, the time and expense of a formal DQO is typically not necessary for establishing a single monitoring location. Application of the general criteria and approaches outlined above resulted in specific criteria that were used in selecting potential areas for placement of a background environmental monitoring station. These criteria were then used to identify specific locations within each area. Criterion A (identifying locations minimally-influenced by facility emissions) was implemented by running an annual average Gaussian dispersion model (CAP88-PC, Rosnick, 2007) with site specific inputs (Table 1). The resultant maximum modeled concentration at the boundary of the PNNL Campus was identified. Then, the modeled concentrations were calculated in each of the 16 compass directions. The distance in each direction where the modeled concentration was equal to 2% (i.e. 1/50th) of the campus boundary maximum was converted to map coordinates, and mapped with geographical information system (GIS) software (ESRI, 2013). Areas inside of this 2% boundary were excluded from consideration for a background air monitoring location (Fig. 1). Criterion B was implemented by excluding areas where air composition was expected to be dissimilar to the PNNL Campus. The upper Yakima Valley was identified as an area to avoid due to different agricultural practices relative to the lower Yakima Valley and Tri-City region. Specifically, the use of smudge pots in Yakima Valley orchards during the spring was of concern. Therefore, a large portion of the western portion of the Yakima Valley was excluded from consideration for a background station (Fig. 1). Similarly, elevation was identified as a consideration for the background station location. It was decided that the background location should be at an elevation not more than 200 m higher or lower than the PNNL Campus elevation. An exclusion area where the elevation is greater than 320 m (i.e., 200 m higher than the PNNL Campus) was created in GIS using a digital elevation model (30-m resolution) for eastern Washington (University of Washington, 2014).
(4)
Table 1 CAP88-PC model inputs.
2.2. Minimum distance calculation It is also important that a background air sampler is located a minimum distance from the site to minimize the potential for biasing the background sample results during short time periods with poor dispersion conditions (Criterion D). These would be conditions where the highest instantaneous concentrations would be present. For example, during inversion conditions, a single sample at the background location could be highly influenced by site emissions if the background location were too close and the wind blew directly from the site to the background location. To calculate a minimum distance for the background sampler, a Gaussian plume dispersion calculation is appropriate (Equations (3) and (4)). The general form of the equation can be reduced to a ground-level centerline receptor case, where the concentration (C [g/m3]) along the plume centerline varies according to the horizontal dispersion coefficient (sy [m]) and the vertical dispersion coefficient (sz [m]). These dispersion coefficients vary as a function of downwind distance (x [m]), and can be interpolated from the moderately stable PasquilleGifford curves (Gifford, 1961; Hunter, 2012). The stack emission rate (q [g/ s]) can be set to unity, and the plume height (H [m]) determined using published plume rise equations (Briggs, 1972). The wind speed (u [m/s]) chosen for this worst-case scenario should be typical of night time wind speeds during winter months, likely between 1 and 2 m/s. These are the conditions where plume dispersion is minimized. The colder weather and shorter days results in less thermal driven mixing, and the low wind speeds minimize mechanical mixing. We recommend that the minimum distance would be the distance where the calculated Gaussian concentration is 20% of the maximum downwind plume centerline concentration. Coupled with the infrequency that these worst-case dispersion conditions might exist, this dilution should provide a sufficient minimum distance for the placement of a background monitoring station. In most directions this minimum distance will be less than the distance established using the annual average dilution factor method. Only in directions where the wind blows infrequently would the minimum distance offset calculated here be farther than calculated using the annual average dilution factor method.
C¼
q exp pusy sz
sy ¼ 0:0792x0:8852 ;
! 1 H 2 2 sz sz ¼ 17:11 LnðxÞ 108:9
Parameter
Data
Reference
Wind speed/direction
temperature precipitation mixing
2002e2011 average meteorology 12 C 160 mm 1000
Hanford Site, Station 11, 10m, Meteorological Data Snyder et al. (2014) Snyder et al. (2014) Snyder et al. (2014)
absolute
8 g/m3
Snyder et al. (2014)
38 m
Effective stack height (Duncan et al., 2014) Snyder et al. (2014) (Effective stack height used)
3. Case study The Pacific Northwest National Laboratory (PNNL, Richland, WA) emits low levels of radioactive materials into the atmosphere under a permit issued by the Washington State Department of Health (WDOH, 2010). In 2010, an environmental surveillance program was established to monitor the concentrations of radioactive materials in ambient air near the PNNL Campus (Barnett et al., 2010). A data quality objectives (DQO) process was used to determine the extent and needs of the initial environmental surveillance network. The PNNL Campus environmental surveillance program currently
Annual average Annual average Annual average height (m) Annual average humidity Stack height Stack diameter Plume rise
1.0 m 0
B.G. Fritz et al. / Journal of Environmental Radioactivity 143 (2015) 52e57
55
Fig. 1. Areas for inclusion and exclusion of potential background air monitoring locations.
