Habitat Conditions in the Gasconade River Watershed
Historically, the Army Corps of Engineers (COE) maintained the Gasconade River for navigation from the mouth of the river to Jerome, Missouri, or approximately 104 miles of stream (Missouri Department of Natural Resources 1986). According to the Missouri Water Atlas (1986), the Gasconade River has no altered segments, meaning it has been neither channelized nor impounded. The Osage Fork of the Gasconade River is also listed as having no altered segments.
Accelerated stream channel changes are possible consequences of in-channel sand and gravel mining. In-channel mining has the potential to artificially accelerate a stream'snatural geomorphic processes by increasing channel slope, water velocity, and sedimentation. A stable stream is in dynamic equilibrium. Gravel improperly removed from a streambed location can result in stream disequilibrium by causing erosion upstream from the nick-point (removal area) and within the nick-point. As the stream seeks new mass-balance equilibrium, the nick-point will eventually erode away and migrate upstream in a process known as "head-cutting" (Patrick, D.M. et al. 1993).
Segments of the Gasconade River and some of its tributaries have been altered by gravel mining activity. Army Corps of Engineer'sRegulatory Analysis Management System database, which encompasses the entire Gasconade River watershed, contained 1-25 permits per 11-digit hydrologic unit (COE 1999). The number of sand and gravel site permits was determined for the period of February 1992 - February 1999 (Figure Lu04, Land Use Chapter). The 8-digit Lower Gasconade River watershed with approximately 500,000 acres had higher densities of permitted sand and gravel sites than the 8-digit Upper Gasconade River watershed with its more than one million acres. Lower gradient and corresponding slower water velocities in the Lower Gasconade River watershed allows more gravel to accumulate in the form of gravel bars, which contributes to the instability of the channel. As demonstrated in the Bourbeuse River Watershed Inventory and Assessment (Blanc 1999), land use and stream reach position in a watershed can influence channel instability; likewise, gravel mining can lead to further stream channel instability. In addition, improperly mined areas may experience side effects that may incur a reclamation liability.
Stream Gravel Mining Recommendations
The MDNR'sLand Reclamation Program strongly encourages that commercial instream gravel miners conduct mining in accordance with the Missouri Department of Conservation's Stream Gravel Mining Removal Guidelines (Missouri Department of Conservation 1991c). These guidelines give general operational recommendations on how, where, and when instream gravel mining should be conducted in order to minimize effects on habitat and biota.
Some essential elements include confining active mining to unconsolidated bars rather than flowing water, leaving buffers around mined areas, restricting damage to streambanks and bank vegetation, preventing the discharge of petroleum products into water. Another operational guideline states that gravel miners should not remove gravel during certain times in several designated reaches to avoid effects on spawning habitats. Prohibiting instream gravel mining seasonally to protect critical spawning habitat is usually incorporated into permits issued by the Army Corps of Engineers for restrictions regulated under Section 404 of the Clean Water Act or by MDNR under Section 401 of the Clean Water Act. Specifically within the Gasconade River watershed, MDC recommends gravel miners observe seasonal spawning closures within the following times and streams:
November 15 to February 15
•Little Piney Creek, from the mouth of Beaver Creek to the first crossing of Dent County line (Section (S) 16, Township (T) 35N, Range (R) 8W) for 15.8 miles to protect trout spawning habitat,
•Mill Creek, from the mouth (S20, T37N, R9W) to the mouth of Deep Hollow Creek (S32, T36N, R9W) for 9.0 miles to protect trout spawning habitat,
•Roubidoux Creek, from the mouth (S14, T36N, R12W) to East Section Line (S6, T35N, R11W) to protect a MDC trout management area; and
March 15 to June 15
•Roubidoux Creek, from the south section line (S3, T34N, R12W) to Highway 32 (S2, T32N, R12W) for 20.2 miles, to protect critical habitat of sensitive endemic aquatic species.
The Missouri Natural Features Inventories are completed for Phelps, Laclede, Pulaski (Ryan 1992), Gasconade, Maries (Currier 1991), Texas, and Wright (Ryan 1993) counties. The objective of the MDC statewide Natural Features Inventory objective was to locate, describe, classify, and rank high quality elements of Missouri's natural habitat. With this knowledge, Missourians protect the state'soutstanding features through inclusion in the state natural-areas system, by voluntary landowner agreements, or by allowing informed decisions in sensitive areas.
