U.S. Department of Energy, Environmental Protection Agency, Nuclear Regulatory Commission and Department of Defense. (December 1997). Multi-Agency Radiation Survey and Site Investigation Manual (MARSSIM). NUREG-1575. EPA 402-R-97-016. U.S. DOE, EPA, NRC and DOD, Washington, DC.

• The final copy of this publication has now been issued.  http://www.epa.gov/radiation/marssim/.  Print copies are available from the Government Printing Office for $51.00, stock number 052-020-00639-3.
• The comments and quotes below are from the draft version.

• "The primary objective of the EPA, NRC, and DOE regulations is to ensure that human health and the environment are protected from radioactive contamination at sites that are to be released to the public. As such, they contain a specific limit, called the release criterion, that pertains to the annual radiation dose to 'any reasonably maximally exposed member of the public' (EPA) or to 'the average member of the critical [population] group' (NRC)." (pg. 1-1).
• "As illustrated in Figure 1.1, the demonstration of compliance is comprised of three interrelated parts:
• I. Translate: translating the cleanup/release criterion (e.g., mSv/y or mrem/y) into a corresponding derived contaminant concentration level (e.g., Bq/kg or pCi/g in soil) through the use of environmental pathway modeling.
• II. Measure: Acquiring scientifically sound and legally defensible site-specific data on the levels and distribution of residual contamination by employing suitable field and/or laboratory measurement techniques.
• Decide: Determining that the data obtained from sampling does support the assertion that the site meets the release criterion, within an acceptable degree of uncertainty, through application of a statistically-based decision rule." (pg. 1-1, 1-3).
• "Because of the large variability in the types of radiation sites, it is impossible to provide criteria that apply to every situation." (pg. 1-3).
• "There are several areas beyond the scope of MARSSIM. These areas include translation of dose or risk standards into radionuclide specific concentrations, or demonstrating compliance with ground water or surface water regulations. ... Other contaminated media (e.g., sub-surface soil, building materials, ground water, etc.) and the release of contaminated components and equipment are also not addressed by MARSSIM." (pg. 1-4).

• "The process described in MARSSIM begins with the premise that a release criterion has already been provided in terms of a measurement quantity. (pg. 2.2).
• "A release criterion is a regulatory limit expressed in terms of dose (mSv/y or mrem/y) or risk (cancer incidence or cancer mortality). The terms release limit or cleanup standard have also been used to describe this term. A release criterion is typically based on total or committed effective dose equivalent (TEDE or CEDE) and generally cannot be measured directly. Exposure pathway modeling is used to calculate a radionuclide-specific predicted concentration or surface area concentration of specific nuclides that could result in a dose (TEDE or CEDE) equal to the release criterion. In this manual such a concentration is termed the derived concentration guideline level (DCGL). Exposure pathway modeling is an analysis of various exposure pathways and scenarios used to convert dose into concentration. In many cases DCGLs can be obtained from responsible regulatory agency guidance based on default modeling input parameters, while other users may elect to take into account site-specific parameters to determine DCGLs. In general, the units for the DCGL are the same as the units for measurements performed to demonstrate compliance (e.g., Bq/kg or pCi/g, Bq/m² or dpm/100 cm², etc.). This allows direct comparisons between the survey results and the DCGL." (pg. 2-2).
• "If the residual radioactivity is evenly distributed over a large area, MARSSIM looks at the average activity over the entire area. The DCGL_W(the DCGL used for the statistical tests, Section 2.5) is derived based on an average concentration over a large area." (pg. 2-3).
• "Decommissioning is the process of removing a site safely from service, reducing residual radioactivity through remediation to a level that permits release of the property, and termination of the license or other authorization for site operation." (pg. 2-3).
• "Areas that have no reasonable potential for residual contamination are classified as non-impacted areas. These areas have no radiological impact from site operations and are typically identified early in decommissioning. Areas with some potential for residual contamination are classified as impacted areas." (pg. 2-4).
• "Impacted areas are further divided into one of three classifications:" (pg. 2-5).
• "Class 1 Areas: Areas that have, or had, a potential for radioactive contamination (based on site operating history) or known contamination (based on previous radiation surveys) above the DCGL_W." (pg. 2-5).
• "Class 2 Areas: Areas that have, or had, a potential for radioactive contamination or known contamination, but are not expected to exceed the DCGL_W." (pg. 2-5).
• "Class 3 Areas: Any impacted areas that are not expected to contain any residual radioactivity, or are expected to contain levels of residual radioactivity at a very small fraction of the DCGL_W, based on site operating history and previous radiation surveys." (pg. 2-5).
• "The characterization survey is the most comprehensive of all the survey types and generated the most data. It includes preparing a reference grid, systematic as well as judgment measurements, and surveys of different media (e.g., surface soils, interior and exterior surfaces of buildings). The decision as to which media will be surveyed is a site-specific decision addressed throughout the Radiation Survey and Site Investigation Process." (pg. 2-24).
• "As previously stated, GCGLs are assumed to be developed with the assumption of a relatively uniform distribution of contamination. Some variability in the measurements is expected. This variability is primarily due to a random spatial distribution of contamination and uncertainties in the measurement process. The arithmetic mean of the measurements taken from such a distribution would represent the parameter of interest for demonstrating compliance." (pg. 2-27).

