U.S. EPA Contaminated Site Cleanup Information (CLU-IN)


U.S. Environmental Protection Agency
U.S. EPA Technology Innovation and Field Services Division

ITRC Internet-Based Training Course Questions Wiki

LNAPLs

ITRC Disclaimer, Privacy, and Usage Policies

The training page, with links to registration or archive, Links page, and feedback, for the 3-part LNAPL course is available at
Part 1 - An Improved Understanding of LNAPL Behavior in the Subsurface
Part 2 - LNAPL Characterization and Recoverability
Part 3 - Evaluating LNAPL Remedial Technologies for Achieving Project Goals

Remaining Simulcast Questions and Answers from October 11, 2016 LNAPL Part 2 class

Question 1: Is there an equation similar to Parker et al for residual saturation? What would you recommend using as a delineation criteria? Residual Saturation?

  • Trainer Name: response

Question 2: You mentioned that a gasoline release stabilizes more quickly than a diesel release. Why is that?

  • Trainer Name: response

Question 3: On the slide 25 data over time, what did you actually test for to get you quantitative data that is then shown on the slide?

  • Trainer Name: response

Question 4: Slide 43 - Do you know of any automated lnapl bail test technique other than bailing using a traditional Bailer and using the Oil/Water Interface meter?

  • Trainer Name: response

Question 5: Slide 46 - What accounts for the disconnect in many wells between the initial product thickness and the volume recovered?

  • Trainer Name: response

Question 6: Slide 47 is b sub o the lnapl saturated thickness of the aquifer thickness?

  • Trainer Name: response

Question 7: Can you discuss how LNAPL recoverability (transmissivity) varies with the type of product or composition of the LNAPL?

  • Trainer Name: response

Question 8: What techniques can one use to estimate natural source depletion? Is NSZD same thing and Natural Attenuation?

  • Trainer Name: response

Question 9: Slide 44 - Why is the piezometric surface between the water surface elevation and the oil surface elevation?

  • Trainer Name: response

Question 10: Slide 60 - LIF/ROST is used in this slide but it is not defined in the notes. What does LIF/ROST stand for?

  • Trainer Name: response

Remaining Simulcast Questions and Answers from October 6, 2016 LNAPL Part 1 class

Question 1: Can your provide a reference for slide 73 pipeline leak plume?

  • Eric Nichols: The information comes from a case study entitled "Field Study of LNAPL and Dissolved-Phase Plume Genesis" presented at the 2004 NGWA Petroleum Hydrocarbons Conference. The abstract may be accessed at http://info.ngwa.org/GWOL/pdf/041980209.pdf

Remaining Simulcast Questions and Answers from January 28, 2016 LNAPL Part 3 class

Question 1: On slide 97, How did you know that 90% of the LNAPL was ~ 120,000 gallons, as shown by red dot on graph?

  • Rick Alhers: The graph shown on slide 97 is output from the LDRM model. The model uses assumptions about in-well LNAPL thickness, LNAPL properties, soil properties, and extent of LNAPL (based on site data) to estimate the volume of recoverable LNAPL. Residual LNAPL in the smear zone above or below the mobile interval and residual LNAPL beyond the lateral extent of the mobile LNAPL is not included in this estimate.

Remaining Simulcast Questions and Answers from September 24, 2015, LNAPL Part 2 class

Answers to questions are available in the archive of the LNAPL Part 3 class at https://clu-in.org/conf/itrc/LNAPLrt_092915/. Questions are answered in the 1st Q&A break at 51 minutes into the archive.

Question 1: Are you suggesting that there are scenarios where one could leave behind residual LNAPL and terminate the remedial efforts?

  • Trainer Name: response

Question 2: What is the relative tradeoff of heating LNAPL to decrease viscosity and enhance recovery? IE what is the temperature you need to increase it vs how much additional recovery do you get?

  • Trainer Name: response

Question 3: I had understood that biodegradation is usually confined to the fringes of a LNAPL plume as the product may be too toxic for the microbial community to survive or metabolize. Is it possible for biodegradation to occur closer to the centre of a plume? Is biodegradation a reasonable remediation strategy in a large plume?

  • Trainer Name: response

Remaining Simulcast Questions and Answers from January 22, 2015, LNAPL Part 1 class

Question 1: Given the determination that pump/treat systems are less effective at actual product recovery than first thought years ago, would you consider pump and treat as a less than desirable approach to cleanup than other options? Do you see pump and treat making a comeback or just a narrower application range?

  • Trainer Name: response

Question 2: At petroleum UST sites, if there is known or suspected LNAPL, the push to 'get remediating' from most regulators is often immediate and not conducive to this more considered approach. Is it possible to condense this approach?

  • Trainer Name: response

Remaining Simulcast Questions and Answers from January 8, 2015, LNAPL Part 1 class

Question 1: What is the pace that these concepts and improved understanding is penetrating the regulatory frameworks being applied about LNAPL stability & recoverability?

