Perennial Grass Covers Affect Long-Term Soil Quality

Project Overview

Project Type: Research and Education
Funds awarded in 2000: $96,100.00
Projected End Date: 12/31/2002
Matching Non-Federal Funds: $99,010.00
Region: North Central
State: Iowa
Project Coordinator:
James Raich
Iowa State University

Annual Reports


  • Agronomic: barley, soybeans, wheat, grass (misc. perennial), hay
  • Additional Plants: native plants


  • Animal Production: feed/forage
  • Crop Production: agroforestry, nutrient cycling, tissue analysis
  • Education and Training: demonstration, extension, on-farm/ranch research, technical assistance
  • Natural Resources/Environment: hedges - grass, grass waterways, riparian buffers, riverbank protection, soil stabilization
  • Production Systems: agroecosystems, holistic management
  • Soil Management: organic matter, soil analysis, soil quality/health


    We investigated soil organic matter accumulations, soil respiration, and soil food webs in riparian grass filters on private farms in northern Story County, Iowa. We specifically compared soils beneath planted prairie grasses to soils beneath non-native forage grasses. Soil organic matter pools varied more between farms than among grassland types. Soil respiration rates were similar among old prairie-grass stands and long-established cool-season grasses. Soil food webs also were similar beneath the two grass types. Planting of native prairie species in filter strips is as effective as is planting non-native forage grasses, but provides greater grass production while perpetuating native species.


    It is widely recognized that grass forages vary in quality as feedstuffs for grazing animals. On-farm research by our group suggests that grass species also differ in their impacts on soils, and that their effects may be predictable based on the well-known dichotomy between cool- and warm-season grasses. Specifically, our previous research suggests that cool-season grasses may generate high-quality, short-lived soil organic matter, and support high rates of nitrogen cycling by a diverse soil food web. Warm-season grasses, in contrast, may promote long-term soil organic matter accumulation, but have slower rates of nitrogen cycling, lower nitrate losses, and less diverse soil food webs. Our previously collected data, however, reflect short-term trends only. A key question that we seek to address on our Bear Creek Project, a National Restoration Demonstration Watershed, is whether warm-season or cool-season grasses are more effective over the long term for use in CRP and riparian buffer plantings.

    The present emphasis on landscape sustainability, supported by such programs as the Conservation Reserve Program, has led to an estimated 14.7 million ha (36 million acres) of U.S. cropland being converted to perennial grasses as of 1995 (Barker et al. 1996). At the same time interest is increasing in providing farmers with carbon credits for increasing soil carbon sequestration. Our goal in this proposed research was to seek a better understanding of how planting of cool- versus warm-season grasses impacts long-term soil quality, as evidenced by rates of soil C cycling and soil organic matter accumulation. Identifying the long-term impacts of grass type on soil quality will help landowners and extension personnel select and manage grasslands to maximize their positive soil impacts. Our team, which includes ten farmers and landowners, has been at the forefront of buffer design, implementation and extension. The research we undertook provided for baseline measurements that can be continued into the future, to provide the long-term data needed to better define best management practices for riparian filter strips in the Midwest.


    The species composition of grassland communities is increasingly influenced by human activities, but the consequences of species changes on soil ecosystems remain poorly understood. Widespread planting of grasses in filter strips, riparian buffers, and Conservation Reserve Program lands has been underway in the Midwest since 1985. The types of grasses that are established in these new grasslands is not a function of climate or soils, but of direct human management. Typically, either native, warm-season prairie grasses or non-native, cool-season forage grasses are established but, we contend, the type of grass being planted has important implications. The types of grasses established can be expected to influence soil properties (e.g., D'Antonio and Vitousek 1992, Vinton and Burke 1995, Burke et al. 1997, 1999). Cool-season forage and warm-season prairie grasses differ in physiology, phenology, phytochemistry, and nutrient use. What effects these functionally different grass types have on soil properties is not adequately known, and apparently varies with time since establishment (e.g., Corre et al. 1999). The information we seek is needed so that landowners, extension personnel, and policy makers can make informed decisions about land use options. It is also needed to improve the effectiveness of planted riparian buffers, and to evaluate the impacts of land use changes on regional biogeochemical cycles. We suggest that cool-season and warm-season grasses represent distinct functional groups whose differing characteristics generate predictable shifts in soil properties that may be important to land owners and landscape managers, and which influence the quality of the environment.