In order to meet Criterion C, which relates to similar weather conditions, the potential background station location was limited to the Lower Columbia Basin Zone Area (National Weather Service, 2014, Fig. 1). This Zone Area is expected to have similar weather patterns, and receive similar amounts of precipitation. The elevation restriction developed with Criterion B also minimized differences in precipitation between the PNNL Campus and the background location. Criterion D was implemented through the use of internet mapping and Gaussian plume modeling. A maximum driving distance of 1 h was determined necessary in order to minimize sample collection labor costs. The 1-h drive time distances in each direction were established using Google Maps (Google Inc. (“Google”), 1600 Amphitheatre Parkway, Mountain View, CA 94043, United States). The minimum distance from the PNNL Campus for establishing a background monitoring location was calculated by modeling the ground level concentration as a function of downwind distance under poor dispersion conditions (Section 2.2). A stable atmosphere, with wind speed 50% of the wintertime average wind speed (Hoitink et al., 2005) were assumed to constitute poor dispersion conditions. The minimum distance was set where the calculated Gaussian concentration was 20% of the maximum (17 km, Fig. 2). When Criteria A through D were implemented, and displayed graphically, it became apparent that four general areas meet criteria AeD. These areas were in the Yakima Valley, Franklin and Walla Walla Counties, an area near the Vernita Bridge, and the Horse Heaven Hills (Fig. 3). Wind roses from around the PNNL Campus (Fig. 4) were used to evaluate these areas against Criterion E (terrain effects and wind patterns). The FranklineWalla Walla Area, while meeting the first four criteria, is in the general “downwind” direction from the PNNL Campus; note wind roses numbers 1, 11, 15, 18, 26, 27, 30. These are the closest wind monitoring locations to the PNNL Campus, and all indicate that wind frequently blows from the
PNNL Campus toward the northwest and southwest. The FranklineWalla Walla Area was, therefore, not considered further for a background air monitoring location. The area near the Vernita Bridge was also not considered for a background air monitoring location; while generally upwind, it does not satisfy the “no nearby sources” specification in Criterion G (Fritz et al., 2014). After excluding these two areas from further consideration, only the Yakima Valley and Horse Heaven Hills areas were left for evaluation against the remaining criteria. After evaluating maps and aerial photographs and considering local knowledge of the area, eleven potential background monitoring locations within the Yakima Valley and Horse Heaven Hills areas were identified. A visit to each potential site was made, and each site was evaluated against Criteria E (wind channeling), F (power availability), G (general siting requirements including
Fig. 2. Results of Gaussian plume model (Eq. (1)) using the specified input values (u ¼ 1.3 m/s, moderately stable stability class, effective plume height 38 m).
56
B.G. Fritz et al. / Journal of Environmental Radioactivity 143 (2015) 52e57
Fig. 3. Areas that meet Criteria A through D for installation of a background air monitoring location.
access, obstructions, and roads), and H (co-located sampling). Criterion I was not considered because of the unlikely prospect of obtaining useful historic information. Once the results of the site inspections were compiled and judged, the eleven sites were
ranked; the top rated site was in the Yakima Valley area at an established air monitoring station that is part of another regional atmospheric monitoring network (Fritz et al., 2014; Poston et al., 2009).
Fig. 4. Wind Roses for the area around the PNNL Campus (Hoitink et al., 2005). Line indicates direction wind is coming from; line length indicates frequency of occurrence.