  • John Menatti: There are several States that have incorporated the ITRC LNAPL Management concepts into their LNAPL Guidance documents. We discuss this in detail in the 2-day classroom training.

Question 2: In the example of confined LNAPL would it not move to unconfined conditions with time since the LNAPL surface would equilibrate to 1 atmosphere in the formation?

  • Trainer Name: (response)

Remaining Simulcast Questions and Answers from July 15, 2014, LNAPL Part 2 class

Question 1: Mentioned in discussion of Slide 32. Please describe how the difference between transmissivity at high seasonal groundwater level and transmissivity at low seasonal groundwater level can be used to determine residual saturation. Is this for the entire smear zone? How well could we expect this to correlate to residual saturation estimates from TPH concentrations or from laboratory methods?

  • Trainer Name: (response)

Question 2: Slide 46 - It appears that there was an almost inverse correlation between measured thickness in the wells and the amount of product recovered, even for similar product types (wells 3 through 7). Can you explain the reason(s)?

  • Trainer Name: (response)

Question 3: how does one determine LNAPL saturated thickness in order to calculate Tproduct?

  • Trainer Name: (response)

Remaining Simulcast Questions and Answers from July 8, 2014, LNAPL Part 1 class

Question 1: Slide 56 - What are the soil types?

  • Trainer Name: (response)

Question 2: Slide 49 - If NAPL is at residual saturation in the vadose zone in an unconfined setting, and water level rises - it seems that NAPL would tend to stay in the soil pores rather than flow into a monitoring well. If we see NAPL flowing into a well under this scenario - are there other conditions other than semi-confining or confining situations which could allow NAPL in the vadose zone feed into a monitoring well?

  • Trainer Name: (response)

Question 3: Slide 72 - Soil type(s)? groundwater table fluctuation over those same time frames?

  • Trainer Name: (response)

Question 4: can you reiterate the lnapl thickness in a well in different water table levels and in an unconfined and confined water table

  • Trainer Name: (response)

Remaining Simulcast Questions and Answers from March 4, 2014, LNAPL Part 1 class

Question 1:Please, define capillary pressure

  • Trainer Name: (response)

Question 2: How do you estimate specific volume when estimating LNAPL volume? From discussion during the class, I understand how to use "specific volume" in in estimating LNAPL volume but not how to arrive at the estimated value for "specific volume".

  • Trainer Name: (response)

Question 3: Modeled v measured LNAPL saturation - is it more common to overestimate equilibrium LNAPL saturation in fine grained relative to coarse grained formations? (e.g. slide 36, modeled soil types 3,4,5)?

  • Trainer Name: (response)

Question 4: Expelling water requires continuity of the pore water column. In some interbedded materials, you could isolate blocks of water by intrusion into the coarser layers. Once the water column is broken, this water cannot leave. (sORRY ABOUT HITTING the Cao lock key

  • Trainer Name: (response)

Question 5: The effective density of the LNAPL column will be affected by its saturation. This will then affect the height of the LNQPL MOUND above the water table. tHIS seems to greatly complicate the calculations of this as you would need an iterative solution to account for this. please comment

  • Trainer Name: (response)

Question 6: By default, LNAPL thickness always refers to the well thickness, not the formation thickness. Is that a uniform criterion everywhere?

  • Trainer Name: (response)

Question 7: What are some specific details that will be covered in the classroom training that is NOT covered within the 3-part internet based training?

  • Trainer Name: (response)

Question 8: What is typical range of soil contaminant concentrations for fuel oil TPH in a fine sand that would exist just below residual saturation

  • Trainer Name: (response)

Question 9: Slide 25: How does one determine values for interfacial tension and wettability?

  • Trainer Name: (response)

Question 10: Movement of LNAPL can also be strongly influenced by a low surface tension. The requirement for a minimum entry pressure becomes irrelevant in such cases. Free phase can then travel in the form of small droplets. ... for example when compounds such as polar organic compounds (glycol antifreeze, alcohols for example) or surfactants are present. Models that do not take this into account fail to predict contaminant migration in such cases.

  • Trainer Name: (response)

Remaining Simulcast Questions and Answers from March 5, 2013, LNAPL Part 1 class

Question 1: For a homogeneous unit with a 10 foot LNAPL column, can you explain why the highest LNAPL saturation is at the top (per shark fin model) when 2 or more feet below the top would have a higher head? - Participant from National Park Service; Fort Collins, CO, United States

  • Eric Nichols: At vertical equillibrium, the highest LNAPL saturation tends to occur near the static water table. This is approximately 1/3 of the distance from the top of the LNAPL-affected zone. Below this point, LNAPL has to compete with water to occupy the same pores. Water pressure increases with depth, thus the LNAPL saturation tends to diminish.

