Decomposition of Forage Species Mixtures in Pasture has Inconsistent Effects on Soil Nitrogen

Benjamin F. Tracy and Matt A. Sanderson, 1102 S. Goodwin Ave., Department of Crop Sciences, University of Illinois, Urbana 61801

Abstract

Decomposition of plant material in pastures influences the amount of plant-available nutrients in soil. Most studies have evaluated decomposition dynamics using single plant species, but different species often occur mixed together in pasture. Mixing plant material may either slow or increase decomposition rates. The ability to predict how different plant materials decompose in the field is important for nutrient management. Our study objective was to determine how mixing plant material from three pasture species (tall fescue, red clover, and chicory) affected decomposition dynamics and soil N availability. Field decomposition was measured at sites in Pennsylvania and Illinois. A laboratory incubation evaluated soil nitrogen availability. Although decomposition dynamics differed at the two sites, mixing plant materials had no effect on decomposition rate. In contrast, soils amended with mixed plant material immobilized more ammonium and nitrate (35 to 60%) than expected based on single species assays. Our results suggest that decomposition and subsequent nitrogen availability from mixed species plant material are not predictable based on decomposition and nutrient release patterns of the individual species. A better understanding of interactions between plant decomposition and nitrogen availability could help producers manage pastures to enhance nutrient retention.

Introduction

Microbial decomposition of plant material is important in functioning of pastures because it determines rate at which nutrients in decaying plant matter are either immobilized or released into soil. In multi-species pasture, plant materials of different species usually decompose together. In such a situation, plant materials that differ in quality (e.g., N concentration) may interact to produce decomposition patterns that differ from those predicted from species that decompose alone (1,2,11). Thomas (9) outlines reasons why this may occur. A mixture containing high and low quality plant material may produce a more uniform resource base for microbial activity, which, in turn, may speed decomposition of poorer quality plant material in the mix. Certain plant material types also may improve the moisture balance of a mixture which might accelerate overall microbial decomposition because of the more favorable microclimate. Plant material quality may affect the pH of a mixture to favor a more balanced ratio of fungal and bacterial populations to increase overall decomposition rate. Lastly, high quality plant materials (e.g., low C:N) may attract microfauna to mixtures where they may consume poorer quality plant materials as well.

Predicting how plant material of different plant species interacts during decomposition is important. In low input, multi-species agroecosystems like pasture, positive or negative interactions among plant material types during decomposition could affect primary productivity and forage nutritive value (12). Reliance on specific plant mixtures to maintain soil fertility could reduce the use of chemical fertilizers and increase pasture sustainability. Some combinations of plant material may also improve nutrient retention in pasture if that material immobilizes nutrients (5) that would otherwise be lost to leaching, surface run off, or other pathways.

In this study, we wanted to determine if mixing plant material from different forage species would affect decomposition dynamics and soil N availability. We used three forage species that are common to many temperate pastures in the United States: Festuca arundinacea (tall fescue), Trifolium pratense (red clover), and Cichorium intybus (chicory). These species were chosen because they represent different functional groups (i.e., plant types with similar morphology, physiology, and function in an ecosystem, such as perennial grasses, legumes, and broadleaf forbs) that characterize the range of species types found in most pastures. We evaluated functional groups because results from recent ecological studies strongly suggest that number of plant functional groups (e.g., perennial grasses, legumes) present in a grassland communities has inordinately large effects on ecosystem processes like decomposition and primary production (6,8,10). These species also were expected to have widely different nutritive values and thus should have the most potential for showing negative or positive effects on decomposition when mixed together (9). We hypothesized that decomposition of low quality (high C:N) tall fescue plant material would be accelerated by addition of higher quality (low C:N) red clover or chicory plant material. We expected the largest effect on decomposition rate would occur when all three species were mixed together.