B.G. Fritz et al. / Journal of Environmental Radioactivity 143 (2015) 52e57
4. Conclusion Those faced with a need to identify a location for background environmental surveillance may find few resources and resort to selecting a location based on a ‘professional judgment’ approach. Here, a detailed methodology was introduced that will provide tools to use when identifying a location to conduct background air monitoring. Additionally, implementation of this methodology for a specific case study was presented. Application of the methodology resulted in the identification of several suitable background air monitoring locations for the PNNL Campus environmental monitoring network. By following the methodology, we were able to document our choice of a preferred background location in a methodical and rigorous manner, as opposed to the ‘professional judgment’ approach that is generally used. This systematic approach will provide useful guidance in the establishment of background monitoring locations for future atmospheric monitoring networks. Acknowledgments The authors would like to thank the Effluent Management Group at Pacific Northwest National Laboratory for funding this effort under U.S. Department of Energy contract DE-AC0576RL01830. Additionally, Julia Flaherty provided an excellent internal peer review which was greatly appreciated. References Barnett, J.M., Meier, K.M., Snyder, S.F., Fritz, B.G., Poston, T.M., Rhoads, K., 2010. Data Quality Objectives Supporting Radiological Air Emissions Monitoring for the PNNL Site. PNNL-19427, rev. 0. Pacific Northwest National Laboratory, Richland, WA. Briggs, G.A., 1972. Chimney plumes in stable and neutral surroundings. Atmos. Environ. 6, 507e510. Duncan, J.P., Sackschewsky, M.R., Tilden, H.T., Barnett, J.M., Su-Coker, J., Ballinger, M.Y., Fritz, B.G., Stoetzel, G.A., Lowry, K.L., Moon, T.W., Becker, J.M., Mendez, K.M., Raney, E.A., Chamness, M.A., Larson, K.B., 2014. Pacific Northwest National Laboratory Annual Site Environmental Report for Calendar Year 2013. PNNL-23523. Pacific Northwest National Laboratory, Richland, WA. ESRI, 2013. ArcMap Version 10.1. ESRI, Redlands, CA. Fritz, B.G., Snyder, S.F., Barnett, J.M., Bisping, L.E., Rishel, J.P., 2014. Establishment of a Background Environmental Monitoring Station for the PNNL Campus. PNNL23930, rev. 0. Pacific Northwest National Laboratory, Richland, WA. Gifford, F.A., 1961. Use of routine meteorological observations for estimating atmospheric dispersion. Nucl. Saf. 2 (4), 47e51. Glantz, C.S., 1990. Meteorological monitoring for dose assessment and emergency response modeling e how much is enough? In: Gray, R.H. (Ed.),
57
Environmental Monitoring, Restoration, and Assessment: What Have We Learned?, 28th Hanford Symposium on Health and Environment. October 1619, 1989, Richland, WA. Pacific Northwest National Laboratory, Richland, WA, pp. 201e206. Hoitink, D.J., Burk, K.W., Ramsdell Jr., J.V., Shaw, W.J., 2005. Hanford Site Climatological Summary 2004 with Historical Data. PNNL-15160. Pacific Northwest National Laboratory, Richland, WA. Hunter, C.H., 2012. A Recommended PasquilleGifford Stability Classification Method for Safety Basis Atmospheric Dispersion Modeling at SRS. SRNL-STI2012-00055, revision 0. Savannah River National Laboratory, Aiken, SC. http:// sti.srs.gov/fulltext/SRNL-STI-2012-00055.pdf (accessed 24.10.14.). International Atomic Energy Agency (IAEA), 2010. Programmes and Systems for Source and Environmental Radiation Monitoring. In: Safety Report Series No. 64. IAEA, Vienna. Kathren, R.L., 1984. Radioactivity in the Environment: Sources, Distribution, and Surveillance. Harwood Academic Publishers, Chur, Switzerland. Keith, L.H., 1991. Environmental Sampling and Analysis; a Practical Guide. Lewis Publishers, Inc., Chelsea, MI. Klement, A.W., 1982. CRC Handbook of Environmental Radiation. CRC Press, Michigan. Meinke, W.W., Essig, T.H., 1991. Offsite Dose Calculation Manual Guidance: Standard Radiological Effluent Controls for Boiling Water Reactors. NUREG-1302. U.S. Nuclear Regulatory Commission, Washington, DC. National Council on Radiation Protection and Measurements (NCRP), 2010. Design of Effective Radiological Effluent Monitoring and Environmental Surveillance Programs. NCRP report no. 169. National Council of Radiation Protection and Measurement, Bethesda, MD. National Weather Service, 2014. Lower Columbia Basin of Washington Zone Forecast. http://forecast.weather.gov/MapClick.php?zoneid¼WAZ028 (accessed 24.10.14.). Poston, T.M., Duncan, J.P., Dirkes, R.L., 2009. Hanford Site Environmental Report for Calendar Year 2008. PNNL-18427. Pacific Northwest National Laboratory, Richland, Washington. Rosnick, R.J., 2007. CAP88-PC Version 3.0 User Guide. Office of Radiation and Indoor Air, U.S. Environmental Protection Agency, Washington, D.C. Snyder, S.F., Barnett, J.M., Bisping, L.E., 2014. Pacific Northwest National Laboratory Campus Radionuclide Air Emissions Report for Calendar Year 2013. PNNL20436-4. Pacific Northwest National Laboratory, Richland, WA. University of Washington, 2014. Washington State e GIS Data. http://wagda.lib. washington.edu/data/geography/wa_state/#elevation (accessed 03.11.14.). U.S. Department of Energy (DOE), 1991. Environmental Regulatory Guide for Radiological Effluent Monitoring and Environmental Surveillance. DOE/EH0173T (DE91-013607). U.S. Department of Energy, Washington, DC. https:// www.orau.org/ptp/pdf/eh0173t.pdf (accessed 08.08.14.). U.S. Environmental Protection Agency (EPA), 2000. Meteorological Monitoring Guidance for Regulatory Modelling Applications. EPA-454/R-99-005. Office of Air and Radiation, Office of Air Quality Planning and Standards, Research Triangle Park, NC. U.S. Nuclear Regulatory Commission (NRC), 2007. Meteorological Monitoring Programs for Nuclear Power Plants. Regulatory Guide 1.23. U.S. Nuclear Regulatory Commission, Washington DC. Washington State Department of Health (WDOH), 2010. RAEL-005. World Health Organization (WHO), 1968. Routine Surveillance for Radionuclides in Air and Water. Switzerland, Geneva.
