Physical Aquatic Habitat Assessment Data, Ozark Plateaus, Missouri and Arkansas

By Robert B. Jacobson, Harold E. Johnson, Joanna M. Reuter, and Maria Panfil Wright

U.S. Geological Survey Data Series DS - 94

Abstract

This report presents data from two related studies on physical habitat in small streams in the Ozark Plateaus Physiographic Province of Missouri and Arkansas. Seventy stream reaches and their contributing drainage basins were assessed using a physical habitat protocol designed to optimize understanding of how stream reach characteristics relate to drainage-basin characteristics. Drainage-basin characteristics were evaluated using geographic information system (GIS) techniques and datasets designed to evaluate the geologic, physiographic, and land-use characteristics of encompassing drainage basins. Reach characteristics were evaluated using a field-based geomorphology and habitat protocol. The data are intended to complement ecological studies on Ozark Plateaus streams.

Introduction

During 1998 – 2003 USGS geomorphologists were engaged in research studies to explore the extent to which stream physical habitat characteristics in the Ozark Plateaus (Ozarks, fig. 1) have been affected by land-use practices. Synoptic assessments can address the spatial relations between a stream and characteristics of its contributing drainage area and develop statistical measures of association. In the Ozarks, a region characterized by relatively high spatial variability and relatively low land-use stress, the measures of association are typically weak (Panfil and Jacobson, 2001). In addition to the spatial variability and low land-use stress, non-uniform and lagged historical effects of land uses on streams have been blamed for the lack of strong contemporary associations (see, Jacobson, in press; Jacobson and Primm, 1997; Jacobson and Gran, 1999).

Many protocols have been proposed for evaluation of physical stream geomorphology and physical habitat (for example, Bain and Stevenson, 1999). Physical habitats typically are evaluated for two related reasons. The first is to provide additional explanatory variables for assessments of stream ecology or water quality. The second is to explore relations between the stream and drainage-basin characteristics, usually with the intention of understanding land-use effects. The first approach is optimized if the habitat variables measured relate spatially and temporally to the specific biological or water-quality samples, both of which can be heavily dependent on the stream discharge at the time of sampling. In contrast, the second approach seeks to minimize the effects of discharge and measure variables that are longer-term integrators of geomorphic responses. The habitat protocol developed for this study is the latter type (Panfil and Jacobson, 2001).

Map of Ozark Plateaus and study drainage basins

Figure 1. Map showing distributions of study basins and reaches within physiographic divisions of the Ozark Plateaus Province.

This report is intended as the repository for data included in Panfil and Jacobson (2001) for tributaries to the Buffalo River, northern Arkansas and the Current River in southern Missouri (fig. 2A, B). Data are also presented for additional drainage basins in northern Arkansas, including some representative of forested areas in the Boston Mountains, and areas characterized by high percentages of cleared and agricultural land outside of the Buffalo River basin (fig. 2B). All habitat data were collected using the protocol defined in Panfil and Jacobson (2001).

Map of study drainage basins, Current River

Figure 2A. Drainage basins and reach-scale data collection locations for selected tributaries to the Current River, southern Missouri. Drainage basins with green location markers have only basin-characteristics data; red location markers indicate both basin and reach characteristics data (Panfil and Jacobson, 2001).

Map of study drainage basins, Buffalo River and vicinity

Figure 2B. Drainage basins and reach-scale data collection locations for selected tributaries to the Buffalo River and adjacent areas, northern Arkansas. MS, mainstem; L, lower; U, upper; M, middle; E., east.

Methods

Two types of data are presented here. Drainage-basin characterization data were derived from topographic, geologic, hydrographic, and land-use spatial data for individual study basins. The drainage basins were defined as the contributing drainage area upstream of the upstream end of the study reach. Drainage-basin characterization datasets, derivation of drainage-basin variables, and prioritization of variables are detailed in Panfil and Jacobson (2001). These variables consist of physiographic, topographic, geologic, and land-use variables (table 1).

