Riparian Buffers: Function, Management, and Economic Implications for Agriculture

Final Report for SW03-115

Project Type: Research and Education
Funds awarded in 2003: $242,035.00
Projected End Date: 12/31/2007
Matching Non-Federal Funds: $33,917.00
Region: Western
State: Washington
Principal Investigator:
Jon Johnson
Washington State University - Puyallup Res. & Ext.
Expand All

Project Information


This project successfully characterized the functioning of 3 riparian buffers over a 4-year period as related to non-point pollution coming from adjoining farms. Factors affecting buffer function included depth to groundwater, species, age and time of year. In addition, we quantified the impact of buffer shade on two crops, silage corn and blueberry. Lifetime costs of buffers on potato farm budget were $11,000 per acre and even higher for blueberry and dairy production in the Skagit River Delta. Four interactive farm enterprise budgets were created for potato, blueberry, raspberry and dairy, and are available to the public at

Project Objectives:

Our goal was to identify what constitutes a functional riparian buffer to protect water quality and improve salmon habitat on agricultural land in western Washington, and to determine the economic impact of such buffers on farm enterprises. Specific objectives include:
1) To determine the effects of buffer width, species composition and management on buffer function including nutrient removal, sediment reduction, shade, and bank stabilization;
2) To conduct economic impact analysis of different riparian buffer designs on individual farm enterprises;
3) To develop and disseminate buffer recommendations and decision-making tools to farmers, farm agencies, regulators and policy makers dealing with farmland along watercourses in western Washington.