• "Historical Site Assessment (HSA) is the first step in the Radiation Survey and Site Investigation Process. The HSA is a detailed investigation to collect existing information (from the start of site activities related to radionuclides) for the site and its surroundings. The necessity for and amount of effort associated with an HSA depends on the type of site, the site's regulatory framework, and availability of documented information." (pg. 3-1).
• "The initial classification of the site involves developing a conceptual model based on the existing information collected during the preliminary investigation. Conceptual models describe a site or facility and its environs, and present hypotheses regarding the radionuclides for known and potential residual contamination (EPA 1987b, 1987c)." (pg. 3-3).
• "An efficient HSA gathers information sufficient to identify the radionuclides used at the site --- including their chemical and physical form. The first step in evaluating HSA data is to estimate the potential for residual contamination by these radionuclides." (pg. 3-11).
• "Information gathered during the HSA should be used to provide an initial classification of the site areas as impacted or non-impacted." (pg. 3-12).
• "The next step in evaluating the data gathered during the HSA is to identify potentially contaminated media at the site." (pg. 3-12).

• "Conducting radiological surveys in support of decommissioning serves to answer several basic questions, including:
• Is there residual radioactive contamination present from previous uses?
• What is the character (qualitative and quantitative) of the residual activity?
• Is the average residual activity level below the established derived concentration guideline level?
• Are there small localized areas of residual activity in excess of the investigation level?" (pg. 4-1).
• "The decommissioning process assures that residual radioactivity will not result in individuals being exposed to unacceptable levels of radiation and/or radioactive materials. Regulatory agencies establish radiation dose standards based on risk considerations and scientific data relating dose to risk. Residual levels of radioactive material that correspond to allowable radiation dose standards are calculated (derived) by analysis of various pathways and scenarios (direct radiation, inhalation, ingestion, etc.) through which exposures could occur. These derived levels, known as derived concentration guideline levels (DCGLs), are presented in terms of surface or volume activity concentrations. DCGLs refer to average levels of radiation or radioactivity above appropriate background levels. DCGLs applicable to building or other structural and miscellaneous surfaces are expressed in units of activity per surface area (typically Bq/m² or dpm/100 cm²). When applied to soil and induced activity from neutron irradiation DCGLs are expressed in units of activity per unit of mass (typically Bq/kg or pCi/g)." (pg. 4-1).
• "The DCGL_W, based on pathway modeling, is the uniform residual radioactivity concentration level within a survey unit that corresponds to the release criterion (e.g., regulatory limit in terms of dose or risk). (pg. 4-3).
• "Identification of radionuclide contaminants at the site is generally performed through laboratory analyses, such as alpha and gamma spectrometry. ... This information is essential in establishing the DCGLs for the site. DCGLs provide the basis for essentially all aspects of designing, implementing, and evaluating the final status survey. The DCGLs discussed in this manual are limited to structure surfaces and soil contamination; the user should consult the responsible regulatory agency if it is necessary to establish DCGLs for other environmental media (e.g., groundwater, and other water pathways). (pg. 4-3).
• "For sites with multiple contaminants, it may be possible to measure just one of the contaminants and still demonstrate compliance for all of the contaminants present." (pg. 4-4).
• "Typically, each radionuclide DCGL corresponds to the release criterion (e.g., regulatory limit in terms of dose or risk). However, in the presence of multiple radionuclides the DCGLs for each radionuclide would in sum total result in the release criterion being exceeded by these DCGLs. In this case, the individual DCGLs need to be adjusted to account for the presence of multiple radionuclides contributing to the total dose. One method for adjusting the DCGLs is to modify the assumptions made during exposure pathway modeling to account for multiple radionuclides. A second method is to use the unity rule to adjust the individual DCGLs." (pg. 4-6).