Measuring Affect of Species Mixtures on Decomposition and N Availability

Plant material decomposition was evaluated using a "litter bag" technique (11) in two field locations. The first study was conducted in 1997-1998 at the Penn State University Haller Farm near State College, PA (40°N, 77°W). The same experiment was repeated in 2001-2002 at the University of Illinois, Crop Sciences Research and Education Center in Urbana, IL (40°N, 88°W). Climate at both sites is continental, humid and temperate. During the field portion of this study, the Illinois site was generally characterized by more extreme temperatures and less precipitation than the Pennsylvania site (Table 1). Soils at the Illinois and Pennsylvania sites were classified as Drummer silt loams (fine-silty, mixed, superactive mesic, Typic Endoaquolls) and Hagerstown silt loams (fine, mixed, semiactive, mesic, Typic Hapludalfs), respectively. Soils at the Illinois site (15-cm depth) indicated a pH pf 6.3 while P and K averaged 41.6 mg/kg and 271 mg/kg, respectively. Soil tests at the Pennsylvania site (to a 15-cm depth) indicated a pH of 5.8, with 29 mg/kg available P, 185 mg/kg available K. In Pennsylvania, stem and leaf material were collected from live plants growing on several local pastures used for beef cattle research. Plant material for the Illinois study was generated from a common garden planted with the three species in Urbana, IL. Plant material was air dried at 20°C for seven days before placement into litter bags to assay decomposition. Plant material in litter bags was coarsely chopped (mean size 20 mm) and consisted of approximately 1:1 ratio of leaf and stem components. Six different plant material treatments were evaluated including three single-species treatments of tall fescue, red clover, and chicory and three mixed treatments that included:

(1) tall fescue + red clover

(2) tall fescue + chicory

(3) tall fescue + red clover + chicory

Table 1. Total precipitation and mean air temperature from Pennsylvania and Illinois field sites during field decomposition trials.

Month	State College, PA		Month	Urbana, IL
Month	Precip. (cm)	Air Temp. (°C)	Month	Precip. (cm)	Air Temp. (°C)
Nov-97	14.5	3.5	Nov-01	4.8	9.6
Dec-97	7.4	0.8	Dec-01	5.6	1.6
Jan-98	5.5	1.9	Jan-02	7.1	1.1
Feb-98	9.1	1.4	Feb-02	7.4	0.6
Mar-98	11.3	4.1	Mar-02	7.8	2.3
Apr-98	12.8	10.6	Apr-02	10.6	11.5
May-98	16.0	17.6	May-02	15.9	14.9
Jun-98	7.2	19.2	Jun-02	7.0	23.4
Jul-98	9.5	21.5	Jul-02	6.9	26.3
Aug-98	8.1	21.8	Aug-02	18.4	24.0
Sep-98	4.6	19.1	Sep-02	4.5	21.2
Oct-98	6.8	11.7	Oct-02	8.0	11.4
Nov-98	2.9	5.8	Nov-02	2.1	4.1
Total/mean	116.3	10.7	Total/mean	106.8	11.7

Each litter bag (10 × 10-cm, 1-mm nylon mesh) received 3 g dry weight plant material divided equally among the respective species. Litter bags were sewn shut and placed in field plots in November 1997 (Pennsylvania) or 2001 (Illinois). The field plots were 1 m² and established near where plant material was originally collected. Four plots were established for the PA trial and this number was increased to six for the IL study. Existing vegetation (mostly tall fescue and Kentucky bluegrass) was cleared from plots by manually pulling plants and smoothing the soil surface. Litter bags then were anchored to the mineral soil surface. Each plot received five replicates of the six treatments and litter bags were arranged randomly within each plot. One set of litter bags representing each treatment was collected from each plot on days 30, 120, 210, 300, and 390. After collection, plant material was removed from each bag, thoroughly washed with water so that no observable soil remained and then air dried at 50°C for 48 h and weighed. Carbon and nitrogen concentrations of plant materials were measured using a Carlo Erba CNS autoanalyzer. Plant material from the three species (n = 4 or 6 per species) was also measured for C and N to determine initial concentrations.