Question 2: Do you know of any studies to quantify degree of saturation with UVIF or ROST data? - Environmental Consultant; Walnut Creek., CA, United States

  • (instructor name): (add text)

Question 3: Refer Slides 47-53. Why are case studies done 20-22 years ago being used for the training. Are there any on-going or recent studies being done to study LNAPL movement? Thanks - State Regulator; Bethlehem, PA, United States

  • Eric Nichols: The case studies are selected to illustrate certain points, regardless of the age of the study. LNAPL occurrence, movement, and depletion are areas of active research. A good journal for finding more recent examples is Ground Water Monitoring and Remediation.

Question 4: Slide 50: It seems like we'd see very high dissolved phase concentrations in water samples collected from wells in each of these scenarios; even if LNAPL is not present in the well. Do you have any information about dissolved-phase concentrations in these different conditions? Environmental Consultant; Portland, OR, United States

  • (instructor name): (add text)

Question 5: We have worked on a refinery plume released in 1945 that is still migrating to a nearby creek. If the "stable plume configuration" includes offsite migration to sensitive receptors, don't we need to do more than monitor the plume footprint? - University Participant; Kalamazoo, MI, United States

  • Eric Nichols: If the LNAPL has already migrated to receptors, or is expected to migrate to receptors based on the LCSM, then active LNAPL management actions are likely needed, such as migration control, hydraulic recovery, etc.

Question 6: Do all labs do total and residual saturation testing? - Environmental Consultant; Hamilton, NJ, United States

  • (instructor name): (add text)

Remaining Simulcast Questions and Answers from September 11, 2012, LNAPL Part 1 class

Question 1: Slide 38 illustrates "shark fin" geometry for differing grain sizes for gasoline LNAPL. Is there a reference document or paper that illustrates the same conceptual graphs and volume estimates for different LNAPL types (diesel fuel, Bunker C, etc)? Thanks - Consultant; Reston, VA, United States

  • (instructor name): (add text)

Question 2: Please suggest a guidance for bail down testing. Thank you. - Consultant; Oakland, CA, United States

  • Eric Nichols: A good guide for LNAPL baildown testing and recoverability evaluation is ASTM E2856-12 Standard Guide for Estimation of LNAPL Transmissivity, available at www.astm.org.

Question 3: Can I calculate the total volume of a LNAPL plume? - Consultant; Sao Paulo, Brazil

  • Eric Nichols: It is possible to estimate total volume of LNAPL given a sufficient number of point measurements of LNAPL saturation, and integrating these measurements vertically to produce resulting LNAPL specific volumes. For each measurement location, you need define a region surrounding the location where you think the LNAPL and soil conditions are similar enough to apply that LNAPL specific volume estimate. Then multiply each specific volume by its area, and sum across the areas. There is considerable uncertainty in any LNAPL plume volume estimate, so be very cautious about using such estimates for making LNAPL management decisions.

Question 4: Slides 49-51 deal directly with a site I am struggling with but don't identify the site. Can you identify it so I can look for more information? - Consultant; Oakland, CA, United States

  • (instructor name): (add text)

Remaining Simulcast Questions and Answers from May 17, 2012, LNAPL Part 2 class

Question 1: I'm looking at a site located in an estuary area. Groundwater is strongly tidally influenced - approximate range over the tidal cyale is 2m. The site has been filled to achieve ground levels so groundwater in the upper part of the tidal range is within variable sand and clay fill, and in the lower part of the tidal range within the natural drift of coarse sand. Have you any tips on obtaining robust transmissivity estimates in this environment? - International participant; UK

  • (instructor name): (add text)

Remaining Simulcast Questions and Answers from December 13, 2011, LNAPL Part 2 class

Question 1: Human Risk Assessment: In Spain, the regional authority for Madrid requires remove LNAPL before evaluating the risk for potential receptors, I would like to know how you proceed to conduct a quantitative risk assessment when there is LNAPL at the site where you are investigating. Is there any software to calculate risk assessment for LNAPL? or considering we hypothetically remove the LNAPL which concentration may introduce in the software if we do not have this analytical data?. In Spain, we are familiarised with RBCA software and Risc software. I would appreciate it if you could provide me some links related to this question - Environmental Consultant; Madrid, Spain

  • Pamela Trowbridge, PG: Sites I am familiar with here in the US have closed with residual saturation levels of LNAPL left in the subsurface. Using characterization and LNAPL transmissivity, you can show that residual LNAPL will not migrate to a new receptor. It may still provide long term dissolved contamination that would need to be addressed in the risk assessment. But if there are no pathways available, then the risk is minimal. You may have no long term risk to the occupier of the property, but you can also calculate the risk to a construction worker who may encounter the residual LNAPL body. We also have institutional controls that we put in place that allow us to deed restrict the use of groundwater on a property. American Petroleum Institute has good documents and models you can use. See attached link. http://www.api.org/ehs/groundwater/lnapl/index.cfm

Remaining Simulcast Questions and Answers from December 8, 2011, LNAPL Part 1 class

Question 1: Tracer Test done on monitoring wells measures flow through well and dependant upon degree of connectivity between well and formation. Slotted PVC pipe and generic gravel pack very poor development. Result is higher water content due to lack of development and hollow stem auger drilling method. How does this test account for these issues? If using data for calculating Transmisivity of product. Development and well design very important is it not? - Environmental Consultant; Millersville, MD, United States

