Neighbourhood-scale diversity, composition and root crowding do not alter competition during drought in a native grassland

1. James F. Cahill Jr Article first published online: 13 JUN 2003 DOI: 10.1046/j.1461-0248.2003.00476.x Issue

Ecology Letters Volume 6, Issue 7, pages 599603, July 2003 Additional Information(Show All) How to CiteAuthor InformationPublication History SEARCH
Search Scope

In this issue

Search String

Advanced > Saved Searches > ARTICLE TOOLS

Get PDF (82K) Save to My Profile E-mail Link to this Article Export Citation for this Article Get Citation Alerts Request Permissions

More Sharing ServicesShare|Share on citeulikeShare on connoteaShare on deliciousShare on www.mendeley.comShare on twitter

Abstract Article References Cited By

Get PDF (82K) Keywords:

Biodiversity; community ecology; diversity-invasion resistance hypothesis; invasion; plant neighbourhood;

root competition

Abstract
The diversity-invasion resistance theory argues that increased diversity results in increased competitive suppression of establishing plants. Although there is support for the pattern of decreased invasion with increased diversity, empirical demonstrations of increased competition are limited. An experiment was conducted during a severe drought in a native grassland community. The drought resulted in minimal shading among neighbours, and in contrast to prior studies, competition here was exclusively belowground. Neither diversity nor community composition influenced root crowding or competition. It appears that when competition is belowground, it is independent of diversity, likely because of fundamental differences in the mechanisms of above- and belowground competition. This suggests that even at the neighbourhood scale, there is no inherently negative relationship between competition and diversity, and lends support to alternative theories which suggest factors of than diversity may more strongly influence community invasibility.

Introduction
For decades ecologists have discussed how plant diversity responds to a variety of factors (e.g. species invasion), and there is current interest in understanding how diversity itself may influence a community's response to these factors (Loreau et al. 2001). Increased diversity may enhance community resistance to species invasion (Elton 1958; Knops et al. 1999; Levine & D'Antonio 1999; Naeem et al.2000; Prieur et al. 2000; Hector et al. 2001; Lyons & Schwartz 2001; Kennedy et al. 2002; Levine et al. 2002), although whether this is a strong or consistent effect relative to other factors is unclear (Davis et al. 2000; Davis & Pelsor 2001; Wardle 2001). The diversity-resistance hypothesis stems from the long-standing debate on the role of competition in natural communities (Levine & D'Antonio 1999), and is centered on issues of resource use, competition, and niche exploitation (MacArthur 1970). Two commonly proposed mechanisms are: (1) Species differ in patterns of resource exploitation, and a diverse community more fully exploits available resources (Naeem et al. 2000) and has increased competitive exclusion (Levine & D'Antonio 1999). Low diversity communities will have open niches susceptible to invasion (Levine & D'Antonio 1999). (2) A related mechanism does not assume differences in resource use and instead argues competition is diffuse, with increased diversity leading to increased crowding and competition (Naeem et al. 2000;Kennedy et al. 2002; Levine et al. 2002). Empirical support for the diversity-resistance hypothesis comes primarily from studies which create artificial communities of varying diversity (Levine et al. 2002). These studies are well suited for showing niche complementarity as at low diversity they are most certainly going to be unsaturated (Diaz et al. 2003), and nearly uniformly support the diversity-resistance hypothesis (Levine et al. 2002). The patterns are less clear in naturally assembled communities, with diverse communities generally most heavily invaded (Levine et al. 2002). A recent attempt to unify these data argues that at the level of plantplant interactions (neighbourhood), there is an

inherently negative effect of diversity on invasion, with larger scale processes (e.g. disturbance) potentially altering the relative importance of diversity (Levineet al. 2002). Alternatively, the implicit assumption that increased diversity results in increased crowding and competition could itself be at fault. The empirical support and theoretical grounding for this assumption come from communities where competition for light is important. For example, there was 90% plant cover in the Eel River (Levine et al. 2002), between 60 80% cover at Cedar Creek (Tilman et al. 1996), and competition for light and nutrients were reported to be associated with increased resistance in a related study (Naeem et al. 2000). However, light is not necessarily always limiting in all communities (or all years), and competition can often be primarily belowground (Casper & Jackson 1997). There are several reasons to doubt the linkage between diversity, crowding, and competition in systems dominated by root competition: (1) Large plants have a disproportionate competitive advantage above(Schwinning & Weiner 1998), but not belowground (Cahill & Casper 2000). Small seedlings will therefore not be at a disproportionate disadvantage in communities with primarily root competition, and strong root competition does not necessarily result in increased competitive exclusion (Newman 1973). (2) Roots are distributed relatively homogeneously in grasslands (Partel & Wilson 2002), the strength of root competition saturates at densities well below natural root densities (Cahill & Casper 2000), and root competition is unrelated to naturally occurring variation in root biomass (Peltzer et al. 1998). Three key questions related to the diversity-resistance hypothesis were tested in a native grassland during a severe drought: (1) Is (root) crowding positively associated with diversity? (2) Does the strength of root competition vary with community composition? (3) Does the strength of root competition vary with diversity? As a result of the drought, aboveground biomass was low (<75 gm2; <20% average), shading was practically nonexistent, and any competition which did occur was belowground.