• "Radiological characterization surveys may be performed to satisfy a number of specific objectives. Examples of characterization survey objectives include: 1) determining the nature and extent of radiological contamination; 2) evaluating remediation alternatives (e.g., unrestricted use, restricted use, onsite disposal, off-site disposal, etc.); 3) input to pathway analysis/dose assessment models for determining site-specific DCGLs (Bq/kg, Bq/m²); 4) estimating the occupational and public health and safety impacts during decommissioning; 5) evaluating remediation technologies; 6) input to final status survey design..." (pg. 5-7).
• "Contamination release and migration pathways, as well as areas that are potentially affected and are likely to contain residual contamination, should be identified." (pg. 5-8).
• "In planning for potential use of characterization survey data as part of the final status survey, the characterization data must be of sufficient quality and quantity, including location information, for that use (see Section 5.5)." (pg. 5-9).
• "Characterization survey activities often involve the detailed assessment of various types of building and environmental media ... The HSA data should be used to identify the potentially contaminated media on-site ... Selection of survey instrumentation and analytical techniques are typically based on a knowledge of the appropriate DCGLs --- because any remediation decisions are made based on the level of the residual contamination as compared to the DCGL." (pg. 5-9).
• "Sample locations should be referenced to reference system coordinates (see Section 4.8.5), if appropriate, or prominent site features. A typical reference system spacing for open land areas is 10 meters. This spacing is chosen to facilitate determining survey unit locations and evaluating areas of elevated radioactivity." (pg. 5-11).
• "Survey data are converted to the same units as those in which DCGLs are expressed (Section 6.2.7). Identification of potential radionuclide contaminants at the site is performed through laboratory and in situ analyses. Appropriate regulatory DCGLs for the site are selected and the data compared to DCGLs." (pg. 5-14).
• "For characterization data that are used to help guide remediation efforts, the survey data are used to identify locations and general extent of residual activity. The survey results are compared with DCGLs, and surfaces/environmental media are differentiated as exceeding DCGLs, not exceeding DCGLs, or not contaminated, depending on the measurements results relative to the DCGL value. Direct measurements indicating areas of elevated activity are further evaluated and the need for additional measurements/samples is determined." (pg. 5-14).
• "Documentation of the site characterization survey should provide a complete and unambiguous record of the radiological status of the site. In addition, sufficient information to characterize the extent of contamination, including all possible affected environmental media, should be provided in the report. This report should also provide sufficient information to support reasonable decontamination approaches or alternatives." (pg. 5-14).
• "The effectiveness of decontamination efforts in reducing residual radioactivity to acceptable levels is monitored as the decontamination is in progress by a remedial action support survey. This type of survey guides the cleanup in a real-time mode." (pg. 5-17).
• "The objective of the remedial action support survey is to detect the presence of residual activity, at or below the DCGL criteria. Although the presence of small areas of elevated radioactivity may satisfy the elevated measurement criteria, it may be more efficient to design the remedial action support survey to identify residual radioactivity at the DCGL_W (and to remediate small areas of elevated activity that may potentially satisfy the release criteria). Survey instrumentation and techniques are therefore selected based on the detection capabilities for the known or suspected contaminants and DCGLs to be achieved." (pg. 5-17).
• "There will be radionuclides and media which cannot be evaluated at the DCGL_W using field monitoring techniques. For these cases, it may be feasible to collect and analyze samples by methods which are quicker and less costly than radionuclide-specific laboratory procedures. Field laboratories and screening techniques may be options to more expensive analyses." (pg. 5-17).
• "Remedial action support surveys may not be feasible for surfaces contaminated with very low energy beta emitters or for soils or media contaminated with pure alpha emitters." (pg. 5-19).
• "The objective of final status surveys is typically to demonstrate that residual radioactivity levels meet the release criterion." (pg. 5-14).

