Oct. 18th, 2006 Conference Call
Great Lakes Regional Toxic
Air Emissions Inventory
Steering
Committee Conference Call Minutes
Participants
Illinois:
Indiana:
Michigan: Dennis
McGeen, Jim Lax and Allan Ostrander
Minnesota: Chun Yi and
Dan Steber
New York: Carlos
Mancilla
Ohio: Tom Velalis
Ontario: Cong Doan
Pennsylvania: John Hulsberg
Wisconsin: Orlando Cabrera-Rivera
GLC: Jon Dettling
EPA:
Other: Mark Young
Update from Mark
Young on database layout
Mark has been working on
developing a prototype data model based on the discussion at the Toronto meeting. He's
been attempting to fit the point, area and mobile source data types into a
single data model that can accommodate each. Mark is hoping to have an initial
version for review as early as next week.
Orlando mentioned that
he's heard something about a draft NIF 4.0. Mark said he had heard mention of
that also. Allan mentioned that he has some very preliminary information on
this but that he doesn't believe the EPA is very far along in releasing this.
It could be a very long way from being completed. Mark will try to keep up on
this as it develops and will look at the current paperwork to see what
differences are in the pipeline at this point. Allan will forward the
information he has.
(this is now posted at http://glin.net/forum/view.php?site=glad&bn=glad_rapids&key=1161272072)
Chun Yi noted that the direction
of the NEI data structure is very different than the original RAPIDS data
structure, such as the ability to have emissions without identified release
points. We may want to adopt some of these differences. Chun Yi
thinks it would help if the whole group looked at the NIF documentation to see
if there are other ideas we want to adopt in a new data model.
Mark commented that on the TTN
website there is a page with accepted codes sets for EPA
Data Standards. Mark will be attempting to keep up with those
standards as much as possible. The group generally agreed with this. Chun Yi
expressed a preference to comply with such standards. There are some cases that
may need to be addressed in more detail in the future, such as pollutant codes,
geographic codes, etc.
Mark had a question about the use
of tribal codes. He's already had a brief email exchange on this with Orlando
and Jon. Tribal codes are now required for NEI, although it doesn't need to be
defined for every record. Mark's question is whether all sources are located
within a county or whether some are on tribal lands, but not within a county.
There was some discussion as to
how the tribes relate to the states and counties. This is a fuzzy issue, but it
can generally be considered that they are similar to states in that they report
directly to EPA. However, some states
also collect their data to include in their inventories or
report the data on behalf of the tribal entity . Orlando noted that in
some cases they are more similar to cities. Mark said he thinks he can work
with this information for now and can raise the issue again if the need arises.
Orlando noted that
there are some facilities in his state that are on county lines and one or the
other county must be chosen. There is also the case of portable sources, which
often present a challenge in determining where to assign these sources.
There has been some discussion
previously on multi-year emission inventories. Mark is attempting to come up
with a design that would store multiple years' data. He is considering an
approach where records for sources, etc., would have an inventory code (e.g.,
EI1999) that would define which inventory it belonged to. This would have some
benefits for querying and wouldn't introduce the complications of having start
and end dates for sources that define which inventory years it pertains to. It
is similar to having multiple databases, but all the records would be in the
same database and tagged with an indicator of which inventory they belong to.
The inventory itself would therefore be the top level of the hierarchy. Of
course, maintaining separate databases for each inventory would still be
possible with this approach.
Mark also mentioned the
possibility of having the source type be more hierarchical. For example,
railroads (type L) could be a type of non-road data (type N). The hierarchy
might follow some of the logic contained in the SCC/AMS codes. There was some
concern about this because some sources can be categorized in multiple ways.
Currently, these codes are not very hierarchical and the codes don't exist
within the database other than as values in the table; that is, there's no
additional table that defines them. Tom noted that the SCC defines what type of
source something is and it can just as easily be used in making reports to
identify various classification of sources.
Mark mentioned that he anticipated
publishing a draft database design for the group to look at as a MS Access
file. Everyone was comfortable with that format. Mark should be able to send
something by the end of next week.
Reference table
updates
A set of new codes were approved
at the recent meeting. Chun Yi has put together an import file and emailed it
to the group. No one has expressed any trouble importing these.
BaP state updates
- data changes due by end of month
Michigan has completed
some updates and will be creating RAPIDS import files. These include prescribed
burning and wildfires. Some BaP factors have been loaded into MAERS. Some
others are still being looked at.
Minnesota's updates are
done.
New York's
updates are being uploaded, with some help from Orlando
Ohio is done.
Ontario is done.
Pennsylvania will be
working on theirs next week. John will talk to Karen.
Wisconsin has completed
their updates and they are all included
The final updates are due by
October 30th.
Discussion of
area source estimations - 2005
There was a brief discussion on the
prospect of handling some area source calculations at the regional level for
2005. Jon pointed out that there may be several cases where this is feasible in
that everyone agrees on a common methodology and emission factors. It would
have some advantages in terms of work efficiency and consistency from state to
state. It was pointed out that there were some source types where this approach
probably will not work well due to differences in methodologies or input data.
This would need to be considered on a per-source basis. Carlos pointed out that
an important first step is to decide what area sources the group will
inventory. We should agree on a certain set that everyone will inventory. We
can then look for some that are best done at a regional level.
Jon has put together an initial
query to determine what Area Source SCCs people are currently estimating. He'll
add some more information to this, such as how many states are calculating
each, and send this out for discussion. Some states are already beginning to
look at their 2005 calculations, so the group should move this discussion along
quickly. Chun Yi mentioned that Suzanne may have done some work in looking at
area sources. It would be wortwhile involving her.
MDI Emissions
At the Toronto meeting, Chun Yi provided some
information regarding MDI emissions. Jon has compiled a summary of the 2002 MDI
emissions. Compared to the information Chun Yi provided, it appears there are
some significant overestimates. Chun Yi said that she worked with the API group
to revise her database for MDI and the changes were much lower emission totals
and more sources included. The main rationale is that many facilities report
usage for MDI rather than calculating emissions. Because the volatility is
fairly low, this causes a large overestimate. The table Jon sent showed many
sources with hundreds to thousands of pounds. Orlando noted that although his 2002 data
showed large emissions for this compound, the more recent reporting years have
not had these large numbers. He will look into why. Chun Yi noted that the API
is willing to do the background work on this on their own. Orlando noted that one needs to be careful to
be critical of information offered by outside groups like this to make sure
there is a good basis for their recommendation. Jon suggested contacting the
API and welcoming them to look at our data and report back to us on
recommendations, as long as we are clear from the beginning that we may or may
not incorporate their changes, depending on our own assessment of them, as well
as some constraints within our organizations (such as the difficulty of
changing past data that has been reported on by facilities and certified). We
would also want to make sure there is an emphasis on future emissions rather
than past emissions so that changes made are permanent and errors do not recur.
Next In-person
meeting
We did not set a date in Toronto for the next
meeting. Orlando has suggested that the meeting
could be held in Madison.
The group was encouraged to look into the feasibility of traveling to Madison for the meeting
prior to the next call. Orlando
can help give some advice on travel arrangements if needed. The dates would be
late January or early February.
The next call will be November
8th.