Great Lakes Binational Toxics Strategy

GLBTS Links

______________________

Dioxin / Furans

October 5, 1999
Workgroup Teleconference Minutes

Workgroup Leaders:
Sandro Leonardelli, EC
Nan Gowda, EPA

Facilitator:
John Menkedick, Battelle

The objectives for the day included reaching agreement on the final form of the dioxin decision tree, deciding on next step’s for it’s application, and discussing the status of outstanding action items.

Decision Tree Discussion

Decision Tree - Changes

The meeting opened with a brief presentation of the changes made to the decision tree. Revisions were made in response to comments made at the September 7 conference call, and included:

the addition of language on pollution prevention (i.e., “P2") in the right column
the addition of the words “or programs” to the box in the second column, second row (i.e., Are there regulations or programs that exist or that are planned for this category?)

A question was asked on how the tree would address ongoing needs for research and gathering data, especially in light of the fact that some sources still have many uncertainties associated with them:

It was noted that this decision tree is primarily intended to focus specifically on identifying actions that the workgroup may want to consider taking now or in the near future. Therefore, although additional emissions estimates are needed from some sources, this decision tree can help the group to decide on which actions (i.e., the top five priorities) it could undertake now, based on existing information.

It was also noted that in the written introduction to the tree, it is stated that the decision tree “is only a tool and it is neither intended to preclude any work group member from actively pursuing projects for any of the remaining sectors nor is it intended to define closure for the work group”.

A question was asked regarding the criteria to be used to decide on a source’s significance in the basin:

As also noted in the introductory text for the tree, “significant sources” will be considered those that account for 2% or greater of emissions in the Great Lakes Basin (as assessed by either the U.S. or Canadian inventories), and/or those sources that have the potential to influence the basin.

The rationale for picking 2% was the fact that in the inventories, the 2% cutoff appeared to separate the major sources accounting for the majority of emissions from a large number of minor sources responsible for only very small percentage of the total emissions.

A question was asked regarding what basis the group would use to estimate whether opportunities for further reductions in emissions exist:

In the U.S., one reference for making these types of decisions might be the background documents prepared for the development of the MACT standards, which contain the analytical basis (assuming full compliance) for the standards.

Concern was expressed about the reliability of the MACT based estimates for MWI. It was also noted that for some sources (e.g., MWI), if the only consideration for further reduction opportunities is based on end-of-the-stack solutions, and if MACT estimates are assumed to be valid, then the group might conclude that the are no other reductions opportunities. However, for MWI there are further reductions opportunities on the waste prevention end.

More discussion and questions followed on the adequacy of the decision tree. Ideas and concerns were also voiced by the group. Issues included:

The placement of P2 in the tree and the requirement to go through the second column boxes first, before reaching P2. It was noted that the intent of the regulatory analysis was to understand the environmental and regulatory context affecting a source, to help understand reduction opportunities. In addition, the intention was not to get stopped in any box, but rather to quickly move to the prioritization boxes with defined actions
Whether the term “significant” should be included in the further reductions opportunities box, and if opportunities should also be expanded to include elimination
The role of virtual elimination versus BNS challenge goals
The group’s commitment to periodically revisit/review certain sectors as necessary over time to re-evaluate the effectiveness of group action, or in light of new information

After the discussion, a decision was reached to add “elimination” so that the opportunities box would read: “Are there opportunities for further reductions / elimination?”

Sandro Leonardelli (EC) pointed out that the dioxin decision tree is meant to be a tool to help the group focus on actions, and suggested that the group not get bogged down in minor details of the tree. The tree can/will be revised as necessary (e.g., additions may become apparent when the group begins to implement the tree). However, he felt that for the group right now, it was probably more important to generate a list of concrete actions and begin constructive work on them.

Decision Tree - Next Steps

The focus of discussion then turned to next steps for implementing the decision tree:

EPA and EC agreed to gather the available data and information that would be needed to discuss each sector of focus. This information will be made available to the group for the next meeting in a matrix format.
At the next workgroup teleconference (i.e., November) the group will plan on using this information to define and prioritize actions for as many sectors as possible using the decision tree process
Workgroup members agreed to submit ideas for potential actions for further reductions / elimination before the next meeting

Presentations / Updates on Action Items

It was agreed that the next steps for the decision tree be added to the list of action items. At this point, the workgroup was asked whether there were any other updates or reports on previous action items. Dwain Winters (EPA) had prepared presentations on backyard burning, and on the uncertainties associated with MWC/MWI emission estimates.