Table 1. Drainage-basin variables, definitions, and data sources

[m, meters; m², square meters; km², square kilometers]


Drainage-Basin Variables		Definition		Data Source
Geology
Formation area, as a proportion	Area of each chronstratigraphic unit summed and divided by drainage area.			1:500,000-scale state geologic map of Missouri (Missouri Department of Natural Resources, 1991) modified 1:500,000-scale state geologic map of Arkansas (Hofer and others, 1995) Map was tiled from 1:24,000-scale and coarser resolution data. Cells were reclassified to match geologic categories on the statewide 1:500,000-scale geologic by Haley and others (1993).
Carbonate bedrock area, as a proportion	Formations regrouped by dominant lithology; area with carbonate bedrock summed and divided by drainage area.
Physiography
Drainage area (m² or km²)	Total area upstream of upper end of study reach; drainage basin boundaries delineated using an ArcView Spatial Analyst Script (http://gis.esri.com/arcscripts/details.cfm?CFGRIDKEY=951497255) and refined by comparison with elevation contours on USGS 1:24,000 digital raster graphics.			30-meter resolution digital elevation model, tiled from 1:24,000 USGS quadrangle sheets (U.S. Geological Survey, 2000a) and 1:24,000-scale digital raster graphics (U.S. Geological Survey, 1999)
Drainage-basin shape factor	Basin length squared divided by drainage area where basin length is the total length of a line bisecting the major river valley, from the upper end of the study reach to the drainage divide.
Elevation range (m)	Highest minus lowest elevation in study drainage basin.
Drainage-basin average slope (degrees)	Average slope for all grid cells within a study drainage basin where slope is calculated by comparison of each cell’s elevation to that of the surrounding eight cells.
Bluff area in stream buffer, as a proportion	Area of cells with slopes greater than 30 degrees within a stream buffer, divided by the buffer area. Buffers had graduated widths based on the Strahler stream order. First order streams had a buffer width of 25 m on each side of the stream. Width increased by an additional 25 m for each sequential stream order up to a maximum of 300 m.
Soils
No variables selected.			1:250,000-scale STATSGO soils coverage (U.S. Department of Agriculutre, 1994a, 1994b)
Stream Network
No variables selected.			1:100,000-scale rf3 river reach files (U.S. Environmental Protection Agency, 1998)

Land Cover
Cleared land area, as a proportion	Sum of area classified as developed, shrubland, transitional, herbaceous upland, or herbaceous cultivated (NLCD categories 33,51,71,81,82,83,84,85), divided by drainage area.	30-meter resolution National Land Cover Data (NLCD) (U.S. Geological Survey, 2000b) Coverage for the state of Arkansas was based on Landsat Thematic Mapper (TM) scenes taken from April 1988 through December 1993. Coverage for the state of Missouri was based on scenes taken from March 1988 through October 1993 (see Panfil and Jacobson, 2001, Appendix 1 for more details).
Steep, cleared land area, as a proportion	Cleared land area on slopes greater than seven degrees divided by drainage area (calculated by reclassifying and merging NLCD and slope grids).
Cleared land area in stream buffer, as a proportion	Cleared land area within stream buffers divided by total drainage area. See definition of bluff area for buffer explanation.
Road Network
Road density (m/m²)	Total road length within a basin divided by drainage area.	1:100,000-scale TIGER/Line files (U. S. Census Bureau, 1992)
Road density in stream buffer (m/m²)	Total road length within a stream buffer divided by buffer area. See definition of bluff area for buffer explanation.
Stewardship and Political
Private land area, as a proportion	Area outside of state, federal, or Nature Conservancy land management areas divided by drainage area.	1:100,000-scale stewardship boundaries (Center for Advanced Spatial Technologies, 1998) and (Missouri Resource Assessment Partnership, 1997)
Cities and towns on reference maps		1:2,000,000-scale city and town locations from the National Atlas (U.S. Geological Survey, 2000c )

Reach-scale data were collected in the field in reaches selected to be representative of the tributary. In the Current River (Missouri) study, reaches were selected at junctions between the tributary and mainstem Current River; reaches were delineated far enough upstream to avoid backwater effects from the mainstem. In the Buffalo River drainage basin and adjacent area, tributary reaches were complemented with satellite, headwater, out-of-basin, and mainstem reaches. At four tributaries in the Arkansas dataset, satellite reaches were chosen 1-3 km upstream of the tributary reach to evaluate mainstem effects on fish community structure. Ten headwater reaches (including one on the mainstem Buffalo River) were selected to sample small, steep, forested drainage basins in the Boston Mountains. Ten out-of-basin reaches were selected outside of the Buffalo River drainage basin to sample an increased range of agricultural land use. Nine additional reaches were selected on the Buffalo River mainstem to explore longitudinal effects. Reach selection was also constrained in part by access; most reaches are on Federal- or State-owned land. Reaches were delineated to include (whenever possible) a sequence of four riffles and three intervening pools (Figs. 3, 4). In all cases, reaches were at least 20 bankfull widths long. The reach-scale data include measures of channel morphology, longitudinal profile, sediment characteristics, and extent of erosion (table 2). The habitat protocol focuses on glide habitat units (fig. 4) for bankfull channel morphology and sediment characteristics in order to minimize within-reach variability. Details of the reach-scale habitat protocol can be found in Panfil and Jacobson (2001).

