Reducing Nitrogen Fertilizer Inputs to Irrigated Pastures and Hayfields by Interseeding Legumes

Final Report for OW10-309

Project Type: Professional + Producer
Funds awarded in 2010: $49,849.00
Projected End Date: 12/31/2012
Region: Western
State: Colorado
Principal Investigator:
Joe Brummer
Colorado State University
Expand All

Project Information


Suppressing grasses with glyphosate prior to interseeding resulted in the most consistent legume establishment. Close mowing to simulate heavy grazing generally did not result in improved establishment. Of the five legumes evaluated, alfalfa established the best in the glyphosate treatment in Colorado, increasing yield by over a ton per acre. In Idaho, establishment was more variable, with red clover establishing well regardless of suppression treatment. No legumes established in Oregon due to heavy rodent activity. This study highlighted the importance of suppressing the existing grasses and choosing a vigorous legume species for interseeding to reduce the risk of seeding failure.


Forage producers that primarily manage irrigated grass pastures or hayfields are struggling with how to maintain yields given the current price of nitrogen fertilizers. The price of nitrogen fertilizers has increased from $0.20 per pound of nitrogen a few years ago to a high of $0.80 to $1.10 per pound in 2008, with current prices in the range of $0.60 to $0.65 per pound. Nitrogen is the number one limiting nutrient for grass production, so it is essential to apply if producers want to maintain yields. Some producers have started to use manures and composts as alternative sources of fertilizer in areas where they are available. The high rates of application coupled with high transportation costs limit the use of these sources.

Another alternative is to introduce various legumes into grass dominated stands by interseeding. Legumes are known for their unique ability to fix nitrogen from the atmosphere through a symbiotic relationship between the plant and specific bacteria that infect the roots, forming nodules. Legume plants themselves benefit from the fixation of nitrogen, but the associated grass plants can also benefit as the nodules periodically sluff, decompose and release nitrogen into the soil for uptake by other plants. Productivity of grass-legume stands will generally never equal what a grass-only stand that is adequately fertilized with nitrogen will produce. However, the loss in productivity is offset by the higher quality forage that is produced by having a legume in the mix.

Interseeding generally involves the use of specialized drills that cut through the existing sod layer and place the seed in contact with the soil at the proper depth. Although seed can be broadcast into the existing stand, as a general rule, establishment success of forages is greater with drilling. With interseeding of legumes, the key is suppressing the existing grasses long enough for the new plants to establish. Only minimal establishment of the seeded species can be expected without suppression of the existing vegetation. Various methods have been tried over the years with varying degrees of success. Suppression methods that have been tried include close grazing with livestock, light disking or rototilling, flail mowing and various herbicides such as Roundup or Paraquat. Even with suppression of the existing grasses, it sometimes takes two to three years before the interseeded legumes reach full productivity. On the positive side, the cost of interseeding is considerably less compared to complete renovation of a stand using tillage. Additionally, a full season's production is generally not lost since the existing plants tend to recover quickly following suppression and can be harvested for hay or lightly grazed during the year of seeding.

Although basic knowledge of how to interseed pastures and hayfields currently exists, producers still have a number of questions on how to successfully implement these techniques. Without answers to their questions, they are reluctant to implement these practices. Questions commonly asked include: 1.) Which brand or type of drill is best to use?, 2.) Which legume species are easiest to establish?, 3.) Once established, which legume species are most persistent?, 4.) What time of year should I interseed to achieve the best results?, 5.) How do I effectively suppress the existing grasses to insure establishment of the legume I am interseeding?, and 6.) How is the yield and quality of my forage stand affected by interseeding a legume? This project attempted to answer a number of these questions posed by producers.

Project Objectives:

The overall objective of this project is to demonstrate to producers how legumes can be interseeded into existing grass-dominated pastures and hayfields, thereby increasing the quantity and quality of forage produced while reducing nitrogen fertilizer inputs. Both on-farm demonstration and small plot trials will be used to achieve the following specific objectives:

Objective 1 - Evaluate the establishment success of various legume species, including alfalfa, birdsfoot trefoil, red clover, white clover and sainfoin, interseeded into grass-dominated pastures and hayfields with and without mechanical, herbicidal and/or animal suppression of the existing vegetation.

Objective 2 - Evaluate the effects of introducing legumes into grass-dominated pastures and hayfields on forage yield and quality compared to fertilizing the existing grass dominated stand with nitrogen.

Objective 3 - Using inputs of seed, herbicide, fertilizer, labor, machinery, etc. associated with objective 1 and yield and quality outputs from objective 2, conduct a basic economic analysis comparing pastures or hayfields that have adequate (>20% by weight) legume composition to straight grass stands fertilized with nitrogen.

Objective 4 - Disseminate results of this project to other producers through such means as field days, workshops, conferences, on-line reports and other media, and the production of a how-to manual.


