#
#
#  FILE_NAME: 		MRB5.ARCHIVAL_FILE_SUMMARY.TXT
#  DATASET_DESCRIPTION:	These data sets were used as part of a nutrient, sediment, and streamflow trend 
#			assessment for the Lower Mississippi, Arkansas-White-Red, and Texas-Gulf Basins.
#  DATA_SET FORMAT:	Tab-delimited ascii file
#  DATA_SET URL:	
#
#-------------------------------------------------------------------------------------------------------- 
#  REPORT TITLE: 	Trends in Nutrient and Sediment Concentrations and Loads in Major River Basins
#			of the South-Central United States, 1993-2004    
#  REPORT SERIES/NUMBER (JOURNAL NAME/VOLUME): USGS Scientific Investigation Report (SIR 2007-5090)     
#  REPORT URL:		http://pubs.usgs.gov/sir/2007/5090/
#
#-------------------------------------------------------------------------------------------------------
#  NAWQA STUDY:		Major River Basin 5
#
#-------------------------------------------------------------------------------------------------------
#
#  METADATA DATE:    	January, 2008
#  METADATA CONTACT: 	Richard A. Rebich
#                    	USGS-WRD, Mississippi Water Science Center 
#                    	308 South Airport Road
#	 	     	Jackson, MS  39208-6649	
#                    	voice: 601-933-2928
#                    	fax:   601-933-2901
#                    	email:  rarebich@usgs.gov
#
#  SECONDARY CONTACT:   Dennis Demcheck
#                    	USGS-WRD, Louisiana Water Science Center 
#                    	3535 S. Sherwood Forest Blvd., Suite 120
#	 	     	Baton Rouge, LA  70816	
#                    	voice: 225-298-5481, ext.3214
#                    	fax:   225-298-5490
#                    	email:  ddemchec@usgs.gov
#
# ------------------------------------------------------------------------------------------------------
#  DATA_SET NAMES:
#
#  The following files document and archive the data sets used for producing this report. Brief descriptions of 
#  each of these report archival files follow.
#
#  Files used for trends analyses:
#  MRB5A_wq_site_list.txt 
#  MRB5A_dv_site_list.txt
#  MRB5A_MRB_model_specs.xls.txt
#  MRB5A.sas7bdat.txt
#  07355500.rdb.txt
#  07373420.rdb.txt
#  07381495.rdb.txt
#  MRB5_trend_results.txt
#  
#  Files used to calculate loads:
#  MRB5_QWdata_for_LOADEST_USGS.txt
#  MRB5_QWdata_for_LOADEST_State.txt
#  MRB5_Qdata_for_loads.txt
#  MRB5_final_loads.txt
#  
#  Files used for determination of trends in ancillary datasets:
#  MRB5.drainagebasins.txt
#  MRB5_ancil_data.txt
#  MRB5_pop_dens.txt
#  MRB5_cons_prac.txt
#
#  Files used for weighted-least-squares regression:
#  MRB5_WLS_input.txt
#  MRB5_WLS.results.txt
#  
#
#--------------------------------------------------------------------------------------------------------