  • (instructor name): (add text)

Question 2: If well thickness cannot be a reliable indicator (for all the reasons discussed) then how can it be reliably used for "shark fin" Pc calculations, and in turn for saturation calculations? Meanwhile field sample residual saturation determinations/measurements are hugely debatable/variable. Theories are sound, but assessing practical site "closure" is difficult. Comments? - State Regulator; Boston, MA, United States

  • Eric Nichols: Indeed, the real world is often much more complex than the simplified examples we use to illustrate LNAPL occurrence and migration. However, if you have well LNAPL fluid level info, a sound LNAPL conceptual site model, including site-specific information such as geology and capillary pressure (moisture retention) data, then you can estimate the ideal "shark fin" distirbution in about 5 minutes using LDRM. But make sure you know what you are doing - the old garbage into a model = garbage out still applies! That model is available at LNAPL Distribution and Recovery Model (LDRM) API Publication 4760 available at : http://www.api.org/ehs/groundwater/lnapl/lnapl-reg.cfm

The key is to understand the uncertainties in making such estimates, and take these into account when making LNAPL management decisions.

Remaining Simulcast Questions and Answers from June 7, 2011, LNAPL Part 1 class

Question 1: Isn't the determination of which material would be the wetting fluid related to the surface tension of the liquid? - Participant; Woodstock, GA, United States

  • Mark Adamski: I think whether a fluid is a wetting fluid on a given surface is mainly a function of the wetting angle between the fluid and the surface. There is a nice explanation on Wikipedia - http://en.wikipedia.org/wiki/Wetting

Question 2: In the volume estimation example, what suggestions do you have for defining the areas represented by the wells? - Local Government Participant; Ann Arbor, MI, United States

  • Mark Adamski: You need to define those areas based on the fact that you think the LNAPL conditions and soil conditions are similar to those at the location where you developed the LNAPL distribution and resulting LNAPL specific volume. This is a considerable source of error in any LNAPL plume volume estimate.

Question 3: Should calculations for groundwater elevations where LNAPL is present need to be adjusted in proportion to the measured thickness (e.g. 1.5 feet vs. 0.2 feet), or any other site specific parameters? - Environmental Consultant; Wilmington, DE, United States

  • Mark Adamski: Not sure I am interpreting the question correctly. But yes, if you want to understand the water table (or potentiometric surface) elevation at a well that has LNAPL you do need to correct for the presence of the LNAPL. That correction is LNAPL thickness multiplied by the LNAPL density. That result is then added to the Oil/water interface elevation that was measured in the well beneath the LNAPL.

Question 4: Could changing aquifer conditions from unconfined to confined conditions assist in LNAPL recovery? e.g. by injection of groundwater underneath LNAPL and thereby 'pushing' it into a recovery well? - Participant; Brussels, Belgium

  • Mark Adamski: Possibly: (1) On the bad side: it would be likely be difficult to pump enough water to create and maintain a confined LNAPL condition. (2) Also on the bad side: You would likely decrease the LNAPL mobility and also increase the residual LNAPL saturation in a confined condition (3) On the good side: in cases where we have seen conditions go from unconfined to confined LNAPL conditions the radius of capture for skimming and low water pumping rate recovery systems has increased to as much as 150ft. (4) Summary: I do not think I would try it. May be a good topic to research

Question 5: What are you doing to educate the regulatory community? - Environmental Consultant; St Paul, MN, United States

  • Mark Adamski: This ITRC Internet-based training is heavily attended by state regulators. In addition the ITRC classroom training is typically hosted by a State regulatory agency. Also, half of the attendees at the classroom training will be state regulators.

Question 6: I didn't understand the concept of Wetting/Non-wetting fluid (Slides 27 & 28), which is drawn upon in Slides 34 thru 36. One other participant posed a related telephone question during the Q&A session (Slide 46). Is this a common lack of understanding; if so, is a more-detailed initial explanation needed? - Participant via feedback form

Question 7: Is there any quick/practical way to establish the shark fin profile from a well data and site specific information? - Navy Participant; Washington; D.C.

  • Mark Adamski: Yes, if you have well LNAPL fluid level info and site specific info such as geology and capillary pressure (moisture retention) data you can get a shark fin in about 5 minutes using LDRM. But make sure you know what you are doing - the old garbage into a model = garbage out still applies! That model is available at LNAPL Distribution and Recovery Model (LDRM) API Publication 4760 available at : http://www.api.org/ehs/groundwater/lnapl/lnapl-reg.cfm

Question 8: Please repeat the web site mentioned for "shark" fin modeling - Environmental Consultant; Leawood, KS, United States

Question 9: Please explain how the LNAPL can move from a source to a confined aquifer....through vertical migration? - Environmental Consultant; Ottawa, , Canada

