GIS Lab Question

[SpiroKorizis]

I need to create a map of risk of PM 2.5 exposure by populations that have asthma. I need to create an additive model that quantifies areas with high and somewhat high concentration of PM 2.5 as well as areas with high and somewhat high areas of asthma rates. I am given a tabular dataset of asthma rates, PM 2.5 point dataset derived using land use regression from EPA monitors, a vector dataset made of up zip codes, and a boundary file. How do I go about doing this? I have never done a lab without instructions and I do not know how to begin? Any help would be greatly appreciated.

[JamesFitzgerald]

Hello Spiro,

Questions: are the zip codes polygons? Are the zip codes your boundary file? Is the boundary file a polyline or polygon? Does the tabular dataset of asthma rates have a unique identifier?

Thanks

James

[SpiroKorizis]

The zip codes are polygons.

They are not my boundary file.

I believe the boundary file is a polygon.

I am not sure what you mean by unique identifier. I can join the asthma table with some of the other layers if I have to.

-Spiro

[SpiroKorizis]

The boundary file is a polygon, and it is based on the zip codes. This layer can represent areas with "no data" and fill in missing areas, such as airports or parks.

[ChrisDonohue__GISP]

A consideration:

Are the specific expectations defined for the results?  I ask as there could be several different ways to do this depending on what the expected final result is.  For example, the expected result might be as simple as the PM2.5 rate spatially joined to each zip code, to a more complex surface analysis where you derive a raster surface from the PM 2.5 points using Spatial Analyst and then maybe even use Spatial Statistics.  If you don't already have it defined, inquire with your instructor as to what the result requirements are.

I bring this up as this is a common issue in the working world.  People say "I need a map", but one often has to dig a bit to discover the particulars of what they really want.  They already have in mind what they need, but don't always provide the many details/complexities that are needed to produce the desired results.  What you want to find out often includes: expectations, methodology, assumptions, quality, accuracy, precision, exceptions, time constraints, etc.  See if you can tease it out of your instructor/T.A.  Also, since it is a class, I would advise making sure you make it clear from the beginning that your are looking for what the expectations of the results are, and that you are not trying to get the exact result out of them.

Chris Donohue, GISP

[SpiroKorizis]

Scenario: The NYC Department of Health and Mental Hygiene (DOHMH) has tasked you with creating a map of risk of PM2.5 (fine particulate matter) exposure by populations that have asthma. PM2.5 is known to exacerbate various cardio-vascular and respiratory diseases, including asthma. All you are given is the data and you must determine how to use it to create the model. The DOHMH wants a simple first pass at this, so all they want is an additive model at the UHF-level (an aggregate of zip codes) that quantifies the following things:

1) UHF zones with very high concentrations of PM2.5 (greater than 15 mg/m3),

2) UHF zones with somewhat high PM2.5 concentrations (between 14 and 15 mg/m3),

3) UHF zones with very high rates of asthma (prevalence > 6 people with asthma per 100 people)

4) UHF zones with somewhat high asthma rates (between 5.5 and 6 people with asthma per 100 people).

These were the instructions. We were also given the weights for the additive model. I already submitted something, which I know isn't correct, but I would like to know how to do this. I have only been using GIS for about 4 months and I decided to take a more advanced course along with an introductory course to graduate sooner. Probably not the smartest idea. I did derive a raster surface for both PM2.5 and asthma layers, reclassified, and did a weighted sum. It didn't come out right, but I don't think the solution is overly complicated.

-Spiro

[JamesFitzgerald]

Hello,

So, below you could create from your PM2.5 point.shp by doing a query using 'Select by Attributes' and export the shapefiles. Then based on zip codes 'Select by Location' and export. This will show you the different concentrations by zip code. Also, once you have the desired results, you could label the symbology by showing different sizes for 1-4.

1) UHF zones with very high concentrations of PM2.5 (greater than 15 mg/m3),

2) UHF zones with somewhat high PM2.5 concentrations (between 14 and 15 mg/m3),

3) UHF zones with very high rates of asthma (prevalence > 6 people with asthma per 100 people)

4) UHF zones with somewhat high asthma rates (between 5.5 and 6 people with asthma per 100 people).

There are many ways you could show 1-4 either by a simple way (quick-The DOHMH wants a simple first pass at this, so all they want is an additive model at the UHF-level (an aggregate of zip codes) that quantifies the following things) or a complex analysis based on both vectors and raster.

Hope this helps!

James