Click linked name(s) to expand/collapse or show everyone's info
  • Curtis Bridges
  • Mark Campell
  • Dr. Chanda Engel
  • Laura Hooper
  • Dr. Glenn Shewmaker
  • Edwin Smith
  • Jack Whittier


Materials and methods:

This project was conducted in cooperation with five producers at sites located near Fort Collins, CO (two sites), Kimberly, ID (two sites) and Klamath Falls, OR (one site). At each location, a small plot study was established in conjunction with one of the cooperating producers. The Colorado site was flood irrigated from gated pipe and was dominated by orchardgrass, meadow brome and smooth brome with minor amounts of tall fescue and perennial ryegrass. There were no existing legumes in the stand. The Idaho site was irrigated with a pivot sprinkler and was dominated by orchardgrass with minor amounts of smooth brome, perennial ryegrass and Kentucky bluegrass. There were some existing white clover and alfalfa plants (< 10%) in the stand. The Oregon site was characteristic of a mountain meadow dominated by meadow foxtail and Kentucky bluegrass. It was flood irrigated and there were no existing legumes present.

In addition to the small plot studies at each location, larger scale demonstration seedings were also established to evaluate the effectiveness of interseeding on a producer scale. The additional two sites in Colorado and Idaho were also used for this purpose.

For the small plot studies, a randomized complete block design with four replications was utilized at each site. Five legume species were tested: ‘Rugged’ alfalfa @ 8 lbs PLS/ac, ‘Norcen’ birdsfoot trefoil @ 6 lbs PLS/ac, ‘Shoshone’ sainfoin @ 18 lbs PLS/ac, ‘Starfire’ red clover @ 5 lbs PLS/ac, and ‘Kopu II’ white clover @ 3 lbs PLS/ac. Each legume was interseeded into plots treated in the following manner: no suppression (seeded directly into the existing vegetation), mowed to a two inch stubble height just prior to seeding to simulate close spring grazing, and sprayed with glyphosate herbicide at 1 qt/ac two weeks prior to seeding when existing vegetation had four to six inches of new growth. To evaluate the effectiveness of interseeding the various legumes, two types of controls were included: no fertilizer/no seed and nitrogen fertilizer applied at 80 lbs N/ac in the spring and 40 lbs N/ac after the first harvest (if harvested twice). Plot size was 6 by 20 ft.

All legume seed was inoculated with the appropriate Rhizobium bacteria prior to planting to insure nodulation. At the Colorado and Oregon sites, planting was done with a 5.6 ft. wide no-till drill equipped with double-disk openers and leading coulters (Model 3P605NT, Great Plains Mfg., Inc., Salina, KS). For seeding small plots, the drill was fitted with a cone seeder attachment (Kincaid Equipment Mfg., Haven, KS). At the Idaho site, a modified John Deere Powr-till drill was used for seeding. It was modified by reducing the width from 8 to 5.3 ft. and adding a cone seeder attachment. Legumes were interseeded in late May or early June of 2010 at all three sites.

Plots were harvested either once or twice in 2010 based on the normal practice for each location. The legumes did not reach sufficient size in 2010 to contribute to yield or quality. In May of 2011, success of legume establishment was measured at each site using a frequency grid method (Vogel and Masters 2001). Presence or absence of the seeded legume was determined in two 25-grid frames per plot and the counts converted to percent frequency. In 2011, yields were taken once in Colorado (July 2) and twice in Idaho (July 7 and October 4). Yields were not taken in Oregon due to poor legume establishment resulting from a major infestation of meadow voles. Forage quality was determined using near infrared reflectance (NIR). Plots were not harvested in 2012 at any of the sites due to poor establishment in Oregon and cooperating producers in both Colorado and Idaho decided to plow out their forage stands and plant corn.

Research results and discussion:

One of the biggest challenges that must be overcome when interseeding legumes into established grass stands is competition from the existing vegetation. There are numerous methods that can be used to suppress the existing grasses. The method that has given the most consistent results is spraying the grasses with a sublethal rate of glyphosate herbicide. Choosing the rate to use is always a tricky proposition. Too light and you don’t get adequate suppression and too heavy results in many of the grasses being killed. Experience has shown that most of the common pasture and hay grasses such as orchardgrass, smooth brome, meadow brome and tall fescue recover rather quickly following application of glyphosate at sublethal rates. Kentucky bluegrass is one grass that is easily killed with even light rates of glyphosate.

Timing of application of the glyphosate can also impact the results. Spring application and seeding are very common. The grasses tend to be more tolerant of glyphosate applied at this time of year because the plants are not storing carbohydrates. The glyphosate will burn the tops but will not be translocated into the root system, killing the plants. Fall application of glyphosate at equivalent or lighter rates compared to the spring tends to have a more severe impact on the grasses because both carbohydrates and the herbicide get translocated into the root system. It is not uncommon for almost 100% of the grasses to be killed with a fall application of glyphosate.