Laboratory study to evaluate soil nitrogen. The same plant material and treatment combinations used for the field experiment were replicated for a laboratory incubation bioassay. The purpose of the laboratory experiment was to evaluate how the different plant material mixtures affected ammonium and nitrate availability in soils. The lab bioassay was conducted at the USDA Pasture Systems and Watershed Management Unit in University Park, PA and not repeated in Illinois. Plant materials were coarsely chopped as before and added to 1-liter mason jars (six per treatment) that contained 60 g of soil collected from the same location as plant materials. Mason jars received 15 g of plant material equally divided among the respective species. The plant material amounts used in the assays were chosen to achieve a constant 1:4 plant material:soil wt ratio. This ratio is higher than would be expected in the field, but was chosen mainly to make sure we had excess plant material in the assay to see clear effects on soil N. One set of jars received no plant material and contained soil only. Plant material and soil were mixed, and the weight of the mixture was increased by 50% using tap water. Jars were covered with parafilm and incubated in the dark at 20 to 22°C for 36 days. At the end of the incubation, 10 g of the soil/plant material mixture was extracted with 50 ml of 2N KCl for ammonium and nitrate determination. Ammonium and nitrate concentrations in the extract were measured using a Lachat Quickchem Autoanalyzer.

Calculations for plant material mixtures. Observed values for mixed plant material treatments were evaluated by comparing them to expected values calculated from measurements made on respective monoculture plant material treatments. For example, observed values for tall fescue + red clover mixtures were compared to values of those species in monoculture when no interactive effects were present. For each collection date, expected values (E) for mass loss and soil ammonium and nitrate concentration were calculated as follows (11):

	S
E =	Σ	V_i / S
	i=1

where V_i = respective variable in monoculture treatment of species i and S = total number of species in mixture.

Differences in mass loss among treatments were evaluated using a split plot ANOVA with plant material treatment and block (plot) as main effects and time (day of sampling) as a subplot effect. Ammonium and nitrate and initial plant material C and N concentrations were analyzed with one-way ANOVA. Differences between observed and expected values in plant material mixtures were tested using t-tests. Statistical significance was considered at P < 0.05.

Plant Material Quality

Plant materials of the three species differed between locations at the start of the field bioassays. In Pennsylvania, red clover, mainly because of its high N concentration, had the lowest C:N followed by chicory and tall fescue (Table 2). Plant materials of all three species in Illinois had relatively high N concentrations compared with PA site (Table 2). This difference may be related to more fertile soils in Illinois where the forages were grown.

Table 2. Plant material nitrogen and carbon concentration of three forage species at start of respective field bioassays. Data are means ± 1 SE.

Site	Species	Nitrogen (g/kg)	Carbon (g/kg)	C:N
Pennsylvania	Chicory	20 ± 0.6	401 ± 0.2	19 ± 0.5
	Red Clover	37 ± 0.4	438 ± 0.3	12 ± 0.1
	Tall Fescue	20 ± 0.2	422 ± 0.6	21 ± 0.3
Illinois	Chicory	35 ± 0.8	425 ± 0.1	12 ± 0.2
	Red Clover	38 ± 0.2	447 ± 0.2	11 ± 0.4
	Tall Fescue	33 ± 0.9	430 ± 0.1	13 ± 0.4

In the field, single species litter bags containing chicory lost the most mass during the 390-day period (Fig. 1, A and B). Decomposition patterns were different among the species at the Illinois site. Tall fescue decomposed more slowly (P < 0.05) than red clover and chicory up to day 300 (Fig. 1B). This result was surprising given the high N concentration of the fescue plant material. The tall fescue cultivars sown at both sites were mostly endophyte free (E-), so it is unlikely that site differences can be attributed to endophyte status. Fescue was not tested for endophyte infection in this study, however, so this remains a possibility. In contrast, species with the lowest C:N ratio at the PA site, red clover, lost the same amount of mass after 390 days as tall fescue, which had the highest C:N ratio. Although we did not measure it in this study, the result may be related to lignin concentration in red clover. Buxton and Russell (3) found that red clover stems had lower cell wall concentrations but higher cell-wall lignin concentrations than orchardgrass (Dactylis glomerata) or smooth bromegrass (Bromus inermis). Overall, the C:N ratio of plant materials was a poor predictor of decomposition rate (1). Other plant material quality indices, like lignin concentration and N bound to cell wall components are likely better predictors of decomposition in pasture systems (5).

