Search this site:

Effectiveness of BMPs in Protecting Water Quality

Silvicultural activities include final timber harvest, intermediate harvests, site preparation, planting, fertilizer application, pest management, road construction and reconstruction, and fire management. Most, but not all, of these activities involve some degree of ground disturbance.

Aquatic conditions most likely to be impacted by forest treatments include water temperature, sediment and nutrient concentrations, stream channel stability, aquatic habitat quality, and toxic contamination. The purpose of silviculture BMPs is to eliminate or mitigate these effects.

Although States report that silviculture is a relatively minor contributor to stream impairment regionally, the pollutant most often associated with silviculture in State section 305(b) reports is sediment (see chapter 19). Forest roads are the greatest source of forestry-related sediment (Waters 1995, chapter 21). Thus, BMPs commonly focus on eliminating or mitigating sediment from forest roads.

Some of the relevant research and operational monitoring conducted in the South are reviewed in the next sections. Some of the cited studies are highly data intensive from instrumented watersheds, while others are less data intensive, employing upstream versus downstream observations of specified parameters. Both study types, if carefully designed and implemented, yield valuable information from which valid conclusions can be drawn.

Early research—BMPs are based on either research results, where available, or scientific principles. USDA Forest Service scientists at the Coweeta Hydrologic Research Laboratory conducted much of the research that formed the basis for BMPs in the South. Coweeta was established by the USDA Forest Service in the Appalachian Mountains of southwestern North Carolina to describe and understand the physical and biological processes that influence water as it moves through forested watersheds. Coweeta studies were and are data intensive.

Coweeta scientists conducted one of the earliest evaluations of effects of practical forest treatments on water quality in 1956 and 1957. A logging operation was conducted in the Stamp Creek drainage of the Tallulah Ranger District on the Chattahoochee National Forest (Black and Clark, no date). Specific operational standards (forerunners of BMPs) were written into the logging contract to test and demonstrate their ability to protect water quality during commercial logging.

Logging practices and road management were designed to control runoff to the adjacent streams. Roads and landings were located away from streams; storm runoff was removed from roads and dispersed onto the forest floor via strategically located broad-based dips; roads were constructed on the contour and limited to less than 10-percent grade; road crossings of streams were minimized and culverts or bridges installed; and road approaches to streams were graveled. Trees were felled downhill and limbed and topped in place; trees were skidded tree-length uphill by cable, butt-end first; skidding was dispersed over the harvest site; and logging slash was left in place except in streams. After the sale, roads and trails were smoothed of ruts and channels, and broad-based dips were restored and maintained to divert road drainage onto the forest floor.

Sediment concentrations in Stamp Creek, monitored throughout the harvest period, averaged 5 parts per million (ppm) as compared to 4 ppm for a nearby control watershed, and 31 ppm for a watershed logged without the applied operational standards. This was one of the first demonstrations that carefully planned and executed commercial logging practices do not degrade water quality. It also demonstrated that water quality can be impacted if protection is not provided.

Other research at Coweeta demonstrated road design considerations that reduce sedimentation from forest roads (Swift 1984). In one study, two sections of an existing logging access road were reconstructed to standards designed at Coweeta. The design called for an outsloped road with no inside ditches, and broad-based dips to divert road drainage. Grades above broad-based dips were kept constant at between 5 and 7 percent, outlets from broad-based dips were directed to undisturbed forest floor, outside berms kept road drainage off fillslopes, and brush barriers were constructed at the toes of fillslopes. Several key observations resulted from this study.

• Soil loss from roadbeds was greatest during winter storms and peak logging truck traffic.
• Lower road grades had lower soil losses.
• Cut-slope erosion was reduced if debris was left undisturbed at the toe of the slope during road maintenance.
• Outsloped roads without inside ditches reduced cut-slope erosion on many light-duty roads.
• Shorter fills, greater compaction, and brush barriers at fillslope toes reduced fillslope erosion.
• Locating fills away from streams reduced direct sediment input from roads to streams.
• Gravel spread on roadbeds and grass cover on slopes minimized soil losses.
• Grass cover on cut slopes reduced winter cut-slope erosion.
• Grass cover reduced downslope movement of slumps on moistened fillslopes.
• Gravel cover reduced roadbed rutting and erosion in wet seasons.
• Minimizing road width and curve radius reduced road erosion.