Besides glyphosate, paraquat herbicide has been used to suppress existing grasses prior to interseeding. This herbicide is less effective because it only burns the tops of the grasses off for a short period of time. They tend to recover too quickly and compete with the emerging legume seedlings. Other methods that have been tried include close mowing just prior to seeding using a rotary or flail-type mower, shallow disking or rototilling, and heavy grazing up until the time of seedling emergence. Producers are reluctant to use the more aggressive suppression methods such as glyphosate because they must sacrifice some of their pasture or hay crop during the establishment year. They prefer to use heavy grazing because their livestock benefit from the forage, but just like the results from using paraquat, the grasses tend to recover too quickly.

In this study, we chose to look at glyphosate since it is generally the most successful suppression treatment, close mowing to simulate heavy grazing because producers are really interested in that approach, and direct interseeding. Based on frequency measurements taken in the spring of 2011, suppression of the existing grasses with glyphosate was the only treatment that allowed for establishment of all five legume species at the Colorado site (Table 1). Alfalfa was the most successful with a frequency of 73% in the glyphosate treatment. Birdsfoot trefoil, red clover and sainfoin were intermediate with an average frequency of 33%. Even though white clover had the lowest frequency of occurrence, it will be able to spread over time due to its stoloniferous growth habit.

Unlike the Colorado site, results from Idaho were more variable (Table 1). Glyphosate did not stand out as the best suppression treatment. Averaged across legume species, frequency of occurrence was essentially the same between direct interseeding and suppression with glyphosate. Red clover and birdsfoot trefoil established well in the direct seeded plots with an average frequency of 33%. In the mowed plots, red and white clover established well. Alfalfa and red clover performed the best in the glyphosate treated plots. Red clover was definitely the most vigorous of the legumes seeded at the Idaho site, establishing well in all three suppression treatments with an overall average frequency of 39%.

A possible reason for the differential response among suppression treatments at Idaho compared to Colorado was that a different type of interseeding drill was used. Although the John Deere Powr-till drill does not disturb a lot of the existing vegetation, it does remove some of the roots in a narrow band (~3/4 of an inch) that the seed is dropped into. A double-disk opener drill with leading coulters like the Great Plains only slice the sod, which has little effect on suppressing the existing vegetation.

Although there were no legumes that survived at the Oregon site in the replicated small plot trial, there were some that did survive in an adjacent larger-scale demonstration area. In the glyphosate treated area, birdsfoot trefoil stood out as the best with a frequency of occurrence of 30%. Only minor amounts of alfalfa (2%) and white clover (3.5%) were present. Red clover was present but did not show up in any of the frames. In the mowed area, only minor amounts of white clover (2%) and birdsfoot trefoil (4%) were present. Unlike the small plot area, this area was exposed to grazing during the establishment year. Although grazing with its associated hoof action can lead to injury and death of seedlings, it also keeps the overstory shorter which reduces light competition. Legume seedlings become spindly and yellow if they are under light stress for too long. Seedlings in this condition are very tender and often die once the overstory is removed and they are exposed to harsher environmental conditions.

The main benefits of interseeding legumes into grass-dominated stands are increased forage yield and quality and a reduced need to apply nitrogen fertilizer. At the Colorado site, very few legumes established in the direct seeded or mowed treatments (Table 1) and this was reflected by no increase in yield (Table 2). Alfalfa established well in the glyphosate treatment and increased yield by over 2000 lbs/ac compared to the unfertilized control. The grass-alfalfa mix even out yielded the control fertilized with 80 lbs N/ac by over 700 lbs/ac. Red and white clover and birdsfoot trefoil that established in the glyphosate treated plots had an intermediate effect on yield falling between the unfertilized and fertilized controls. Even though sainfoin established well in the glyphosate plots (Table 1), the plants were small, did not compete well with the grasses and did not contribute to total yield.

As discussed earlier, establishment of the various legumes in relationship to the suppression treatments was quite variable at the Idaho site (Table 1). Red clover established well across all thre suppression treatments, and this was reflected by increased yield in those plots. In the glyphosate treatment, the grass-red clover mix yielded over 1600 lbs/ac more forage compared to the unfertilized control. Generally, the better a particular legume established, the greater the impact on total yield. Just like at the Colorado site, sainfoin did not perform well at the Idaho site.

When it comes to forage quality, the one parameter that tends to increase when legumes are interseeded into existing grass stands is crude protein content of the forage. Alfalfa, red clover and birdsfoot trefoil tended to establish the best at the Colorado site in the glyphosate suppressed plots as evidenced by the increased yield (Table 2) and had the greatest impact on crude protein (Table 3). Alfalfa was the most successful, increasing crude protein content of the forage by three percentage points compared to the control. Brummer (1999) found a similar response when birdsfoot trefoil was interseeded into a mountain hay meadow, increasing crude protein content of the hay by three percentage points (from 7.5 to 10.5%) when it made up about 25% of the composition.

