Using MARXAN for Landscape Analysis in the Vicinity of Chocolate Bayou, Texas Pamela S. Showalter, Director, James and Marilyn Lovell Center for Environmental Geography & Hazards Research, Department of Geography Warren Pulich, Jr., River Systems Institute The authors also gratefully acknowledge the work contributed to this project by Corrie Colvin, MAG 2007. Funding for this research has been provided by the Galveston Bay Estuary Program Showalter © 2007 Research Goal: Establish the relationship between existing Land Use/Land Cover (LU/LC) to water quality in the larger study area. LU/LC classes were created by manually interpreting 1:24,000 color aerial photography (2004). Initial findings revealed the spatial relationships between LU/LC and other landscape features, as well as the restricted distribution/low percentage of remaining native habitat. This poster focuses on the sub-watershed of Chocolate Bayou, located about 25 miles SE of Houston The portion of the research displayed here illustrates an initial evaluation of "MARXAN"--an ArcView® extension that employs spatial annealing--as a tool for identifying locations that may be those best suited for helping maintain and/or improve the area’s water quality and wetland functions. Guiding Principle: LU/LC impacts water quality in either a positive, negative, or neutral fashion The Need: Spatial analysis of the study area’s many data layers in relationship to water quality parameters Desired Output: A new study area map showing where the combination of activities taking place on the landscape are suspected of impacting water quality in a positive, negative, or neutral fashion The Basic Steps: LU/LC classes are assigned different “weights,” which are then “embedded” into hexagons that have been created to cover each class. Weights are summed across the landscape, producing a new map of color gradients showing where the sum of the weights is high, low, etc. High weights display as reds, low weights display as grays. MARXAN was developed by the University of Queensland & the Australian Antarctic Division (http://www.ecology.uq.edu.au/index.html?page=29780) References Ball, I.R. and H.P. Possingham (2000). MARXAN (V1.8.2): Marine Reserve Design Using Spatially Explicit Annealing, a Manual; and Possingham, H.P., I.R. Ball, and S. Andelman (2000). Mathematical methods for identifying representative reserve networks. In, S. Ferson and M. Burgman (eds.), Quantitative methods for conservation biology, New York: Springer-Verlag, pp. 291-305. Results: The extension “drops” the conserved and excluded areas, performing its analysis solely on the remaining “available” data. As expected, reds appear in locations where the combination of classes would logically most enhance water quality (high weights) while the reverse holds true for areas least likely to enhance water quality (grays). To confirm our understanding of how the extension processed the data, a test run was performed by creating criteria that had a logical, expected outcome: 1) the FEMA floodplain was set as “conserved”, and 2) highly urbanized/industrial areas and hexagons found outside the sub-watershed were set as excluded. (Other default values offered by the software were accepted.) Conclusion: Although created as an aid for biologists to use for identifying marine conservation reserves, MARXAN shows promise as a tool for analyzing LU/LC classes vis a vis water quality issues. Experimentation with MARXAN continues, which includes changing the weights, adding layers (e.g., roads, railroads), and including the “earmarked” parameter.

Using MARXAN for Landscape Analysis in the …gato-docs.its.txstate.edu/jcr:bba7bae9-a0c4-46d7-96dc-dd...Using MARXAN for Landscape Analysis in the Vicinity of Chocolate Bayou, Texas

Download PDF Report

Upload
others
View
2
Download
0

Embed Size (px)

Citation preview

Page 1: Using MARXAN for Landscape Analysis in the …gato-docs.its.txstate.edu/jcr:bba7bae9-a0c4-46d7-96dc-dd...Using MARXAN for Landscape Analysis in the Vicinity of Chocolate Bayou, Texas

Warren Pulich, Jr., River Systems InstituteThe authors also gratefully acknowledge the work contributed to this project by Corrie Colvin, MAG 2007. Funding for this research has been provided by the Galveston Bay Estuary Program Showalter © 2007

Research Goal: Establish the relationship between existing Land Use/Land Cover (LU/LC) to water quality in the larger study area. LU/LC classes were created by manually interpreting 1:24,000 color aerial photography (2004). Initial findings revealed the spatial relationships between LU/LC and other landscape features, as well as the restricted distribution/low percentage of remaining native habitat.

This poster focuses on the sub-watershed of Chocolate Bayou, located about 25 miles SE of Houston The portion of the research displayed here illustrates an initial evaluation of "MARXAN"--an ArcView® extension that employs spatial annealing--as a tool for identifying locations that may be those best suited for helping maintain and/or improve the area’s water quality and wetland functions.

Guiding Principle:LU/LC impacts water quality in either apositive, negative, or neutral fashion

The Need:Spatial analysis of the study area’s manydata layers in relationship to waterquality parameters

Desired Output:A new study area map showing where thecombination of activities taking place onthe landscape are suspected of impactingwater quality in a positive, negative, orneutral fashion

The Basic Steps:LU/LC classes are assigned different “weights,” which are then “embedded” into hexagons that have been created to cover each class.

Weights are summed across the landscape, producing a new map of color gradients showing where the sum of the weights is high, low, etc.

High weights display as reds, low weights display as grays.

MARXAN was developed by the University of Queensland & the Australian Antarctic Division(http://www.ecology.uq.edu.au/index.html?page=29780)ReferencesBall, I.R. and H.P. Possingham (2000). MARXAN (V1.8.2): Marine Reserve Design Using Spatially Explicit Annealing, a Manual; andPossingham, H.P., I.R. Ball, and S. Andelman (2000). Mathematical methods for identifying representative reserve networks. In, S. Ferson and M. Burgman (eds.), Quantitative methods for conservation biology, New York: Springer-Verlag, pp. 291-305.

Results:The extension “drops” the conserved and excluded areas, performing its analysis solely on the remaining “available” data.

As expected, reds appear in locations where the combination of classes would logically most enhance water quality (high weights) while the reverse holds true for areas least likely to enhance water quality (grays).

To confirm our understanding of how the extension processed the data, a test run was performed by creating criteria that had a logical, expected outcome:

1) the FEMA floodplain was set as “conserved”, and

2) highly urbanized/industrial areas and hexagons found outside the sub-watershed were set as excluded.

(Other default values offered by the software were accepted.)

Conclusion:Although created as an aid for biologists to use for identifying marine conservation reserves, MARXAN shows promise as a tool for analyzing LU/LC classes vis a vis water quality issues. Experimentation with MARXAN continues, which includes changing the weights, adding layers (e.g., roads, railroads), and including the “earmarked” parameter.