Contents lists available at ScienceDirect
Journal of Environmental Radioactivity journal homepage: www.elsevier.com/locate/jenvrad
Development of criteria used to establish a background environmental monitoring station Bradley G. Fritz*, J. Matthew Barnett, Sandra F. Snyder, Lynn E. Bisping, Jeremy P. Rishel Pacific Northwest National Laboratory, 902 Battelle Blvd., Richland, WA 99354, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 9 January 2015 Received in revised form 9 February 2015 Accepted 10 February 2015 Available online 2 March 2015
It is generally considered necessary to measure concentrations of contaminants-of-concern at a background location when conducting atmospheric environmental surveillance. This is because it is recognized that measurements of background concentrations can enhance interpretation of environmental monitoring data. Despite the recognized need for background measurements, there is little published guidance available that describes how to identify an appropriate atmospheric background monitoring location. This paper develops generic criteria that can guide the decision making process for identifying suitable locations for background atmospheric monitoring station. Detailed methods for evaluating some of these criteria are also provided and a case study for establishment of an atmospheric background surveillance station as part of an environmental surveillance program is described. While the case study focuses on monitoring for radionuclides, the approach is equally valid for any airborne constituent being monitored. The case study shows that implementation of the developed criteria can result in a good, defensible choice for a background atmospheric monitoring location. © 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Keywords: Background monitoring Environmental monitoring Airborne Radioactivity Air Sampling
1. Introduction Background monitoring data is generally considered a necessary component of atmospheric environmental surveillance (NCRP, 2010; IAEA, 2010; U.S. DOE, 1991; Keith, 1991; Kathren, 1984; Klement, 1982; WHO, 1968). This is because it is recognized that measurements of background concentrations can enhance interpretation of environmental monitoring data. For example, background concentrations provide a point of reference for other measurements on or near a site with emissions. If on-site samples were reported to have elevated concentrations, the initial assumption would be that the elevated concentrations resulted from on-site releases. However, results from samples collected at a background location could provide evidence for another explanation (e.g., regionally elevated concentrations). While many published works identify and stress the need for background atmospheric monitoring locations when establishing monitoring networks (i.e., IAEA, 2010; NCRP, 2010; Meinke and Essig, 1991), there is little published guidance provided about how to identify an appropriate background location. How far away
* Corresponding author. Tel.: þ1 509 371 7119. E-mail address: [email protected] (B.G. Fritz).
is far enough? How far is too far? These are questions not adequately addressed in available literature. This paper develops generic criteria that can guide the decision making process for identifying suitable locations for background atmospheric monitoring station. Additionally, some detailed methods for evaluating potential locations against the criteria are provided. Finally, a case study is presented that focuses on the establishment of an atmospheric background surveillance station for the measurement of radionuclides associated with an environmental surveillance program. 2. Background siting criteria Various definitions of background values and locations have been published. The NCRP (2010) defines background radiation as “the level of radiation from sources other than the source of interest”. Control samples are defined by Keith (1991) as being collected near the time and place where the analytes of interest may exist, and used to determine if concentrations measured on a site are truly different from background concentrations. The IAEA (2010) notes that “A reference sampler might be located in an area where the natural background levels are similar to those at the site, but where the influence of discharges from the facility is negligible”.
