Riparian buffers are vegetated areas next to water resources that protect water resources from nonpoint source pollution and provide bank stabilization and aquatic and wildlife habitat. The formal definition of riparian buffer is diverse and depends on the individual or group defining the term.

The USDA Forest Service defines a riparian buffer as follows:

Leading experts (Lowrance, Leonard, and Sherida, 1985) on riparian buffers define them as follows:

Natural riparian buffers are composed of grasses, trees, or both types of vegetation. If riparian buffers are maintained or reestablished, they can exist under most land uses: natural, agricultural, forested, suburban, and urban.

What Do They Do and How Do They Work?: Introduction
Since riparian buffers in North Carolina are predominantly forested, discussion in this manual will focus on riparian forest systems. Forested riparian buffer systems in North Carolina are typically comprised of two integrated streamside riparian buffers (forest and grass or shrub) that are designed to intercept surface runoff and subsurface flow (Figure 2). Riparian buffers have been shown to be effective in controlling nonpoint source pollution by removing nutrients, especially nitrogen and sediment (USDA, 1997).

There are many factors that determine the effectiveness of riparian buffers for any given pollutant. To understand these factors, it is necessary to understand how riparian buffers work to reduce pollutant movement into surface waters. Movement of water from agricultural land through riparian buffers is illustrated in Figure 2. Sediment and sediment-associated pollutants, such as some pesticides and phosphorus, move to surface waters almost exclusively through surface runoff. Thus, to remove sediment and its associated pollutants, surface runoff water must be intercepted.   Figure 2. Schematic of the two zone Riparian Forest Buffer System (modified from Lowrance et al., 1995).

The most important factor controlling effectiveness of riparian buffers is hydrology: how the water moves through or over the buffer. For example, removal of contaminants from surface runoff requires that runoff water be sufficiently slowed to allow sediment to settle out. If the runoff water does not spread over the buffer, it will move through the buffer in channels. Channelized water moves almost as quickly through a buffer as it does from the field, thereby making the buffer ineffective at pollutant removal (Dillaha et al., 1989).

Most nitrogen from agricultural fields reaches surface water as nitrate in the ground water below the soil surface. In order for nitrate to be removed from ground water before it reaches surface water, the ground water must enter a zone where plant roots are or have been active. These plant roots may either absorb the nitrate for use in plant growth or, more importantly, may provide an energy source for bacteria that converts nitrogen in nitrate to a gas, which then escapes to the atmosphere. This process, denitrification, occurs almost exclusively in water-saturated zones where abundant organic matter is present.

Within all of the riparian buffer sites in North Carolina that were measured, nitrate concentrations in shallow ground water were significantly reduced as the water flowed through the riparian buffer (Gilliam et al., 1997). However, it is possible for nitrate to pass below the riparian buffer at depths far enough below the root zone that very little nitrate removal occurs (Correll et al., 1994). It is also possible for ground water to move through the riparian buffer so quickly that removal is limited (Haycock and Pinay, 1993). To quantitatively predict nitrate removal in riparian buffers, it is necessary to understand the hydrology of each site (Hill, 1996). Because riparian buffers are effective in reducing nitrogen under most conditions in North Carolina, we have no hesitation in recommending their use wherever practical.

What Do They Do and How Do They Work?: Buffer Design
Scientists agree that a corridor of vegetation can be effective at buffering valuable aquatic resources from the potential negative impacts of human use of the adjacent land. The streamside vegetated buffer filters nonpoint source pollutants from incoming runoff and provides habitat for a balanced, integrated, and adaptive community of riparian and aquatic organisms (Welsch, 1991). These filtering and habitat functions are often best provided by natural vegetation such as trees and associated woodland or forest plants in the zone directly adjacent to the waterway. While there is general agreement about the benefits of buffers, the specific design criteria, such as buffer width, types of vegetation, and management, are the subject of considerable debate.