#  REPORT ARCHIVAL DATA SET:  MRB5_wq_site_list.txt
#
#	This data set lists the water-quality sites used in trend analyses of nutrient and sediment
#	concentrations and loads for streams in the Lower Mississippi, Arkansas-White-Red, Texas-Gulf
#	Basins as presented in Rebich and Demcheck, 2007 (SIR 2007-5090). This investigation was 
#	completed as part of the USGS National Water-Quality Assessment Program. Trend analyses were 
#	completed using Fluxmaster, which is a load estimation program that is typically used in 
#	conjunction with the Spatially Referenced Regression on Watershed Attributes (SPARROW) model. 
#	The Fluxmaster program uses a multiple regression technique to estimate loads.
#	The regression equation includes a time term, and the coefficient for time is the trend in 
#	percent per year for the study period analyzed. Coefficients of time were determined for
#	streamflow, total trends of nutrients and sediment unadjusted for flow, and flow-adjusted 
#	nutrient and sediment trends. The complete description of the theory and equations for Fluxmaster
#	are documented in Schwarz and others, 2006 (USGS Techniques and Methods report, Book 6,Section B, 
#	Chapter 3). The Fluxmaster program is run using SAS software (Statistical Analysis System 
#	Institute, 2000). Software and documentation for Fluxmaster/SPARROW can be downloaded from the 
#	USGS SPARROW web page at URL: http://pubs.usgs.gov/tm/2006/tm6b3/. The file listed here is a 
#	text file and is ready for use in SAS for Fluxmaster/SPARROW in its current format.  Analyses 
#	were performed for all sites listed, but not all results were published in Rebich and Demcheck, 
#	2007 (SIR 2007-5090) due to poor model fit or large errors.
#
#--------------------------------------------------------------------------------------------------------
#  REPORT ARCHIVAL DATA SET:  MRB5_dv_site_list.txt
#
#	This data set lists the daily flow sites used in trend analyses of nutrient and sediment
#	concentrations and loads for streams in the Lower Mississippi, Arkansas-White-Red, Texas-Gulf
#	Basins as presented in Rebich and Demcheck, 2007 (SIR 2007-5090). This investigation was 
#	completed as part of the USGS National Water-Quality Assessment Program. In most cases, daily flow
#	sites were co-located with water-quality sites. In some cases, alternate daily flow sites were used 
#	that were located near water-quality sites. The alternate site ID's are listed in
#	MRB5A_wq_site_list.txt.  If no alternate site ID is listed there, then the SAS routines use the
#	flow station ID listed in this file.  Trend analyses were completed using Fluxmaster, which is a 
#	load estimation program that is typically used in conjunction with the Spatially Referenced 
#	Regression on Watershed Attributes (SPARROW) model. The Fluxmaster program uses a multiple
#	regression technique to estimate loads. The regression equation 
#	includes a time term, and the coefficient for time is the trend in percent per year for the study 
#	period analyzed. Coefficients of time were determined for streamflow, total trends of nutrients 
#	and sediment unadjusted for flow, and flow-adjusted nutrient and sediment trends. The complete 
#	description of the theory and equations for Fluxmaster are documented in Schwarz and others, 2006 
#	(USGS Techniques and Methods report, Book 6,Section B, Chapter 3). The Fluxmaster program is run 
#	using SAS software (Statistical Analysis System Institute, 2000). Software and documentation for 
#	Fluxmaster/SPARROW can be downloaded from the SPARROW web page at 
#	URL: http://pubs.usgs.gov/tm/2006/tm6b3/. This file is used by subroutines in the Fluxmaster 
#	program to retrieve daily flow data at the listed sites from the USGS National Water Information 
#	System (NWIS). The web-based version of NWIS can be found at URL: 
#	http://waterdata.usgs.gov/nwis. The file listed here is a text file and is ready for use in SAS 
#	for Fluxmaster/SPARROW in its current format.  Analyses were performed for all sites 
#	listed, but not all results were published in Rebich and Demcheck, 2007 (SIR 2007-5090) due to 
#	poor model fit or large errors.
#
#--------------------------------------------------------------------------------------------------------