http://dx.doi.org/10.1016/j.jenvrad.2015.02.010 0265-931X/© 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
B.G. Fritz et al. / Journal of Environmental Radioactivity 143 (2015) 52e57
While these definitions provide some context for background locations, none of them are ideal or specific to background levels of contaminants in ambient air. Using these definitions as a guide, the following definition of an ideal background air monitoring location is proposed: “An ideal background air monitoring location is a point where the measured concentrations of analytes of interest are equal to the concentrations that would be measured at the site if operational emissions did not occur.” Based on this definition of an ideal background monitoring location, a list of general criteria, and approaches for evaluating potential sites against those criteria, were developed. These are generic requirements that could be applied anywhere to assist in establishing an environmental surveillance background air monitoring station, and are presented in order of importance. A. Air concentration of each constituent of concern measured at a background location should be relatively uninfluenced by facility emissions. The acceptable level of influence a facility emission has on the concentration measured at a background location will be different for different programs. However, the increase in concentration at the background location caused by facility emissions should be less than the total acceptable error associated with the measurement. 1. Atmospheric modeling can be used to estimate the dilution of emissions, and the corresponding impact to background concentrations, at varying distances away from the source. 2. The estimated change in measured background concentrations caused by influence of facility emissions can be evaluated relative to the program's stated acceptable error. B. The air sampled at a background location should be typical of the air sampled at or near the facility (except for those constituents of concern [COCs] emitted from the facility). That is to say, analytes other than the COCs should have similar concentrations at the background location and the facility. 1. Qualitative assessment of the source facility and potential background locations are sufficient to meet this criterion. Background monitoring locations should be in an area with comparable land use and cover, similar anthropogenic emissions, etc. C. Typical weather conditions (e.g., inversions, dust storms, precipitation, prevailing wind patterns) at the facility should also occur at the background station (Glantz, 1990). 1. Knowledge of current and historic local weather patterns can be sufficient to qualitatively assess the representativeness of the background location with respect to weather. For examples, wind roses and precipitation maps could be useful. D. The background location should be established at a reasonable distance away from the emission source (i.e., not too close or too far away). A reasonable distance is a function of the size of the emission source and magnitude of emission, but generally should be as close as possible while still meeting the other requirements. The reasonable distance should also consider a worst-case scenario with wind blowing directly from the source to the background location. 1. Gaussian plume dispersion modeling under worst case dispersion conditions is sufficient for determining the minimum distance for a background location. 2. Project resources should be considered in determining the maximum acceptable distance (e.g. cost of driving to station for sample collection). E. Terrain should be a secondary consideration in this evaluation, considered after the initial modeling effort (for models that do
53
not consider terrain in the dispersion calculation). This consideration is related to Criterion D. 1. Atmospheric modeling with terrain effects or an evaluation of wind patterns and topographic maps can be used to qualitatively assess the representativeness of the background location. F. All necessary infrastructure must be available (i.e., power, pavement, communications) 1. Once a general area is identified as meeting the large scale requirements (Criteria AeE), potential specific locations within that area can be identified. G. The sampling location must meet general siting requirements for an air sampling location (e.g., minimal obstructions, no nearby sources, minimal impact to environment, adequate security and safety provisions, accessible by staff). 1. Potential sampling locations should be evaluated against siting requirements. If projects do not have established siting criteria, refer to published meteorological tower siting requirements for guidance (i.e. U.S. EPA, 2000; U.S. NRC, 2007). 2. Consider if there are any unique siting requirements specific to the sampling equipment used. Some optional considerations include: H. Co-located sampling by other agencies can be useful to provide backup data in the event of equipment failure, and for QA purposes. 1. Local regulatory agencies should be able to provide a list of other active and relevant monitoring programs in the area. I. Historic data from previous/other sampling program(s) can be useful for comparison and QA purposes. 1. A literature review should provide information about historic projects in the area. 2.1. Source to background dilution factor Atmospheric dispersion models are used to estimate the dilution factor at varying distances from the source (Criterion A). The dilution factor is used to identify the distance away from the COC source at which concentrations would be diluted enough to be negligible. One consideration is the impact that overestimation of the true background can have on other measurements made on or near the site. If the background monitoring station is ‘too close’ to the site, then site emissions will be collected by the background monitor, and the reported background will be higher than the true background (as defined above). Therefore, it is necessary to locate the background station sufficiently far from the site such that the systematic error in the measured background created by collection of site effluent at the background location is less than the total acceptable error. For example, consider a program where the required accuracy of the reported concentration is ±20%. If the estimated random errors in the sample volume and analytical measurements are ±10 and 15% respectively, then the total combined error (calculated as the root mean square of the individual error terms for random errors) is 18% (Equation (1)). An additional 2% systematic error could then be contributed by collection of site emissions at the background location and still result in the total combined error being 20% (Equation (2)). Therefore, if a potential background location has an annual average concentration 1/50th of the concentration estimated at the site boundary (or less), then that location might be considered acceptable for use as a background location because the small amounts of effluent collected at the background station will be indistinguishable from the random sampling error. For programs with lower tolerance for error, a lower dilution factor may be necessary
54
B.G. Fritz et al. / Journal of Environmental Radioactivity 143 (2015) 52e57
error ¼
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 0:152 þ 0:102 ¼ 0:18
(1)
error ¼
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 0:152 þ 0:102 þ 0:02 ¼ 0:20
(2)
(3)
consists of air sampling stations at several on-site locations. In 2013, it was determined that a background air monitoring station should be established as part of the PNNL air monitoring program. Prior to identifying the need for a background station, potential options for acquiring background concentration data for constituents of concern were considered (Fritz et al., 2014). A summary of the advantages and disadvantages of each option was developed, and costs associated with each option were estimated. After considering the various options for obtaining background data, it was decided that installation and operation of a background air monitoring station was necessary. It was the only option that could provide data of the quality and pedigree necessary to fully meet the needs of the project. While a formal DQO process could be used to establish a suitable background monitoring location, the time and expense of a formal DQO is typically not necessary for establishing a single monitoring location. Application of the general criteria and approaches outlined above resulted in specific criteria that were used in selecting potential areas for placement of a background environmental monitoring station. These criteria were then used to identify specific locations within each area. Criterion A (identifying locations minimally-influenced by facility emissions) was implemented by running an annual average Gaussian dispersion model (CAP88-PC, Rosnick, 2007) with site specific inputs (Table 1). The resultant maximum modeled concentration at the boundary of the PNNL Campus was identified. Then, the modeled concentrations were calculated in each of the 16 compass directions. The distance in each direction where the modeled concentration was equal to 2% (i.e. 1/50th) of the campus boundary maximum was converted to map coordinates, and mapped with geographical information system (GIS) software (ESRI, 2013). Areas inside of this 2% boundary were excluded from consideration for a background air monitoring location (Fig. 1). Criterion B was implemented by excluding areas where air composition was expected to be dissimilar to the PNNL Campus. The upper Yakima Valley was identified as an area to avoid due to different agricultural practices relative to the lower Yakima Valley and Tri-City region. Specifically, the use of smudge pots in Yakima Valley orchards during the spring was of concern. Therefore, a large portion of the western portion of the Yakima Valley was excluded from consideration for a background station (Fig. 1). Similarly, elevation was identified as a consideration for the background station location. It was decided that the background location should be at an elevation not more than 200 m higher or lower than the PNNL Campus elevation. An exclusion area where the elevation is greater than 320 m (i.e., 200 m higher than the PNNL Campus) was created in GIS using a digital elevation model (30-m resolution) for eastern Washington (University of Washington, 2014).
