Raster data and Suitability Modeling

[DeweyEvans]

I am trying to learn how to perform site suitability analysis and work with raster on ArcGIS Pro 2.7.1.  I have a completed a couple of ESRI trainings (Suitability Modeling: Introduction and Suitability Modeling: Creating a Weighted Suitability Model) but still feel clueless.

Some of my first questions include how to transform vector > raster since I know you have to have raster to perform suitability analysis.  Determining cell size is another place where I get hung up.

I have around 1 year experience with GIS and feel comfortable with a lot of the ESRI workflows and software, but am hitting a huge roadblock here.  The terminology seems so confusing.

Thanks to anyone who can help. 

[SteveLynch]

If your cell size is too large some features won't get rasterized and if it is too small your output raster will be too large and it might slow down processing.

You should decide what cell size is acceptable for the phenomena that you are modeling.

[HeatherSmith]

Hi,

I don't think it's unusual to still be unsure about GIS terminology after 1 year (or many more!)

The lesson Site a sustainable shrimp farm is probably a good resource for you: https://learngis.maps.arcgis.com/sharing/rest/content/items/e97469f991f84da38c13dcad1a0e6dbf/data
It uses the suitability modeler, which is a new tool in ArcGIS Pro. If you have not used it before, it may help to make the process easier to understand. On page 15 there is also a note that describes how to create a distance raster from a roads vector layer. This is one way to convert vector data to raster. Some other tools you can use to convert are:
- https://pro.arcgis.com/en/pro-app/latest/tool-reference/conversion/polygon-to-raster.htm
- https://pro.arcgis.com/en/pro-app/latest/tool-reference/conversion/feature-to-raster.htm
Another lesson about raster suitability analysis is Build a model to connect mountain lion habitat:
https://learn.arcgis.com/en/projects/build-a-model-to-connect-mountain-lion-habitat/

It's possible to do suitability analysis with vector as well, by codifying suitabiltiy in attributes and using the intersect and union tools. However, I think that this may become complicated quickly and so many people prefer to work in raster.

It also depends on what you mean and what you need from suitability analysis. For example, the following lessons use vector data to locate suitable locations (for a new hospital, or to target funding), although they don't do any weighting:

- https://learn.arcgis.com/en/projects/site-a-new-hospital/
- https://learn.arcgis.com/en/projects/help-end-homelessness/
- https://learn.arcgis.com/en/projects/find-potential-hurricane-shelters/

I'm not an expert on suitability analysis, so hopefully others will chime in here with more suggestions!