Is it possible to replicate focal regions like circuitscape?

31 views
Skip to first unread message

Robby

unread,
Nov 3, 2011, 5:48:52 PM11/3/11
to Connectivity Analysis Toolkit
Hello everyone,

I've been experimenting with CAT, I was wondering if it was possible
to create focal regions like forest patches which would have multiple
cells. In circuitscape this option is available, how would you do this
in CAT?

Thanks

carlos

unread,
Nov 3, 2011, 6:09:48 PM11/3/11
to connectiv...@googlegroups.com
Yes, it is possible in the sense that the textfile which lists source/target hexes can list the hexes that fall within those focal regions. You would just overlay your focal regions on the hex shapefile in GIS and query the hexids of the hexes falling within those regions. But I believe this is somewhat different to how Circuitscape treats a focal region. In the CAT, a focal region is just treated as a bunch of hexes rather than having any unique identity as a region. Let me know if any more info would be useful.

Robby Marrotte

unread,
Nov 4, 2011, 3:29:32 PM11/4/11
to connectiv...@googlegroups.com
I'm just having a hard time trying to wrap my head around how I can represent my data like in the example.

I have Forest patches that are suitable habitat in a fragmented landscape and i have a resistance raster for the entire landscape where these patches are located.

For Circuitscape I just had to rasterize my forest patch and square my resistance raster. For Cat how would I create the habitat hexmap? Would I simply turn my forest patch raster values to all the same value and use the resistance raster to create the cost Hexmap?

Thanks alot for the Help! Btw just read the paper which compares shortest path, current flow and Min-cost-max-flow, great piece of work! It's very helpful, it enlightened me for my masters project.

2011/11/3 carlos <klamathco...@gmail.com>

carlos

unread,
Nov 10, 2011, 4:48:49 PM11/10/11
to connectiv...@googlegroups.com
Your raster needs to be expressed in conductance not resistance.
You would overlay your forest patch coverage on the shapefile of hexes created by the CAT when you create your conductance .hxn file, to identify those hexes falling within the forest patches.

Robby Marrotte

unread,
Nov 14, 2011, 11:31:49 AM11/14/11
to connectiv...@googlegroups.com
At the moment what I tried was just to use the habitat model which ranges from 0-1000 and I created a hexmap with it. Then I created conductance and distance edges and ran the connectivity analysis with it. I'm just wondering how I can make CAT know where the real suitable habitat is and just map between those  habitat patches using the habitat model.

2011/11/10 carlos <klamathco...@gmail.com>

carlos

unread,
Nov 14, 2011, 12:23:06 PM11/14/11
to connectiv...@googlegroups.com
Exercise 2 on subset centrality may help to show the process for doing what you are trying to accomplish. You need to create a source/target text file by using a GIS program such as ArcGIS.

Robby Marrotte

unread,
Nov 30, 2011, 12:56:22 PM11/30/11
to connectiv...@googlegroups.com
Any plans on adding other Betweenness Centrality Metrics?

2011/11/15 Robby Marrotte <robbym...@gmail.com>
Alright Thanks!


2011/11/14 carlos <klamathco...@gmail.com>

Robby Marrotte

unread,
Nov 15, 2011, 2:30:18 PM11/15/11
to connectiv...@googlegroups.com
Alright Thanks!

2011/11/14 carlos <klamathco...@gmail.com>
Exercise 2 on subset centrality may help to show the process for doing what you are trying to accomplish. You need to create a source/target text file by using a GIS program such as ArcGIS.

Reply all
Reply to author
Forward
0 new messages