Width is considered the most important controllable variable in determining the effectiveness of buffers in reducing pollutants and protecting stream health. Buffers that are too narrow may not be sustainable or effective at protecting stream banks. Conversely, buffers that are wider than needed limit the use of adjacent land and are unpopular with landowners. Complicating the determination of design buffer widths are the effects of varying site characteristics associated with topography, hydrology, geology, and land use. Additionally, other factors, such as the value of the water resource and adjacent land, must be considered when determining widths.

The width of most existing riparian forest buffers was established by leaving the area adjacent to the stream as forest. This area was generally too wet or too steep to be used conveniently for agricultural or urban purposes. Welsch (1991) recommended a widely acclaimed riparian buffer system that was 95 feet wide on both sides of the stream. There is little debate among riparian buffer experts that the system he described is very good as an idealized stream. However, the senior author of this document does not agree that this width should be required along every stream. The width necessarily depends upon what functions are expected of the riparian buffer and the site characteristics.

Most decisions about buffer widths will be a compromise between ideal widths based on environmental goals (wildlife corridors, bank stabilization, water quality protection) and sociologic or economic constraints. Science-based criteria, for which research data may be available to support an informed decision, include the functional value of the water resource; watershed, site, and buffer characteristics; adjacent land use; and buffer function. The functional value of the water resource is important for determining buffer width in that a highly valued resource may merit a wider buffer for increased protection.

Watershed, site, and buffer characteristics are most important when evaluating pollutant-filtering effectiveness. The size and topography of the watershed determine the amount and rate of surface and ground water passing through the buffer. Site characteristics, such as soil type, slope steepness, microbial populations, and vegetation, determine the amount of pollutants that are filtered out of the water before it enters the waterway. Buffer characteristics, such as the types of vegetation and their location in the buffer, can also influence pollutant removal effectiveness.

 One of the most widely recognized buffer planning models is the three-zone buffer that was developed by the USDA Forest Service (Welsch, 1991). Zone one of the model begins at the normal water level or at the edge of the active channel and extends a minimum of 15 feet along a line perpendicular to the watercourse . Dominant vegetation consists of existing or planted woody vegetation suitable for the site and intended purpose. This zone should remain undisturbed; therefore, tree removal is generally not permitted. Zone two begins at the edge of zone one and extends a minimum of 60 feet perpendicular to the watercourse. While vegetation in zone two should be similar to zone one, removal of tree and shrub products is permitted on a regular basis provided the tree and shrubs are replaced. The third zone begins at the outer edge of zone two and has a minimum width of 20 feet. Vegetation in this zone can be grazed or ungrazed grass or other plant communities as long as it facilitates sediment filtering, nutrient uptake, and the conversion of concentrated flow to uniform, shallow, sheet flow through the use of structural practices such as level spreaders (Lowrance et al., 1995).

The current proposed buffer standards in North Carolina use a two-tiered riparian buffer: forested areas near the streams and grassed areas away from the stream. The proposed buffer width is 50 feet: 30 feet of forest and 20 feet of grass (NCDEHNR, 1997). Some streams, however, may need greater and some streams need less buffer width, depending not only on site location but also on the pollutant that is being controlled. For optimal performance, riparian forest buffer systems must be designed and maintained to maximize sheet flow and infiltration and impede concentrated flow.

The design also depends on the stream order and the land area that drains the riparian buffer. The larger the drainage area, the wider the buffer width. The Neuse River Basin consists of over 4,000 miles of streams, although the Neuse River itself is only 200 miles long. Since most nonpoint source pollution enters the river system through these first order streams, it is very important to protect the smaller streams with buffers.

Stream networks are designated by using stream orders. First-order streams have no tributaries. A second-order stream starts at the confluence of two first-order streams. The confluence of two second-order streams is a third-order stream and so on (Dingman, 1994).
What Do They Do and How Do They Work?: Protect Stream Health
The most general function of riparian forest buffer systems is to provide control of the stream environment. This function includes moderating fluctuations in stream temperature and controlling light quantity and quality; enhancing habitat diversity; modifying channel morphology; enhancing food webs and species richness; and protecting water resources from nonpoint source pollutants, such as sediment and nutrients (USEPA, 1995b).

The design specifications for forested riparian buffers should provide the desired function of the buffer at the particular location – whether it is being used to control nonpoint source pollution so that downstream waters do not deteriorate or to protect aquatic organisms.