Fig. 1. Percent plant material mass remaining on days 30, 120, 210, 300, and 390 for single (A, B) and mixed species (C, D) plant materials in the field. Each data point is a mean with error bars 1 SE. Abbreviations: tf = tall fescue, rc = red clover, and ch = chicory.

Decomposition of Mixed Plant Material

Among the mixed species treatments, no differences in decomposition rate were noted during either 390-day period at Pennsylvania or Illinois (Fig. 1, C and D). We found few consistent interactions, positive or negative, among plant materials from the three plant functional groups evaluated in this study. At both sites, decomposition of mixed plant material was generally similar to what would be expected from those species decomposing alone (Table 3). Similarly, nitrogen disappearance of mixed plant materials showed no consistent trend between observed and expected values (data not shown). The decomposition dynamics of mixed plant material has been difficult to predict from nutritive value indices (1,2,4,11). For example, Wardle et al. (11) evaluated decomposition and N concentration of 32 plant species that represented four functional groups (grasses, forbs from grasslands and cropping systems and trees) mixed together in various combinations ranging from 2 to 8 species. Mixtures of different plant functional groups affected decomposition and litter N, but mainly in ways that could not be predicted from the singles species treatments. Increasing the number of species represented in the plant material from 2 to 8 did not affect the measured plant material variables in any consistent way.

Table 3. Observed and expected values for mixed plant material treatments. Columns indicate mass remaining in litter bags for day 30, 120 and 210. Data from days 300 and 390 are not included but showed similar patterns. Initial mass was 3.0 g for all bags. P values are based on individual t-tests. NS refers to P < 0.05. Values are means with 1SE.

	Pennsylvania site, Mass remaining (g dry wt)			Illinois site, Mass remaining (g dry wt)
	Observed	Expected	P	Observed	Expected	P
Day 30
Tall fescue- Red clover	1.9 ± 0.06	1.9 ± 0.02	NS	2.1 ± 0.06	2.3 ± 0.06	NS
Tall fescue- Chicory	1.7 ± 0.05	1.7 ± 0.02	NS	2.2 ± 0.07	2.3 ± 0.07	NS
Tall fescue- Red clover- Chicory	1.7 ± 0.05	1.7 ± 0.03	NS	2.4 ± 0.02	2.4 ± 0.07	NS
Day 120
Tall fescue- Red clover	1.3 ± 0.02	1.3 ± 0.03	NS	1.5 ± 0.12	1.7 ± 0.07	NS
Tall fescue- Chicory	1.1 ± 0.05	1.1 ± 0.05	NS	1.3 ± 0.09	1.5 ± 0.13	NS
Tall fescue- Red clover- Chicory	1.2 ± 0.003	1.2 ± 0.04	NS	1.1 ± 0.10	1.5 ± 0.10	0.02
Day 210
Tall fescue- Red clover	1.1 ± 0.06	1.1 ± 0.07	NS	1.6 ± 0.04	1.3 ± 0.05	0.02
Tall fescue- Chicory	0.9 ± 0.03	0.9 ± 0.04	NS	1.1 ± 0.08	1.3 ± 0.05	NS
Tall fescue- Red clover- Chicory	0.9 ± 0.03	0.10 ± 0.05	NS	1.1 ± 0.07	1.2 ± 0.02	NS

Mixed Plant Material Effects on Soil N

In contrast to decomposition, ammonium and nitrate availability was affected by mixing plant species. In our lab incubation experiment, soils containing mixed plant material consistently had less available ammonium + nitrate than expected even though differences in nitrate concentration were not as consistent (Table 4). Overall, mixed plant material immobilized more N, particularly ammonium, than would have been expected from the single species incubations. These results suggest that interactions among different plant functional groups during decomposition could improve nitrogen retention in pastures by increasing N immobilization. Because substantial N may be lost from pastures via nitrate leaching in the fall and early spring in temperate, humid climates (7), finding ways to manage the types and quantity of plant material left remaining on pasture may warrant increased attention. If ammonium immobilization limits nitrate production, mixed species plant material may have a role in reducing N losses from the pasture system.

