Patterson Awarded 2010 Jesse Rossbach Memorial Scholarship

NMGIC Jesse Rossbach Memorial Scholarship winner Anna Patterson

Anna Klimaszeski Patterson is the 2010 Jesse Rossbach Memorial Scholarship Winner. She is a Graduate Assistant/Teaching Assistant at New Mexico State University in Las Cruces, NM, pursuing a Master in Applied Geography, with Dr. Michael N. DeMers serving as her advisor. She received her B.A. in Geography from San Diego State University in 2005. Her submittal was entitled:

"Verification of vegetative boundaries based on an arid-environment geomorphological rapid assessment model in the northern Chihuahuan Desert"

Anna says she has two primary interests. One is to "reconstruct" pre-colonization environments to determine the extent of ecological modification and model what the environment would look like today without excessive anthropogenic impact. This knowledge could then be used for ecological restoration/mitigation efforts to restore balance to impacted ecological systems. The second is the application of mobile devices (i.e. smartphones) as geographic tools.

Master's Research

Ecosystem degradation research has not focused explicitly on either the accuracy of boundaries drawn, or on taking climate-induced vegetation progression into account. Arid and semiarid environments, by nature, have sparse, patchy vegetation and a predominant land cover of bare soil. Remote images in these ecosystems allow for examination of geomorphology without the impedance of vegetation. Considerable research has been done on the correlation between geomorphology and vegetation types, opening the possibility of using geomorphology as a proxy for expected vegetation given current climate conditions in an arid/semiarid environment. Using Landsat 7 ETM images and visual analysis, DeMers et al. (2003) developed a rapid bioassessment model for the northern Chihuahuan Desert, with geomorphology predicting expected vegetation; extensive field verification was not performed.

Anna's Master's thesis is to establish and perform field verification techniques for an accuracy assessment of vegetation boundaries based on DeMers et al.’s (2003) model. The accuracy assessments will be calculated both overall and broken down by individual boundary types.

If proven, the rapid bioassessment model could be used by researchers to rapidly create expected vegetation maps based on current climate conditions, allowing for comparison between actual and expected vegetation communities within the greater biome for which the model is calibrated.

This research will further science by providing an improvement in the use of field verification of categorical boundaries, including more technical processing such as image segmentation. It will also provide a sampling scheme to remote sensors for field validation. Beyond researchers, this research will improve the ability for range managers to assess and manage their land based on accurate management boundaries.