#  REPORT ARCHIVAL DATA SET:  MRB5_MRB_model_specs.xls.txt
#
#	This data set lists the model specifications used in trend analyses of nutrient and sediment
#	concentrations and loads for streams in the Lower Mississippi, Arkansas-White-Red, Texas-Gulf
#	Basins as presented in Rebich and Demcheck, 2007 (SIR 2007-5090). This investigation was 
#	completed as part of the USGS National Water-Quality Assessment Program. Trend analyses were 
#	completed using Fluxmaster, which is a load estimation program that is typically used in 
#	conjunction with the Spatially Referenced Regression on Watershed Attributes (SPARROW) model. 
#	The Fluxmaster program uses a multiple regression technique to estimate loads. The 
#	regression equation includes a time term as well as terms for flow and season (sine and 
#	cosine functions).  Numbers listed in this file correspond to a specified equation that includes 
#	combinations of all or some of the possible terms.  The numbered equations are listed below. 
#	The coefficient for time is the trend in percent per year for the study 
#	period analyzed. Coefficients of time were determined for streamflow, total trends of nutrients 
#	and sediment unadjusted for flow, and flow-adjusted nutrient and sediment trends. The complete 
#	description of the theory and equations for Fluxmaster are documented in Schwarz and others, 2006 
#	(USGS Techniques and Methods report, Book 6,Section B, Chapter 3). The Fluxmaster program is run 
#	using SAS software (Statistical Analysis System Institute, 2000). Software and documentation for 
#	Fluxmaster/SPARROW can be downloaded from the SPARROW web page at 
#	URL: http://pubs.usgs.gov/tm/2006/tm6b3/. The file listed here is a text file but must be entered 
#	and saved as an Microsoft Excel file (.xls) before using in the SAS software for the Fluxmaster/
#	SPARROW program.  Analyses were performed for all sites listed, but not all results were published 
#	in Rebich and Demcheck, 2007 (SIR 2007-5090) due to poor model fit or large errors.
#
#--------------------------------------------------------------------------------------------------------
#  REPORT ARCHIVAL DATA SET:  MRB5.sas7bdat.txt
#
#	This file contains the complete water-quality data set used in trend analyses of nutrient and 
#	sediment concentrations and loads for streams in the Lower Mississippi, Arkansas-White-Red, 
#	Texas-Gulf Basins as presented in Rebich and Demcheck, 2007 (SIR 2007-5090). This investigation 
#	was completed as part of with the USGS National Water-Quality Assessment Program. The original water 
#	quality data were compiled mostly from the USGS National Water Information System (NWIS). The web- 
#	based version of NWIS can be found at URL: http://waterdata.usgs.gov/nwis. Non-USGS data are also 
#	included in this file, which were compiled individually from State Environmental Agencies in the 
#	MRB5 study area. The data listed in this file are modified according to documentation in Rebich 
#	and Demcheck, 2007 (SIR 2007-5090). Trend analyses were completed using Fluxmaster, which is a 
#	load estimation program that is typically used in conjunction with the Spatially Referenced 
#	Regression on Watershed Attributes (SPARROW) model. The Fluxmaster program uses a multiple
#	regression technique to estimate loads. The regression equation 
#	includes a time term, and the coefficient for time is the trend in percent per year for the study 
#	period analyzed. Coefficients of time were determined for streamflow, total trends of nutrients 
#	and sediment unadjusted for flow, and flow-adjusted nutrient and sediment trends. The complete 
#	description of the theory and equations for Fluxmaster are documented in Schwarz and others, 2006 
#	(USGS Techniques and Methods report, Book 6,Section B, Chapter 3). The Fluxmaster program is run 
#	using SAS software (Statistical Analysis System Institute, 2000). Software and documentation for 
#	Fluxmaster/SPARROW can be downloaded from the SPARROW web page at 
#	URL: http://pubs.usgs.gov/tm/2006/tm6b3/. The file listed here is a text file and can be read 
#	directly into the SAS software.  However, the data must be saved with the extension .sas7bdat 
#	before being ready for use in Fluxmaster/SPARROW.  Analyses were performed for all sites 
#	listed, but not all results were published in Rebich and Demcheck, 2007 (SIR 2007-5090) due to 
#	poor model fit or large errors.
#
#--------------------------------------------------------------------------------------------------------