Total fiber content of the forage as measured by NDF tends to decrease with the addition of legumes to grass stands. Alfalfa had the greatest impact on total fiber, lowering it by 9.5 percentage points in the glyphosate suppressed plots compared to the control (Table 3). The interseeded birdsfoot trefoil and red clover lowered total fiber by about half that amount (4.4 percentage points) in the plots suppressed with glyphosate compared to the grass control.

Although total fiber decreased with the interseeding of alfalfa, birdsfoot trefoil and red clover in the glyphosate suppressed plots, digestibility of the NDF fraction tended to decrease, especially for plots containing alfalfa and red clover (Table 3). This is due in part to the higher lignin content associated with legumes as they mature. Most of the legumes were in full flower at the time of harvest.

The recent development and use of the relative forage quality (RFQ) index takes NDF digestibility (NDFD) into account (Undersander and Moore 2002). Although total fiber (NDF) was lower in plots where the legumes established well, because NDFD of the forage was lower in those plots, we measured very little impact to RFQ (Table 3). The most noticeable change was a slight reduction in RFQ in the plots suppressed with glyphosate and seeded with red clover.

The biggest positive impacts to forage quality were in the plots fertilized with nitrogen in which crude protein increased, total fiber (NDF) decreased and digestibility of the NDF fraction increased in comparison to the unfertilized grass control, which resulted in an overall relative forage quality index of 152.

Literature Cited

Brummer, J.E. 1999. Where does birdsfoot trefoil fit in Colorado? Colorado State University Cooperative Extension, Agronomy News 19(4):1-4, Fort Collins, CO.

Undersander, D., and J.E. Moore. 2002. Relative forage quality. Univ. Wisconsin, Focus on Forage 4(5):1-2.

Vogel, K.P., and R.A. Masters. 2001. Frequency grid-a simple tool for measuring grassland establishment. J. Range Manage. 54:653-655.

Participation Summary

Research Outcomes

No research outcomes

Education and Outreach

Participation Summary:

Education and outreach methods and analyses:

Brummer, J.E., G.E. Shewmaker, and C.L. Engel. 2011. Challenges and benefits of interseeding legumes into grass dominated stands, p. 72-78. In: Proceedings of the 2011 Western Alfalfa & Forage Conference, Las Vegas, NV, 11-13 December 2011, Univ. Calif. Cooperative Extension, Davis, CA.

Brummer, J.E., G.E. Shewmaker, and C.L. Engel. 2012. Challenges and benefits of interseeding legumes into grass dominated stands. In: Proceedings of the 5th National Conference on Grazing Lands, 9-12 December 2012, Orlando, FL.

Shewmaker, G.E., J.E. Brummer, and C.L. Engel. 2013. Interseeding legumes into grass dominated pastures. Abstract #25, Society for Range Management 66th Annual Meeting, 4-7 February 2013, Oklahoma City, OK.

In conjunction with the two proceedings papers and one abstract cited above, results from this study were presented at those meetings as an oral presentation at the Western Alfalfa and Forage Conference (~700 attendees), as a poster at the 5th National Conference on Grazing Lands (~450 attendees) and as an oral presentation at the International Meeting of the Society for Range Management (~1200 attendees).

The principles of interseeding that resulted from conducting this study have also been incorporated into the following classes taught at Colorado State University by Joe Brummer: Forage and Pasture Management (~65 undergraduate students per year) and Integrated Forage Management (~16 master’s level students per year). The master’s level class is part of the Western Center for Integrated Resource Management Program and many of the students end up managing farms and ranches where they directly apply the type of information generated from this study.

Although the field days planned for 2012 did not happen because the cooperators plowed the test plots under to plant corn, each of the Extension Specialists involved in this project routinely utilize the information from this project in their dealings with producers. Pasture renovation, of which interseeding is one approach, is a common topic that all of us deal with, and having up-to-date information and examples to share with producers aids them in making informed decisions on which approach is best for their particular situation.

A manual on how to successfully interseed which incorporates findings from this and other studies is currently being drafted and will hopefully make it through internal review so that it can be published by early next year.

Education and Outreach Outcomes

Recommendations for education and outreach:

Areas needing additional study

Since producers are reluctant to use herbicides such as glyphosate to suppress the existing vegetation prior to interseeding, there needs to be a concerted effort among forage specialists to improve outreach that highlights this practice as the one that results in the most consistent establishment. In addition, there needs to be further development of methods to suppress the existing vegetation that don’t require the use of herbicides, such as mechanical methods similar to the John Deere Powr-till drill that is no longer manufactured.

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.