These data are used in the book "Computational Statistics" 
by G.H. Givens and J.A. Hoeting.

Name: 
Chapter: 1
Discussed in: Examples 12.4-12.6 and Problem 12.5

Source: U.S. Environmental Protection Agency through its Environmental
Monitoring and Assessment Program (EMAP), http://www.epa.gov/emap.  
Special thanks to Alan Herlihy (Oregon State University and EPA) who
compiled the data from Jennifer Hoeting (Colorado State University).

Description: These data are a subset of those collected by the
  Environmental Protection Agency as part of a study of 353 sites in
  the Mid-Atlantic Highlands region of the eastern United States from
  1993 to 1998.  In this case, the goal is to predict the response, an
  index of biotic integrity, from a set of predictors.  Some of the
  predictors were collected at the site (the site habitat and
  chemistry measures listed in problem 12.5) and other variables are
  watershed-based measured.  The geographic and watershed measures
  were estimated using GIS.  For more information on these and other
  MAHA data, see the following websites.  For an overview of MAHA
  project, see http://www.epa.gov/maia/html/maha.html.  For details on
  these and other MAHA data, see
  http://www.epa.gov/emap/html/dataI/surfwatr/data/


Variable labels: 
 
 STREAM.ID	EPA's identification number
 IBI.BUG	Index of biotic integrity 
 YEAR		year the data were collected
 LAT.DD		latitude
 LON.DD		longitude
 PHSTVL		pH
 ANC		Acid neutralizing capacity
 log.COND	specific conductance
 log.CL		chloride
 log.SO4	sulfate
 log.PTL	total phosphorus
 log.NTL	total nitrogen
 xcdenbk	canopy density above mid-channel
 pct.fast	percent fast water
 xslope		channel slope
 lsub.dmm	substrate diameter
 log.WSAREA	watershed area above site
 ELEV.X		elevation
 SLOPMEAN	mean slope in watershed above site
 log.POPDENKM   human population density
 FOR.NLCD	percent land forest
 AG.NLCD	percent land in agriculture
 log.URB.NLCD   percent land urban
 log.MINENLCD   percent land mined

Note: "log." indicates the variable was transformed logarithmically 