Landscape Stressor Summaries for GLCWC study sites

Background

Understanding the relationship between human disturbance and ecological response is essential to the process of indicator development. In addition, understanding and quantifying the full gradient of stress acting upon ecosystems throughout a region is critical for the identification of reference areas.  As part of the Great Lakes Environmental Indicators (GLEI) project, we delineated watersheds for coastal areas throughout the USside of the Great Lakes as summary units for a wide variety of anthropogenic stressors. For these delineations we used three approaches 1) segmentation of the shoreline at halfway points between adjacent 2^nd order and higher streams, 2) specific watershed delineations for GLEI sampled wetlands and embayments, and 3) a Great Lakes basin-wide catchment approach (using ArcHydro) where sub-basins can be agglomerated into contributing areas for specific sites. All three approaches proved useful, but each had specific applications and geomorphic types for which they were the most appropriate.  The third approach has the highest resolution and provides a framework for developing "multi-scale" watershed metrics that can be tailored for specific sites using area weighted averages of stressor scores from the intersecting sub-basins with a particular site.  These also cover the entire Great Lakes basin which allows us to identify the reference (best 20%) and degraded (worst 20%) watersheds in reference to the whole basin, particular lakes, ecoregions or states.

Working  with the Great Lakes Wetland Consortium we applied the third set of stressor summaries, described above, to GLWC study sites to further everyone's efforts developing effective ecological indicators for the Great Lakes.  We have also  developed  watershed stressor summaries for the Canadian side of the Great Lakes to quantify continuous gradients of anthropogenic stress for the entire Great Lakes basin.  This high resolution, spatially explicit, summary of Great Lakes anthropogenic stress will prove useful to many other researchers now and into the future.  Yet working with these data is not simple.  Working with the GLWC to apply these stressor summaries specifically to GLCW sites will help us develop step by step instructions for future researchers to effectively coalesce these sub-watersheds into larger watersheds specific to a particular site or ecosystem of interest.

GLC Wetland Consortium Study Sites

A request was made to all GLCWC investigators for study site locations they used in developing Great Lakes coastal wetland indicators.  With help from the GLC a set of points  and a set of polygons were compiled for relevant GLCWC study sites and data sets, based on this request.  Because these were compiled from several different files a master set of point and polygon attributes was compiled containing all of the data contained in the component data sets.  These were uniquely named and numbered and linked on the field "key1" to the related points and/or polygons.  The points and polygons were all classed by US or Canada, and whether or not they existed on islands (we currently do not have stressor summaries for islands). 

Shapefiles representing the points and polygons and the combined attribute tables are contained the attached zipped file (GLCWC_site_locations-attribs.zip)

Data Sets 

United States (GLEI) Stressor Summaries 

The attached zipped file (GLCWC_GLEI_stressors.zip) contains stressor values originally derived for the Great Lakes Environmental Indicators Project (GLEI), which we've now related to GLCWC wetland sites (for the locations requested via e-mail on 3/1/07). 

For the GLEI project 3,591 coastal watersheds were delineated for the entire US Great Lakes Basin.  These watersheds consist of stream watersheds flowing to the coast, and the small areas between stream mouths without streams, that flow directly to the shore (coastal interfluves).   Once these watersheds were delineated they were ordered along the coast from west to east, allowing adjacent watersheds for a particular area to be agglomerated, and their stressor summaries area weighted by catchment area.  For the GLEI project we summarized many different variables related to Land cover, atmospheric deposition, agricultural land use, human population and other types of stress for each of these 3,591 watersheds.  For the GLCWC sites the associated watersheds were identified and area weighted means (by catchment area) were calculated when more than one watershed intersected a particular site.  The stressors included in these summaries include Pete Wolter's (Wolter et al. 2006) enhanced 2001 National Land Cover data, and Principal component summaries of the stressor categories (Agricultural land use, atmospheric deposition, urban land & pop density, land cover, point sources, and a combined score or "sum_index"). For a more complete description of the Stressor PCs (but applied to a different set of watersheds and variables see Danz et al. 2007).  Both publications are included in the attached zip file.

The stressor scores for the GLCWC sites are contained in two summary output files (contained in the zipped file mentioned above) ""polys_weighted_nlcd01_out1.csv", "polys_weighted_pcs_out.csv", and "pts_pcs_nlcd_out.csv"  They relate on the field "key1" to the tables "pts_merge_attribs.dbf" and "polys_merge_attribs.dbf" for more complete descriptions of the GLCWC sites (with site names and project names). To put the scores in perspective with the entire population of US Great Lakes watersheds, frequency histograms were developed of the PC scores for the entire population (n = 3591).  These are contained in the attached power point file (catchment_pcs_histograms.ppt)

The associated watershed shapefiles available.  Contact Tom Hollenhorst (thollenh"at"nrri.umn.edu) for these.

US/Canada Integrated Stressor Summaries 

A similar set of watersheds have been developed for the Canadian portion of the Great Lakes Basin and integrated with the US watersheds.  This integrated set contains 5,971 unique watersheds.  Initial stresser summaries using these watersheds have been quickly developed, specifically for the GLCWC, to describe continuous gradients of anthropogenic stress quantified for the entire Great Lakes basin.  This includes 1990's integrated landcover, road density and population density.  Once again, for the GLCWC sites the associated watersheds were identified and area weighted means (by catchment area) of stressor scores were calculated when more than one watershed intersected a particular site.  The stressor values for calculated for the GLCWC site locations are also attached and linked below:

GLCWC Points (landcover) us-can_lc_allpts-fin.csv
GLCWC Polygons (landcover) us-can_lc_prop-calced_polys.csv

GLCWC Polygons (road density) us-can_rd_density_polys.dbf

GLCWC Polygons (population density)us-can_pop_polys-summed-fin.dbf

GLCWC Points (road density & population density) corrected_fin-pts-road-density_human_density.dbf

GLCWC Points Sum-Rel (overall stressor score)  corrected_fin-pts-sumrel.dbf

GLCWC Polygons Sum-Rel (overall stressor score)   polys_sum-rel.csv

Integrated Overall Stressor Score (click to enlarge)

 
These summaries all have the field "key1" which relates back to the GLCWC site locations attribute files mentioned in the GLC Wetland Consortium Study Sites section above.