Several Pacific Northwest (PNW) salmonid runs were listed as threatened or endangered under the Endangered Species Act (ESA) in 1999. Scientific research is lacking on how to manage floodplain agricultural areas to improve salmon habitat. Because of the lack of data, and the immediacy of action required by the ESA listing, regulatory agencies are using upland riparian research to guide regulations for downstream agricultural riparian areas. Wide, forested riparian buffers that the agencies are considering may or may not be suited to agricultural areas, and may threaten the economic viability of PNW agriculture by eliminating thousands of acres of productive land.
Skagit County, and all Washington State counties, is required by law to review their critical areas ordinances (CAOs) every five years. Skagit County is currently making major revisions to its CAOs, particularly concerning the protection of habitat and water quality on waterways that cross agricultural land. One of the protection options under consideration includes wide riparian buffers on all water bodies that cross agricultural land. There has been a great deal of contention and litigation concerning this option, both from environmental groups that want adequate protection of waterways for endangered salmonids and from farmer groups who are deeply concerned about the potentially large economic. There are currently 115,000 acres of land used for agricultural in Skagit County. Under one proposed buffer scenario, 11% or over 12,000 acres of this land would be taken out of production and put into no-touch buffers.
We propose to address this information gap by examining the environmental and economic implications of establishing and managing riparian buffers on agricultural lands in Skagit County, Washington. This rural county is largely dependent on commercial agriculture. The economic impact of installing and maintaining riparian buffers on farms is currently unknown. A major obstacle to landowner adoption of sustainable agricultural practices is the uncertainty surrounding the cost and economic impact of the practice to be adopted. The project addresses this obstacle “head on” by giving farm owners tools to see how different buffer widths and species composition affect their budgets and land value. We will study buffers by evaluating them in the context of whole-farm budgets. By testing buffer designs that can be actively managed, we will evaluate the ability of buffers to provide diversified income sources, which may help offset buffer establishment costs. This project seeks to identify efficient buffers that will enable farm operators to maintain their economic viability while enhancing salmon habitat and water quality, both of which benefit society as a whole.
Riparian buffers are widely considered to be a good land stewardship practice because of their ability to conserve soil, and improve the quality of surface and ground water, and provide fish and wildlife habitat. However, scientific literature on riparian buffer function and design is limited for low gradient streams, rivers and associated floodplains, and is non-existent for low gradient areas in western Washington watersheds. This project entails the establishment of replicated experimental riparian buffers and use of exiting riparian buffers at three commercial farms located in Skagit River delta representing typical farming practices in this area: Annual cropping, perennial cropping and dairy operation. The goal of this project is to design a riparian buffer system that is functional while being economically viable for the farmer. By determining buffer functionality, we will be able to provide management recommendations for the efficient use of arable land. Information gathered in the Skagit watershed will be directly transferable to other watersheds in western Washington and Oregon.
The use of fast growing trees, such as red alder, black cottonwood, and hybrid poplar, in the forested areas of the buffers will accelerate site occupation and development of a functional buffer. Once established, management options will include creating suitable planting sites for shade tolerant native conifers and/or sustainable harvesting of the trees for supplemental farm income. In addition to its economic benefit, judicious removal of biomass from the buffer will stimulate new growth and provide greater rate of nutrient uptake, thus improving buffer function.
The use of vegetated “filter strips” or “buffers” along agricultural watercourses is a well documented means of controlling pollution from adjoining agricultural fields (Dillaha, Reneau et al. 1989) (Lowrance, Todd et al. 1984) (Lowrance, Leonard et al. 1985) (Muskat, Harris et al. 1993) (Snyder 1998). Vegetated buffers achieve this function through intercepting and up taking nutrients, especially nitrogen and phosphorous, (Groffman, Axelrod et al. 1991) (Haycock and Pinay 1993) (Hill 1990) (Hubbard and Lowrance 1997) (Jacobs and GIlliam 1985) (Lowrance 1992) (Pinay and Decamps 1988), and by blocking sediment runoff into streams (Cooper, Gilliam et al. 1987) (Sheridan, Lowrance et al. 1999), among many other functions. Populating the buffer with woody species will further improve its filtering capacity (Haycock and Pinay 1993), as well as provide additional salmon habitat benefits such as shade and woody debris in the stream (Knutson and Naef 1997) (Bishaw, Rogers et al. 2001) (US Geological Survey 2000) (Washington State Conservation Commission 1999). Despite the known benefits of riparian buffers, much of the work has been done in eastern regions of the US, and little is known about their function in low gradient areas such as the Skagit River flood plains of Western Washington. Our project seeks to address this knowledge gap. Furthermore, while work to date has pointed out the benefits of buffers, little comparative work addressing the issue of what constitutes a “sufficient” buffer in terms of function has been conducted. This is a central question in the current study.
By virtue of their rapid height growth and accumulation of biomass, fast growing species such as poplars and alders are more effective than grasses or slower growing woody species at rapidly establishing buffer function (Haycock and Pinay 1993) (Bishaw, Emmingham et al. 1998). The poplars have average annual height growth approaching 15 feet and the alders, approaching 6 feet. A number of authors have evaluated the nutrient removal ability of fast growing woody species (Dix, Klopfenstein et al. 1997) (Brockway and Urie 1983) (Heilman, Stettler et al. 1990), demonstrating their efficacy. Haycock and Penay (1993) found that nitrate moving downslope was retained in the first 5 m of a buffer consisting of poplar trees. In addition, poplar trees retained 99% of the nitrate whereas grass buffers retained only 84% during the winter (this study was conducted in England that has a similar climate to western Washington). Lowrance et al. (1985) reported that certain pesticides including picloram, aldicarb and dalapon, move through the environment dissolved in runoff and leachate and are retained in riparian zone aquifer where deep-rooted species like poplar and alder could enhance retention. They also found that 99% of the nitrate moving from the agricultural plots did so by subsurface water, representing 80% of total water movement off the plots. Denitrification and storage in woody vegetation removed 6 times the amount of nitrogen that reached the stream.
Although the wisdom of actively managing riparian buffer areas is still hotly debated in Washington State, a number of studies show the importance of regular biomass removal from the buffer in order to maintain its nutrient uptake capacity (Merwin, Powers et al. 1999) (Omernik, Abernathy et al. 1981) (Peterjohn and Correll 1984). The present study seeks to determine if there are differences in function between managed and unmanaged buffers for Western Washington sites. This issue is not only important from a buffer function point of view, but also affects the economic viability of the buffer. If management could be successfully achieved, the net cost to both landowners and government agencies of installing buffers would be greatly reduced. A handful of small studies in other parts of the US have explored the income generation capacity of riparian areas (Tjaden and Klapworth 1998) (Licht 1992). A Recent study in Western Oregon suggests that forested buffer management can be as profitable on a per acre basis as dairy and grass seed production (Merwin, Powers et al. 1999). No studies that we know of have looked at buffers from the perspective of economic impact on individual farm enterprises. This is a key objective of the present study.