Figure 3. Two scales of data collection

Table 2. Reach-scale variables, definitions, and measurement techniques


Reach-scale Variable	Definition	Measurement Technique
Channel Geometry
Reach gradient	Slope of a best-fit line through water surface points surveyed along the thalweg.	Calculated from the geometry of the longitudinal profile survey, see Figure 4.
Total residual pool length (m)	Total length of reach within residual pools.
Residual pools, as a proportion	Total residual pool length divided by total reach length.
Average residual pool length (m)	Total residual pool length divided by the number of residual pools.
Average residual pool depth (m)	Residual pool area (measured along longitudinal profile) divided by total residual pool length.

Pools, as a proportion of reach length	Total reach length classified as lateral, bluff, mid-channel or obstruction pools divided by total reach length.	Calculated from visual identifications of habitat type made at each survey point along the longitudinal profile. See Figure 6 for habitat classification criteria.
Glides, as a proportion of reach length	Total reach length classified as glides divided by total reach length.
Obstruction pools, as a proportion of reach length	Total reach length classified as obstruction pools divided by total reach length classified as pools.

Average bankfull channel width (m)	Total distance across channel at bankfull elevation; average from 3-6 cross sections.	Calculated from the geometry of surveyed cross sections. Bankfull elevation was projected into cross sections from indicators identified throughout the study reach.
Average bankfull channel depth (m)	Bankfull channel area divided by bankfull channel width; average from 3-6 cross sections.

Substrate
Mud/sand along thalweg, as a proportion of reach length	Dominant particle size <2 mm; total reach length classified as mud/sand divided by total reach length.	Calculated from visual estimates of dominant particle size and embeddedness at each survey point along the longitudinal profile. Estimate made within a one meter diameter circle around the base of the surveyor’s stadia rod. Embeddedness reported as the proportion of the circle covered with mud or sand, in intervals of 0.1.
Gravel along thalweg, as a proportion of reach length	Dominant particle size 2-64 mm; total reach length classified as gravel divided by total reach length.
Cobbles and boulders along thalweg, as a proportion of reach length	Dominant particle size >64 mm; total reach length classified as cobbles/boulders divided by total reach length.

Thalweg embeddedness index	Summation of embeddedness class times the proportion of reach length within each embeddedness class.	Visual estimation

Glide D16 (mm)	16^th percentile of particle size distribution; average from three glides.	Calculated from cumulative particle size distributions from pebble counts of 100 particles.
Glide D50 (mm)	50^th percentile of particle size distribution; average from three glides.
Glide D84 (mm)	84^th percentile of particle size distribution; average from three glides.
Glide sorting (phi)	(D84 – D16)/4 + (D95-D5)/6); where particle sizes were transformed to phi (-log₂(diameter, mm)) and D84, D16, D95, and D5 are equal to 84^th , 16^th, 95^th, and 5^th percentiles of particle size distribution in glides.

Glide embeddedness, as a proportion	Average of embeddedness from two locations in each of three glides.	The proportion of a 60 cm quadrant covered with mud or sand, reported in intervals of 0.05.
Channel Stability
Bank vegetation index	Summation of vegetation class times the proportion of reach length within each embeddedness class; average of left and right banks.	Calculated from visual estimates made at each survey point along the longitudinal profile. Observations made of vertical banks below bankfull elevation.
Severely eroding banks, as a proportion of reach length	Total reach length classified as severely eroding divided by total reach length; average of left and right banks.
Moderately and severely eroding banks, as a proportion of reach length	Total reach length classified as moderately or severely eroding divided by total reach length; average of left and right banks.

Reach sinuosity	Total reach length divided by straight line distance between endpoints.	Calculated from planview of longitudinal profile survey.

Glide canopy cover	Average from densiometer readings at both ends of 3-6 cross sections.	Calculated from concave spherical densiometer readings near water’s edge on each cross section. Methodology followed Fitzpatrick and others (1998).

Figure 4. Scheme for data collection at the reach scale, showing planform, longitudinal profile, and hydraulic habitat units.