  • Mark Adamski: (1) Confined LNAPL does not have to be in a 'confined aquifer' to be confined. As mentioned in passing during the webinar LNAPL can be confined in a medium grained sand lens capped by fine sand. It is the difference in capillary properties that creates the confined LNAPL. (2) I have seen LNAPL in a confined sand beneath a shallow confining clay layer. This is actually fairly common if you have a shallow clay over a high perm layer and the water table a few feet up into the confining clay. Macro pores, preferential pathways such as sand lenses, animal burrows, old geotechnical borings have all been theorized as pathways allowing the LNAPL to migrate into the sand beneath the clay. This is also facilitated by a typical downward hydraulic gradient in the clay layer.
  • Suggested references on LNAPL in fine grained soils and confined LNAPL:
  • Applied NAPL Science Review, Volume 1, issue 5 http://www.h2altd.com/knowledge-center
  • Adamski, M., Kremesec, V., Kolhatkar, R., Pearson, C., and Rowan, B. 2005. LNAPL in Fine-Grained Soils: Conceptualization of Saturation, Distribution, Recovery, and Their Modeling. Ground Water Monitoring and Remediation, 25, No.1.
  • Charbeneau, R.J., 2007. LNAPL Distribution and Recovery Model (LDRM) Volume 1: Dirstribution and Recovery of Petroleum Hydrocarbon Liquids in Porus Media. API publication 4760. API Publications, Washington, DC

Question 10: What role does the partitioning of LNAPL into water play when evaluating how stable the plume is? State Government Participant; Culpeper, VA, United States

  • Mark Adamski: A steady dissolved phase plume is one line the multiple line of evidence approach for determining LNAPL plume stability.

Question 11: In what ways are specific properties of LNAPL's important for determining LNAPL distribution and mobility? - University Participant; Utrecht, Netherlands

  • Mark Adamski: There is a fairly long list and I would refer you to the references below, but to name several: LNAPL density; LNAPL / Air surface tension; LNAPL / water interfacial tension; Air/water surface tension; LNAPL viscosity

Remaining Simulcast Questions and Answers from March 24, 2011, LNAPL Part 3 class

Question 1: Slide 94 shows that the well screen is exposed to air. This is obviously necessary due to the vertical LNAPL distribution (Slide 84). Given that the LNAPL constituents are biodegradable, did you encounter significant difficulties in maintaining well performance ( i.e., did output decline significantly over time and were frequent efforts to rehabilitate the well(s) required? - Participant; Milford, OH, United States

Rick Ahlers: This case study comes from another team participant, so those details aren't available. However, it is not uncommon for rehabilitation to increase performance of LNAPL recovery wells.

Question 2: Have instructors remediated any sites with Bunker C Oil using six-phase heating? Was it successful. What type of costs? - Environmental Consultant; New Baltimore, MI, United States

Nichols: Not personally. Given the low solubility, low volatility, and high viscosity of residual fuels such as Bunker C, these types of fuels often have very little mobility, and a low risk of exposure to volatile or soluble constituents of concern. A six-phase heating remedy would be very aggressive in these circumstances.

Remaining Simulcast Questions and Answers from March 23, 2011, LNAPL Part 2 class

Question 1: Slide 25 shows NAPL saturation as unitless. How is this used if it is unitless and what is it compared to? - New York State Regulator; Albany, NY, United States

Terrence Johnson: NAPL saturation is not really unitless, it is dimensionless. NAPL saturation has units of volume of NAPL per unit volume of pore space. NAPL saturation is compared to other NAPL saturations spatially, or to other fluid saturations, such as water or air, in the spore space.

Question 2: During periods of drought and lowering of the water table, smearing occurs in the porous media. When ground water levels are recharged and the LNAPL phase ascends, does this phenomena facilitate more LNAPL into the dissolved phase causing elevated levels of dissolved phase ground water concentrations? New Jersey State Regulator; Trenton, NJ, United States

Terrence Johnson: Yes, this phenomenon should increase the dissolution of LNAPL constituents over a greater depth as the water table rises, resulting in a larger dissolved plume in the vertical dimension. Dissolved concentrations in the subsurface may not necessarily increase, but the amount of dissolved mass in the subsurface certainly will. Will higher dissolved concentration be observed in a monitoring well? It depends on your well and method of sampling. For example, if you are sampling from a well with a large screened interval by purging and sampling; you may see higher concentration as the water table rises, as there is less in well dilution as compared to the former. However, if you sample using diffusion bags, you may not see an increase in dissolved concentrations.

Question 3: How do results of the NAPL saturation calculation (slide 25) compare to NAPL saturation that is determined by laboratory analyses (i.e. soil samples collected in the field)? - Environmental Consultant; Regina , Canada

(first last name): (response)

Question 4: On Slide 47, displacement is plotted vs. time. What is being plotted - the oil/water interface or the air/oil interface? Is the estimated K the oil conductivity? - Participant; Ashland, VA, United States

Terrence Johnson: Plotted as displacement is the difference in oil-air interface at time zero and the oil-air interface at time t as the LNAPL in the well recovers.