Click linked name(s) to expand/collapse or show everyone's info
  • James Dobrowolski
  • Carolyn Henri


Materials and methods:

Farm sites
We evaluated 2 existing buffers and established a new replicated experimental buffer on two commercial farms in the Skagit River delta representing typical farming practices in this area: 1) Dairy farm (DeVries farm on the Nookachamp River; manure application to pasture) and; 2) Blueberry farm (Bayview farms on Joe Leary Slough; a perennial cropping system). The sites were designated Blueberry, Mature and Experimental for reporting purposes.

1) To determine the effects of buffer width, species composition and management on buffer function including nutrient removal, sediment reduction, shade, and bank stabilization

Experimental Buffer Site: A new experimental buffer was established on the DeVries dairy farm along the Nookchamps River using a design consisting of a) 50’ of hybrid poplar (7’ x 7’ spacing) + 25’ grass filter strip; b) 50’ red alder (7’ x 7’ spacing) + 25’ grass filter strip, and: c) 75’ perennial grass filter strip. Each plot measured 100’ long the river by 75’ wide and was replicated twice. Buffer treatments were randomly assigned within each of the 2 replicated blocks. The site was cleared in August 2003 and competing vegetation controlled with Rodeo applications before and during the establishment period. The trees were planted in early spring (April 2004) at 7’ x 7’ spacing in an offset (diamond) pattern. The hybrid poplar variety, 15-29, came from the WSU Poplar program and has been grown successfully in the Skagit River delta. The poplars were planted as dormant 12” stem cuttings, purchased from a commercial nursery. The red alder were planted as 1 year-old bare-root seedlings that came from the Weyerhaeuser red alder nursery. Competing weeds were controlled by mowing and spot treating with Rodeo throughout the growing season with additional Rodeo application during the winter. A four-foot high fence was erected along the river, approximately 650 feet long, to keep beavers out of the tree plots. During flood events in late 2004 and early 2005, beaver floated over the fence and removed nearly all of the poplars. In April 2005, the poplar plots were replanted with 8-foot tall dormant whips planted to a depth of 2 to 3 feet. Each stem was fitted with a 3-foot tall plastic tree stem protector. The following winter in 2006, again during flood events, beaver felled about 40% of the trees at the top of the protector. These “stumps” have subsequently sprouted to provide tree coverage in these plots.
Two existing buffers were included for comparison: 1) a natural 60-70 year old riparian forest dominated by a mixture of black cottonwood and red alder adjacent to pasture receiving manure application at the DeVries farm, but along the Skagit River and; 2) an 9-year-old riparian hybrid poplar plantation at Bayview farms along Joe Leary Slough and adjacent to existing blueberry fields.

Measurements: Sediment, surface and subsurface nutrient movement from the farm fields through the buffers were measured using ground water wells, suction lysimeters and sediment traps placed along transects from the field edge to channel perpendicular to stream flow (3 transects per plot and up to 7 sampling points per transect). The sampling wells were installed as follows: Nookachamps Experimental buffer – starting at the grass-tree interface (0’), 10’, 25’ and 50’ (at river edge) – additional single wells were installed on the field side of a berm coinciding with the middle transect of each plot; Skagit Mature buffer – 0’(field edge), 10’, 25’, 50’ and 75’(middle of buffer); Bayview Farm – blueberry only – 0’(in blueberry row), 25’(mid-way in a grass strip) and 50’ (at edge of slough), blueberry + poplar – 0’, 25’, 50’(at edge of poplar stand), 60’, 75’, 100’ and 200’(at edge of slough). Sampling took place when both ground water and surface flow were likely to occur (November, December, March, April, May and June beginning in 2004 and ending with one sampling in May 2007). Sampling procedure we followed is detailed in Appendix A. Groundwater well and suction lysimeter samples were analyzed for N and P (except the last sample date when only N was analyzed for lack of funds) using Alpkem Flow Solutions 3000 with the cadmium reduction method at the WSU Animal Nutrition laboratory. Tensiometer readings rarely showed any soil water deficits resulting from our sampling being confined to times of the year when the soil was saturated. Overland sediment flow was determined by Low Impact Flow Event (LIFE) samplers, but due to the low quantity of individual rainfall events and nearly flat terrain, no overland flow was ever detected in any of the samplers at any of the 3 sites.
We contracted with the University of Washington’s Rural Technology Institute to adapt an Excel-based Buffer Shade Program to allow for the determination of cumulative energy input on both sides of a buffer, the stream and the farm crop. To test the effect of buffer shade on crop production, we located two established natural buffers on the DeVries Dairy farm that were adjacent to silage corn fields. One buffer was the mature buffer described above that was oriented north-south with the field located east of the buffer. The buffer thus provided shade in the afternoon. The other buffer on the farm was oriented east-west with the corn field located north of the buffer, which shaded the crop for most of each day. In September 2006, we harvested duplicate plots consisting of two adjacent rows of linear distance of 10 feet. In addition to total biomass, number of stalks and ears were quantified along with ear biomass. Plots were located on two parallel transects located perpendicular to the buffer. Plots were sampled at distances of 25, 75, and 150 feet with an additional 2 plots at 500 feet sampled at the mature buffer site. Similar sampling transects were established in the blueberry rows located northwest of the hybrid poplar buffer that ran southwest-northeast at the Blueberry site. The blueberry plants were shaded from early morning to mid-afternoon. At each sampling location in the blueberry field, 2 adjacent plants were measured for number of stems, and height and diameter of each stem.

