The Landscape Toolbox

From the outset of the Landscape Toolbox project, one of our core principles has been that the Landscape Toolbox should be built upon a foundation of peer-reviewed science. While we've certainly made a lot of headway building on the work of others, we have also generated quite a bit of interesting research ourselves as part of the Toolbox project.

I'm quite excited to announce that the first scientific paper to come from the Landscape Toolbox was published today in the journal Landscape Ecology. You can find the paper at SpringerLink. If you'd like a PDF of the paper, email me.

This paper deals with how information in remotely-sensed imagery changes as scale becomes coarser - one of the fundamental concepts in the Toolbox's multi-scale framework. We looked at how the relationship between field and image data changed as the data were scaled up, and we compared the typical approach of scaling by aggregating pixels into larger squares to the new object-based approach that we have adopted for the Toolbox. Our results clearly demonstrated that not only did the object-based approach give higher correlations between the field data and imagery, but that the correlations were consistent across scales. Pixel-based methods produced unpredictable relationships between the field data and imagery.

This paper is a great validation of several key aspects of the Landscape Toolbox project. It's also just the first of many papers to be published. Two more papers are in different stages of publication - one in press with Rangeland Ecology and Management, and one in review for Ecological Informatics. Two additional manuscripts are in preparation, and several more are on the drawing board.

I'll make sure and let you know as additional publications come out.

Cheers-

Jason Karl, Ph.D.
Landscape Toolbox Project Lead
USDA-ARS Jornada Experimental Range