    The research we undertook was driven largely by our preliminary work in riparian buffers, with previous funding from the EPA, USDA, USGS, SARE, Iowa Department of Natural Resources and Iowa's Leopold Center for Sustainable Agriculture. In evaluating the effects of different riparian plant communities on soil properties and N processing, we have obtained intriguing data from our replicated stands of switchgrass (Panicum virgatum, a warm-season prairie grass) and mixed cool-season grasses (primarily Bromus inermis, Poa pratensis, Dactylis glomerata, Phleum pratense, and Phalaris arundinacea). End-of-season aboveground biomass was nearly three times greater in switchgrass plots (1600 g/m2) than in the cool-season plots (550 g/m2) (Tufekcioglu 2000). Live fine root biomass was also consistently greater in the switchgrass than in the cool-season plots (Tufekcioglu et al. 1999). However, total soil respiration was significantly greater in the cool-season plots than in the switchgrass plots (Tufekcioglu et al. 2001). We therefore had the paradox of apparently higher productivity and root biomass, but less soil respiration, in switchgrass than in cool-season grasses. These findings suggested substantial soil C sequestration in the switchgrass stands, but soil C pools were lower in switchgrass than in the cool-season plots (Marquez et al. 1999). Furthermore, surface-soil (0-15 cm) microbial biomass averaged >200 mgC/kg soil higher in the cool-season plots than in switchgrass (Pickle 1999), and protozoan populations were substantially higher on two sampling dates (unpublished data). These findings suggested faster rates of soil C and N turnover beneath cool-season grasses than beneath switchgrass. Differences among plots in soil ammonium and nitrate concentrations have not been consistent, but switchgrass soils have shown a propensity for net N immobilization during the winter months.

    These different findings may reflect the fundamental differences between cool-season, C3, and warm-season, C4, grasses, and temporal differences in their effects on soils. The C4 photosynthetic system that typifies many of the dominant grasses of our native prairies includes a suite of biochemical, physiological, and morphological adaptations that influence the way that plants respond to environmental conditions (Pearcy and Ehleringer 1984, Ehleringer et al. 1997). Many publications compare the characteristics of C3 and C4 plants, with a focus on their aboveground physiological differences and habitat preferences (e.g., Black 1971, Teeri and Stowe 1976, Tieszen et al. 1979, Pearcy et al. 1981, Pearcy and Ehleringer 1984). Overall, the C4 pathway appears to be most favorable at high temperatures (Ehleringer 1978, Doliner and Jolliffe 1979, Boutton et al. 1980), and C4 prairie grasses are therefore called warm-season grasses. Cool-season, C3 grasses, in contrast, are favored under cooler conditions (e.g., Teeri and Stowe 1976, Boutton et al. 1980, Rundel 1980). In mixed grasslands, the relative dominance of cool-season grasses declines, and that of warm-season grasses increases, in late summer, when conditions favor C4 growth (Williams and Markley 1973, Redmann 1975, Boutton et al. 1980, Ode et al. 1980, Barnes et al. 1983).

    One result of the CO2-concentrating mechanism present in C4 plants is their tendency to have less Rubisco in their leaves, and the nitrogen use efficiency of photosynthesis is typically higher in C4 (warm-season) than in C3 (cool-season) plants (Brown 1978, Pearcy and Ehleringer 1984). This, coupled with differences in leaf anatomy and greater C allocation to non-leaf structures (Akin 1989), results in warm-season grasses generally having lower-quality tissues than do cool-season grasses. In our sites, lignin concentrations were higher and N contents were lower in switchgrass than in the cool-season grasses. The mean N content of cool-season grass forages in the U.S. is 1.8% (n=43) and that of warm-season grasses is 1.3% (n=53). The mean lignin contents of cool-season and warm-season grasses are 7.7% (n=22) and 11.1% (n=21) (derived from data in Miller 1958). These values translate into lignin:N values that are almost twice as high in warm-season grasses than in cool-season grasses (8.4 vs. 4.3, respectively). Cool-season grasses are favored over warm-season grasses as forages because of their higher nutritive values. Lignin contents, and lignin:N, are also primary factors controlling rates of organic matter decomposition in terrestrial ecosystems (Meentemeyer 1978, Kaplan and Hartenstein 1980, Melillo et al. 1982, McClaugherty and Berg 1987, Parton et al. 1987). The same factors that allow cool-season grasses to be more easily digested by ruminant microbes also favor their consumption by soil organisms. We suggest that these differences have important ecosystem consequences. Specifically, we propose that physiological, morphological, phenological and phytochemical differences between cool-season and warm-season grasses modify soil detrital dynamics and alter soil properties, and that their effects on soils vary with time since establishment. These issues are currently important because grassland establishment is driven in part by their potentials for soil improvement, C sequestration, and the immobilization of nitrate in agricultural runoff.