Results

Table 3. Drainage-basin and reach characteristic data file in Excel format

Table 3. Drainage-basin and reach characteristic data file in comma delimited format

Literature Cited

Bain, M. B., and N. J. Stevenson, 1999, Aquatic Habitat Assessment: Common Methods, American Fisheries Society Bethesda, MD, 216 pp.

Center for Advanced Spatial Technologies, 1998, AR-GAP Land Stewardship of Arkansas, University of Arkansas: http://www.cast.uark.edu/gap/ .

Fitzpatrick, F.A., Waite, I.R., D’Arconte, P.J., Meador, M.R., Maupin, M.A., and Gurtz, M.E., 1998, Revised Methods for Characterizing Stream Habitat in the National Water-Quality Assessment Program; U.S. Geological Survey Water-Resources Investigations Report 98-4052, 67 pages.

Haley, B.R., Glick, E.E., Bush, W.V., Clardy, B.F., Stone, C.G., Woodward, M.B., Zachary, D.L., 1993, Geologic map of Arkansas: U.S. Geological Survey and Arkansas Geological Commission, map revised from 1976 version

Hofer, K.R. Scott, H.D., and McKimmey, J.M., 1995, Spatial distribution of the surface geology and 1992 land use of the Buffalo River Watershed: Arkansas Water Resources Center Publication, no. 174, pg.43.

Jacobson, R.B., in press, Watershed Sustainability: Downstream Effects of Timber Harvest in the Ozarks of Missouri, in, Flader, S.J., ed., Toward Sustainability for Missouri Forests, USDA Forest Service North Central Experiment Station General Technical Publication. 27 ms. pages.

Jacobson, R.B. and Gran, K.B., 1999, Gravel routing from widespread, low-intensity landscape disturbance, Current River Basin, Missouri: Earth Surface Processes and Landforms, v. 24, 897-917.

Jacobson, R.B. and Primm, A.T., 1997, Historical land-use changes and potential effects on stream disturbance in the Ozark Plateaus, Missouri: U.S. Geological Survey Water-Supply Paper 2484, 85 p.

Missouri Department of Natural Resources, 1991, Geologic map of Missouri, 1:500,000-scale: http://msdisweb.missouri.edu/datasearch/metadata/utm/st_geol_utm.xml.

Missouri Resource Assessment Partnership, 1997, Stewardship: http://www.cerc.cr.usgs.gov/morap/projects.asp.

Panfil, M.S., and Jacobson, R.B., 2001, Relations among geology, physiography, land use, and stream habitat conditions in the Buffalo and Current River systems, Missouri and Arkansas: U.S. Geological Survey Biological Sciences Report, USGS/BRD/BSR-2001-0005, 111 p. On CD-ROM or online at: http://www.cerc.usgs.gov/pubs/center/pdfDocs/bsr2001-0005.pdf.

U.S. Census Bureau, 1992, TIGER/Line files for the Continental United States, scale 1:100,000: http://www.census.gov/mp/www/rom/msrom12f.html.

U.S. Department of Agriculture, 1994a, State Soil Geographic (STATSGO) data base for Missouri, scale 1:250,000: http://www.ncgc.nrcs.usda.gov/branch/ssb/products/statsgo/index.html.

U.S. Department of Agriculture, 1994b, State Soil Geographic (STATSGO) data base for Arkansas, scale 1:250,000: http://www.ncgc.nrcs.usda.gov/branch/ssb/products/statsgo/index.html.

U.S. Environmental Protection Agency, 1998, USEPA/OW River Reach File 3 for the Continental United States, scale 1:100,000: http://www.epa.gov/waters/doc/rfindex.html .

U.S. Geological Survey, 1999, Digital Raster Graphics, Fact Sheet 070-99: http://mac.usgs.gov/mac/isb/pubs/factsheets/fs08801.html.

U.S. Geological Survey, 2000a, US GeoData Digital Elevation Models, Fact Sheet 040-00: http://mapping.usgs.gov/mac/isb/pubs/factsheets/fs04000.html.

U.S. Geological Survey, 2000b, National Land Cover Data, 30-meter resolution: http://edcsgs9.cr.usgs.gov/programs/lccp/nationallandcover.html.

U.S. Geological Survey, 2000c; National Atlas: http://www.nationalatlas.gov/atlasftp.html.

Webpage content contact: Robert Jacobson. Webmaster: Chris Henke