Question 5: Do you have an ethanol variables for the modeling that considers concentrations. - Pennsylvania State Regulator; Wilkes-Barre, PA, United States

Terrence Johnson: Ethanol is completely miscible in water and would no phase separate as LNAPL does. As a result modeling ethanol would be analogous to modeling chemical transport such as benzene in groundwater. Depending on the amount of ethanol that gets to the groundwater, the ethanol/groundwater mix may have a lower density than pure water. In that case density, density dependent flow may have to be considered. Ethanol parameters that would be needed for ethanol transport include organic carbon distribution coefficient, Henrys and biodegradation rate constants.

Remaining Simulcast Questions and Answers from March 22, 2011, LNAPL Part 1 class

Question 1: Is it possible to model the movement of LNAPL on the soil? If yes, which model can I use? - Environmental Consultant; Brazil

Terrence Johnson: Yes, it is possible to model LNAPL movement or migration. Models such as ARMOS, MARS, MOTRANS and MOVER can be used. However, LNAPL migration is relevant mostly for recent spills. Older spills tend to stabilize at a fixed footprint. In rare cases, a change in groundwater hydraulics (such as the installation of municipal well nearby) could cause a stable plume to re-migrate.

Question 2: Can you elaborate on how tidal changes influence lateral spread of LNAPL and dissolved components? - Participant

(first last name): (response)

Question 3: Refer to slide 74, what if any site conditions affected the horizontal stability of the LNAPL plume? did the empirical data support any modeling, if performed? New Jersey State Regulator; Trenton, NJ, United States

(first last name): (response)

Question 4: i.e., if the borehole acts as a sump? As follow up to previous question: How do boreholes filled with sand packs that are relatively coarser than the soils surrounding the borehole affect LNAPL thicknesses observed in a well as water tables fluctuate? Massachusetts State Regulator; Springfield, MA, United States

(first last name): (response)

Question 5: During periods of drought, and lowering of the water table, smearing occurs in the porous media. When ground water levels are recharged and the LNAPL phase ascends, does this phenomena facilitate more LNAPL into the dissolved phase causing elevated levels of GW concentrations? - New Jersey State Regulator; Trenton, NJ, United States

Terrence Johnson: Yes, this phenomenon should increase the dissolution of LNAPL constituents over a greater depth as the water table rises, resulting in a larger dissolved plume in the vertical dimension. Dissolved concentrations in the subsurface may not necessarily increase, but the amount of dissolved mass in the subsurface certainly will. Will higher dissolved concentration be observed in a monitoring well? It depends on your well and method of sampling. For example, if you are sampling from a well with a large screened interval by purging and sampling; you may see higher concentration as the water table rises, as there is less in well dilution as compared to the former. However, if you sample using diffusion bags, you may not see an increase in dissolved concentrations.

Remaining Simulcast Questions and Answers from December 9, 2010, LNAPL Part 1 class

Question 1: How does one determine how long it takes for gasoline LNAPL to convert to the dissolved phase? What literature is available regarding this issue? Nevada State Regulator; Carson City, NV, United States

Ian Hers: While we understand the mechanisms for LNAPL depletion, it is very difficult to accurately predict how long it would take for the LNAPL body to deplete given the the uncertainty in LNAPL source zones, rates of processes that deplete LNAPL, and hydrogeologic factors. API has published a model (LNAST) that evaluates LNAPL depletion and researchers that are looking at this issue include David Huntley and Gerry Beckett (see for example David Huntley and G. D. Beckett. Persistence of LNAPL sources: relationship between risk reduction and LNAPL recovery. Journal of Contaminant Hydrology. Volume 59, Issues 1-2, November 2002, Pages 3-26). There is also an on-going SERDP research project that is looking at this issue (Improved Field Evaluation of NAPL Dissolution and Source Longevity, project ER-200833). The ITRC LNAPL team also published a document on natural source zone depletion (NSZD) in 2009 that provides useful information on this issue: Evaluating Natural Source Zone Depletion at Sites with LNAPL (LNAPL-1, 2009) available from http://www.itrcweb.org/guidancedocument.asp?TID=59

Question 2: If there is a crude oil release 3 years old. There are still lighter ends on the site. Does this indicate continued release or could confining layers slowing degradation? - Louisiana state regulator; New Orleans, LA, United States

Ian Hers: It is not possible to answer this question, but can comment that when developing a conceptual site model for a site such as this, recognize that natural weathering processes that result in depletion of lighter hydrocarbons are relatively slow processes. The ITRC NSZD guidance also speaks to this issue. Evaluating Natural Source Zone Depletion at Sites with LNAPL (LNAPL-1, 2009) available from http://www.itrcweb.org/guidancedocument.asp?TID=59

Question 3: Throughout the presentation, especially to the end, there was some references to LNAPL plume, meaning pure phase or mixed pure phase/dissolved phase. Isn't plume a term that we ascribe to dissolved phase? - Local Government Participant from Canada

Ian Hers: The emerging practice is to refer to a LNAPL "body" as opposed to a dissolved groundwater plume.