2) To conduct economic impact analysis of different riparian buffer designs on individual farm enterprises
Riparian buffer designs were evaluated for economic impact on the farm enterprise, as well as long-term farmland value. Software programs, written in Microsoft Excel, and a user’s manual were developed for distribution and use by landowners (Appendix). The programs allow the landowner to evaluate the short and long-term economic effects on his or her specific enterprise of installing and maintaining riparian buffers.
Short-term (annual) impacts were measured by incorporating costs and revenues associated with riparian buffers into existing farm enterprise annual budgets. Existing sector-specific budgets for the major farm types included in our study (dairy, field crops, vegetable crops, and perennial berries) were used as the analytical base. Components were added to the budgets that to allow the user to compare his or her budget, line for line with and without riparian buffers. The programs highlighted the major changes in sources of revenue and cost associated with riparian buffers. Longer-term economic impacts were evaluated using standard Soil Expectation Value (SEV) equations. SEV identifies land value based on productive capacity and probably of earnings as agricultural land for a period of time into the future. SEVs for the different farm types identified above were developed, and a comparison of SEVs with and without riparian buffers was analyzed.

3) To develop and disseminate buffer recommendations and decision-making tools to owners of farmland along watercourses in western Washington.
Extension Activities. This project played a major role in Washington State University’s Water Quality Management Teams’s (WQMT) extension activities (Dobrowolski was the WQMT representative). Our outreach activities focused on extending our results to our broad clientele base in a manner that could be immediately used to make more informed decisions about agricultural land management. We found that the credibility of our project results were greatly enhanced if clientele are aware of, and given an opportunity to comment on, the intent, scope and method of the work early in the process. For that reason we developed an interactive web page ( early in the project to allow viewers to send questions and comments directly to our team. We also held regular field days—timed to correspond with important steps in the project to inform our various clientele groups of the progress being made on the project. We hosted one field review, two formal presentations to county commissioners and the Skagitonnians for the Protection of Agricultural Land and several small group and individual discussions. Several workshops were conducted at Skagit Community College to allow farmers and other interested parties to learn how to use the farm enterprise budget software developed during the project.

Educational Activities. The proposed research project provided opportunities across the K-12 through university continuum, and for train-the-trainer programs for extension and government agency personnel. The buffer monitoring infrastructure and vegetation/water quality record will provide a valuable teaching resource for university courses. We partnered with Skagit High School’s Vocational Educational Program including two formal classroom presentations and two field trips to the Nookachamps and Skagit buffer sites to allow hands-on experience by the students.

Evaluation of Outreach Efforts. All extension and educational products (website, fact sheets, etc.) from this project were subject to peer-review through the WSU CAHE Extension Publication Peer Review process to ensure technical and educational quality as well as suitability for the intended audiences. Participant entrance and exit surveys were conducted for all field days, short courses, and workshops associated with this project to provide feedback on program quality, applicability, and potential improvements.

Research results and discussion:

Objective 1
In order for a buffer to remove nutrients from the groundwater, the roots must be able to penetrate into the groundwater, so depth to groundwater is an important factor for determining buffer function. Each buffer site had different depths to groundwater which was affected by season (the greatest depth was usually recorded in the fall after the normal summer drought while the shallowest depth coincided with early spring after the winter rains). The range of depths for the blueberry site was 43-106 cm in the no buffer area and 62-141 cm in the hybrid poplar buffer (the deeper groundwater in the poplars was attributed to transpirational water removal of the groundwater). At the mature buffer site the groundwater ranged from 192 – 468 cm. At the experimental site the groundwater depth in the field ranged from 317- 484 cm, at the beginning of the buffer it ranged from 129-293, and at the river edge from 27-177cm. In general, the groundwater was deepest in the grass plots and shallowest in the poplar plots.