#  REPORT ARCHIVAL DATA SETS:  07355500.rdb.txt, 07373420.rdb.txt, and 07381495.rdb.txt
#
#	This file contains daily flow data for three U.S. Army Corps of Engineers sites used in trend 
#	analyses of nutrient and sediment concentrations and loads for streams in the Lower Mississippi, 
#	Arkansas-White-Red, Texas-Gulf Basins as presented in Rebich and Demcheck, 2007 (SIR 2007-5090).
#	This investigation was completed as part of the USGS National Water-Quality Assessment Program. 
#	These three sites were located where water-quality data were compiled from the National Water 
#	Information System (NWIS) and were likely sampled by the USGS, but no flow data were available 
#	in NWIS. Daily flow data for these three sites were requested and received from the U.S. Army 
#	Corps of Engineers. The file name includes the 8-digit station ID for the water-quality site. 
#	Trend analyses were completed using Fluxmaster, which is a load estimation program that is 
#	typically used in conjunction with the Spatially Referenced Regression on Watershed Attributes 
#	(SPARROW) model. The Fluxmaster program uses a multpile regression technique to estimate loads. 
#	The regression equation includes a time term, and the coefficient for time is the trend in 
#	percent per year for the study period analyzed. Coefficients of time were determined for 
#	streamflow, total trends of nutrients and sediment unadjusted for flow, and flow-adjusted 
#	nutrient and sediment trends. The complete description of the theory and equations for Fluxmaster 
#	are documented in Schwarz and others, 2006 (USGS Techniques and Methods report, Book 6,Section B, 
#	Chapter 3). The Fluxmaster program is run using SAS software (Statistical Analysis System 
#	Institute, 2000). Software and documentation for Fluxmaster/SPARROW can be downloaded from the 
#	SPARROW web page at URL: http://pubs.usgs.gov/tm/2006/tm6b3/. The files listed here are text 
#	files, however, the data must be saved with the extension .rdb before being ready for use in 
#	Fluxmaster/SPARROW.  Analyses were performed for all sites listed, but not all results were 
#	published in Rebich and Demcheck, 2007 (SIR 2007-5090) due to poor model fit or large errors.
#
#--------------------------------------------------------------------------------------------------------
#  REPORT ARCHIVAL DATA SET:  MRB5_trend_results.txt
#
#	This file contains results of trend analyses of nutrient and sediment concentrations and loads 
#	for streams in the Lower Mississippi, Arkansas-White-Red, Texas-Gulf Basins as presented in 
#	Rebich and Demcheck, 2007 (SIR 2007-5090). This investigation was completed as part of the USGS 
#	National Water-Quality Assessment Program. Trend analyses were completed using Fluxmaster, which 
#	is a load estimation program that is typically used in conjunction with the Spatially Referenced 
#	Regression on Watershed Attributes (SPARROW) model. The Fluxmaster program uses a multiple 
#	regression technique to estimate loads. The regression equation includes a time 
#	term, and the coefficient for time is the trend in percent per year for the study period analyzed. 
#	Coefficients of time were determined for streamflow, total trends of nutrients and sediment 
#	unadjusted for flow, and flow-adjusted nutrient and sediment trends. Reference concentrations 
#	and loads were also computed. The complete description of the theory and equations for Fluxmaster 
#	are documented in Schwarz and others, 2006 (USGS Techniques and Methods report, Book 6,Section B, 
#	Chapter 3). The Fluxmaster program is run using SAS software (Statistical Analysis System 
#	Institute, 2000). Software and documentation for Fluxmaster/SPARROW can be downloaded from the 
#	SPARROW web page at URL: http://pubs.usgs.gov/tm/2006/tm6b3/. This file is an abbreviated version
#	of Fluxmaster output streamlined to be pertinent to published results. This file is a text file 
#	but is recommended to be read into spreadsheet software such as Microsoft Excel to view. Analyses 
#	were performed for all sites listed, but not all results were published in Rebich and 
#	Demcheck, 2007 (SIR 2007-5090) due to poor model fit or large errors.
#
#--------------------------------------------------------------------------------------------------------