Backyard Burning

Dwain presented an overview of the current research being conducted by EPA on dioxins/furans emissions associated with backyard / open barrel burning:

EPA has completed a second round of testing intended to examine factors that may affect dioxins/furans emissions. Currently, EPA is reconfirming the data collected from the second round, and in addition, has begun a third round of testing. Dwain predicted that a fourth round of testing would also probably be necessary.
Tests were conducted using a standard waste that was burnt in a 55 gallon barrel housed within a test chamber. Samples of smoke with in the chamber were analyzed for a number of pollutants, including dioxins/furans and particulates. Different waste treatments were used to examine which factors affect emissions.
Factors that were tested included: chlorine content of waste, copper content of waste, density of the waste, and moisture content of waste and variability of emissions within the same standard operating parameters.
Results of burns that were of the same conditions using the same base case waste formulation varied by a factor of 10 between high and low measurements, indicating that even amongst very similar burns, the chaotic nature of combustion strongly effects emissions.
Correlation of emissions with temperature measurements indicated that within the barrel, there was likely to be narrow envelope of conditions in which the majority to dioxin/furans formation was likely to take place. The factors that had the highest correlation with the level of dioxin formation were, copper and HCl concentration in the combustion gases and temperature measurements within the fire.
Based on this and future data, EPA will work to develop emissions factors for different types of open burning .
In summary, Dwain indicated that although it may be a while before reliable quantitative estimates can be made, data collected so far clearly prevents us from dismissing opening burning as a trivial source of dioxins. He concluded that information available so far probably supported a workgroup effort to examine options to reduce open burning, particularly in light of the fact that open burning is also a source of other pollutants of concern, such as polycyclic aromatic hydrocarbons (PAHs) and fine particulates. However Dwain cautioned that it is premature to conclude that backyard burning of waste is a major source of dioxins.
Dwain suggested that appropriate actions might include an assessment of the prevalence and geographical distribution of backyard burning, and an analysis of the role institutional barriers play in encouraging open burning (e.g., unavailability of trash service and transfer stations in rural areas, regional legal and governmental structure). This type of information on the prevalence and nature of backyard burning will be critical in identifying potential actions, particularly if we find a high degree of regional variation.
Dwain volunteered to make available the paper that was recently presented at the international dioxin meting.

Workgroup members discussed the importance of sharing this type of new information on emerging sources with their respective groups.

Uncertainties Associated with MWC/MWI Emission Estimates

Next, Dwain presented a brief overview of the uncertainties associated with MWC/MWI dioxins/furans emissions estimates in the U.S. Inventory:

Overall, the inventory contains some sources with estimates that are considered reliable enough to be added into the total estimate, and some that are not. There are five categories of uncertainty:
- A, B, and C levels of uncertainty, which are included quantitatively in the inventory
- D, which is identified and quantitatively estimated, but too uncertain to be included in the inventory, and
- E, which is identified but not quantified
The level of uncertainty has two major components:
- The emission factor, which is an estimate of dioxin emissions per unit of particular activity, and
- The activity level estimate
Each component involved in determining uncertainty receives a letter rating, and the overall rating of a source category receives a total reliability estimate which is the lowest of the two.
Dwain said that non-standard operating conditions (e.g., startup, upset) are not assessed or reflected in the U.S. inventory. Dioxin formation does not necessarily behave like other pollutants in non-standard combustion conditions, this means there is no sound basis for estimating dioxin emissions for non-standard conditions.
Dwain provided a few current reliability estimates as examples:
- Pulp and Paper facility emission estimates have a reliability rating of A
- MWC and hazardous waste incinerator estimates are rated a B
- MWI are currently rated C, which means the estimate has more uncertainty, but is still good enough to be added into the inventory
In relation to predicting decreases of emissions across the inventory’s two reporting years (1997,1995), the MWC/MWI estimates are considered fairly reliable due to several factors:
- half of the MWI emissions reductions are due to closing of facilities
- for MWC, most of the reductions took place in the 14 facilities that accounted for about 80% of the emissions from that source category
Dwain explained that one of the additional benefits of the implementation of the MACT standards is that additional monitoring and reporting will be required. This additional testing and reporting will lead to reduced uncertainly in future emission inventories and will allow us to better assess the significance of degradation of performance over time

Concern was raised regarding the accuracy of the assumption that the closing of facilities actually translated to less burning of waste, or if it was possible that the waste may just be shipped to other facilities.

Dwain related that EPA had reasonable good estimates the amount of waste burned in the 1995 which is after the closing of facilities.

A question was asked about the range in the MWC estimates, and if these estimates included nonstandard conditions:

Dwain explained that the inventory only accounts for uncertainty in estimates for standard conditions, which is derived from the nature of the calculation in each individual source category.

He noted that, for all sources, whether the calculated uncertainty was related to the emission factor or the activity level may be important to the workgroup when deciding on significance of a particular source.

Dwain volunteered to check on the confidence levels and confidence intervals associated with the sources in the inventory for the group to use in determining significance in the decision tree process. The current inventory, as part of the final reassessment, should be in peer review in December of this year, under review by the SAB in the spring, and released in early summer of next year. Dwain also offered to check on the availability of a summary paper for the workgroup to use now.

This led to a brief update on action items as well as the addition of several new action items:

Next Steps / New Action Items

Minutes from the October 5 conference call will be drafted and distributed
EPA/EC will revise the decision tree per comments made by the workgroup at today’s meeting (revised version to be distributed before next meeting)
EPA/EC will gather the available information on sources and organize it in an information matrix
The Action Items list will be updated and distributed. Workgroup members will review this list, and where possible, offer to report out on progress at the next conference call.

The workgroup will continue work on current action items identified in the action item update. Goals for the next conference call will include presentation of the sector information gathered and application of the decision tree to as many sectors as possible. The next call is scheduled for November 2.