Nitrate levels in groundwater were affected by site, date and distance into the buffers whereas phosphate was only affected by date (and therefore is not reported). Nitrate levels in the soil solution were affected by site, distance and depth, but phosphate in the soil solution was unaffected by nearly all main effects (and therefore is not reported). Since each site was distinct, the results will be presented separately.

Blueberry Farm Site

Analysis of variance showed significant (P<0.0001) main effects of date, buffer type and distance from the blueberries for nitrate. For phosphate, only buffer type was significant (P<0.017) with the mean levels ranging from 0.23 in the slough to 0.65 in the no buffer to 3.5 ppm in the poplar buffer. The highest concentrations measured were for December 2004 at 437 ppm and 1297 ppm for the no buffer and poplar buffer, respectively in the blueberry field. For most months the nitrate levels were below 40 ppm. With the exception of December 2005, the nitrate levels in the groundwater decreased from the blueberry field to the middle of the grass filter strip and then usually increased slightly at the edge of the slough. The low concentrations in the filter strip can be attributed to high denitrification rates resulting from anaerobic conditions and a carbon source from the grass roots. The increase in nitrate at the slough edge is probably due to movement of N from the slough into the bank (see April 2005 and 2006). The nitrate levels in the blueberry field adjacent to the poplar stand showed greater variation from over 1200 (shown as 200 on the graph) to near 0 ppm. Once into the grass filter strip and poplar buffer, nitrate levels dropped significantly to usually less than 10 ppm. The nitrate levels within the poplar stand remained at 1 ppm or less until the well at the edge of the slough (200 feet). It is unclear why the nitrate levels increased for the November and December 2005 sample date at 100 and 200 feet. Mature Buffer Site Analysis of variance showed significant main effects of date (P <0.0001) and distance into the buffer (P<0.0072). Nitrate levels at this site were consistently higher than the blueberry buffer site ranging from 10 to 265 ppm. Phosphate showed only significant date effects (P<0.0006), but no statistical effect of distance into the buffer. The general trend was a decrease in nitrate levels the farther into the buffer suggesting that the trees in the buffer were removing at least some nitrate from the groundwater. The reason the nitrate was not completely removed is probably a function of the depth to groundwater, which was close to 5 m during the driest sampling months. It is unlikely that many roots from the trees in the buffer extended this deep into the soil. Any nitrate removal was probably from the capillary fringe which would be less efficient in nutrient removal than having the roots in the groundwater. The low levels recorded for May and June 2006 can be attributed to unusually high groundwater levels when depth to groundwater increased from 4.4 m in May to 2.9 m in June (depth to groundwater was not measured for November 2004). Experimental Buffer Site Analysis of variance showed significant effects by date (P<0.0001), buffer type (P<0.0001) and distance (P<0.001) for nitrate. Only date (P<0.008) was significant for phosphate. For presentation purposes, buffer types were separated. The grass filter strip showed relatively high nitrate levels, up to 140 ppm, in the groundwater sampled at the field’s edge. By the time the groundwater reached the upper edge of the buffer, nitrate levels were observed to drop to levels consistently below 20 ppm and remained low across the buffer. The alder buffer showed a different response. First, the nitrate levels at field’s edge was generally lower except (<60 ppm) for November 2005 which had a level of 160 ppm. Once again the well at the upper edge of the buffer showed levels less than 20 ppm except for November 2004 and 2005. From November 2005 into 2006, the nitrate levels were higher in the buffer proper (10 and 25 feet), possibly from N input from the alders’ roots and decomposing leaves. The high levels recorded in November 2004 may also be attributed to the presence of the alder, but the trees were still small having been planted that April. The poplar buffer showed a relatively flat change in nitrate levels across the buffer except for November 2005. Even at the field’s edge, nitrate levels were below 40 ppm and decreased only slightly moving across the buffer. Soil Solution
Blueberry Farm Site