Within counties of the Gasconade River watershed, identifying sites and adjacent areas involved surveying seven categories: natural communities (undisturbed assemblages of plants and animals), state-listed species habitats (rare and endangered species), habitats of relict species, outstanding geologic features, areas for nature studies, other unique features, and aquatic communities. The natural community, geologic feature, and aquatic community sites were further classified using the Terrestrial Natural Communities of Missouri (Nelson 1987), the Geologic Natural Feature Classification System for Missouri (Hebrank 1989), and the Aquatic Community Classification System for Missouri (Pflieger 1989). Following the classification, biologists graded sites for their natural quality, and ranked sites to provide a means of comparing similar features for the preservation value (Currier 1991; Ryan 1992, 1993). Ranking assignments were: significant, exceptional, and notable. According to Ryan (1992, 1993) and Currier (1991), areas that he defined as significant natural features should receive a form of protection (possible inclusion in the Missouri natural areas system), and areas that he defined as exceptional were not of natural area quality but deserving of some protection. Lastly, notable areas on private land did not merit special management or protection.
The focus of this inventory was to identify high-quality natural communities. In the Currier (1991) survey, Spring Creek Gap Glades Natural Area, owned by MDC and located within Spring Creek Gap Conservation Area (Figure Lu07, Land Use Chapter), was ranked as significant. The site is 12 acres but is considered the best glade system on Jefferson City Cotter dolomite in the upper Ozarks. The Clifty Creek Natural Area, located within the Clifty Creek Conservation Area, contains exceptional limestone and dolomite cliffs, notable sandstone forests, and a rare dry-mesic chert forest. The entire natural area is ranked as significant.
A total of 14 significant natural communities was identified in the Ryan (1992) survey. In fact, several of these communities were found within a few miles of each other. The US Forest Service (USFS) owns two sandstone glades that are located in Phelps County. The first sandstone glade is a string of glades within close proximity of each other. The second glade, the Kaintuck Hollow sandstone glade, is about 2.5 acres and is near several rare species, an unique forest, and a deep muck fen (Table Hc01). Contained near this site is an exceptional 15-acre dry-mesic sandstone forests with 100-year-old pines. The deep muck fen, Kaintuck Hollow Fen, is about 10 acres in size but is low quality. The largest of these communities, a mesic bottomland forest, is found on private land and is 30 acres.
Aquatic communities were ranked based on recommendation from William Pflieger of Missouri Department of Conservation (Currier 1991). Currier (1991) commented that the Gasconade River is one of the few unimpounded rivers in the Ozarks and is one of only three rivers in the Mississippi Valley where the anadromous Alabama shad still spawns. Currier (1989) surveyed the Webster County portion of the Osage Fork of the Gasconade River and ranked it as a significant Ozark-Missouri headwater creek and small river that supports numerous sensitive species. In Ryan's1992 survey of Laclede, Phelps, and Pulaski counties, Little Piney Creek was identified as a notable creek and small river, and Gasconade River and Osage Fork of the Gasconade River were identified as significant small rivers. Multiple sections of the Gasconade River from T37N to T38N, R9W to R10W were listed as a significant large river. In the Ryan (1991) survey of Texas County, Ryan mentioned Roubidoux Creek (T31N-T33N, R11W-R12W) as a significant small creek and headwater Ozark-Missouri stream, supporting a diverse fish fauna. In the same survey of Webster County, once again the Gasconade River was mentioned as a significant large river with numerous endemic fish species. Other portions of the Gasconade River watershed were described in the Natural Features Inventories, some mentioning heron rookeries and others, backwater pools.
Ryan (1992) described other special aquatic communities in the survey. Pulaski County has a large number of springs (Figure Hy01, Hydrology Chapter) and caves. These include two spring branches, Howell Spring and Prewett Spring, and two springs, Boiling Spring and Roubidoux Spring. Ryan noted no disturbance at Howell Spring and moderate plant diversity, but Prewett Spring was grazed. The solution cave was an added feature of the Howell Spring community. Another cave that has a small population of Myotis sodalis is Great Spirit Cave, owned by Missouri Department of Conservation. A slough in Pulaski County was described as having 0.6 miles of the old river channel, cliffs, a spring, and wooded streambanks.