#  REPORT ARCHIVAL DATA SET:  MRB5_QWdata_for_LOADEST_USGS.txt
#
#	This file contains the water-quality data set used to calculate nutrient and 
#	sediment loads for streams in the Lower Mississippi, Arkansas-White-Red, 
#	Texas-Gulf Basins as presented in Rebich and Demcheck, 2007 (SIR 2007-5090). This investigation 
#	was completed as part of the USGS National Water-Quality Assessment Program. These water 
#	quality data were compiled from the USGS National Water Information System (NWIS). The web- 
#	based version of NWIS can be found at URL: http://waterdata.usgs.gov/nwis. The data listed in 
#	this file are modified according to documentation in Rebich and Demcheck, 2007 (SIR 2007-5090).
#	Load calculations were completed using LOADEST [documented in Runkel, R.L., Crawford, C.G., and 
#	Cohn, T.A., 2004, Load Estimator (LOADEST):A FORTRAN Program for estimating constituent loads 
#	in streams and rivers: U.S. Geological Survey Techniques and Methods Book 4, Chapter A5, 69 p.].
#	LOADEST is a load estimation program that is part of a USGS "plug-in" in the S-Plus Statistical
#	Software suite (Insightful Corporation, 2005, S-PLUS version 7.0 for Windows Professional
#	Developer: Seattle, Wash., Insightful Corporation). The LOADEST program uses a multiple
#	regression technique to estimate loads. The regression equation can
#	include time, flow and seasonal terms.  
#	The "best" equation was determined using the Akaike Information Criterion (AIC) as documented
#	in Runkel and others, 2004.  The file listed here is a text file and can be read 
#	directly into the LOADEST software.  To calculate annual loads, LOADEST also requires mean 
#	daily streamflow data, which can be found in metadata file MRB5_Qdata_for_loads.txt . Not all 
#	load results were published in Rebich and Demcheck, 2007 (SIR 2007-5090) due to poor model fit 
#	or large errors.
#
#--------------------------------------------------------------------------------------------------------
#  REPORT ARCHIVAL DATA SET:  MRB5_QWdata_for_LOADEST_State.txt
#	This file contains the water-quality data set used to calculate nutrient and 
#	sediment loads for streams in the Lower Mississippi, Arkansas-White-Red, 
#	Texas-Gulf Basins as presented in Rebich and Demcheck, 2007 (SIR 2007-5090). This investigation 
#	was completed as part of the USGS National Water-Quality Assessment Program. These water 
#	quality data were compiled from databases from State agencies in the study area. Station ID's 
#	for these sites were modified to reflect USGS flow gage Station ID's and station names.
#	Load calculations were completed using LOADEST [documented in Runkel, R.L., Crawford, C.G., and 
#	Cohn, T.A., 2004, Load Estimator (LOADEST):A FORTRAN Program for estimating constituent loads 
#	in streams and rivers: U.S. Geological Survey Techniques and Methods Book 4, Chapter A5, 69 p.].
#	LOADEST is a load estimation program that is part of a USGS "plug-in" in the S-Plus Statistical
#	Software suite (Insightful Corporation, 2005, S-PLUS version 7.0 for Windows Professional
#	Developer: Seattle, Wash., Insightful Corporation). The LOADEST program uses a multiple
#	regression technique to estimate loads. The regression equation can
#	include time, flow and seasonal terms.  
#	The "best" equation was determined using the Akaike Information Criterion (AIC) as documented
#	in Runkel and others, 2004.  The file listed here is a text file and can be read 
#	directly into the LOADEST software.  To calculate annual loads, LOADEST also requires mean 
#	daily streamflow data, which can be found in metadata file MRB5_Qdata_for_loads.txt . Not all 
#	load results were published in Rebich and Demcheck, 2007 (SIR 2007-5090) due to poor model fit 
#	or large errors.
#
#--------------------------------------------------------------------------------------------------------