Habitat: Aquatic Organisms.(The following information on habitat was largely taken from USDA’s Riparian
Organisms Forest Buffer Handbook for the Chesapeake Bay Watershed, 1997. Thanks to Al Todd for letting us use the draft version of this document.)

The riparian buffer is an important feature of stream habitat. The vegetation of the riparian buffer affects the type and amount of organic matter food sources available for stream organisms. Streamside vegetation also affects the amount of sunlight that reaches the stream and, in turn, the temperature of the water. In addition, the physical structure of the stream, such as the extent of pools and riffles, is affected by riparian vegetation. Climate and watershed characteristics also affect aquatic life habitat. All of these factors influence species diversity and abundance.

Food. Food sources for macroinvertebrates include detritus and algae. Detritus is organic matter such as leaves, stems, sticks, and logs that falls into the stream. Because their mouth parts are adapted for a particular food source, some macroinvertebrates eat primarily detritus and others eat only algae. Two types of algae found in streams are diatoms and filamentous algae.

The vegetation in the riparian buffer affects the type and quantity of detritus that occurs in the stream. It is likely that vegetation that falls into the stream generally does not move very far away so that the food benefits are highly localized to the immediate stream corridor. Older stratified forests may provide the greatest variation in quality of detritus food for macroinvertebrates.

Vegetation also affects the amount of light that reaches the stream, but this is a function of stream order and stream width as well. For first-, second-, and third-order streams, the riparian canopy of trees can block sunlight from reaching the water. A shaded stream is likely to have more diatoms and less filamentous algae. A stream that runs through a cleared riparian buffer or one that has meadow vegetation is likely to have more filamentous algae. The detritus food source from the clearing of a riparian buffer is only temporary as detritus rapidly decays. For grassed riparian buffers, filamentous algae is likely to dominate. Also, large streams and rivers will receive a large portion of direct sunlight which encourages filamentous algal production in open areas. Nearshore areas bordered by mature vegetation are likely to have diatoms and sufficient detritus.

Temperature and light. Vegetation type, canopy development, and directional orientation of the stream controls light energy and impacts stream temperature. A north-south oriented stream is less affected by buffer canopy shading. The vegetation on the north side of an east-west oriented stream may also have little effect on light penetration. For first-, second-, and third-order streams, the majority of water flows through a shaded riparian buffer. For higher order streams, which are wide and open in cross-section, shading has less of an impact on water temperature. However, the loss of the buffer canopy on any stream, due to clearing, can increase water temperature substantially, causing a shift in macroinvertebrate and fish species.

Physical habitat (pools, riffles, etc.). Roots of riparian vegetation stabilize the stream bank and prevent stream bank erosion and sedimentation. Stabilized stream banks also help maintain the geometry of the stream, including characteristics such as the meander length and profile. Preventing excess sedimentation helps prevent silt from covering large rocks and stones in the stream bed which serve as habitat for some macroinvertebrates. Pools can be vital parts of stream habitat for fish. Excess sediment can fill pools and eliminate habitat. Tree roots and woody debris are also important habitat features for macroinvertebrates and fish. Overhanging stream banks, stabilized by tree roots and large woody debris, can be important habitat for fish.

Large woody debris provides critical macroinvertebrate habitat. Large woody debris can also create dams and trap sediment and detritus. Riparian forests may have the greatest enhancing effect on fish habitat on mid-order streams (i.e., stream order 3-6), with sufficient large woody debris structure and flow to support diverse fish and macroinvertebrate populations.

Habitat: Wildlife Wildlife species require food, water, and cover. Well managed riparian buffers generally support larger populations of wildlife because the buffer provides many habitat requirements. In a stratified forest, different habitat zones exist vertically, including the soil-air interface, herbs and shrubs, intermediate height trees, and the canopy. Included with the leaf litter and rotting logs at the soil-water interface are insects, isopods, spiders, and mites. These organisms are a food source for reptiles, mice, and birds. The herbs and shrubs provide habitat for insects, birds, and mammals. The intermediate zone and the canopy serve as habitat for birds, bats, squirrels, opossums, and raccoons. Bird habitat may be highly stratified and birds generally show a preference for certain layers that differ in habitat characteristics and food sources.

