SUMMARY OF THE REMEDIATION TECHNOLOGIES DEVELOPMENT FORUM
PHYTOREMEDIATION ACTION TEAM
TOTAL PETROLEUM HYDROCARBON IN SOIL SUBGROUP
CONFERENCE CALL

June 3, 1999
1:00 p.m.-2:00 p.m.

On June 3, 1999, the following members of the Phytoremediation Action Team, Total Petroleum Hydrocarbon (TPH) in Soil Subgroup, met in a conference call:

Lucinda Jackson, Chevron Corporation (RTDF Action Team Co-Chair and Subgroup Co-Chair)
Phil Sayre, U.S. Environmental Protection Agency (EPA) (Subgroup Co-Chair)
Evelyn Drake, Exxon Research and Engineering
Steve Geiger, ThermoRetec, Inc.
Peter Kulakow, Kansas State University (KSU)
Ernest Lory, Naval Facilities Engineering Service Center
Bud Prevatt, Phillips Petroleum Company
C. M. (Mike) Reynolds, U.S. Army Cold Regions
Steve Rock, EPA
David Tsao, BP Amoco
Duane Wolf, University of Arkansas

Also present were Andrew White of Microbial Insights, Inc., and Christine Hartnett of Eastern Research Group, Inc. (ERG).

UPDATE ON FIELD DEMONSTRATION SITES

The TPH in Soil Subgroup has created a field study program to evaluate how effectively plants degrade petroleum hydrocarbons across a range of test sites. Conference call participants provided the following site reports:

• Site A. Lucinda Jackson said that the phytoremediation study at Site A is well under way. Time zero (T₀) samples were collected in early December 1998 and time one (T₁) samples were collected at the end of California's growing season (i.e., early June 1999).

• Site B. Steve Rock said that T₀ samples have been collected and plants (including poplars and hackberries) have been established at Site B. He summarized the sampling results, noting that no contaminants were detected in the top 12 inches of soil, but that they were detected below that depth.

• Sites C, D, and E. Mike Reynolds said that he had little new information to report on Sites C, D, and E. He briefly noted that T₁ samples were collected from Site E in early June and that samples will be collected from Site D in mid- to late June.

• Site F. Steve Geiger said that time initial (T_i) samples have been collected from Site F. The results indicated that polycyclic aromatic hydrocarbon (PAH) concentrations are fairly low. As a result, Geiger's client obtained permission from the state to spread a more contaminated soil over the top 6 inches of the test treatment plots. (This soil was contaminated as a result of onsite MGP processes. It has been stored onsite in bins since it was excavated.) Geiger said that T₀ sampling and planting activities will be completed by mid-June.

• Site G. Peter Kulakow said that sampling and planting have not yet been initiated at Site G. Kulakow asked Reynolds for information on forming a research agreement with EPA. Reynolds agreed to send the information to him.

• Site H. David Tsao said that he hopes to initiate field activities at Site H in the near future. The site owners, Tsao said, have applied for the appropriate permits and are in the process of removing concrete pads from the site. Tsao said that he hopes to collect T₀ samples in July 1999 and to receive a finalized Cooperative Research and Development Agreement (CRADA) from EPA in the near future. Phil Sayre agreed to ask EPA's Larry Fradkin for an update on the CRADA's status.

• Site J. Duane Wolf said that the University of Arkansas plans to initiate a phytoremediation demonstration project at Site J; the University has signed a CRADA and forwarded it to EPA for finalization. Wolf said that field activities will be initiated after his work plan has been thoroughly reviewed. He was not optimistic that seeds will be spread before the end of the spring growing season, feeling that planting will more likely be performed in early September. Wolf said that a mix of Bermuda and crab grass will be used in one of the treatments plots. Evelyn Drake expressed interest in seeing his results, saying that the Petroleum Environmental Research Forum (PERF) got good results when they experimented with Bermuda grass.

FERTILIZATION ISSUES

Fertilized Control Plots versus Unfertilized Control Plots

Jackson said that conference call participants have had several discussions about whether unplanted control plots should be fertilized. In the Subgroup's original protocol, she said, members recommended (1) fertilizing planted plots to optimize growing conditions, and (2) leaving nonvegetated control plots unfertilized. Over the last several months, Jackson said, the decision to leave the nonvegetated control plots unfertilized has been challenged. In fact, during the previous conference call (dated May 4) participants decided to change the protocol to recommend that all plots be fertilized at the same rate. (This would mean that nonvegetated control plots would be fertilized if their counterparts, planted plots, are also fertilized.) Jackson said that she is uncomfortable with this decision; she reopened the topic for debate.

All of the conference call participants agreed that it would be best if investigators could include two nonvegetated control plots (one that is fertilized and one that is not), in addition to two planted plots (one that is planted with a standard mix and one that is planted with a native species). Reynolds said that both control types have been established at Sites C, D, and E, but he acknowledged that most researchers will not be able to include both. (Jackson confirmed this, saying that Sites A and B will only have one control. Also, Geiger said that it is unlikely that his client can afford to include both types at Site F. Likewise, Wolf said that Site J is too small to include both types.) In cases where only one control type can be established, conference call participants debated whether it should be fertilized or unfertilized.

