Creating a Risk Index? for a polygon layer based on various parameters.

[PhilBeilin1]

Hello all,

I am analyzing Soft Story buildings for my city in Northern California using parcel data and earthquake data and I want to create a risk index based on various parameters, like number of stories in the parcel layer and whether or not the parcels are in liquefaction/shaking zones. 

So if it has a desired number of stories I would give it a 1. If it's in a liquefaction zone I would give it a 1. If it's in both it would receive a 2. And so on with my other parameters. 

I don't think Risk Index is the right phrase as it leads me to nothing of use on Google, so if someone could at least give a name to what it is I'm trying to do that would also be very helpful.

[DanPatterson_Retired]

if the attribute(s) exist in the tables of the featureclasses it can be done within one file, but if they exist within separate files, then a Union would carry over the attributes and intersect the geometry.  Perhaps some further information on how your data are structured are in order.

[PhilBeilin1]

I was already planning on making a union of the different features, ie the earthquake data to the parcel data.

[DanPatterson_Retired]

Then the rest will entail how you want to ascribe risk and whether you want to use definitive sharp boundaries or whether there needs to be offsets (buffers) from delineated zones.

Perhaps a fuller list of attributes and how you want to deal with their spatial patterns and/or associations would help with other suggestions.

[PhilBeilin1]

I'll explain a little more in detail:

First, I do not know whether or not parcels contain soft stories. 

Using the parcel data I am able to predict which parcels might have buildings with soft stories based on a few criteria: year built, land use codes, garages, number of stories. If a parcel meets any of these criteria it would be given a 1 per each.

I also will union earthquake data: Alquist Priolo zones (yes=1 and no=0) and liquefaction susceptibility (high=3, moderate=2, low=1, none=0).

I would like to see these added up in a single column of my parcel data and parcels with higher numbers would be focused on first.

[DanPatterson_Retired]

a union will still provide the intersection geometry and it will carry over the attributes during the process.  Open the table from the Union and see what you have.

If the results are there, then you would need to add a new field and calculate your final value, which you indicate would be the sum of the class values from the intersecting layers (Alquist Priolo and liquefaction) If desired you could weight these during the calculation.  ie Yes for alquistPriolo and low for liquefaction both have a 1... are they equally weighted? or is one more important than the other?)

[PhilBeilin1]

At this point in time adding weighted values may be extra work that is not needed

yet at the same time I understand that weighted values gives a better understanding of the final values - so how would I go about adding weighted values?

[DanPatterson_Retired]

A .... your first field  ( 0 or 1

B ..... your second field ( 0, 1, 2, 3)

if A is 3 times more important than B

C is the new field you are doing the calculation in

C =

     (A * 3) + B

just algebra... but you can get real fancy if you can provide a model that you would follow.  For instance... if the liquefaction index resulted in a value of 3... I would sortof be concerned regardless of whether it was on a particular zone... for instance of course

[RichardDaniels]

Check out https://www.fema.gov/hazus/ for additional ideas on doing this.