One purpose of these surveys was to rank bottomland forests within the respective counties. No bottomland forests were surveyed in the Gasconade River watershed portion of Texas and Wright counties or Webster County, but in the Gasconade River watershed portion, bottomland forests were surveyed in Gasconade, Maries, and Osage counties, three in each county. Only eight bottomland forests were surveyed in the Laclede County, Phelps County, and Pulaski County Natural Features Inventory. Most bottomland forests were mesic bottomland forest of young to mature second growth and not high quality. Within the Gasconade, Maries and Osage county surveys these bottomland forests were mid-successional and moderately to heavily disturbed communities, which may have included moderate recent disturbance or heavy past disturbance. The sizes ranged from seven to 40 acres. Within the Laclede, Phelps, and Pulaski county surveys, three of the eight survey sites were lightly or heavily grazed. Three of the eight survey sites were recently logged. One of the eight survey sites was the Strawhaun Bottomland Forest that was ranked as significant.
The habitats of some state-listed species are found on sites within the watershed. These sites are identified as information becomes available. Individual state-listed species that located in the watershed are identified in the Rare and Endangered Species subsection.
To control streambank erosion, improve water quality, and establish fish habitat, MDC fisheries biologists use cedar tree revetments, corridor reforestation, streambank re-vegetation, willow staking, and rock blankets (riprap). Table Hc02 lists some projects in the Gasconade River watershed that make use of these techniques. Eight of the stream improvement sites used the cedar tree revetment technique to stabilize streambank. Cedar tree revetment involves the use of eastern red cedar trees anchored along a streambank to protect the toe of the bank and to slow water velocity (Fantz et al. 1993). This low cost bank stabilization technique involves using refined methodology that must be tailored to the erosion site. Not all streambank erosion sites are conducive to cedar tree revetments because of watershed influences. The numerous projects on Mill Creek make use of a variety of techniques to stabilize streambanks and provide instream habitat on Forest Service as well as private land.
Corridor improvements are an important part of streambank erosion reduction and fish habitat enhancement. The future ecological benefits to the aquatic community are reduced sediment supply, shade from the sun, temperature reductions, and leaf litter inputs for the aquatic food web.
Stream Habitat Assessment
Using Arc/Info (Environmental Systems Research Institute'sGeographic Information Systems software), the Gasconade River 1:100,000 scale stream network, and Missouri Resource Assessment Partnership's(MORAP 1997) Phase II Land Use / Land Cover (LULC) individual stream segments were classified by the percentages of surrounding land use types (for GIS methodology contact the Missouri Department of Conservation). For example, stream segments were classified by the ranges of the percentage of the forest class contained within the stream buffer area to identify those segments that had the highest probability of direct exposure of forest to stream a channel (Figure Hc01 and Hc03). To highlight the forested corridors within the Lower Gasconade River watershed, the lowest and the highest percentage of forested corridors was, respectively, the Lower Gasconade 11-digit HU at 48.8% and the 11-digit Lower Gasconade River Hills at 55.3%. Values within the Lower Gasconade River HU were probably somewhat higher than the 20-40% forested corridor presented, because within this 90-meter buffer the 8-digit Lower Gasconade River watershed had approximately 5% of the pixels as water (Figure Hc01). LULC satellite imagery was dated 1992-93 during years of high water, which likely influenced the resulting forested segments.
The entire 8-digit Upper Gasconade River watershed was poorly forested along major segments of its tributaries and main stem compared to the 8-digit Lower Gasconade River watershed (Table Hc03 and Figure Hc03). A total of 38.2% of the major segments (main stem river and tributaries segments with permanent flow) within the Upper Gasconade had forested corridors, and 46.1% of the major segments in the Lower Gasconade supported forested corridors. To highlight the forested corridors in the Upper Gasconade River watershed, the lowest and highest percentages were the 11-digit Upper Osage River HU at 38.5% forested corridor and the 11-digit Upper Gasconade River HU at 48.9% forested corridor, which was a spread of 10.4% (Table Hc03). The Mark Twain National Forest influences the quantity of forested corridors within the 11-digit Upper Gasconade River HU. In reality, its corridor quality was good in comparison to other watersheds (Figure Hc03).
Corridor quality was determined to assess the stream segments within 11-digit hydrologic units (Figure Hc02 and Figure Hc04). While many factors impact the quality of the corridor, stream channel stability, and water quality, forest and woodland land uses improve stream quality because of their soil holding capacity, where as grassland, cropland, and urban land uses do not improve stream water quality (Jacobson and Primm 1994; Blanc, Caldwell, and Hawks 1998). Grassland, cropland, and urban areas are known to have higher soil erosion and runoff rates. To determine where land uses were influencing corridor quality, the following ratio was developed:
(% grassland and cropland and urban)/(% forest and woodland).
