April 12, 2007 Conference Call
Great Lakes Regional Toxic
Air Emissions Inventory
Steering
Committee Conference Call Minutes
Participants
Illinois:
Indiana:
Michigan: Dennis McGeen
and Allan Ostrander
Minnesota: Chun Yi Wu
and Michael Smith
New York: Carlos
Mancilla
Ohio: Tom Velalis
Ontario: Cong Doan
Pennsylvania: Karen Gee
Wisconsin: Orlando Cabrera-Rivera
GLC: Jon Dettling
EPA:
Other: Mark Young
Chun Yi introduced Michael Smith,
who works on MPCA's CAP inventory. MPCA is considering some new directions in
their inventory work and Michael is working with Chun Yi to evaluate how RAPIDS
may fit into this.
April 10th was the cut-off date
for the reservations in Madison.
If people haven't yet made their room reservations, they should try to do so
very soon.
Jon sent an email earlier in the
day regarding changing the time for calls to best accommodate everyone's
schedule. Please provide input on this if you haven't already so that we can
choose the best time.
Mark posted some new ERDs on the
discussion forum site prior to the call. The first page shows what the data
type codes refer to. The top level ERD shows the nation, state, and county
tables as well as the source table, inventories table and the periods and scenarios
tables.
Within most tables, there are
description fields allowing 40 characters. These could be lengthened, but that
might lead to them being truncated during conversion of data between systems.
Mark inquired whether anyone had any concerns about at 40-character limit on
the naming fields.
There is also a comment text field
in each table that allows 1024 characters.
It was agreed that an 80 character
description field would cover all needs for NIF and for RAPIDS.
Regarding the sources table, Mark
inquired whether the address field is sufficient as it is shown. This version
is based on NIF. Chun Yi expressed that she needs to have a contact table, as
in the current RAPIDS. Having a separate table would allow multiple contacts
for each source. There could be a reference code to define the types of
contacts.
Mark noted that the RAPIDS legal module has been entirely left out. This
doesn't appear to have ever been used.
Tom suggested that modules like the
contacts list, which aren't required in NEI and don't affect the emissions
calculation, could be lower priorities and implemented at a later stage. The
current focus should be on what is needed for the emissions calculation and
reporting.
The group agreed to cover this item
after the new system is in place.
Regarding the stacks table, Mark
inquired whether it made sense to change the terminology to "release
points" as is done in NEI. Allan suggested that this was a good idea. Chun
Yi agreed that "release points" is better, especially if fugitives
are included.
The group discussed the field
lengths for numeric fields. There have been issues in the past where the use of
exponents (e.g., E-10) have reduced the ability to store or export numbers properly .
Coordinates are needed for
sources, devices, and possibly release points. They are not needed for
processes or controls. The group agreed that there is no need to have
coordinates on any other units.
Mark discussed that there are
numerous stack parameters that aren't needed for emissions estimation. These
include things like height and diameter. These often have set units within NIF.
Do we want to have the flexibility to use multiple units, or use the NEI
standards? Orlando pointed out that flexibility
may be important for Canada,
as they use different units. Mark noted that if flexibility is allowed, a
conversion can be programmed on import and export.
Mark asked whether it is the
group's preference to have the stack attributes in the release point table, or
whether these should be in a separate table. For import, export and reporting,
it is generally easier to have these in one table. For organization, it can
help to have them in a separate table, as well as if there are multiple sets of
parameters for the same release point (this doesn't appear to be the case). It
was agreed to keep this information in the release points table.
Regarding coordinates, the greater
precision in the new RAPIDS will allow greater accuracy. The prior limitation was
two decimal places (about 10m accuracy). The new system will allow better
accuracy than this and in many cases, the locational data is accurate enough to
take advantage of this.
Regarding the schedule, Mark will
be creating scripts to build the database on the server. Mark tried connecting,
but had difficulties and will need to contact Jon and Pete to resolve this. He
will also be ironing out the final details on the GIS tables. At that point,
around the end of the month, we should have a core system to start working with
estimation routines. Mark will also be working to convert data from the CAROL
system to PostgreSQL. Once the data structure is set up, there will be the task
of also dealing with reference tables.
The next call will be May 2nd
(Wednesday) at 1:30 pm EDT.