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

May 4, 1999
12:00 p.m.-1:00 p.m.

On May 4, 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)
Pam Davis, Exxon Product Research
Evelyn Drake, Exxon Research and Engineering
Steve Geiger, ThermoRetec, Inc.
Peter Kulakow, Kansas State University (KSU)
Ernest Lory, Naval Facilities Engineering Service Center (NFESC)
Bob Menzer, EPA
Steve Rock, EPA
David Tsao, Amoco Research Center
Albert Venosa, EPA
Duane Wolf, University of Arkansas

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


FERTILIZATION ISSUES

Fertilized Control Plots versus Unfertilized Control Plots

Conference call participants noted that the April 1999 version of the Subgroup's field study protocol states that planted plots should be fertilized to optimize growing conditions, but that the unplanted control plots should not be fertilized. The participants debated the latter point. They agreed that it would be best to have two control plots, one that is fertilized and one that is not, but they acknowledged that researchers at many sites will only be able to afford one type of control plot. Conference call participants debated whether the control plot should be fertilized or unfertilized in such cases.

Evelyn Drake argued in support of having an unfertilized control plot, noting that this setup would allow investigators to compare the effects of phytoremediation against doing nothing. She said that fertilizer dramatically increases hydrocarbon degradation by promoting bioremediation. She said that the Subgroup should not be concerned with comparing the effects of fertilized phytoremediation (which would be demonstrated in the planted/fertilized plots) against bioremediation (which would be demonstrated in the unplanted/fertilized plot.) She said that comparisons of this type have already been made and do not need to be repeated in the Subgroup's field study. (Drake said that the Petroleum Environmental Research Forum [PERF] conducted a study comparing plots that were [1] fertilized and planted, [2] unfertilized and unplanted, and [3] fertilized, unplanted, and tilled. The first and third treatments yielded fairly comparable hydrocarbon degradation results. Drake said that these results showed that using phytoremediation is better than using tilled bioremediation [the fertilized, unplanted, and tilled treatment] because it is more cost-effective: plants till the soil naturally, eliminating the need to bring in costly machinery or manpower to do the job.)

Drake encouraged participants not to get bogged down in the effort of teasing out plant effects from fertilizer effects. She said that phytoremediation needs to be promoted as an entire package, as a technology that involves plants and all of the components that are required to optimize growth (e.g., fertilizer, irrigation, lime). She said that regulators will likely approve of the technology if the Subgroup can show that using phytoremediation initiates more environmental improvement than doing nothing. Albert Venosa disagreed, pointing out that regulators will want to know whether adding plants enhances remediation above and beyond adding fertilizer. He said that this can only be determined if researchers fertilize their control plots. If the unplanted control plot is not fertilized, he continued, the plots that are vegetated and fertilized will differ from the control plot in two areas rather than just one. He said that the presence or absence of plants should be the only variable that differs between test plots and control plots. Venosa and Phil Sayre agreed that it is important to have an unconfounded experimental design when trying to publish data and to persuade regulators to accept a new technology. Several conference call participants agreed that experiments involving fertilized planted plots should feature an unplanted control that is fertilized. In fact, this is the approach that is already being used at Sites A and B.

Steve Geiger reminded participants that the protocol instructs researchers to fertilize their planted plots as needed to optimize growth. At some sites, he said, soils may be fertile enough to support optimal plant growth without fertilization. He warned participants against changing the protocol to state that control plots should be fertilized. Rather, he recommended, the protocol should be changed to state that all sites should be managed identically. (This would mean that unplanted control plots would be left unfertilized at sites where fertilizer is not needed on planted plots. Likewise, fertilizer would be added to the control at sites where fertilizer is needed to optimize plant growth.) Conference call participants agreed with this approach, saying that it would also be wise to use identical management practices with irrigation and liming practices. Peter Kulakow agreed to modify the protocol.

Although conference call participants were comfortable using Geiger's suggested approach, they still agreed that the best option is to have two unplanted plots per site, one that is fertilized and one that is not. Lucinda Jackson said that Mike Reynolds is using this setup at Sites C, D, and E. She said that she will consider evaluating an unfertilized control plot as a side experiment at Site A. She agreed to talk to Drake about this topic offline.

Fertilizer Amounts and Distribution

Conference call participants discussed the amount of fertilizer that should be used. Drake said that plants pull fertilizers out of the soil, leaving less fertilizer to support bioremediation activities. As a result, she said, unplanted fertilized plots might have more fertilizer in their soil than planted plots do. Drake said that such a situation would promote more aggressive bioremediation in the control and might give it an unfair advantage over the planted plots. Venosa said that this problem can be avoided if enough fertilizer is added to account for plant uptake and bioremedial activities.