Riparian areas may also serve as corridors linking dryer, less diverse uplands to more moist, more diverse bottom lands. The width of riparian buffers needed for wildlife is not clear. This may be a function of the type of wildlife and their vegetation requirements. Upland game birds such as pheasant and bobwhite quail benefit from grasses. A stratified forested may be needed to maintain wrens and robins in a herbaceous zone and tree-creeping birds and robins in the canopy.

What Do They Do and How Do They Work?: Reduce Nitrogen
In the Coastal Zone of North Carolina, most nitrogen, as stated previously, enters surface waters from ground water as nitrate-nitrogen. As the shallow ground water moves through the riparian buffer, microorganisms change the nitrate-nitrogen to gaseous nitrogen via a process known as denitrification (Figure 4). When the soil is poorly aerated (anaerobic conditions), some microorganisms reduce nitrates to the gaseous components of nitrous oxide, nitric oxide, or free nitrogen gas. Figure 4 Figure 4. Conceptual model of below-ground processes affecting ground water nutrients in riparian forest (from Correll and Weller, 1989).

Denitrification is most effective in root-zone soil layers where carbon sources are available for the denitrifying bacteria. Numerous researchers have reported that it is the complex interaction between vegetation and below-ground environment that provides the appropriate conditions for denitrification to occur (Lowrance et al., 1995). The area of interaction within the riparian buffer is generally quite narrow —10 to 50 feet (or 3 to 15 meters) — from the field through the riparian buffer (Figure 5). The majority of denitrification that has been observed in riparian buffers occurred within the first 15 feet of the forested riparian buffer. Figure 5 Figure 5. Nitrate concentrations in ground water beneath riparian forests.

Denitrification has measured in Coastal Plain forested riparian buffer areas has removed as much as 263 lb N acre-1yr-1. Typically, though, denitrification rates are generally between 18 and 55 lb N acre-1 yr-1. Most studies indicate that denitrification takes place throughout the year (Lowrance et al., 1995).

Vegetation in riparian buffers also removes nitrogen and phosphorous through uptake. Some of these nutrients are sequestered in woody vegetation, whereas the nutrients absorbed into herbaceous materials, generally, are recycled as the vegetative matter dies. Several studies have indicted that uptake by above-ground woody vegetation removes various amounts of nitrogen and phosphorus, depending on the riparian conditions (Table 1).

Table 1. Above-ground woody vegetation uptake of nitrogen and phosphorus in Coastal Plain riparian forests (from Lowrance et al., 1995).


Nitrogen Phosphorus


Reference  Location  Total 
Correll & Weller, 1989 Rhode R., MD ND*  12 to 20 ND 3 to 5
Peterjohn & Correll, 1984 Rhode R., MD 77 12 10 1.7
Fail et al., 1986, 1987(mean)  Little R., GA 114 52 7.5 3.8
Fail, 1986 (maximum) Little R., GA 194.4 97.6 12.6 6.9
Fail, 1986 (maximum) Little R., GA 80 34.6 4.5 1.9

*ND =not Determined


Although nitrogen uptake by the vegetative portion of the riparian buffer buffer contributes to nitrogen reductions, denitrification is the primary process that removes nitrate from the shallow ground water that flows through riparian buffers.

What Do They Do and How Do They Work?: Reduce Sediment and Phosphorus
Riparian buffers, both the grassed and forested portions, serve to slow water velocity, thus allowing sediment to settle out of the surface runoff water. The grassed portion of the buffer functions as a grass vegetated filter strip. There is extensive research demonstrating the effectiveness of vegetated filter strips for sediment removal (Lowrance et al., 1995).