    This SARE-funded project provided funds to address three important questions concerning the comparative dynamics of grassland soils as influenced by either cool-season or warm-season grasses: (1) Do warm-season grasslands have a greater long-term potential for soil C sequestration than do cool-season grasslands? (2) Do cool-season grasslands support a more diverse soil food web than do warm-season grasslands? and (3) Are total soil respiration rates higher in cool-season than in warm-season grasslands? Our overarching goal was the continued development of the information base needed to provide meaningful recommendations to land owners and land managers interested in long-term soil improvement and soil quality restoration in the central United States.


    Tilman and Wedin (1991) grew three cool-season species and two warm-season species in monoculture along a N gradient. They found no significant differences in aboveground mass after 3 yr, but the warm-season species had substantially higher root biomass than did the cool-season species. Wedin and Pastor (1993) found higher rates of net N mineralization beneath cool-season than beneath warm-season grasses, but Pickle (1999), using laboratory assays, found no such differences in our sites. Wedin (1995) measured mass loss and N dynamics in decomposing litter of one warm-season (Schizachyrium scoparium) and three cool-season grasses. Schizachyrium litter and roots had initially higher C:N than all cool-season species, but these differences disappeared after about 30% mass loss, and the C:N of all species approached 20 after 60% mass loss. However, 60% of the Schizachyrium litter remained after 2 y and immobilized N, whereas the cool-season species decomposed faster and did not immobilize N. These data are consistent with our own findings of faster C and N cycling in cool-grass sites, but lower detrital turnover rates in warm-season grasslands should promote C sequestration over the long term.

    Federal government incentives have initiated large-scale conversion of cropland to grass filter strips, grass waterways and long-term pasture largely dominated by perennial grass species. Native warm-season grasses are being planted in riparian areas and CRP ground because their extensive root systems could potentially lead to substantial increases in SOC (Corre et al. 1999). A number of studies in the past 10 to 15 years have attempted to estimate new soil C inputs from warm-season vegetation growing in predominantly cool-season soils. One of the first to use this tool, Balesdent et al. (1987) estimated that 22% of the total soil C turned over after 13 years when corn (Zea mays, warm-season) was planted in soil with a predominantly cool-season signature. Gregorich et al. (1995) confirmed these data in their report that 30% of the total soil C in the plow layer (0-27 cm) had turned over after 13 years of corn growth in a forest-derived C3-soil. In perennial Australian warm-season grasslands planted into soils that had developed under subtropical rainforest vegetation, 33% of the original soil C had been replaced by grass-derived C after 35 years (Skjemstad et al. 1990). Wedin et al. (1995), working with monocultures of native tallgrass prairie bunchgrass (warm-season) growing in soils with a predominantly cool-season signature, reported that 22% of the total soil C was replaced by new C within 4 years. In contrast, only 10% of the total soil C turned over in 4 years for perennial cool-season species growing in the same soil. This suggests that native warm-season prairie grasses are more efficient at sequestering new C than cool-season grass species. Corre et al. (1999) utilized a change in vegetation to quantify the rate of accumulation of warm-season SOC in a C3-labeled soil. After 16-18 years, 53-72% of the total soil C had turned over. The potentially slow accumulation of warm-season-derived SOC is an important consideration when using warm-season grass to restore riparian and conservation areas.

    Soil food webs consist of organisms that live all or part of their lives belowground. Energy and nutrients are transferred through the various trophic levels as organisms feed on one another. Many scientific studies in the past 25 years have demonstrated that microbial and faunal interactions have significant impacts on the cycling of carbon, nitrogen, phosphorus and sulfur in soils (Coleman et al. 1977, 1978, 1983; Parker et al., 1984; Whitford et al., 1983; Hunt et al., 1987; Moore et al., 1988; Gupta and Germida 1989). Soil protozoans have rapid generation times, turning over an average of 10 to 12 times in a growing season, which is significantly more rapid than other soil biota (Clarholm, 1985). Protozoa are important in N cycling because they excrete bacterially immobilized N as ammonia (Anderson et al. 1981). Numerous studies have shown that protozoan grazers stimulate C turnover and ammonium-N release in the soil (Coleman et al 1977; Woods et al. 1982; Anderson et al., 1985). Kuikman et al. (1990) further observed that N uptake by plants also increased from 9 to 17% when protozoa were present. Our preliminary sampling (unpublished data) suggests that protozoan populations are more than twice as high beneath cool-season grasses than beneath switchgrass.

    Preliminary studies on our sites also showed that total microbial biomass-C was twice as high under cool-season grass filters (327 mg C kg soil-1) than under switchgrass (161 mg C/kg soil) (Pickle, 1999). It is believed that the continual litter input found under cool-season grass was the reason for the elevated biomass-C. The high C:N switchgrass roots are slow to decompose, and microbial growth is limited when compared to that found under cool-season grass. We have also found higher macroaggregation under the cool-season grass (Marquez et al. 1999). Macroaggregates are complexes of soil particles and humic compounds that are by-products of microbial biomass. Soil respiration rates were also higher under the cool-season grass than under the switchgrass (see below). These data are also consistent with our working hypothesis that cool-grass stands promote rapid soil C and N cycles, potentially at the expense of long-term C storage.

