Monitored natural attenuation (MNA) will only be considered as a remedial action if the following four conditions are found to hold:

1. There is a clear indication that the site currently poses no unacceptable risk to human health or the environment.
2. There is no active source term.
3. Plume contours are static or retreating.
4. Geochemical and/or hydrological data suggest a strong likelihood that attenuation processes are operative at the site, and that they may assure attainment of remedial goals in an acceptable time frame.

Acceptable risk, and static or retreating plumes, are often the direct result(s) of ongoing biogeochemical attenuation in the subsurface. However full-scale implementation will ultimately require a clear understanding of the specific processes that lead to those conditions by decreasing contaminant availability. Not only must the specific processes be identified, they must also be quantified to the extent that their long-term reliability can be assured. Natural attenuation pathways are typically contaminant- and site-specific. Moreover, their identification and quantification will involve a non-trivial investment of time and effort. It is, therefore, reasonable to focus MNA implementation efforts on only those sites where natural attenuation is likely to occur, and where it will account for a significant reduction in contaminant availability over time.

MNAtoolbox is a web-based tool that has been developed by DOE to assist site environmental managers and their staff and contractors to determine if MNA may be an appropriate remedial action for environmental restoration sites prior to collecting extensive characterization data. MNAtoolbox acts as a database for contaminant chemistry and degradation pathways and can be used to identify which phase transfer and degradation pathways are likely to be important.

MNAtoolbox helps focus consideration of MNA by:

1. Outlining the most likely attenuation pathways.
2. Pointing out the factors that will mitigate against MNA.
3. Identifying data needs for demonstrating attenuation.
4. Providing examples of regulatory acceptance of MNA for specific contaminants.

MNAtoolbox uses a scorecard that relies on user input site-specific data to provide a reasonably quick assessment of whether MNA is sufficiently effective to warrant its closer examination as a remedy for a particular site. It should be noted that, although site-specific, non-technical objections to reliance on MNA must be considered before implementation, the primary focus of MNAtoolbox is technical. Full documentation for MNAtoolbox is found in the technical guidance document that is linked to the MNAtoolbox home page. Below is a brief description.

The MNA pathways section of MNAtoolbox builds on an extensive examination of the soil geochemistry literature and indicates the likely biogeochemical process(es) providing significant attenuation of mobile/bioavailable contaminant. For organic contaminants, degradation rates, breakdown pathways, and sorption coefficients (K_d’s) are provided. For inorganics, the latter is provided. The extent of irreversible sorption of contaminants has been estimated for both organics and inorganics from either literature surveys (metals/radionuclides), or empirical correlation (organics) and is included in MNAtoolbox. Each contaminant module is also linked to the EPA CERCLA records of decision (RODs) involving that contaminant. MNA is not a panacea for all contaminants under all environmental conditions. The MNA mitigators section of MNAtoolbox provides information about the environmental conditions that may serve to hamper MNA for a specific contaminant.

For each contaminant, summaries of the CERCLA RODs that contain natural attenuation as a remediation method are listed in the MNAtoolbox by EPA region and by state. These summaries contain information on the site description, history, and owner; contaminants of concern; and selected remedy. (Note that a complete listing of non-CERCLA sites relying on MNA would be much larger than that in MNAtoolbox.) Note also that, although the RODs are segregated by contaminant type on the basis of word searching, natural attenuation may have been relied on for contaminants other than the comingled metals. While the decision to select natural attenuation as a remediation alternative must be made on a site-specific basis, the RODs provide an indication of the level of technical detail that has been used to support past decisions to use MNA within each EPA region and state.

A Site Screening Scorecard is associated with each contaminant to help the site manager rapidly determine if natural attenuation may be possible under site-specific conditions prevailing at the site of interest. The Scorecard is subdivided into hydrologic and geochemical sections. Credit is given, or taken away, based on the presence of, respectively, favorable or unfavorable conditions. A high score approaching the maximum of 100 indicates that MNA is likely to be effective and should be investigated. A low score does not necessarily discourage consideration of MNA at a particular site; it may simply indicate that a greater level of effort to collect characterization data and conduct modeling is required to support MNA for a particular contaminant at a particular site.