In other research, Swift (1986) tested a number of regionally recommended stream buffer widths and an array of other road BMPs for sediment reduction effectiveness. His findings were:

• Grassed fillslopes reduced sediment travel distance to half of that below mulched only and bare slopes.
• Undisturbed forest floor reduced sediment travel distance to half that on a forest floor with litter consumed by prescribed fire.
• Sediment travel distance below forest roads was related to forest floor slope.
• Sediment travel distance from outsloped roads with broad-based dips was not as great as that discharged via culverts.
• Grassed fillslopes and forest floor roughness reduced sediment travel distance by more than 20 feet below forest roads, and brush barriers reduced it more.
• The presence of brush barriers essentially removed the percent slope relationship for sediment travel distance from grassed and ungrassed roadways.
• Ninety-four percent of the soil deposition distances were less than stream buffer widths recommended by the USDA Forest Service Appalachian Guide standards of 1973 for “slight erosion hazard” soils. Thus, the buffer widths were largely adequate.
• A combination of tested practices can be used to reduce the width of required buffer strips for control of sediment from roads.

Swift recommended that filter strip widths between roads and streams in the Appalachians be based on site conditions and construction and stabilization factors such as grassing slopes, out-sloping roads, broad-based dips, cross drains, brush barriers, and forest floor cover.

Examination of State BMPs reveals strong similarities to the previously mentioned practices that were tested at Coweeta. Indeed, this research has been widely used as the scientific basis for BMPs in Southern States. It also demonstrates that BMPs complement one another when employed as a system of practices.

Operational effectiveness monitoring—Several studies have been conducted in the South to test the effectiveness of State BMPs or national forest water-quality standards and guidelines. A variety of water-quality parameters has been evaluated in a variety of locations, testing the effectiveness of differing practices. All provide valuable insight into the topic and several are summarized in the following paragraphs.

Clinginpeel (1989) and Neihardt (1992) measured the effectiveness of BMPs on the Ouachita National Forest in Arkansas and Oklahoma. Clinginpeel focused on BMPs for streamside management areas (SMAs) and for road crossings at streams; Neihardt evaluated BMPs for temporary road crossings of intermittent and ephemeral streams. The measured parameters in both studies were sediment, turbidity in Jackson turbidity units (JTUs), conductivity, alkalinity, pH, nitrites, nitrates, sulfates, and chlorides. Additional parameters in Neihardt’s study were total dissolved solids, hardness, turbidity in nephelometric turbidity units (NTUs), acid, and several metals.

Clinginpeel found that sulfates differed significantly above and below stream crossings, but actual differences were small (1.84 mg per liter and 1.94 mg per liter, respectively). Above and below measurements at SMAs were statistically different for turbidity (16.1 and 19.5 JTUs, respectively) and pH (6.13 and 6.32 pH, respectively), but remained within State standards. All the other parameters were unchanged. Neihardt found that turbidity measured in JTUs was statistically different, but turbidity measured in NTUs was not.

Both investigators concluded that forestry BMPs, as implemented on the Ouachita National Forest, effectively maintained water quality within State standards.

In a separate monitoring effort, Clinginpeel (1993) evaluated the effectiveness of BMPs for silvicultural herbicide application on the Ouachita National Forest from fiscal years 1989 through 1993. Again, stormwater samples were collected above and below treated areas from streams in potentially impacted areas, and analyzed for positive readings of Garlon, Velpar, and Roundup. In all, 348 water samples were collected from 168 sites. Sixty-nine samples, or 19.8 percent, tested positive for herbicides, but all positive samples were less than one-fourth the U.S. Environmental Protection Agency (EPA) limit for the specific herbicide and the toxic limit for fish. He concluded that the BMPs tested effectively protected water quality and fisheries.