The quality of a stream corridor varied as the sum of the percentage of forest and woodland changed with respect to the changes in the sum of grassland, cropland, and urban. As values of the numerator increased and the denominator decreased, the quality of the corridor within the buffer zone declined. These areas were shown as the "poor" ratio values from 5-100%. Conversely, within a 90-meter buffer zone, as grassland and cropland declined and forest corridor increased, this translated into better quality corridors. More poor quality stream segments were found in the tributaries to major order segments. An "acceptable" corridor had, depending on the stream order, 15-35 meters of corridor (Wehnes 1996), which was approximately 17-40 % or greater forest and woodland within the buffer zone. Hence, the quality ratio of "acceptable" had to be within the range of 1.5-5. A quality ratio value of 0.0-1.49 had better corridor conditions and were rated as "good."
The results of the quality ratio show the differences between the 8-digit Upper Gasconade River watershed and the 8-digit Lower Gasconade River watershed. Using the limitations of the 1:100,000 scale stream network, which did not have many 1st- and 2nd-order streams, stream segments within the Lower Gasconade River watershed had 81% (6,752) as good (quality ratio range of 0.0-1.49), another 12.5% (1,041) as acceptable, and the remaining 0.6% (526) of the segments as poor. There were 8,319 stream segments within the Lower Gasconade River watershed and 14,404 stream segments within the Upper Gasconade River watershed that had an average length of 361 meters. The Lower Gasconade River watershed had more good quality segments than the Upper Gasconade River watershed, which had 63.8% (9,199) rated as good, 17.5% (2,518) as acceptable, and the remaining 18.7% (2,687) segments as poor.
Several 11-digit hydrologic units could be targeted for private lands incentive programs. Lower Gasconade River HU below Highway 68 Bridge and the confluence with Spring Creek could be targeted for stream incentive programs (Figure Hc02). The upper portion of the main stem Little Piney Creek HU has much cropland and grassland that should be surveyed for possible restoration. Third Creek HU has some troubled tributaries that need attention. Second Creek within the Lower Gasconade River Hills has stream segments near the confluence with the main stem Gasconade River that may need attention. Roubidoux Creek HU has in the past received attention but does merit further emphasis because of its unique combination of land uses (Figure Hc04). The water quality challenges within this watershed were identified by Imes et al. (1996) in the USGS water quality assessment of the Fort Leonard Wood military base. Groundwater resources are particularly sensitive in this region of the Upper Gasconade River watershed. Within the Upper Gasconade River Tributaries HU, Whetstone Creek and Woods Fork had stream segments with extreme amounts of grassland land uses. Forested corridor was limited in selected portions of both Whetstone Creek and Woods Fork (Figure Hc04).
Identifying other degraded or healthy streams, narrowing the list of potential causes of degradation within stream segments, and selecting the most pristine or degraded reaches will be done interactively within ArcView by MDC East Central Region personnel. Measures to be taken by personnel within the Gasconade River watershed to improve riparian corridors include offering financial assistance to help landowners fence cattle from riparian corridors and re-vegetate riparian zones. Studies have shown that fencing cattle from a stream and its riparian corridor can reduce soil losses (Ownes et al. 1996, Magilligan and McDowell 1997). Researchers observed two consistent stream channel changes with the restoration of riparian corridor: a decrease in channel widths and development of more channel pools. The researchers concluded that the regrowth of streambank vegetation added stream channel roughness, which increased channel scour holes or pools during floods.
Land Use Conditions
Using Arc/Info Geographic Information Systems (GIS), MORAP Phase II Land Use / Land Cover (LULC) Classification and the Gasconade River watershed boundaries were combined (for GIS methodology contact the Missouri Department of Conservation). A rating system was developed to determine the overall impact of land uses to the each hydrologic unit within the Upper and Lower Gasconade River watersheds. Beneficial to stream health were the forest and woodland classes, because watershed roughness components from vegetative land cover were a vital part of the stream'serosion protection and the water filtering capacity. Also, the forest and the woodland classes were land uses that were positively correlated with biotic integrity (Wang et al. 1997). These percentages were added to make another field called percentage of forest and woodland. Other classes such as urban and cropland tend to have detrimental effects on stream habitat and water quality. Likewise, these percentages were added to make an additional field called percentage of urban and cropland. These combined percentages are negatively correlated with biotic integrity (Wang et al. 1997). The urban and cropland land uses were subtracted from the percentage of forest and woodland to obtain a third field, called impacted. Working with the resulting range of values, the highest value was given a value of "100" and the lowest, a value of "zero." The value of "zero" represented the most impacted area and the value of "100," the least impacted. A range of rating values was developed from this third field range, impacted, and subsequently assigned to the remaining impacted values (Table Hc04).