Analysis of variance showed significant main effects of date (P<0.0001) and distance (P<0.0001) from the blueberries for nitrate, but buffer type and depth were not significant. For phosphate, date and distance were significant (P<0.0001 and P<0.027, respectively). For nitrate, November and December 2005 had concentrations of 60 and 56 ppm, respectively, while the rest of the dates were between 12 and 26 ppm. For phosphate, June 2005 had a level of 0.8 ppm compared to the rest of the dates that ranged from 0.16 to 0.44 ppm. As with the groundwater levels, nitrate concentrations were highest in the blueberries at both 18 and 24 inches deep. Increasing nitrate levels with increasing depth is indicative of movement from the upper soil towards the groundwater. In the poplar buffer, nitrate levels dropped to less than 5 ppm at 25 feet from the blueberries and remained at or below this level until the sample point next to the slough. In contrast, nitrate levels were 37 and 27 ppm for the 18 and 24 inch depths, respectively. At the slough’s edge, nitrate level was at 3 ppm at 18 inches, but was 26 ppm at 24 inches. Mature Buffer Site Analysis of variance showed significant main effects of date (P <0.0001) and distance into the buffer (P<0.056) for nitrate, and for phosphate, date (P <0.0001) and distance into the buffer (P<0.029) were significant. Date was significant for nitrate due to a low value recorded in June 2006 (as was seen in groundwater levels) and a high phosphate level recorded in June 2005. Nitrate and phosphate levels at distances into the buffer showed no discernable trend. Experimental Buffer Site Analysis of variance showed significant effects by date (P<0.0001), buffer type (P<0.0001), distance (P<0.001) and depth (P<0.058) for nitrate. Only date (P<0.0001) and depth (P<0.05) were significant for phosphate. For nitrate, November and December 2005 had concentrations of 86 and 73 ppm, respectively, while the rest of the dates were between 9 and 36 ppm. For phosphate, June 2005 had a level of 0.8 ppm compared to the rest of the dates that ranged from 0.20 to 0.35 ppm. Changes in nitrate levels across the buffers showed the grass plots at both depths started low and remained relatively constant. In contrast, nitrate levels began high in the poplar buffer and then declined dramatically to the river’s edge at both soil depths. The alder buffers showed the possible effect to nitrogen fixation on input to the system. At both 18 and 24 inches, nitrate levels increased from the upper edge of the buffer to mid-buffer and then declined at river’s edge. Discussion For tree buffers to function in nutrient uptake, their roots must be near or in the groundwater. In this project, we had 3 sites that ranged in depth to groundwater. The shallowest groundwater was at the blueberry farm where the groundwater was periodically observed to be at the surface, but no deeper than 1.4 meter. Consequently, the greatest nitrate removal was observed at this site. It is speculated that both plant uptake and denitrification played important roles in reducing nitrate loading. On the other extreme, the mature buffer site had the deepest groundwater and we documented marginal nitrate attenuation by the buffer. At the experimental buffer site, the grass filter strip appeared to not affect nitrate moving across the buffer even though high concentrations were observed at the field’s edge. In terms of tree buffer function, neither the alder nor poplar trees were big enough at the end of the project to remove significant amounts of nitrate from the groundwater. However, red alder, a nitrogen-fixer, did show an indication that nitrogen was being added to the system, especially in 2006 as the trees occupied the site. Continued monitoring at this site over several years could have documented these effects, unfortunately attempts at securing additional funding from SARE were not successful. Shade Effects of Buffers Corn Analysis of variance showed that shading of corn silage by an adjacent tree buffer significantly affected both mean stem and ear dry weight, but had no effect on stem number or ear number. For stem dry weight, distance from buffer (P<0.0012) and orientation (P<0.0042) were highly significant as were distance from buffer (P<0.0004) and orientation (P<0.0284) significant for mean ear dry weight. Corn located 25 feet from the buffer showed reduced stem weight by 30 and 47% compared to corn growing greater than 150 feet for the N-S and E-W buffers, respectively. Stem decreased by 14 and 5% at 75 and 150 feet from the buffer, respectively, for Field 1 and by 4% at 75 feet in Field 2. Ear weight was impacted more by buffer shade, exhibiting decreases of 41 and 64% for the N-S and E-W buffers, respectively, at 25 feet from the buffer. At 75 feet from the buffer, ear weight decreased by 11 and 10% in Field 1 and 2, respectively. At 150 feet in Field 1, ear weight decreased by 9%. Orientation of the buffer relative to the corn crop also impacted stem weight. In Field 2, the crop at 25 feet from the buffer was in shade for most of the day whereas in Field 1, the entire crop received full sun from sunrise until mid-day. Stem weight in Field 1 had 65 more grams compared to Field 2. The difference lessened as the distance from the buffer increased. Ear weight in the 2 fields differed at 25 feet from the buffer by 27 grams, but no difference farther from the buffer. Blueberry Buffer shade significantly affected blueberry plant height and cumulative stem cross-section, but had no effect on the number of stems per plant or mean stem cross-section of each stem. Unlike the corn, transect at this site was found to be significant (Height: P<0.002; Cumulative cross-section: P<0.0172). Transect 1 was located in a field where the plants were established on hilled rows whereas plant in transect 2 were planted in rows without hilling. Because of the shallow groundwater at this site, hilling in transect 1 provided better root aeration and hence better plant growth, especially cumulative stem cross-section. Distance from buffer was significant for plant height (P<0.0134) with a decrease of 19 and 10% in transect 1 and 2, respectively for the plants closest to the buffer. For cumulative stem cross-section (P<0.0268), plants closest to the buffer exhibited a 57% reduction while in transect 2, this reduction was only 17% when compared to the plants beyond 170 feet Discussion The effect of buffers shading adjacent crops was found to be significant in the two crops our farmers grow. The magnitude of effect depends on the orientation of the buffer relative to the crop, distance from the buffer, crop type and the response variable. These findings have significant ramifications if a farmer is required to establish tree buffers that achieve tall stature as they mature. Yield reductions resulting from buffer shading can have an impact on an individual farm income. Objective 2
Installation and maintenance costs of the riparian buffers installed specifically for this research project were collected and adjusted to reflect costs that would be incurred on private farms. The total cost of installing and maintaining the buffer area over the three year period was approximately $7,112 on a per acre basis. 26% of these costs were associated with site preparation prior to establishing the buffer, 44% with actual establishment of the buffer, and 30% with its maintenance for three years after establishment. The majority of costs (48%) were for labor to perform work in all phases of creating the buffer. Thirty-eight percent (38%) of the total install and maintenance costs were for material, and 14% were for custom services associated with site clearing and preparation. Cost details can be found at: These results were corroborated with similar studies conducted by the economic PI in other parts of Washington State and independent buffer cost estimates for the Puget Sound region published by other groups in the state (Shared Strategy for Puget Sound).