The scorecard seeks to estimate the potential for natural attenuation by summing contaminant attenuation due to six factors: (1) hydrologic dilution, (2) sorption, (3) irreversible uptake by the soil matrix, (4) mineral formation (see below), (5) biodegradation, and (6) radioactive decay. In essence, the natural attenuation factor (NAF) is the sum of a hydrologic dilution factor (HDF), a sorption factor (SF), an irreversible uptake factor (R_irv), and a biodegradation/chemical transformation factor (BF):

NAF = HDF + SF + R_irv + BF 1

The hydrologic dilution and sorption terms are both based, to the greatest extent possible, on the EPA soil screening procedures. Specifically, dilution factors are calculated according to the simple water balance and dilution model outlined in the EPA soil screening guidance. Calculation of the HDF requires the input of several site-specific hydrologic parameters including: the hydrologic conductivity, the hydrologic gradient, the mixing zone depth, the recharge rate, and the length of the source parallel to flow. The infiltration rate multiplied by the source area is the contaminant flux into the aquifer. Site managers can estimate the infiltration rate in two ways; by simply using the natural value (if the site is uncapped) or by lowering this number accordingly if the site has some hydrologic barrier on top of it.

Typically, biodegradation obeys a first order rate law; that is, breakdown rates are proportional to the amount of available contaminant. However, rate constants can vary by orders of magnitude depending upon redox state (nature and abundance of electron donors), nutrient supply, etc. In other words, degradation rate constants are site-specific. For the breakdown of most organic contaminants, the default rate constants in MNAtoolbox are set, where available, to the default values used in Bioscreen, an EPA-supported natural attenuation screening tool for fuel hydrocarbons. Default degradation rate constants for chlorinated organic contaminants, and others, have been taken from the literature. Alternatively, the AFCEE protocol can be used in MNAtoolbox to provide what we believe to be bounding estimates of biodegradation rates. The desirability of site-specific degradation rate constants cannot be overemphasized. The uncertainty associated with using generic, default values is likely to be very large.

Integration of a first-order degradation rate law for contaminant concentration C(t) gives the expression for the biodegradation/chemical transformation factor: BF = C_o/C(t) -1 = e^kt -1 = e^kx/v-1; where C_o is the initial concentration of contaminant present at time t = 0 yr; k is the degradation rate constant (yr^-1); and x is the distance (m) the contaminant travels in groundwater at velocity v (m/yr) to the nearest receptor. The latter substitution allows time-dependent breakdown to be expressed in terms of contaminant movement. Input is consequently taken either as an estimated travel time, or velocity and distance, to the nearest receptor. This approach assumes constant subsurface fluid velocities, hence constant potentiometric gradients and infiltration rates. For radionuclides the BF term tracks radioactive decay. Rate constants and supporting documentation are linked to their respective contaminant modules in MNAtoolbox. The user can override the BF values in the toolbox with site-specific values. A more extensive discussion of the origins of the scorecard are in the technical guidelines document that is linked to the MNAtoolbox home page.

Each of the terms in the NAF expression (1) is equal to zero if there is no attenuation; i.e. HDF = 0 if there is no dilution; SF = 0 if the K_d is zero because there is no sorption, etc. Each term becomes greater than zero if attenuation is predicted to occur. In other words, if no attenuation is predicted, NAF is zero. If attenuation is predicted, NAF is greater than zero. The score that is initially calculated in the scorecard is:

Score = NAF(1-NAF/100); 2

The object is to provide a score that linearly scales with the NAF at relatively low values of the latter, but asymptotically approaches 100 at very high values of NAF. This score is subsequently modified for many of the inorganic contaminants, to take into account the formation of insoluble solids, leading to a higher score. In the end, sites that score near 100 possess hydrogeochemical characteristics that are predicted to favor MNA. Low scores predict the opposite.

By outlining the pathways, as well as obstacles, to attenuation on a contaminant-specific basis, usage of MNAtoolbox provides site managers a clear identification of likely data needs for MNA implementation. Specifically, the data used as input for the scorecard (e.g., K_d’s, degradation rates) are also the input needed to construct full conceptual models that are used by regulators to assess implementation of MNA. The sensitivity of the score to the various input factors consequently gives a rough indication of the importance of the respective pathway, and supporting data, to a full conceptual model.