Hydrologic Unit LULC Ratings
Within Upper and Lower Gasconade River watersheds, the percentage of forest and woodland and percentage of urban and cropland for each 14-digit hydrologic unit provided a means of comparing among HUs (Table Hc04). The three highest ratings and three lowest ratings were compiled in Table Hc04for each 8-digit watershed.
These ratings provided a useful means of assessing the watershed and gave insight to potential problem areas to be better managed with the best available practices. Hydrologic units that have poor ratings can be earmarked for further investigation, and landowners within these units targeted for possible landowner incentive programs.
Within the Upper Gasconade River watershed (Table Hc05), averaging all 14-digit HUs within each 11-digit HU indicated that the Middle Gasconade River HU had the highest mean value of 85.91. The Roubidoux Creek HU had the lowest average rating because two of its 14-digit HUs had relatively low ratings. However, the Roubidoux Creek 11-digit HU had the third highest rating 14-digit HU within the 8-digit Upper Gasconade River watershed (Table Hc05; 95.9). Sections of the upper Roubidoux were within the Mark Twain National Forest and the private holdings were forest or woodland land use, which explains the higher 14-digit rating. The Upper Osage Fork had a fairly low rating that may merit attention given its present status as a NRCS Conservation Priority Area to target water quality problems (Missouri Unified Watershed Assessment Steering Committee 1998). While the Upper Gasconade River Tributaries HU rates as a relatively pristine environment due to the presence of the Mark Twain National Forest, the Upper Gasconade River HU was more impacted. In fact, areas within the 11-digit Upper Gasconade River HU are NRCS Conservation Priority Areas (Missouri Unified Watershed Assessment Steering Committee 1998). Within the Lower Gasconade River watershed (Table Hc06), averaging all 14-digit HUs within each 11-digit HU indicated that the Little Piney Creek HU had the highest mean rating of 70.10. The lowest mean value of 48.11 was found in the Lower Gasconade River HU.
Based on this analysis, priority for improvement should be given to those hydrologic units that were rated low. The Lower Gasconade River HU (#10290203-020) was rated poor due to the lack of forested stream corridor (Table Hc06). But the present land use information may have under-represented the amount of forest in that HU, however no other information is available. A cross referencing with helicopter videos (Missouri Department of Conservation 1993) of the Lower Gasconade HU, filmed from the confluence with Little Piney Creek down Paydown Access on the Gasconade River, showed that, in general, the corridor varied from forested areas intermixed with pastured areas to one or two rows of trees progressing toward Paydown Access. These narrow corridors may not have been detected by image analysis. Still, the results showed that relative to other HUs, the Lower Gasconade River HU remained in poorer condition. An additional HU, the Lower Roubidoux Creek HU, should be given priority management attention because of its sensitive springs and fisheries (Figure Hy01, Hydrology Chapter) and the presence of a growing human population (Figure Lu04, Land Use Chapter).
Erosion and Deposition
Contributions of woody vegetation to streambank stability and to stream energy dissipation have been supported by researchers (McKenney, Jacobson, and Wertheimer 1995). Woody vegetation imparts overall strength to the streambed and streambank and greater erosion resistance, and as a result, greater channel stability. Based on this information, land and stream managers have advocated increased stream corridor widths and densities of streamside vegetation to decrease streambank erosion (Missouri Department of Conservation 1997; Reno, Pulliam, and Priesendorf 1995; Roell 1994). Recent photogrammetric/GIS studies on Little Piney Creek (a 12-kilometer 5th-order segment extending from, approximately, Yancy Mills to Hickory Point) have determined that the benefits derived from vegetation in the maintenance and recovery of stream channels were influenced by watershedwide factors and land cover and land use characteristics of individual reaches (Jacobson and Pugh 1997). In this GIS analysis, Jacobson and Pugh assumed that woodland had a greater chance of being eroded than grassland/cropland, which were positioned farther away from the stream channel. To determine erosion and deposition susceptibility of the Little Piney Creek study segment, Jacobson and Pugh performed calculations in a digital GIS format using a polygon identity map (intersection of two maps) from each pair of successive maps, i.e., transition periods 1938-48, 1948-55, 1955-64, 1964-76, and 1976-89. Jacobson and Pugh (1997) concluded that the results of their GIS analysis were applicable to other 4th - 6th order Ozark streams with similar physiographic controls and land use histories.