#  REPORT ARCHIVAL DATA SET:  MRB5_Qdata_for_loads.txt
#
#	This file contains the mean daily flow data used to estimate nutrient and 
#	sediment loads for streams in the Lower Mississippi, Arkansas-White-Red, 
#	Texas-Gulf Basins as presented in Rebich and Demcheck, 2007 (SIR 2007-5090). This investigation 
#	was completed as part of the USGS National Water-Quality Assessment Program.
#	Load calculations were completed using LOADEST [documented in Runkel, R.L., Crawford, C.G., and 
#	Cohn, T.A., 2004, Load Estimator (LOADEST):A FORTRAN Program for estimating constituent loads 
#	in streams and rivers: U.S. Geological Survey Techniques and Methods Book 4, Chapter A5, 69 p.].
#	LOADEST is a load estimation program that is part of a USGS "plug-in" in the S-Plus Statistical
#	Software suite (Insightful Corporation, 2005, S-PLUS version 7.0 for Windows Professional
#	Developer: Seattle, Wash., Insightful Corporation). The LOADEST program uses a multiple
#	regression technique to estimate loads. The regression equation can
#	include time, flow, and seasonal terms.  
#	The "best" equation was determined using the Akaike Information Criterion (AIC) as documented
#	in Runkel and others, 2004.  To calculate annual loads, LOADEST also requires 
#	water quality data, which can be found in metadata files MRB5_QWdata_for_LOADEST_USGS.txt and
#	MRB5_QWdata_for_LOADEST_State.txt . 
#	The file listed here is a text file that includes all of the flow data for every site.  This
#	file may be too large to read into LOADEST as a stand-alone file.  It may need to be broken
#	into smaller files using a text editor prior to loading into LOADEST.  Not all 
#	load results were published in Rebich and Demcheck, 2007 (SIR 2007-5090) due to poor model fit 
#	or large errors.  
#
#--------------------------------------------------------------------------------------------------------
#  REPORT ARCHIVAL DATA SET:  MRB5_final_loads.txt
#
#	This file contains the final estimated nutrient and 
#	sediment loads for streams in the Lower Mississippi, Arkansas-White-Red, 
#	Texas-Gulf Basins as presented in Rebich and Demcheck, 2007 (SIR 2007-5090). This investigation 
#	was completed as part of the USGS National Water-Quality Assessment Program.
#	Load calculations were completed using LOADEST [documented in Runkel, R.L., Crawford, C.G., and 
#	Cohn, T.A., 2004, Load Estimator (LOADEST):A FORTRAN Program for estimating constituent loads 
#	in streams and rivers: U.S. Geological Survey Techniques and Methods Book 4, Chapter A5, 69 p.].
#	LOADEST is a load estimation program that is part of a USGS "plug-in" in the S-Plus Statistical
#	Software suite (Insightful Corporation, 2005, S-PLUS version 7.0 for Windows Professional
#	Developer: Seattle, Wash., Insightful Corporation). The LOADEST program uses a multiple
#	regression technique to estimate loads. The regression equation can
#	include time, flow, and seasonal terms.  
#	The "best" equation was determined using the Akaike Information Criterion (AIC) as documented
#	in Runkel and others, 2004.  To calculate annual loads, LOADEST also requires mean 
#	daily streamflow data, which can be found in metadata file MRB5_Qdata_for_loads.txt . 
#	The file listed here is a text file and can be read 
#	directly into the LOADEST software.  The file contains the mean annual load at each
#	site per consituent for the study period of 1993-1994, the standard error of prediction, 
#	the lower 95 percentile confidence limit estimate, and the upper 95% confidence limit
#	estimate, all of which were converted to metric tons.  Not all 
#	load results were published in Rebich and Demcheck, 2007 (SIR 2007-5090) due to poor model fit 
#	or large errors.  
#
#--------------------------------------------------------------------------------------------------------