Remaining Simulcast Questions and Answers from October 12, 2010, LNAPL Part 1 class

Question 1: Refers to question asked about how many borings are needed (slide 45): On a similar note, how much can you tell from one boring if you know the thickness of product in the well and soil type. - Local Government Participant; WI, United States

Sanjay Garg: One boring only tells you what is in the immediate vicinity of that boring. The distribution/spacing of borings can be a site-specific decision. The decision for spacing of borings is similar to any other traditional site data (e.g., soil or groundwater data). In-well LNAPL thickness can tell you something about the vertical extent of impacts. In scenarios where there is not a large vertical gradient or significant groundwater fluctuation and the LNAPL is unconfined, the well thickness represents the vertical extent of impacts. But if any of these conditions are not met (e.g., large vertical gradients that exceed buoyancy, significant water table fluctuation, perched or confined LNAPL), the in-well LNAPL thickness may under or over represent the impacted zone in the formation.

Question 2 From a participant; Minneapolis, MN, United States:

A) How is the polygon representing each monitoring well computed in Slide 45?

Sanjay Garg: The interpolation can be done in several ways. One approach is to draw Thiessen polygons. Another is to use a contouring software like Surfer to contour the specific volumes. The grid file of surfer will contain the interpolated specific volume for each grid cell; the size of the grid cell is set by the user at the start of the contouring. The area of the grid cell x interpolated specific volume for that individual cell gives the volume for that one cell. One can then add up all the individual volumes for each of the cells to come up with the total volume.

B) Does the volume estimation method apply to well of the same depth and screened at the same interval?

Sanjay Garg: Not sure that I understand the question. If one is estimating the volume based on predicted profiles (sharkfins) for in-well thicknesses, then they would just be estimating the LNAPL in that interval. There are limitations associated with using LNAPL in-well thickness to estimate the smear thickness, which were discussed in the webinar (e.g., water table fluctuation, perched or confined LNAPL, etc.)

Question 3: From Slide 45. Data will be from multiple historical measurements from wells, which can vary considerably with time. How do you decide which thickness to use? - Environmental Consultant; Raleigh, NC, United States

Sanjay Garg: Good question, and always a challenge. Two approaches that can be taken are (i) historical highest thickness (ii) The distance between the highest LNAPL-air interface and the lowest LNAPL-water interface (aka Smear zone). Limitation with approach i is that the highest thickness may not represent the entire smear zone and the limitation with approach ii is that you end up modeling a thickness that exceeds what the formation was ever subject to, and you may end up predicting a larger shark fin. It is always a good idea to get site-specific saturation data if a volume estimate is important.

Question 4: so based on the pressure issues we discussed is it safe to assume that most LNAPL will be caught up in the vadose zone pore spaces - Environmental Consultant; Bloomfield , CT, United States

Sanjay Garg: No. Vadose zones pores can only hold LNAPL up to their residual saturation. If the LNAPL volumes exceeds this capacity of the vadose zone, it will reach the capillary fringe/water table, continue to accumulate there until it overcomes pore-entry pressure and then push itself below the water table and move laterally.

Question 5: Can you explain how LNAPL may move horizontally under the scenario when you have fine grain native material with a saturated LNAPL condition next to an old excavation area or tank farm area that has been backfilled with a coarser material (i.e., gravel). Is there a surface tension phenomenon where the native material will "hold" LNAPL or will it drain to coarser material? - Private Sector Participant; Wayne, NJ, United States

Ian Hers: The question appears to be referring to the potential for mobile LNAPL to migrate from a source zone within fine-grained deposits to a nearby backfilled area with gravel. If the LNAPL source does not intersect the gravel deposits then the ability for LNAPL to migrate laterally will depend on the capillary properties of the fine-grained material. As indicated in the training slides, a relatively large displacement head would be required for LNAPL to overcome capillary resistance (the surface tension described above) and migrate into unimpacted soil pores. Once in gravel, much less displacement head would be needed.

Remaining Simulcast Questions and Answers from August 3, 2010, LNAPL Part 1 class

Question 1: Please send me some literatures how to measure residual NAPL saturation level in soil and API database mentioned during the Q/A session. - Environmental Consultant

Mark Adamski: The API database and other API LNAPL tools can be obtained from the API website at http://www.API.org/LNAPL

The Database is included in the interactive guide. It can be found here: http://www.api.org/ehs/groundwater/lnapl/lnapl-guide.cfm

The LNAPL Part I Links to Additional Resources page includes two papers that I presented at the NGWA hydrocarbons conference in 2003 and 2005, both I believe were in Costa Mesa California. The 2003 paper titled Residual Saturation: What is it? How is it measured? How should we use it? is a light discussion about the issues around the use of residual saturation on our environmental problems. It also has some good references on residual papers that did most of the early work. The 2005 paper by Colin Johnston of CSIRO presents some work the CSIRO did for BP on determining an f-factor for soil types other than the sands that Dr. Kueper and Steffy determined f-factors for in ~1993 and 1997. Note, I think Waddill and Parker were the first to use the term f-factor in his 1997 paper (reference for that one is below).