Evidence presented in Jacobson and Primm (1994) supports the theory that streams were destabilized by historic land-use practices and their present state of instability is the result of decreased riparian vegetation. The results of the GIS analysis performed by Jacobson and Pugh (1997) indicate that erosion or deposition susceptibilities are not solely controlled by riparian vegetation. Reaches are susceptible to disturbance by mechanisms such as valley wall geometry, bank height greater than root depth, upstream changes, and sediment size changes, that are quite complex. Finally, Jacobson and Pugh (1997) believe that before a biologist attempts a stream improvement project, as listed in Table Hc02, he or she should have additional information on disturbance history, streambank soil cohesion, channel gradient, and if possible, runoff rates and stream bed load.
National Wetland Inventory
The U.S. Fish and Wildlife National Wetland Inventory (NWI) data for the Gasconade River watershed was summarized for like wetland polygons within each hydrologic unit. To interpret the NWI coding system, several sources were used. Translating the wetland types from the Cowardin System (Cowardin et al. 1979) to the Missouri Wetland system was done with the aid of the Epperson (1992) (Table Hc07). A database containing the all polygonal (the cartographic representation of a wetland'sgeometry) wetland types, identified using the Cowardin System code, in the Gasconade River watershed was translated into systems, subsystems or classes, modifiers, and descriptions (for GIS methodology contact the Missouri Department of Conservation).
The NWI dataset is the most detailed information available for water bodies. The existing 1:100,000 scale water body file, extracted from the USGS digital line graph files (DLGs), had fewer water bodies represented. The NWI database had a total of 17,795 polygonal wetlands in the Upper Gasconade River watershed and 8,071 polygonal wetlands in the Lower Gasconade River watershed. Percentage of total wetland acres for each wetland system/class and description within the each 8-digit Gasconade River watershed illustrates the distribution of wetland types over the changing topography (Table Hc08).
Several Riverine subsystems and classes are describing temporary-semipermanent pool within the river. Pools are important for fish population growth and production. In the Upper Gasconade River watershed, 13.1% of the total polygonal wetland acreage (TPWA) are temporary-semipermanent pools (Table Hc08). Riverine wetlands comprise 54.8% of total wetland acreage in the Upper Gasconade River, representing the largest wetland system. A total of 30% of the TPWA is the pool/riffle complex in the Upper Gasconade River watershed. Because this watershed is the headwater of the Gasconade River, 0.36% is permanent pool.
The predominant wetland types in the Lower Gasconade River watershed are Palustrine. Palustrine wetlands represent the largest wetland system with 43% of the TPWA (Table Hc08). The lower gradient and larger order stream system has setup conditions for more Palustrine wetlands. The lower watershed has a large percentage (42.4%) of deciduous bottomland forests. Many of these bottomland forests are temporarily or seasonally flooded, which makes them unavailable to cropland conversion without substantial diking. However, many of the wetland polygons have special modifiers identifying them as drained, diked, impounded, or excavated. These modified bottomland areas may be providing a buffer from flooding for cropland that is farther upland. A very large percentage of the total acreage of wetlands in Gasconade River watershed are farm ponds (NWI code: PUBGh, PUBFh), 33.4% for the upper watershed and 28.2% for the lower watershed.
Mark Caldwell of MDC Fisheries Research used NWI data to identify potential nursery wetlands for fish in the Meramec River Watershed Inventory and Assessment (Blanc, Caldwell, and Hawks 1998). Using the criteria that the classes had to be Palustrine (non-channel, non-lake, perennial, or nearly perennial, and be a natural wetland, i.e., not excavated or impounded) and connected to a perennial stream, he identified polygonal wetlands that had potential to function as fish nursery habitat. A similar procedure was used for the Gasconade River watershed (for GIS methodology contact the Missouri Department of Conservation).