(4)
Table 1 CAP88-PC model inputs.
2.2. Minimum distance calculation It is also important that a background air sampler is located a minimum distance from the site to minimize the potential for biasing the background sample results during short time periods with poor dispersion conditions (Criterion D). These would be conditions where the highest instantaneous concentrations would be present. For example, during inversion conditions, a single sample at the background location could be highly influenced by site emissions if the background location were too close and the wind blew directly from the site to the background location. To calculate a minimum distance for the background sampler, a Gaussian plume dispersion calculation is appropriate (Equations (3) and (4)). The general form of the equation can be reduced to a ground-level centerline receptor case, where the concentration (C [g/m3]) along the plume centerline varies according to the horizontal dispersion coefficient (sy [m]) and the vertical dispersion coefficient (sz [m]). These dispersion coefficients vary as a function of downwind distance (x [m]), and can be interpolated from the moderately stable PasquilleGifford curves (Gifford, 1961; Hunter, 2012). The stack emission rate (q [g/ s]) can be set to unity, and the plume height (H [m]) determined using published plume rise equations (Briggs, 1972). The wind speed (u [m/s]) chosen for this worst-case scenario should be typical of night time wind speeds during winter months, likely between 1 and 2 m/s. These are the conditions where plume dispersion is minimized. The colder weather and shorter days results in less thermal driven mixing, and the low wind speeds minimize mechanical mixing. We recommend that the minimum distance would be the distance where the calculated Gaussian concentration is 20% of the maximum downwind plume centerline concentration. Coupled with the infrequency that these worst-case dispersion conditions might exist, this dilution should provide a sufficient minimum distance for the placement of a background monitoring station. In most directions this minimum distance will be less than the distance established using the annual average dilution factor method. Only in directions where the wind blows infrequently would the minimum distance offset calculated here be farther than calculated using the annual average dilution factor method.
C¼
q exp pusy sz
sy ¼ 0:0792x0:8852 ;
! 1 H 2 2 sz sz ¼ 17:11 LnðxÞ 108:9
Parameter
Data
Reference
Wind speed/direction
temperature precipitation mixing
2002e2011 average meteorology 12 C 160 mm 1000
Hanford Site, Station 11, 10m, Meteorological Data Snyder et al. (2014) Snyder et al. (2014) Snyder et al. (2014)
absolute
8 g/m3
Snyder et al. (2014)
38 m
Effective stack height (Duncan et al., 2014) Snyder et al. (2014) (Effective stack height used)
3. Case study The Pacific Northwest National Laboratory (PNNL, Richland, WA) emits low levels of radioactive materials into the atmosphere under a permit issued by the Washington State Department of Health (WDOH, 2010). In 2010, an environmental surveillance program was established to monitor the concentrations of radioactive materials in ambient air near the PNNL Campus (Barnett et al., 2010). A data quality objectives (DQO) process was used to determine the extent and needs of the initial environmental surveillance network. The PNNL Campus environmental surveillance program currently
Annual average Annual average Annual average height (m) Annual average humidity Stack height Stack diameter Plume rise
1.0 m 0
B.G. Fritz et al. / Journal of Environmental Radioactivity 143 (2015) 52e57
55
Fig. 1. Areas for inclusion and exclusion of potential background air monitoring locations.