    Soil respiration refers to the total amount of CO2 produced by a soil, including that produced by soil macrofauna, microbes, and live plant roots (Schlesinger 1977, Raich and Schlesinger 1992, Rustad et al. 2000). It is, therefore, a measure of the total biological activity within an aerobic soil. Measurements of soil respiration rate are particularly valuable as they reflect the interactions among a suite of characters that influence overall soil quality, including soil aeration, soil water and temperature regimes, root dynamics, and decomposer activities. Soil respiration is also the single greatest flux of carbon out of soils (Schlesinger 1977, Raich and Schlesinger 1992), and is therefore a key component of soil carbon budgets (e.g., Schlesinger 1977, Raich and Nadelhoffer 1989). Bremer et al. (1998) measured soil respiration in a variety of Kansas grasslands, and found that total soil respiration correlated positively with aboveground productivity. Among grasslands of the world, total soil respiration rates average 70% higher than aboveground productivity (Raich and Tufekcioglu 2000). In our Bear Creek study sites, soil respiration rates were significantly higher beneath cool-season grasses than beneath switchgrass, but aboveground biomass production and fine root biomass were greater in the switchgrass plots (Tufekcioglu et al. 1999).

    The lower rates of soil respiration observed in our sites may reflect their lower soil C pools (Marquez et al. 1999). We suspect that the cool-season grasses produce much more labile detritus that decomposes very quickly, stimulating CO2 production, whereas the switchgrass produces long-lived, low-quality detritus that is more slowly processed by soil organisms but which, in the long term, sustains higher levels of soil organic matter. In other words, we believe that cool-season and warm-season grassland soils differ in their temporal dynamics. By producing high-N, low-lignin detritus, cool-season grasses may quickly stimulate soil C formation but, because of rapid C cycling, have low rates of soil C sequestration over the long term. Warm-season prairie grasses, in contrast, with long-lived root tissues, may more slowly contribute to the soil C pool, but may achieve overall higher levels of soil organic matter, as found in our native prairie soils. If this is true, both soil OM storage and soil respiration rates would be higher beneath cool-season grasses during the first years following establishment, but would be higher beneath warm-season grasses over the long term.

    These ideas are consistent with data from the northeastern United States, where cool-season grasses were found to have higher SOM contents during the first decade after establishment, but switchgrass was believed to have higher soil C sequestration rates after 15 years (Corre et al. 1999). In Saskatchewan, 50-year-old successional prairies dominated by warm-season grasses had higher SOM, total soil N, and available soil N pools than did similarly aged cool-season grasslands (Christian and Wilson 1999). They speculated that lower soil C pools beneath the cool-season grass, which were planted over millions of hectares, left about 4 * 10^14 g of C in the atmosphere which would have been stored in soils had native, warm-season species been planted. Identifying the temporal dynamics of soil C storages and cycling are fundamental questions we wish to address in this current research. Unlike the NE or Saskatchewan, central Iowa's environment is very favorable to warm-season grasses, and we expect that their effects on soils will be more quickly observable. Because most of the C lost from soils is in the form of CO2 (Schlesinger 1977; Raich and Nadelhoffer 1989), measurements of soil respiration are integral to the development of the long-term soil C budgets that are needed to address these issues.

    Project objectives:

    Objective 1: To determine if cool-season and warm-season grasses differ in their soil organic carbon (SOC) sequestration potentials.
    Hypothesis 1a: Soil organic carbon (soil organic matter) accumulations are greater beneath warm-season grasses than beneath cool-season grasses when the entire soil profile, to 1 m depth, is considered.
    Hypothesis 1b: Surface-soil (0-35 cm) organic carbon pools are greater beneath cool-season grasses than beneath young warm-season grasses, but this difference declines as the warm-season grass stands mature.
    Hypothesis 1c: Soil organic matter pools increase through time in warm-season grasslands.

    Objective 2: To compare the soil food webs present beneath cool-season and warm-season grasses.
    Hypothesis 2: Surface soils beneath cool-season grasses support larger populations of more diverse soil organisms than do those beneath warm-season grasses.

    Objective 3: To compare overall soil quality, as quantified by total soil respiration rate, in cool-season and warm-season grasses.
    Hypothesis 3: In situ soil respiration rates are higher beneath cool-season grasses than beneath warm-season grasses.

    Objective 4: To develop best management practices for perennial cropping systems in reserved lands, filter strips, and riparian buffers.

    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.