In the early 1990s the North Carolina Division of Water Quality and the USDA Forest Service examined the effectiveness of BMPs on a forest road in the Appalachians (North Carolina Division of Water Quality 1994). A long-existing road, which closely paralleled Timbered Branch and its tributaries for about 2 miles and had been a chronic source of road sediments to the stream, was retrofitted with a number of measures designed to reduce sediment loading. They included ditch outlets, sediment traps, berms, weeps, outslopes, humps, and relief culverts. Sediment reduction was assessed qualitatively, and biological monitoring was conducted on the affected streams to determine effects on aquatic species. Improvements in taxa richness and diversity in the aquatic community were attributed to the sediment reduction practices.

The Georgia Forestry Commission, under a CWA section 319 grant and with quality assurance and quality control provided by the Georgia Environmental Protection Division, monitored 1-year-old harvested sites in all physiographic regions of that State and tested for State turbidity standard violations (Green 1995). Selected sites were 90 to 100 percent compliant with forestry BMPs, and all included timber harvests and road construction. Turbidity measurements in NTUs were taken upstream and downstream monthly and immediately after runoff-generating storm events. Neither violations of State turbidity standards nor significant increases in turbidity were found.

The Florida Division of Forestry and the Florida Department of Environmental Protection conducted a biological assessment of four commercially harvested sites before and after harvest (Vowell 2001). Sites selected were on forest industry land and were scheduled for harvest as part of normal ongoing company operations. Management activities at all sites involved clearcut timber harvest, intensive mechanical site preparation, herbicide and fertilizer application, and replanting. Florida’s silviculture BMPs were strictly adhered to during all operations. Upstream and downstream habitat and biological assessments were conducted before and immediately after activities were performed, and were continued for 2 years. Investigators found no statistically significant differences in parameters measured between the reference and treated sites. Hence, the authors concluded that Florida’s silviculture BMPs were effective in protecting water quality, aquatic habitat, and overall stream ecosystem health.

The South Carolina Forestry Commission, in cooperation with Clemson University and the South Carolina Department of Health and Environmental Control, evaluated the effectiveness of silviculture BMPs in protecting water quality in all physiographic regions in South Carolina (Adams and others 1995). Twenty-seven harvested sites from the Coastal Plain to the mountains were selected. BMP compliance on the sites ranged from inadequate to excellent, thus bracketing the full range of potential effects. BMP effectiveness was determined by Stream Habitat Assessment (SHA) and benthic macroinvertebrate monitoring. Upstream reference sites were used for comparison. Ten sites that rated inadequate for BMP compliance experienced negative SHA impacts, but only one site experienced moderate macroinvertebrate impairment. On sites where BMP compliance was rated as adequate or excellent, SHA indicated that streams were not impacted. The study did not look at an incremental comparison in SHA or bioassessment with incremental BMP compliance. Sites either passed or failed BMP inspection. Sites that passed BMP compliance inspection scored well on the bioassessment. The authors concluded that BMP compliance inspections appeared to be a reliable and economical surrogate for monitoring BMP effectiveness in South Carolina.

Williams and others (1999) evaluated BMP effectiveness in the South Carolina Piedmont, which they considered the most sensitive physiographic province in the State. The authors studied three harvest, site preparation, and regeneration alternatives (with BMPs) for changes in flow, sediment, and nutrients, and compared results to a control watershed. They observed statistically significant increases in observed parameters in all alternatives, but all waters met State water-quality standards. Further, they demonstrated that forestry BMPs reduced sediment yield to one-tenth of that occurring without BMPs.

A report published by the National Council of the Paper Industry for Air and Stream Improvement, Inc. (NCASI 1992), presented numerous documented studies of buffer-strip effectiveness in protecting water quality from silvicultural impacts. It concluded that buffers are effective in reducing transported sediment and pesticides and generally effective in reducing soluble nitrogen and, to a lesser extent, phosphorus delivery to streams.

The above body of scientific literature and monitoring results consistently demonstrates that forest management practices are capable of impacting surface water quality. However, it also demonstrates that appropriate BMPs fully implemented as designed and adapted to the site effectively protect water chemistry, aquatic habitat, and aquatic biota.