In order to meet Criterion C, which relates to similar weather conditions, the potential background station location was limited to the Lower Columbia Basin Zone Area (National Weather Service, 2014, Fig. 1). This Zone Area is expected to have similar weather patterns, and receive similar amounts of precipitation. The elevation restriction developed with Criterion B also minimized differences in precipitation between the PNNL Campus and the background location. Criterion D was implemented through the use of internet mapping and Gaussian plume modeling. A maximum driving distance of 1 h was determined necessary in order to minimize sample collection labor costs. The 1-h drive time distances in each direction were established using Google Maps (Google Inc. (“Google”), 1600 Amphitheatre Parkway, Mountain View, CA 94043, United States). The minimum distance from the PNNL Campus for establishing a background monitoring location was calculated by modeling the ground level concentration as a function of downwind distance under poor dispersion conditions (Section 2.2). A stable atmosphere, with wind speed 50% of the wintertime average wind speed (Hoitink et al., 2005) were assumed to constitute poor dispersion conditions. The minimum distance was set where the calculated Gaussian concentration was 20% of the maximum (17 km, Fig. 2). When Criteria A through D were implemented, and displayed graphically, it became apparent that four general areas meet criteria AeD. These areas were in the Yakima Valley, Franklin and Walla Walla Counties, an area near the Vernita Bridge, and the Horse Heaven Hills (Fig. 3). Wind roses from around the PNNL Campus (Fig. 4) were used to evaluate these areas against Criterion E (terrain effects and wind patterns). The FranklineWalla Walla Area, while meeting the first four criteria, is in the general “downwind” direction from the PNNL Campus; note wind roses numbers 1, 11, 15, 18, 26, 27, 30. These are the closest wind monitoring locations to the PNNL Campus, and all indicate that wind frequently blows from the
PNNL Campus toward the northwest and southwest. The FranklineWalla Walla Area was, therefore, not considered further for a background air monitoring location. The area near the Vernita Bridge was also not considered for a background air monitoring location; while generally upwind, it does not satisfy the “no nearby sources” specification in Criterion G (Fritz et al., 2014). After excluding these two areas from further consideration, only the Yakima Valley and Horse Heaven Hills areas were left for evaluation against the remaining criteria. After evaluating maps and aerial photographs and considering local knowledge of the area, eleven potential background monitoring locations within the Yakima Valley and Horse Heaven Hills areas were identified. A visit to each potential site was made, and each site was evaluated against Criteria E (wind channeling), F (power availability), G (general siting requirements including
Fig. 2. Results of Gaussian plume model (Eq. (1)) using the specified input values (u ¼ 1.3 m/s, moderately stable stability class, effective plume height 38 m).
56
B.G. Fritz et al. / Journal of Environmental Radioactivity 143 (2015) 52e57
Fig. 3. Areas that meet Criteria A through D for installation of a background air monitoring location.
access, obstructions, and roads), and H (co-located sampling). Criterion I was not considered because of the unlikely prospect of obtaining useful historic information. Once the results of the site inspections were compiled and judged, the eleven sites were
ranked; the top rated site was in the Yakima Valley area at an established air monitoring station that is part of another regional atmospheric monitoring network (Fritz et al., 2014; Poston et al., 2009).
Fig. 4. Wind Roses for the area around the PNNL Campus (Hoitink et al., 2005). Line indicates direction wind is coming from; line length indicates frequency of occurrence.
B.G. Fritz et al. / Journal of Environmental Radioactivity 143 (2015) 52e57
4. Conclusion Those faced with a need to identify a location for background environmental surveillance may find few resources and resort to selecting a location based on a ‘professional judgment’ approach. Here, a detailed methodology was introduced that will provide tools to use when identifying a location to conduct background air monitoring. Additionally, implementation of this methodology for a specific case study was presented. Application of the methodology resulted in the identification of several suitable background air monitoring locations for the PNNL Campus environmental monitoring network. By following the methodology, we were able to document our choice of a preferred background location in a methodical and rigorous manner, as opposed to the ‘professional judgment’ approach that is generally used. This systematic approach will provide useful guidance in the establishment of background monitoring locations for future atmospheric monitoring networks. Acknowledgments The authors would like to thank the Effluent Management Group at Pacific Northwest National Laboratory for funding this effort under U.S. Department of Energy contract DE-AC0576RL01830. Additionally, Julia Flaherty provided an excellent internal peer review which was greatly appreciated. References Barnett, J.M., Meier, K.M., Snyder, S.F., Fritz, B.G., Poston, T.M., Rhoads, K., 2010. Data Quality Objectives Supporting Radiological Air Emissions Monitoring for the PNNL Site. PNNL-19427, rev. 0. Pacific Northwest National Laboratory, Richland, WA. Briggs, G.A., 1972. Chimney plumes in stable and neutral surroundings. Atmos. Environ. 