The enterprise budget models showed that the economic impacts of the buffers extended beyond the basic installation and maintenance. In addition, the models demonstrated that the sources of increased buffer costs differed depending on the farming enterprise. In the case of potato farms the other largest impact on the enterprise, accounting for over 60% of the buffer’s total economic impact, was the lost revenue from previously farmed lands placed in buffers. Red and yellow potatoes are high per-acre value crops in the Skagit Valley. Total lifetime costs of the buffers were approximately $11,000 per acre on average for Skagit potato enterprises. A more detailed presentation of the Potato results can be found at the following web link:

Blueberries are an even higher value crop in the Skagit Valley, and this “income loss” effect was even more significant for this crop. In the case of perennial berry crops, impacts varied greatly depending on the age and agronomic practices used (e.g. machine vs. hand harvesting or organic vs. non-organic production methods). Other sources of buffer costs include increased fungicide treatment and increased harvest costs due to uneven fruit maturation on shaded plants. Other results can be found at the WSU Riparian Buffer website:

In the case of dairies, other major economic impacts included increased feed costs and waste management costs associated with removing land from silage production.

All the enterprise models developed allow the user to account for buffer revenues from the State-Federal buffer cost sharing program, Conservation Reserve Enhancement Program (CREP). The results of this research showed that this is program, when full payments were received, was effective at offsetting significant portions of buffer costs, particularly in lower value crops, such as silage. However, eligibility requirements for the program prevent many farms from accessing these funds, and major maintenance costs caused by beaver damage (a significant issue in this research project) are not covered. Instruction manual created for these models can be found in Appendix B-E.