Overall, total nursery wetland acreage was 107.8 acres for the Upper Gasconade River watershed and 43.8 acres for the Lower Gasconade River watershed. Of the total wetland acreage within the Upper Gasconade River watershed, 0.9% met the nursery wetland criteria, and within the Lower Gasconade River watershed another 0.6% met the criteria. Connectivity to streams was not tested.
Habitat for fish, especially smallmouth bass, is best where there is good pool development. Using the NWI data sets, several classes of the Riverine system identify stream reaches that have suitable fish habitat. Several Riverine system classes were summarized into the groups that help interpret the Cowardin System code: temporary-semipermanent pool, temporary pool, pool/riffle complex, and permanent pool (Table Bc01, Biotic Community Chapter). For instance, grouped into the temporary-semipermanent pool description, the R3UBF, R3UBG, and R2UBG attributed polygons are upper (3) and lower (2) perennial stream segments with 25% of particles smaller than stones, vegetative cover less that 30%, and unstable bottoms that can be sand, mud, gravel or organic materials. As described by Cowardin, L. M. et al. (1979), modifiers F or G describe these habitats as semipermanent flooded or intermittently exposed. These water regime modifiers are an important feature of the wetland classification because they indicate the hydrologic characteristics of the wetland. Indicating how long water stands in the habitat, a Riverine system habitat can be classified within the range temporarily flooded to saturated to intermittently exposed to permanently flooded.
Having a more rugged topography and higher gradient than the Lower Gasconade River watershed, the Upper Gasconade River watershed has 2.56% of the total polygonal wetland acreage as temporary pool, and the Lower Gasconade River watershed has 0.11% as temporary pool (Table Hc08). This subsystem is intermittent and contains flowing water only part of the year (Cowardin et al. 1979). Intermittently, habitat classified within this subsystem will have temporary pools for extended periods. This habitat type is found within the Upper Osage Fork HU (0.06% TPWA), Roubidoux Creek HU (22% TPWA), and Little Piney Creek HU (0.48% TPWA) of the Upper Gasconade River watershed. To a lesser extent, some tributaries to Little Piney Creek (0.48% TPWA), Lower Gasconade River Hills HU (0.08% TPWA), and Lower Gasconade River Tributaries (0.06% TPWA) have temporary pool habitat. Poor in terms of sport fish habitat, these wetland types are likely to have, at a minimum, frogs and a few non-game fish species.
More acres of permanent habitat types, like the temporary-semipermanent pool, are found within the Upper Gasconade River watershed (13.13%) and fewer acres in the Lower Gasconade River watershed (4.42%). Density of streams is much greater in the upper watershed areas as compared to the lower, which explains the large difference in percentages. All hydrologic units have this habitat type. It represents the largest Riverine habitat. When its percentage of the TPWA is small, a habitat with a permanent water regime is present. The 8-digit Upper Gasconade River watershed had numerous acres of the temporary-semipermanent Riverine pool habitat types that are replaced by a permanent water regime in the form of very long pool/riffe complexes (Table Hc08). In contrast, the 8-digit Lower Gasconade River watershed had more permanent pools. The largest expanse of the pool/riffle complex habitat was found in the Middle Gasconade River HU (67.4%). Lower gradient and many tributaries lend to the development of this habitat. Somewhat lower in gradient than the Upper Osage Fork HU, the Lower Osage Fork HU had 32.1% of TPWA as pool/riffle complex Riverine habitat. Fifty-seven acres of permanent pool or 6.26% of TPWA were found in the Little Piney Creek HU. While lower perennial streams of the Lower Gasconade River Hills HU made up 84 acres of the 97 total temporary-semipermanent pool acres, their water regime was not classified as permanent but rather intermittently exposed. Permanent pool became more a feature of the Lower Gasconade River HU with 0.16% of the TPWA.
Channel condition of streams within hydrologic units of the Gasconade River watershed was characterized by evaluating the gravel bar status. The total acreage of gravel bars can be a good indicator of overall watershed and stream channel health. Channel stability, as well as fish habitat, is influenced by a variety of factors, such as bed load and gradient.