Table 4. Ammonium (NH₄) and nitrate (NO₃) extracted from soils at the end of lab incubation. Values are means with 1SE. P values are based on two sample t-tests.

Nitrogen type	Plant material mixes	N concentration (m g N - g soil)
Nitrogen type	Plant material mixes	Observed	Expected	P
Total mineral nitrogen (NO₃ + NH₄)	Tall fescue-red clover	57 ± 8	118 ± 6	< 0.0001
	Tall fescue-chicory	26 ± 2	41 ± 7	0.0008
	Tall fescue-red clover-chicory	61 ± 6	93 ± 4	0.0016
Nitrate	Tall fescue-red clover	24 ± 2	16 ± 2	0.04
	Tall fescue-chicory	8 ± 0.9	14 ± 1	0.003
	Tall fescue-red clover-chicory	16 ± 4	11 ± 2	NS
Ammonium	Tall fescue-red clover	30 ± 6	102 ± 7	< 0.001
	Tall fescue-chicory	19 ± 1	27 ± 2	0.004
	Tall fescue-red clover-chicory	44 ± 6	82 ± 5	0.006

Conclusions

Overall, we found that plant materials mixed together decomposed no differently than those species that decomposed alone. Mixed plant materials had unexpected effects on levels of soil N availability, however. When mixed plant material decomposed it immobilized more N than we predicted from single species assays. We suggest that decomposition rates in pasture can be predicted by evaluating single species plant materials, but this information may tell us little about how multi-species plant material decomposition affects soil N availability. More field research on mixed plant materials and their effect on soil N are needed in pasture systems to better clarify potential mechanisms behind our results. In particular, decomposition studies that evaluate mixed plant material at different stages of senescence should help shed more light on potential interaction between decomposition and nitrogen availability.

Literature Cited

1. Bardgett, R. D., and Shine, A. 1999. Linkages between plant litter diversity, soil microbial biomass and ecosystem function in temperate grasslands. Soil Biol. Biochem. 31:317-321.

2. Blair, J. M., Parmelee, R. W., and Beare, M. H. 1990. Decay rates, nitrogen fluxes, and decomposer communities of single and mixed species foliar litter. Ecology 71:1976-1985.

3. Buxton, D. R., and Russell, J. R. 1988. Lignin constituents and cell wall digestibility of grass and legume stems. Crop Sci. 28:553-558.

4. Chapman, K., Whittaker, J. B., and Heal, O. W. 1988. Metabolic and faunal activity in litters of tree mixtures compared with pure stands. Agric. Ecosys. Environ. 24:33-40.

5. Dubeux, J. C. B., Jr., Sollenberger, L. E., Interrante, S. M., Vendramini, J. M. B., and Stewart, R. L. 2006. Litter decomposition and mineralization in bahiagrass pastures managed at different intensities. Crop Sci. 46:1305-1310.

6. Hooper, D. U., and Vitousek, P. M. 1998. Effects of plant composition and diversity on nutrient cycling. Ecolog. Monog. 68:121-149.

7. Stout, W. L., Fales, S. A., Muller, L. D., Schnabel, R. R., Priddy, W. E., and Elwinger, G. F. 1997. Nitrate leaching from cattle urine and feces in northeast USA. Soil Sci. Soc. Amer. J. 61:1787-1794.

8. Symstad, A. J., Tilman, D., Willson, J., and Knops, J. M. H. 1998. Species loss and ecosystem functioning:effects of species identity and community composition. Oikos 81:389-397.

9. Thomas, W. A. 1968. Decomposition of loblolly pine needles with and without addition of dogwood leaves. Ecology 49:568-571.

10. Tilman, D., Knops, J., Wedin, D., Reich, P., Ritchie, M., and Siemann, E. 1997. The influence of functional diversity and composition on ecosystem processes. Nature 277:1300-1302.

11. Wardle, D. A. 1997. Biodiversity and plant litter: experimental evidence which does not support the view that enhanced species richness improves ecosystem function. Oikos 79:247-258.

12. White, T. A., Barker, D. J., and Moore, K. J. 2004. Vegetation diversity, growth, quality and decomposition in managed grasslands. Agric. Ecosys. Environ. 101:73-84.