Jackson said that the protocol recommends using agronomic analyses to determine how much fertilizer is needed, but Kulakow asked participants whether they would prefer to base fertilizer amounts on hydrocarbon composition. Conference call participants agreed that this would be a good approach and recommended aiming for the following ratios: 50 to 1 for carbon to nitrogen and 100 to 1 for carbon to phosphorus. Kulakow agreed to modify the protocol to reflect the Subgroup's decision on this matter. Drake said that all the fertilizer can be applied during the first year or over a 3-year period. Using the latter approach, Venosa said, investigators can make sure they do not overfertilize or "burn out" their plants. Sayre advised researchers to think carefully about when to add their fertilizer if they are planning to perform microbial analyses. He noted that some studies have shown that microbial populations experience a short-lived boost in their numbers immediately after fertilizer is applied.. Sayre encouraged participants to keep accurate records of how they apply their fertilizers.


MICROBIAL ANALYSES

Sayre said that Microbial Insights, Inc., gave a presentation at the April 21, 1999, Subgroup meeting (which was held in San Diego, California). Sayre said that the Subgroup thought the presentation was informative and wanted to learn more about how changes in microbial community structures relate to changes in hydrocarbon bioavailability. (Sayre said that Microbial Insights, Inc., posited that the bioavailability of hydrocarbons can be estimated based on whether a microbial population returns back to its normal state [i.e., the state that it was in before the site was contaminated].) At the Subgroup's request, Sayre said, Microbial Insights, Inc., prepared a proposal and had it distributed to Subgroup members before the conference call.

Andrew White, the president of Microbial Insights, Inc., gave a brief overview of the company's proposal and summarized the information that could be gained if the Subgroup used his company for microbial analyses. He described three of the analyses offered:

White recommended performing the PFLA analysis on Subgroup samples to get an overall characterization of the community, and then performing DGGE if a more detailed view of the microbial community is required. Venosa said that he thinks both tests are informative and he encouraged participants to have PFLA and DGGE performed on their samples.

White said that the costs for the analyses are outlined on the second page of his proposal and that the total cost was calculated assuming that 16 samples will be analyzed over a 3-year period. He obtained this sample number by assuming that microbial analyses will be performed on two treatments and that one composite sample will be collected from two different depths per treatment during all four sampling times (T0, T1, T2, and T3). (Kulakow noted that White's proposal also assumes that the composite will be collected over four replicates of the same treatment. Conference call participants said that they would prefer the sample to be a subsample of a composite collected from a single replicate. If this approach is used, they said, the sample that is analyzed for microbes will have corresponding hydrocarbon data.) Sayre noted that some grant money might be available to offset some of the microbial analysis costs.

White said that Subgroup participants do not need to make an immediate decision about whether to perform microbial analyses. He said that he is willing to store their samples until they decide whether to proceed; samples can be analyzed up to 1 year after collection if they are kept frozen. White told participants that fresh samples may be sent on ice by overnight mail. He said that he can also accept samples that were collected months ago as long as they have been frozen. He stressed the importance of avoiding freezing and thawing episodes and encouraged participants to send frozen samples in such a manner that they will not be able to thaw. He said that 30 to 70 grams are needed to perform the microbial analyses. White encouraged participants to contact him (423-573-8188 or microbe@microbe.com) with any further questions. He also agreed to send ERG a copy of the presentation that he and David White gave in San Diego, California.


MISCELLANEOUS

Kulakow noted that the first page of the Subgroup's April 1999 protocol has a discrepancy regarding the number of subsamples that must be collected per plot to create a composite sample. Jackson said that eight cores should be collected to create a composite sample.

Kulakow noted that mowing will be required at some sites, but that it should only be performed once a year at the end of the growing season. (More frequent mowing stunts root growth.) Kulakow said that he will modify the protocol to reflect this recommendation.

Steve Rock noted that unplanted controls are supposed to be kept weed-free. In previous conference calls, he said, the Subgroup identified using RoundUp as the preferred weed-control technique. He said that this is not an option at a site that he is working on because the site owner will not allow any chemicals to be added to the property. Rock asked whether tilling is a good alternative to RoundUp. Conference call participants did not think that this was the best approach because tilling an unplanted fertilized control is too similar to land farming. Drake suggested using a perforated clear plastic to prevent weed growth.

A couple of conference call participants provided site updates. David Tsao said that he is still waiting to receive regulatory approval for a demonstration project at Site H and that he is investigating the possibility of doing a demonstration project at Site I.


ACTION ITEMS