• "There are certain radionuclides which will be essentially impossible to measure at the DCGLs in situ using current state-of-the-art instrumentation and techniques because of the types, energies, and abundances of their radiations. Examples of such radionuclides include very low energy, pure beta emitters such as ³H and ⁶³Ni [⁹⁰Sr, ⁹⁹Tc] and low-energy photon emitters such as ⁵⁵Fe and ¹²⁵I. Pure alpha emitters dispersed in soil or covered with some absorbing layer will not be detectable because alpha radiation will not penetrate through the media or covering to reach the detector." (pg. 6-1).
• "Scanning surveys are performed to locate radiation anomalies indicating residual gross activity that may require further investigation or action. In other words, scanning is used to locate small areas of elevated activity that exceed the investigation level. ... sampling on the commonly used grid spacing may have a low probability of identifying such small areas. For this reason scanning surveys are typically performed before direct measurements or sampling." (pg. 6-3).
• "DCGLs for residual activity are typically stated in units of net activity (i.e., above the level occurring in background)." (pg. 6-6).
• "... levels of direct radiation (exposure rates) and some naturally occurring (uranium and thorium decay series, ⁴⁰K) or man-made (¹³⁷Cs, ^238-240Pu) radionuclides are typically present in the environment at levels that are easily quantifiable and may have background levels that are significant relative to the DCGLs (Wallo, et al. 1994). As background levels approach, or even exceed, the DCGLs, the number of measurements estimated to demonstrate compliance using the statistical tests may increase." (pg. 6-7).
• "As with laboratory-based gamma spectrometry, in situ gamma spectrometry provides the means to discriminate among various radionuclides on the basis of characteristic gamma and x-ray energies and thus constitutes a nuclide-specific measurement." (pg. 6-7).
• "Traditionally, gamma-ray spectrometry performed in the field for low-level contamination was limited to relatively strong gamma emitters. Recent availability of large high-efficiency germanium detectors means that in some cases rather low intensity gamma emitters can be measured -- i.e., those with emission intensities of a fraction to a few percent. ... Using arrays of detectors to increase sensitivity, even highly attenuated low-energy emitters such as ²⁴¹Am (60 keV) are measurable (Reiman 1994). Using other types of detectors, such as large area proportional counters, it is also possible to measure the x-rays associated with certain alpha emitters, such as ²³⁸Pu, ²³⁹Pu, and ²⁴⁰Pu. Photon spectrometry is not possible for pure beta emitters such as ⁹⁰SR." (pg. 6-8).
• "Radation survey data are usually obtained in units, such as the number of counts per unit time, that have no intrinsic meaning relative to DCGLs. For comparison of survey data to DCGLs, the survey data from field and laboratory measurements should be converted to DCGL units." (pg. 6-9).

• "Liquid scintillation is the preferred method for counting low-energy beta-emitters (e.g., ³H, ¹⁴C, and ⁶³Ni) and is excellent for counting high energy beta (e.g., ³²P) and low-energy photon-emitters (e.g., ⁵⁵Fe and ¹²⁵I)." (pg. 7-20).
• "It is prudent to subject at least a representative number of soil or sediment samples to gamma spectral analysis, even if no gamma emitters are expected, as a check on the reliability of the identification of potential contaminants. Either solid-state germanium detectors or sodium iodide scintillation detectors may be used. However, the solid-state detector has an advantage because of its ability to resolve multiple gamma photopeaks that may differ from each other by as little as 0.5 to 1 keV." (pg. 7-20).
• "Spectra should also be reviewed for gamma-photopeaks not previously identified as principal facility contaminants of concern." (pg. 7-20).
• "Radionuclides emitting primarily alpha particles are best analyzed by wet chemistry separation followed by counting to determine amounts of specific alpha energies present." (pg. 7-21).
• "Analysis of soil/sediment samples for most pure beta radionuclides, such as ⁹⁰Sr, ⁹⁹Tc, and ⁶³Ni generally involves wet chemistry separation, followed by counting using liquid scintillation or beta proportional instruments." (pg. 7-21).

• "Radiological survey data are usually obtained in units, such as the number of counts per unit time, that have no intrinsic meaning relative to DCGLs. For comparison of survey data to DCGLs, the survey data from field and laboratory measurements should be converted to DCGL units." (pg. 8-2).
• "Note that some individual survey unit measurements may exceed the DCGL_W even when the survey unit as a whole meets the release criterion. In fact, a survey unit that averages close to the DCGL_W might have almost half of its individual measurements greater than the DCGL_W. Such a survey unit may still meet the release criterion." (pg. 8-13 - 8-14).
• "The assumption is that the survey unit measurements are independent random samples from a symmetric distribution." (pg. 8-14).
• "Unusually high readings should be flagged and measurements that exceed the larger of either 3 standard deviations above the mean of the survey unit or the DCGL_W, should be investigated further." (pg. 8-24).