1) Completion of four enterprise budgets that include the economic impacts of riparian buffers. These tools can be downloaded from the project website:

2) Numerous presentations to local groups related to agriculture, project presentation at national and statewide meetings, publication of results in professional journal, industry newspaper, and on project website. Many of the presentations are found on the project website:

3) Two workshops teaching hands-on use of the buffer budget tools, and three field days, demonstrating economic impacts. Materials and photos from the July 19, 2005 field day can be found at:

Research conclusions:

Buffer function in removing non-point pollutants coming from adjacent fields depends on the depth to the groundwater and the ability of the roots of trees in the buffer to interact with the groundwater to remove mobile nutrients like nitrates. Depth of groundwater could be used to target riparian areas for buffer establishment. If the groundwater is deep, the buffer will not provide proper function and other areas may be better suited to establish buffers. Of course, buffers do provide additional function such as shading of the watercourse, soil and bank stabilization and input of detritus to the stream system. As for species to use in buffers, both red alder and hybrid poplar were easy to establish and closed canopy by the second growing season which is important in reducing maintenance. Beaver predation on the trees significantly impacted the hybrid poplar, but this appears to be partly a result of the extent of flooding episodes we experienced on the Nookachamps River. The water rose to levels that allowed the beaver to float over the fence we installed along the river. A similar type fence on another river that doesn’t flood as much has protected the poplars from predation. Individual tree protectors, if tall enough (>4 feet tall) might also provide beaver protection at least until the stem becomes too big or the protector disintegrates. Using red alder in buffers may contribute additional nitrogen input to the stream through leaf litter and root exudation. Depending on the amount of nitrogen coming from the farm operation, red alder may not be the best species to use in a riparian buffer.

The finished economic models and user instructions are available for download from the project website. Individual farmers can use these models to assess their own farm and the viability of adding conservation buffers to their operation. Policy makers and regulators have access to the models to better understand the interactions of riparian buffers with the farm enterprise. Publication of the research article contributes to the (scarce) body of literature concerning the economic viability of riparian buffers.

Field visits and workshops at the research sites have provided the local farm and regulatory community with real data on how riparian buffers function and how they interact economically with the adjacent farm enterprise. The availability of this information has enabled farmers as well as policy makers to make more informed choices about conservation buffers. Integration of the buffer research sites into the High School education curriculum is giving students a basic knowledge of riparian buffer principles based on actual field data and visits to the research sites. Presentations by PIs to the farming community have created better awareness of the information being generated by the project.

Injecting objective research data and results into the riparian buffer policy debate has improved understanding by all parties including farmers, regulators, policy makers and environmental groups as to the important functions of riparian buffers and the magnitude and ways in which they can affect farm enterprises. This improved understanding was evident among field day and workshop participants. How this information is ultimately used to resolve this agricultural policy issue is unknown at this time.

Participation Summary

Research Outcomes

No research outcomes

Education and Outreach

Participation Summary:

Education and outreach methods and analyses:

Computer lab workshops covering use of the enterprise budget models were developed and delivered to approximately 30 farm owners and managers in November 2005 and February 2006.

Four field days were held in 2005 and 2006 for agricultural producers and high-school agricultural students in Skagit County, covering project methods and results, as well as hands-on demonstrations of project methods, such as water quality sampling.

PIs Johnson and Henri gave project presentations at several local farm group meetings, including the Skagit Agricultural Leadership Group, Skagit County Agricultural Advisory Board, and the Skagit Potato Growers Annual Meeting. Field sampling results and economic results were presented using PowerPoint. In addition, PIs Johnson and Henri gave presentations at the AWRA national riparian buffer conference in Lake Tahoe, CA in June, 2004 regarding general project design, and economic analyses, respectively.

Project website was updated and expanded for the final time in July, 2008 to include all relevant project documents. Expansion included addition of new and updated buffer economics information, project presentations, materials presented at field workshops, and additional buffer references. The website is:

Henri and Johnson’s article on the economics of hybrid poplar buffers (Bayview research site) was published in the Journal of Soil and Water Conservation in August, 2005.

An article was published in the regional agricultural newspaper (Capital Press) in April, 2005 on the research project, focusing on economic results.

Education and Outreach Outcomes

Recommendations for education and outreach:

Areas needing additional study

1) How to make small farm buffers profitable through selection of high value vegetation, adaptation of management practices, innovative product marketing and sale, and better access to cost sharing programs. 2) Identify state and federal regulatory barriers to buffer profitability, and propose changes and solutions. 3) Long-term studies of buffer function as it relates to management practices including harvesting of trees.

Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture or SARE.