6, 507e510. Duncan, J.P., Sackschewsky, M.R., Tilden, H.T., Barnett, J.M., Su-Coker, J., Ballinger, M.Y., Fritz, B.G., Stoetzel, G.A., Lowry, K.L., Moon, T.W., Becker, J.M., Mendez, K.M., Raney, E.A., Chamness, M.A., Larson, K.B., 2014. Pacific Northwest National Laboratory Annual Site Environmental Report for Calendar Year 2013. PNNL-23523. Pacific Northwest National Laboratory, Richland, WA. ESRI, 2013. ArcMap Version 10.1. ESRI, Redlands, CA. Fritz, B.G., Snyder, S.F., Barnett, J.M., Bisping, L.E., Rishel, J.P., 2014. Establishment of a Background Environmental Monitoring Station for the PNNL Campus. PNNL23930, rev. 0. Pacific Northwest National Laboratory, Richland, WA. Gifford, F.A., 1961. Use of routine meteorological observations for estimating atmospheric dispersion. Nucl. Saf. 2 (4), 47e51. Glantz, C.S., 1990. Meteorological monitoring for dose assessment and emergency response modeling e how much is enough? In: Gray, R.H. (Ed.),
57
Environmental Monitoring, Restoration, and Assessment: What Have We Learned?, 28th Hanford Symposium on Health and Environment. October 1619, 1989, Richland, WA. Pacific Northwest National Laboratory, Richland, WA, pp. 201e206. Hoitink, D.J., Burk, K.W., Ramsdell Jr., J.V., Shaw, W.J., 2005. Hanford Site Climatological Summary 2004 with Historical Data. PNNL-15160. Pacific Northwest National Laboratory, Richland, WA. Hunter, C.H., 2012. A Recommended PasquilleGifford Stability Classification Method for Safety Basis Atmospheric Dispersion Modeling at SRS. SRNL-STI2012-00055, revision 0. Savannah River National Laboratory, Aiken, SC. http:// sti.srs.gov/fulltext/SRNL-STI-2012-00055.pdf (accessed 24.10.14.). International Atomic Energy Agency (IAEA), 2010. Programmes and Systems for Source and Environmental Radiation Monitoring. In: Safety Report Series No. 64. IAEA, Vienna. Kathren, R.L., 1984. Radioactivity in the Environment: Sources, Distribution, and Surveillance. Harwood Academic Publishers, Chur, Switzerland. Keith, L.H., 1991. Environmental Sampling and Analysis; a Practical Guide. Lewis Publishers, Inc., Chelsea, MI. Klement, A.W., 1982. CRC Handbook of Environmental Radiation. CRC Press, Michigan. Meinke, W.W., Essig, T.H., 1991. Offsite Dose Calculation Manual Guidance: Standard Radiological Effluent Controls for Boiling Water Reactors. NUREG-1302. U.S. Nuclear Regulatory Commission, Washington, DC. National Council on Radiation Protection and Measurements (NCRP), 2010. Design of Effective Radiological Effluent Monitoring and Environmental Surveillance Programs. NCRP report no. 169. National Council of Radiation Protection and Measurement, Bethesda, MD. National Weather Service, 2014. Lower Columbia Basin of Washington Zone Forecast. http://forecast.weather.gov/MapClick.php?zoneid¼WAZ028 (accessed 24.10.14.). Poston, T.M., Duncan, J.P., Dirkes, R.L., 2009. Hanford Site Environmental Report for Calendar Year 2008. PNNL-18427. Pacific Northwest National Laboratory, Richland, Washington. Rosnick, R.J., 2007. CAP88-PC Version 3.0 User Guide. Office of Radiation and Indoor Air, U.S. Environmental Protection Agency, Washington, D.C. Snyder, S.F., Barnett, J.M., Bisping, L.E., 2014. Pacific Northwest National Laboratory Campus Radionuclide Air Emissions Report for Calendar Year 2013. PNNL20436-4. Pacific Northwest National Laboratory, Richland, WA. University of Washington, 2014. Washington State e GIS Data. http://wagda.lib. washington.edu/data/geography/wa_state/#elevation (accessed 03.11.14.). U.S. Department of Energy (DOE), 1991. Environmental Regulatory Guide for Radiological Effluent Monitoring and Environmental Surveillance. DOE/EH0173T (DE91-013607). U.S. Department of Energy, Washington, DC. https:// www.orau.org/ptp/pdf/eh0173t.pdf (accessed 08.08.14.). U.S. Environmental Protection Agency (EPA), 2000. Meteorological Monitoring Guidance for Regulatory Modelling Applications. EPA-454/R-99-005. Office of Air and Radiation, Office of Air Quality Planning and Standards, Research Triangle Park, NC. U.S. Nuclear Regulatory Commission (NRC), 2007. Meteorological Monitoring Programs for Nuclear Power Plants. Regulatory Guide 1.23. U.S. Nuclear Regulatory Commission, Washington DC. Washington State Department of Health (WDOH), 2010. RAEL-005. World Health Organization (WHO), 1968. Routine Surveillance for Radionuclides in Air and Water. Switzerland, Geneva.