#  REPORT ARCHIVAL DATA SET:  MRB5.drainagebasins.txt
#
#	This file documents the location of drainage area boundaries delineated for surface water quality 
#	stations used in the report, "Trends in Nutrient and Sediment Concentrations and Loads in Major 
#	River Basins of the South-Central United States, 1993-2004," by Rebich and Demcheck, USGS 
#	Scientific Investigation Report (SIR 2007-5090), located at http://pubs.usgs.gov/sir/2007/5090/.  
#	The drainage area boundaries were used as digital polygon overlays to partition GIS-derived 
#	ancillary data at each station.  Station-specific ancillary data were then used as explanatory 
#	information to help understand notable water-quality trends that were documented in SIR 2007-5090.
#
#--------------------------------------------------------------------------------------------------------
#  REPORT ARCHIVAL DATA SET:  MRB5_ancil_data.txt
#
#	This file documents trend values for selected ancillary data for water-quality 
#	stations used in the report, "Trends in Nutrient and Sediment Concentrations and Loads in Major 
#	River Basins of the South-Central United States, 1993-2004," by Rebich and Demcheck, USGS 
#	Scientific Investigation Report (SIR 2007-5090), located at http://pubs.usgs.gov/sir/2007/5090/.  
#	The information presented here is summarized as a Thiel slope for each dataset at each station.
#	The first step to determine trends in the ancillary data was to delineate drainage area 
#	boundaries at each site.  The drainage area boundaries were used as digital polygon overlays 
#	to partition the GIS-derived ancillary data at each station.  The ancillary data were then 
#	entered into a spreadsheet, and then a Thiel slope was calculated for each station.  A Thiel 
#	slope is the median of all pairwise slopes as specified in Helsel and Hirsch (1992).  Once 
#	calculated, the Thiel slope was then normalized by dividing by drainage area.  The final values 
#	were then plotted for the study area and then related to trends in water-quality data using 
#	weighted-least-squares (WLS) regression analysis.  Input and results for WLS analysis are found 
#	in files MRB5_WLS_input.txt and MRB5_WLS_results.txt, repsectively.
#	The drainage basin boundaries and ancillary datasets for each water-quality station were obtained 
#	from one of 2 NAWQA websites (http://water.wr.usgs.gov/pnsp/gis/data/swancil/mrb.html, or 
#	http://water.wr.usgs.gov/pnsp/gis/data/swancil/cycle2.html).  These two websites contained 
#	data that had originally been compiled from other external sources.  Datasets and analysis 
#	techniques are further described in Rebich and Demcheck, 2007 (SIR 2007-5090).  
#
#--------------------------------------------------------------------------------------------------------
#  REPORT ARCHIVAL DATA SET:  MRB5_pop_dens.txt
#
#	This file documents trend values for population density ancillary data for water-quality 
#	stations used in the report, "Trends in Nutrient and Sediment Concentrations and Loads in Major 
#	River Basins of the South-Central United States, 1993-2004," by Rebich and Demcheck, USGS 
#	Scientific Investigation Report (SIR 2007-5090), located at http://pubs.usgs.gov/sir/2007/5090/.  
#	The information presented here is summarized as a decadal difference for population density 
#	at each station.  Population density data were used as a surrogate to estimate potential
#	trends in point sources for the study area.
#	The first step to determine trends in the population density data was to delineate drainage area 
#	boundaries at each site.  The drainage area boundaries were used as digital polygon overlays 
#	to partition the GIS-derived data at each station.  The population data were then 
#	entered into a spreadsheet, and a difference was calculated for each station.  The
#	population density data were already normalized by drainage area.  The differences 
#	were then plotted for the study area and then related to trends in water-quality data using 
#	weighted-least-squares (WLS) regression analysis.  Input and results for WLS analyses are found 
#	in files MRB5_WLS_input.txt and MRB5_WLS_results.txt, repsectively.
#	The drainage basin boundaries and population datasets for each water-quality station were obtained 
#	from one of 2 NAWQA websites (http://water.wr.usgs.gov/pnsp/gis/data/swancil/mrb.html, or 
#	http://water.wr.usgs.gov/pnsp/gis/data/swancil/cycle2.html).  These two websites contained 
#	data that had originally been compiled from other external sources.  Datasets and analysis 
#	techniques are further described in Rebich and Demcheck, 2007 (SIR 2007-5090).  
#
#--------------------------------------------------------------------------------------------------------
