Importing shape files

Im setting up BenMapCE for use in NSW, Australia, and have a shape file of local government areas, similar size to US counties. The source of my shapefile provided 4 files, with extensions .shp, .prj, .dbf, and .shx. when I load it into BenMap Im asked for only one of these, and my map shape appears on screen but has no attributes such as the names of the LGA, or the row and column references needed to match this to incidence and population data. I assume these details are in the other files, but there is no opportunity to load them.
What is the way forward for someone with epidemiology skills but not GIS skills?

Hi Ben,

The shapefile preview window won’t show you the full attribute table. Rather, the program is just displaying a preview of the shapefile so you can confirm that the polygons are plotting correctly. It’s also confirming that you’ve included a column/row index. As you note, this column/row index should correspond to the column/row index used to grid your incidence and population data.

You may also find the self-paced training materials helpful. Here you’ll learn more about importing shapefiles. You can find these materials here. Hope this helps!


Hi Neal
When I upload the shapefile for Mexico City from the training materials, it arrives with 5587 columns and 915 rows for demographic units, which I assume are going to match the population data. When I upload the shape file for NSW BenMap creates one column and 197 rows. My population data is listed by local government area name, so how do I create a link between my LGA names and the 197 rows of the shape file?

Hi Ben,

This is a great question. Before I answer it, let me provide some contextual information.

The population, air quality and baseline health data can each be gridded to the same shapefile or can each be gridded to a different shapefile. Here’s an example for the U.S.:

  • Air quality is gridded to 12km by 12km cells

  • Death rates are gridded to U.S. counties

  • Population are gridded to census tracts

Each of the shapefiles above differ in size and shape. When calculating impacts, the BenMAP-CE program will use an area weighting algorithm to assign the death rates and population to the air quality grid cell; this is because the program calculates impacts at the air quality grid cell level.

In your project, are all of your input data at the same spatial scale (the LGA), or are they gridded to different scales?

In your project, your population are gridded to LGAs. Do you have a way of assigning your table of LGA-level population back to the LGA shapefile? That is, can you either:

  • Assign the BenMAP-CE-created index of one column and 197 rows to your LGA population table; or

  • Identify a unique numerical column/row index in your LGA population table that you could assign to the LGA shapefile?

Yes, I liked the comment in the training manual for the Mexico exercise, that if all the data is based on the same geography you could just calculate the heath burdens by hand (or more likely in Excel). The sophistication in BenMap is matching across different geographical levels.
I have mortality at LGA scale, but other outcomes at health district level, so will need this functionality.
How do I get BenMap to display which row it has given to which LGA?
Or conversely, how do I write LGA names into the shapefile?
My shapefile: NSW_LGA_polygon.shp is accompanied by NSW_LGA_polygon.dbf and others with extensions .prj and .shx Would these other files contain attributes of the LGAs, such as their name?

My current problem is that my population data is for LGA by name, and the shape file has LGA by row number allocated by BenMap, and I dont have any way of matching them.

Im sure there is an easy answer to this.

Hi Ben,

The GIS in BenMAP-CE is pretty rudimentary and is best used to display results, rather than pre-processing input data. I

I think I could give you step-by-step instructions for performing this crosswalk if you send me your data; I can then post this to the user forum. I suspect that other users would find this helpful to know. Are your data small enough to e-mail? Or, perhaps you could link to the source of the LGA shapefile and tabular population counts?

Hi Neal
I think I found a way to do this. I opened the .dbf file in Excel, added 2 new columns called ROW and COL , gave the COL variable the value 1 for all areas, and the ROW value sequential numbers. Excel these days doesnt support exporting as .dbf so i I saved this as a .csv file and opened it in MS Access and exported it from there as a dBASEIV format. Remembering that that is limited to 8 character file names. My BenMap project now has a map of NSW and LGAs identified by a ROW value that I can look up in Excel to find the LGA name.

New problem: The monitors are clustered in the Eastern part of NSW, where most people live. The values displayed for the LGAs that have a monitor seem to be correct, but out west many LGAs have negative values for PM 2.5. I selected the Voroni method for attributing values. What happened to generate these incorrect negative values?

Hi Neal
When I create a baseline air quality surface using the Nearest Monitor method, I get many areas with extreme negative values, like minus 10 ^34. It looks like its trying to write minus infinity. Why does it generate them? Am I doing something wrong? How do I suppress these values? They occur in areas distant from the monitors.