Channel condition may be poorest in those HUs with a high percentage of gravel bar acres per HU acres. R2USA (Riverine, Lower Perennial Unconsolidated Shore, Unaltered wetland type) and R3USA (Riverine, Upper Perennial Unconsolidated Shore, Unaltered wetland type) were represented as gravel bars in the summarization of polygon acreage (Table Hc09). To compare the quantity of gravel bars between 11-digit hydrologic units (HUs), the total 11-digit unit acreage was used to normalize the gravel bar acreage within each HU. The percentage of gravel bar acres per HU area was highest in three of the eleven 11-digit hydrologic units, Upper Gasconade River Tributaries at 0.149%, Third Creek at 0.149%, and Upper Osage Fork at 0.128%, respectively. Third Creek has the smallest HU area of all 11-digit units but had the highest percentage of gravel bar acres relative to its small size. Upper Gasconade River Tributaries has the fourth smallest HU area and the Upper Osage Fork HU, the third smallest HU area. Other HUs that were larger in size, such as the Upper Gasconade River HU and the Little Piney Creek HU, were low in gravel bar acres, and the Little Piney Creek HU was the lowest in the total gravel bar acreage.
Sources of gravel may not have been from with the tributaries of each HU but from upstream adjoining HUs. The presence of a forested corridor (Table Hc03) may have contributed to slower water velocity and subsequent bed load deposition. To illustrate, a continuum of gravel bars along the main stem Gasconade River indicated decreases in percentage of gravel bar acres/HU area from the Upper Gasconade River HU (#10290201-010) to the Lower Gasconade River Hills HU (#10290203-040). Within five main stem HUs starting with the Upper Gasconade River HU and ending with the Lower Gasconade River Hills HU, the percentage of gravel bar acres per HU area was 0.041, 0.149, 0.088, 0.081, and 0.077. The total gravel bar acreage in the Upper Gasconade River HU was only 61.2 acres, which was considerably less than those HUs along the river continuum such as the following Upper Gasconade River Tributaries HU with 224.3 acres and the adjoining Middle Gasconade River HU with 136.5 acres. The Upper Gasconade River HU had more land uses that contribute to sediment loading and streambank erosion (see Hydrologic Unit LULC Ratings, Table Hc04), while the Upper Gasconade River Tributaries HU had mostly forest and woodland land uses ? beneficial to streams. The forested corridor areas of the Upper Gasconade River Tributaries may have slowed water velocity and allowed gravel deposition. This suggests that the source of much of the sediment loading to Upper Gasconade River Tributaries HU (LULC rating of 85.9) may be areas within the Upper Gasconade River HU (LULC rating of 59.4), the largest HU of the eleven 11-digit HUs, and possibly from an additional major tributary, Beaver Creek HU (LULC rating of 68.4). Furthermore, the sources may be upland areas because the percentage of forested stream corridors within the Upper Gasconade River HU was the highest within the 8-digit Upper Gasconade River watershed (Table Hc03). A large number of gravel mining permits were issued for the Upper Gasconade River HU and Beaver Creek HU (Figure Lu04, Land Use Chapter), which indicated that gravel was available, and possibly the information contained within the NWI dataset was deficient of those smaller gravel bars less than one-tenth acre in size. These smaller gravel bars would have been common in upper watershed areas. Also, the forested corridor canopy may have made gravel bars invisible to the stereoscopic analysis of the high altitude aerial photographs performed by the National Wetland Inventory.
The several 11-digit units within the 8-digit Upper Gasconade River watershed may be the source of sediment for gravel bars of the Middle Gasconade River HU. The Upper and Lower Osage Fork (61.0 and 66.4, respectively) were rated relatively low in the LULC rating, compared to the Middle Gasconade River watershed (Table Hc05). Also, these HUs had several acres of gravel bars. Lower in percentage of forested corridor than other HUs, the gravel bars in the Upper and Lower Osage Fork may have been more visible to the stereoscopic analysis of aerial photographs or channel instability was contributing to their presence. Once this sediment load arrived in the Middle Gasconade River HU, the better forested corridor and possibly the drop in channel gradient slowed water velocity and deposited the bed load. Also, a tributary to the Middle Gasconade River HU, Roubidoux Creek, although it scored low in LULC rating, had few gravel bars, which likely may have been attributed to its more stable forested upper watershed.
In conclusion, Jacobson and Primm (1994) support the theory that historic land-use practices destabilized streams and their present state of instability is the result decreased riparian vegetation. Channel stability is not solely controlled by riparian vegetation; other mechanisms such as valley wall geometry and upstream changes can significantly affect channel stability (Jacobson and Pugh 1997). As demonstrated within relatively low impacted HUs with healthy forested stream corridors, potentially healthy channels may be adversely affected by poorer upper watershed conditions. Channel condition and stability are a complex combination of variables, of which several variables, such as those previously mentioned, play an important role.