#  REPORT ARCHIVAL DATA SET:  MRB5_cons_prac.txt
#
#	This file documents trend values for conservation practice ancillary data at water-quality 
#	stations used in the report, "Trends in Nutrient and Sediment Concentrations and Loads in Major 
#	River Basins of the South-Central United States, 1993-2004," by Rebich and Demcheck, USGS 
#	Scientific Investigation Report (SIR 2007-5090), located at http://pubs.usgs.gov/sir/2007/5090/.  
#	The information presented here is summarized as a difference in acreage for specific conservation
#	practices between the years 1992 and 1997 at each station.  
#	The first step to determine trends in the conservation practices data was to delineate drainage 
#	area boundaries at each site.  The drainage area boundaries were used as digital polygon overlays 
#	to partition the GIS-derived data at each station.  The aggregated conservation practices data 
#	were then entered into a spreadsheet, and a difference between the 1992 and 1997 data was 
#	calculated for each station .  The differences were normalized by drainage area and multiplied 
#	by 100 to express the value as a percent change.  Percentages greater than 1 percent or lower 
#	than -1 percent, which represented increasing and decreasing trends, respectively, were tabled 
#	in SIR 2007-5090.
#	The drainage basin boundaries and conservation practices datasets for each water-quality station 
#	were obtained from one of 2 NAWQA websites 
#	(http://water.wr.usgs.gov/pnsp/gis/data/swancil/mrb.html, or 
#	http://water.wr.usgs.gov/pnsp/gis/data/swancil/cycle2.html).  These two websites contained 
#	data that had originally been compiled from other external sources.  Datasets and analysis 
#	techniques are further described in Rebich and Demcheck, 2007 (SIR 2007-5090).  
#
#--------------------------------------------------------------------------------------------------------
#  REPORT ARCHIVAL DATA SET:  MRB5_WLS_input.txt
#
#	This file contains input data used for weighted-least-squares regression (WLS) analyses of water
#	-quality trends and ancillary data trends as 
#	reported for streams in the Lower Mississippi, Arkansas-
#	White-Red, Texas-Gulf Basins presented in Rebich and Demcheck, 2007 (SIR 2007-5090). This 
#	investigation was completed as part of the USGS National Water-Quality Assessment Program.
#	WLS regression analysis were completed using standard routines devloped for the S-Plus Statistical
#	Software suite (Insightful Corporation, 2005, S-PLUS version 7.0 for Windows Professional
#	Developer: Seattle, Wash., Insightful Corporation). 
#	Trends in water-quality data are found in MRB5A_trend_results.txt.  Both significant and
#	non-significant trends were used in the WLS analyses using weights calculated as the inverse
#	of the variance of the trend results.  Trends in ancillary data are found in files 
#	MRB5_ancil_data.txt and MRB5_pop_dens.txt.  The file listed here is a text file and can be 
#	read directly into the SPLUS software.  WLS results were considered statistically
#	significant at a p-value < 0.05.  Resutls are published in Rebich and Demcheck 2007
#	(SIR 2007-5090).
#
#--------------------------------------------------------------------------------------------------------
#  REPORT ARCHIVAL DATA SET:  MRB5_WLS.results.txt
#
#	This file contains results for weighted-least-squares regression (WLS) analyses of water-quality 
#	trends and ancillary data trends as reported for streams in the Lower Mississippi, Arkansas-
#	White-Red, Texas-Gulf Basins presented in Rebich and Demcheck, 2007 (SIR 2007-5090). This 
#	investigation was completed as part of the USGS National Water-Quality Assessment Program.
#	WLS regression analysis were completed using standard routines devloped for the S-Plus Statistical
#	Software suite (Insightful Corporation, 2005, S-PLUS version 7.0 for Windows Professional
#	Developer: Seattle, Wash., Insightful Corporation). 
#	Trends in water-quality data are found in MRB5A_trend_results.txt.  Both significant and
#	non-significant trends were used in the WLS analyses using weights calculated as the inverse
#	of the variance of the trend results.  Trends in ancillary data are found in files 
#	MRB5_ancil_data.txt and MRB5_pop_dens.txt.  The file listed here is a text file and can be 
#	read directly into spreadsheet software.  WLS results were considered statistically
#	significant at a p-value < 0.05.  Resutls are published in Rebich and Demcheck 2007
#	(SIR 2007-5090).
#
#
#--------------------------------------------------------------------------------------------------------