As far as a quick summary on how to determine residual LNAPL saturation, here is my personal take (not BPs, not APIs):

1) Residual in concept is simple and elegant; in practice at real sites it is infinitely variable and hence infinitely complicated. From a detailed technical perspective I don?t think we can (or ever will) speak in absolutes. However, we do know when we are close to residual in a general sense or below it in a general sense.

2) Residual is a function of the soil type and the maximum oil saturation that soil has contained (initial oil saturation). Therefore, with the Shark-fin LNAPL distribution, even if you had a homogeneous soil you would have a varying LNAPL residual saturation as a result of the different LNAPL initial saturations.

3) Methods for determining site specific residual saturations:

a. In the past I have used/proposed using a reasonable f-factor for the soils at your site. Serious problems with this approach include:

i. You almost never have an idea of the initial (max) LNAPL saturation. So, if you apply the f-factor you will always be above residual when you start. We know in many cases that is not true.

ii. Soils vary, so will your f-factor

iii. It costs a fortune to determine an f-factor, so I would just say use an f-factor that is published for a soil that is similar to the soil at your site

b. Some (me included) have used laboratory tests including the 1000-G Centrifuge test and Water imbibition tests. Serious problems with this approach include:

i. Lab conditions are not field conditions and can exert pressures higher than possible in the field (conservative) or can potentially deform the soils inhibiting oil flow (potentially non-conservative)

ii. Sites are heterogeneous. Can you be sure you collected a core at field conditions and submitted to the lab for testing that is representative of the zones that produce the oil at you sites. Generally is hard to verify this. I would maintain that this is the biggest problem with this method and is non-conservative. Example of this is can be seen on slide 57 of the ITRC LNAPL Part one training. Tests for residual saturation were run at 3 vertical locations: approximately 15, 16 and 29 feet. Where measured field LNAPL saturations appear to be ~ 9, 9, and 7 percent respectively. While no residual sample was collected at 13, 14, 37, 38, and 41 ft where the measured field saturation was ~15.5, 18, 13.5, 15.5, and 15 percent respectively.

c. I feel most comfortable in calling some LNAPL saturation a residual saturation if it is a measured saturation from an interval below the water table and give long monitoring periods, the Oil / Water interface does not equilibrate down to the depth of that sample. So that can provide a good measure of an LNAPL saturation that is below residual for that sand. If you have similar sand above, you can infer that the residual would be higher than the value you measured at depth.

d. Recovery system performance. If you have a recovery system that has been hydraulically recovering LNAPL for an extend period of time and it is approaching asymptotic recovery, and you can account for water table fluctuations, then the formation in general is likely close to residual. You could then sample the area and say these are representative of residual field conditions.

This may be a lot more than you asked for, but that is what happens when you start looking into the simple concept of residual saturation. It gets real hairy really fast. So you can see why we did not take 3 hours of the 2 hour ITRC training to explain residual LNAPL saturation.

Dr. Garg and I have many friendly discussions on this topic and I welcome him to provide additional context.

Also, Ian Hers, who is the normal instructor for the mobility section that I presented yesterday, presented at the Battelle Conference this last May in Monterey, CA with the title: A Review of Methods and Recent Data for Estimation of Residual NAPL Saturation. I. Hers and B. Koons. (Golder Associates)

Waddill, D.W. and Parker, J.C. (1997). Simulated recovery of light, nonaqueous phase liquid from unconfined heterogeneous aquifers, Ground Water, 35(6), 938-947.

Sanjay Garg: Thanks to Mark for providing an excellent overview and listing some of the problems with the methods available. A couple of additional points for your consideration are:

  • There are obvious challenges with point sampling (as Mark discussed). However, in this industry, we commonly make decisions based on point data (e.g., soil concentration data) and accept certain uncertainty. There are some additional issues associated with the centrifuge test, e.g., (i) whether the test itself has an impact on the soil matrix/packing, that causes the LNAPL to redistribute and not drain (ii) the test is normally done under atmospheric conditions, i.e., air is the displacing fluid, and whether that is representative of saturated conditions. Residual saturation is an incredibly important concept but an absolute value/answer is elusive. Bottomline is, if we do not try to over-interpret the residual saturation (Sor) data from the labs and use it in conjunction with other site information, we get useful information from these tests, which is, is there significant hydraulically recoverable LNAPL or not. The problem arises when we start arguing whether it is 15% or 25% (for example) that is recoverable and do not focus on the fact that it will be ~>75% that will remain after hydraulically recovery. We are focused on the wrong problem.
  • Figure at http://www.itrcweb.org/Documents/TrainingProgram_ToolsTemplates/LNAPL1.jpg shows what Mark discussed in item 3c. The two figures show the range of water-table fluctuation. (This will be in the classroom training)

Top of Page