Smathermather's Weblog

Remote Sensing, GIS, Ecology, and Oddball Techniques

Topographic Position Index and Ecological Land Type (warning completely unrefined not quite Geologic dribble– with bad maps :) …)

Posted by smathermather on January 10, 2010

Warning.  What follows is somewhat informed, but I’m no geologist.  I just play one on wordpress.

Understanding the basic underlying geology and associated topography plus site history helps us achieve a basic understanding of a sites ecological potential.  At the most basic level, we expect different wildlife and vegetation dynamics in a floodplain vs. a mountain ridge.  Classification of digital elevation models can be done with a concept called topographic position index (of which Jensen Enterprises has a pretty good explanation).

I’ve been playing with TPI for two areas in the Allegheny Plateau in Ohio– one in an unglaciated portion, and one in a glaciated portion.  The Allegheny Plateau developed as the once very tall Appalachians wore down to the nubs they are today.  That plateau has since eroded into little hill-like remains, with stream valleys redefining a hill and hollow landscape, and most of the ridge-tops being near the same elevation.  In the glaciated portion, this is overlayed with glacial till, a re-flattening of these “hilltops”, and the definition of gorges in places where sub-glacial rivers subdivided the terrain.

The TPI attempt I did for the unglaciated portion roughly defines the ridge tops in red, stream valleys and slope bases in blue, and slopes as yellow/yellow green.  The map that follows is the overlay of TPI performed in gdaldem at multiple scales, but we are somewhat limited in what we can do with gdaldem, because we can’t control our window size.

Then I performed a proper TPI analysis at 80 feet and 300 feet for a glaciated portion of the plateau in Northern Ohio, following Jensen’s guidlines.  What follows is a photo of a printed map, cause I forgot to bring an jpeg of pdf home… .  Here, upland flat portions are dark green.  They are adjacent to light blue areas that are also upland, but exceed 5% slopes.  We can see upland areas adjacent to cliff areas as well as peaks in red, lower areas of high relief in orange, gorge valleys in dark blue, U-shaped valleys in a lighter blue (stream polygons and lines are also overlayed, adding to color confusion– hence the warning in the title).

What intrigues me about this map are the areas around the streams just northwest of the center of the map.  Capturing the gorges and upland areas successfully was not a surprise.  What is surprising is that we capture and map the areas that seem to me likely contribute directly to surface flow into the streams as distinct areas from the upland plains.  Next, I think I’ll compare this analysis with the soil map for the area, as well as the detailed vegetation cover we have for the area.

About these ads

3 Responses to “Topographic Position Index and Ecological Land Type (warning completely unrefined not quite Geologic dribble– with bad maps :) …)”

  1. Jerome said

    Hi,

    it’s very interesting,
    do you use gdaldem for the computation?
    how do you obtain the colored image ?
    what software do you use to display it ?

    regards,
    Jerome

  2. smathermather said

    Display for the first map is in Quantum GIS. The second map is actually all done in (gasp!) ArcGIS.

  3. [...] Topographic Position Index and Ecological Land Type (warning completely unrefined not quite Geologic… January 2010 2 comments 5 [...]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

 
Follow

Get every new post delivered to your Inbox.

Join 792 other followers

%d bloggers like this: