RE: RAPIDS Data Model: Beginning Design

["Wu, Chun Yi" <Chun.Yi.Wu@state.mn.us>]

Mark,

 

I reviewed your design and have the following questions and suggestions.

 

1. All fields from strInventoryTypeCode down in rap_inventories table: although these fields may be MANDATORY for reporting to EPA, they are not the same as intInventoryYear in nature.  For example, we have one source for both CAP EI and AT EI.  We can report to EPA for only CAP, only AT, or both.  However, do we want to have three recodes for the same source with different INVENTORY_RIDs?

 

2. SOURCE_SUB_TYPE in rap_sources: also needed for allocated point sources such as POTW, landfills, airports…

 

3. Do you still want to have the geographic tables?  There are rids in rap_sources.  We also want to have locational coordinates for sources.

 

4. How do you deal with one process with multiple control devices?

 

5. The EPA DRAFT Reporting Content For EPA's Emissions Inventory System states:

“An Emissions Process has relationships with Facility Site, Emissions Units, and

Emissions Release Points defined in the Configuration Structure collection. Different scenarios

include: Facility Site only (as in the case of a landfill that has no gas collection system), which

would be reported as having no Emissions Unit and no Emissions Release point); Facility Site

and an Emissions Unit with no definable Emissions Release point (quarry or mining operation

with a rock crusher); and a Facility Site with an Emissions Unit and an Emissions Release point

(stationary combustion source).” 

 

The proposed structure will not work for the scenario of Facility Site only (as in the case of a landfill that has no gas collection system), which would be reported as having no Emissions Unit and no Emissions Release point).

 

6. Do we need a STACK_RID in the rap_emissions table?  Rap_processes has a stack_rid there.

 

 

Chun Yi Wu, Ph.D., P.E.
Environmental Analysis & Outcomes Division
Minnesota Pollution Control Agency
520 Lafayette Road N.
St. Paul, 55155-4194
Phone: (651)282-5855
Fax: (651)297-7709
E-mail: chun.yi.wu@state.mn.us

 

-----Original Message-----
From: owner-airtoxics@great-lakes.net [mailto:owner-airtoxics@great-lakes.net] On Behalf Of Mark Young
Sent: Monday, October 30, 2006 4:58 PM
To: Airtoxics
Subject: RAPIDS Data Model: Beginning Design

 

Hello All,

 

I've put together an Access file containing a very basic set of core tables for beginning the RAPIDS data modelling process.  These tables are based on the NIF 3.0 hierarchy discussed at the Toronto meeting.  The design follows closely with Figure 6 in the discussion document from the meeting.

 

The link to the forum post is:

http://glin.net/forum/view.php?site=glad&bn=glad_rapids&key=1162248158

 

The contents of the tables can be seen in Design View by right clicking a table name and selecting Design View.

 

A rudimentary relationship diagram can be seen by selecting Tools/Relationships from the main menu bar.  You'll have to scroll to the right to see all the tables and relationships that are defined.

 

This is a very basic presentation of the idea.  There are many steps to this design process, and this is just the first.  The data element names in all caps are the fields that define primary key and unique key constraints.  Some of them add to or replace fields that exist in the NIF 3.0 formats.  The remaining fields are just copies of the data elements in the EPA's NIF 3.0 Access database.

 

This should give an idea of the core data model and a starting point for integrating existing RAPIDS data elements.

 

We can discuss at the next conference call.

 

Thanks,

Mark Young