Jackson, Drake, Reynolds, and Wolf argued in support of leaving the nonvegetated control plot unfertilized, pointing out that this setup will allow investigators to answer the Subgroup's original question: How does using phytoremediation technologies (which include plants, fertilizer, lime, and irrigation) compare to doing nothing? Jackson said that some Subgroup members want to change the focus of the question and ask instead, "Does adding plants to a system enhance contaminant degradation over that which would occur if fertilizer was used alone?" She acknowledged that fertilizing the nonvegetated control would be essential if the Subgroup wanted to answer this second question, but she strongly encouraged members to hold to their original focus. Jackson said that she did not want to belittle the importance of differentiating between plant and fertilizer effects, but said that other research teams are already focusing on this. (The University of Alberta has conducted a short-term greenhouse study to compare the degradation capabilities of systems that have plants plus fertilizer against those with just fertilizer. Jackson said that the results suggest that the plants have not enhanced biodegradation beyond what the fertilizer has done alone. Reynolds said that this result is not surprising given the short-term nature of the study. Many researchers believe that plants must be established for several years before their complete benefits become apparent.)

Sayre summarized one of the arguments that has been offered in support of fertilizing the nonvegetated control: peer reviewers will look unfavorably on the Subgroup's work if the control plots differ from the planted plots in two variables (i.e., plants and fertilizer) rather than just one. Sayre questioned whether the Subgroup's data will be publishable if their experimental design is considered to have too many confounding variables. Drake did not think that this would be a problem if the Subgroup is clear about what question they are trying to answer. In addition, she reminded the Subgroup that the original intent of the project was to collect information for potential users rather than to create a perfectly sound scientific experiment.

Kulakow feared that some potential users may choose to use fertilizer as a stand-alone treatment unless the Subgroup proves that adding plants to a system enhances contaminant degradation. Jackson and Reynolds disagreed, saying that it is unlikely that anyone would use fertilizer by itself as a treatment. Drake noted that the other benefits associated with plants, such as dust and erosion control, will likely compel users to add plants to their remediation systems even in the absence of hard evidence of enhanced biodegradation. Sayre questioned this, noting that dust and erosion can be controlled by simply letting weeds grow on an area that has been treated with fertilizer. Therefore, he said, some users might need to be convinced that establishing a managed plant system will provide additional benefit over just throwing fertilizer on a field and letting volunteer species run their course. Rock said that he would be surprised if the regulatory community got bogged down in teasing out the effects of fertilizer versus plants. He said regulators will probably look at phytoremediation as a holistic system and will simply ask for instructions on how to use the technology if the Subgroup shows that it is more effective than doing nothing. Drake said that Royal Nadeau, a regulator from New Jersey, echoed this sentiment in an e-mail.

Conference call participants debated the topic at great length. In the end, Jackson proposed encouraging researchers to include two types of controls if possible, but specifying that the nonvegetated control should remain unfertilized if only one type of control plot can be established. Most of the conference call participants agreed with this approach, but Sayre said that it was not an optimal solution. In an attempt to strike a compromise, Tsao noted that the protocol lists the native plant treatment plot as optional. Assuming that most teams will only be able to afford three types of plots, he advised that some set up their design to include one nonvegetated control plot (unfertilized) and two planted plots (both fertilized, one with standard mix and the other with native species), while other teams set up their design to include two nonvegetated control plots (one with fertilizer and one without) and one planted plot (fertilized, with the standard mix). Conference call participants liked his suggestion.

REDEFINING SAMPLING DEPTHS

Jackson said that the Subgroup's protocol suggests collecting samples at a depth of 0 to 6 inches and 0 to 18 inches. She asked whether Subgroup members thought "inches" was the best way to define depth and questioned whether it would be more useful to use a definition based on plant physiology, such as "areas with dense roots" and "areas below the root zone." Reynolds argued in support of using a physiological definition, noting that root depths for the standard plant mix will differ dramatically across sites depending on climate and soil type. Drake agreed that this was a good point, but said that she does not want to completely abandon the idea of using inches to define depth. From a human health risk perspective, she said, regulators require cleanups to extend to specific prescribed depths.

MISCELLANEOUS

Kulakow said that he updated the Subgroup's protocol and distributed it to Jackson, Sayre, and Rock. The new version incorporates many of the Subgroup's recent decisions about when to mow, how many cores to collect, and whether to fertilize the nonvegetated control. (This latter point will need to be edited based on the new decisions made during this conference call.) Kulakow said that information has also been added on Microbial Insights Inc.'s Phospholipid Fatty Acid Analysis and Denaturing Gradient Gel Electrophoresis procedures.

ACTION ITEMS

• Reynolds agreed to send Kulakow information on how to form a research agreement with EPA.

• Sayre agreed to ask Fradkin for an update on the CRADA that is being established between EPA and BP Amoco.

• Conference call participants agreed that Subgroup's protocol needs to be updated to reflect their decision about fertilizing nonvegetated control plots.

• Kulakow agreed to create a questionnaire and to distribute it before the next Subgroup conference call. (The questionnaire will ask Subgroup members to provide information about their sites and study design.)

• Jackson and Rock agreed to forward cost-tracking spreadsheets to ERG for distribution to the entire Subgroup.

• ERG agreed to set up the next conference call and to notify participants of the phone number and access code. (The call is scheduled for July 8, 1999, from 1:00 p.m. to 3:00 p.m. EST.)