The effectiveness of well maintained grass riparian buffers for sediment may be as high as high as 90–95%. Likewise, nitrogen and phosphorus attached to the sediment and, to a lesser extent, dissolved nitrogen and phosphorus are abated. Frequently, the concentration of dissolved nutrients in the runoff passing over a grass filter does not change or may slightly increase. However, because some of the runoff water infiltrates in the buffer, less runoff water leaves the buffer than enters it. For example, we have observed that approximately half of field runoff events have no runoff that passes through a 24-foot grass filter. Thus, there is a reduction in total amount of dissolved nutrients that leaves the filters even though nutrient concentration may not change. These filter strips are not designed for high velocity flow but, rather, are used to slow flows so that sediment drops out. Because grass riparian buffers are designed to trap sediment, they REQUIRE MAINTENANCE to remain effective (Dillaha et al., 1989).

In experiments conducted to determine optimum width of a grass riparian buffer adjacent to a forested riparian buffer, Parsons (personal communication, 1997) determined sediment reduction for different grass riparian buffer widths. Approximately 100 data points were collected for storms that produced >1000 g of sediment loss at the edge of the field. The percent sediment reduction is calculated as 1.0 – [(grass buffer loss)/(field edge loss)]. In the Piedmont, 28 feet of buffer width retarded sediment such that there was 86–90% reduction (Table 2), whereas the narrower buffer width of 14 feet reduced sediment loss by 70%. By contrast, the difference in sediment reduction between grass buffer widths of 14 and 28 feet was not as marked at a Coastal Plain location: 86% reduction for buffer widths of 28 feet and 76.5% reduction for 14-foot grass riparian buffers (Table 3).

Table 2. Sediment reduction by grass riparian buffers on a Piedmont site.

Grass Buffer Width


 % Reduction 

14 ft 1 71
14 ft 2 68
28 ft 1 90
28 ft 2 86

Table 3. Sediment reduction by grass riparian buffers on a Coastal Plain site (Kinston).

Grass Buffer Width Plot  % Reduction 
14 ft 1 70
14 ft 2 83
28 ft 1 82
28 ft 2 90

Grass riparian buffers in combination with forested areas appear to do the best job of reducing both sediment and phosphorus, as can be seen from the following table. The effects of different riparian buffer widths in reducing sediment, nitrogen, and phosphorus are presented in Table 4.

Table 4. Effects of different size riparian buffers on reductions of sediment and nutrients from field surface runoff (from Lowrance et al., 1995).

    Sediment     Nitrogen     Phosphorus    
Buffer Width Buffer Type Input 
Output Conc. Reduction Input Conc. Output Conc. Reduction Input Conc. Output Conc. Reduction
m   --mg L--   % --mg L--          
4.61 Grass 7284 2841 61.0 14.11 13.55 4.0 11.30 8.09 28.5
9.21 Grass 7284 1852 74.6 14.11 10.91 22.7 11.30 8.56 24.2
19.02,3 Forest 6480 661 89.9 27.59 7.08 74.3 5.03 1.51 70.0
23.65 Grass/ Forest 7284 290 96.0 14.11 3.48 75.3 11.30 2.43 78.5
28.26 Grass/ Forest 7284 188 97.4 14.11 2.80 80.1 11.30 2.57 77.2

1Calculated from masses of total suspended solids, total N, total P, runoff depth, and plot size (22 x 5 m) from Magette et al. (1989)
2Input concentrations from Table 2, Peterjohn & Correll (1984). Nitrogen = Nitrate-N + exch. part. ammonium + diss. ammonium + part. organic N + diss. organic N. Phosphorus = part. P + diss. P.
3Surface runoff concentrations at 19 m into forest reported by Peterjohn & Correll (1984). N and P constituents same as input (footnote 2).
4Percent reduction = 100 * (Input-Output)/Input.
54.6 m grass buffer plus 19 m of forest.
69.2 m grass buffer plus 19 m of forest.
conc. = concentration.
Because most phosphorus loss is so closely tied to erosion, the above discussion on riparian buffers and sediment is relevant to the control of phosphorus.

As noted previously, vegetative uptake of phosphorus is another reduction mechanism. Researchers have shown (see Table 1) that between 2 and 12 lb P acre-1 year-1 are absorbed by the above-ground woody vegetation in Coastal Plain riparian forests (Lowrance et al., 1995). Top of Page / Table of Contents / Next Section