Material and methods

Ten 9 16 m blocks containing ten 2 3 m subplots were established throughout a 20 ha area of rough fescue prairie at the Kinsella Research Ranch, Alberta, Canada. The study year was a severe drought, with about 50% average rainfall. Vegetation (by biomass) is primarily graminoid, although the majority of plant species are forbs (Cahill in press). Ten focal species of varying levels of natural abundance were selected (Table 1), with individual plants grown with or without competition (see below) at a variety of naturally occurring levels of diversity. Four seedlings of one species were transplanted into one subplot in June 2002 (separated by >1 m), with each species present in one subplot per block. Transplants were watered for 7 days, during which time there was <10% mortality (they were replaced). Species Family Life history Occurrence Competition Terms retained in logistic regression

Diversi

1.

Species information and results from stepwise logistic regressions. The occurrence of each focal species is described as comm

to encounter), rare (present but infrequently encountered), and absent (has not been seen it at least 10 years). The significance of the e

competition on plant survival (F-statistic from GLMM) are indicated by superscripts [*P < 0.05; **P < 0.005 (Bonferroni adjusted fo

comparisons)]. Also presented are the results from the stepwise forward logistic regression for both the aggregated and species specif

Species

Family

Life history

Occurrence

Competition

Terms retained in logistic regression

Diversi

sets. In each analysis, diversity, species composition, and neighbour root biomass served as independent variables and plant status (al was the response variable. Only terms included in the final model are presented. Although constants are included in the models, they

presented here. To reduce the probability of Type II errors, an alpha of 0.10 was used as the cutoff for entry of a variable into the regr

Because of concerns over high type II error rates in logistic regressions, the direction of the effect of diversity on mortality is presente

each species and for the aggregated data, regardless of the significance of that term in the model. A + indicates mortality incr eased w diversity (supporting diversity invasion hypothesis), and a indicates mortality decreased with increased diversity. Aggregated data Abutilon theophrasti Achillea millefolium Agropyron trachycaulum Bouteloua gracilis Coreopsis tinctoria Festuca hallii Geum triflorum Koeleria macrantha Linum lewisii Potentilla pensylvanica Malvaceae Asteraceae Poaceae Poaceae Asteraceae Poaceae Rosaceae Poaceae Linaceae Rosaceae Annual Perennial Perennial Perennial Annual Perennial Perennial Perennial Perennial Perennial Absent Common Common Common Absent Common Common Common Rare Common 120.18** 11.10** 9.48
**

None None None None Diversity None Diversity, composition Composition None

7.25* 2.46 14.05** 5.99* 0.96 0.33 15.50

**

None None

14.79**

Table 1. Species descriptions and effects of competition, diversity, species composition, and neighbour root biomass on plant mortality Of particular concern in using natural variation in diversity is whether the observed variation in diversity is due to a confounding extrinsic variable. The cause of the variation in diversity found here is unknown, however evidence suggests it is not because of an underlying abiotic gradient: (1) Blocks were chosen based upon homogeneity in slope and aspect and all appeared typical fescue grassland. (2) In 40 plots measured in 2001, soil moisture was not related to diversity (F1,38 = 1.09; P = 0.30). (3) At the end of this study, limited sampling indicates that soil moisture was extremely low (<4% by weight) and consistent across plots. Although natural and artificial variation in diversity are created through different mechanisms (Diaz et al. 2003), the underlying assumption of the diversity-invasion hypothesis remains the same: within a plant neighbourhood, increased diversity should be associated with increased crowding and competition. Competition was manipulated with PVC exclusion tubes (15 cm diameter 20 cm length) installed the previous year (Cahill in press). In each subplot, one seedling was placed inside each of two tubes, preventing their roots from intermingling with the surrounding vegetation. The remaining two seedlings had no tube (sham control) and experienced competition. I had originally also included a shoot competition treatment, although neighbours did not influence light levels reaching the transplants (Shoots tied back vs. left in place: F1,395 = 0.31;P = 0.58). Any competition which occurred was belowground, and I have pooled the data across the shoot competition treatments. Seedling survival was measured 21 days after transplanting, with the difference in survival between the +/ competition treatments indicating the strength of competition. Mortality is often used as an indicator of competition (Goldberg et al. 1999), and initial patterns of mortality can be a predictor of invasion a year later (Davis & Pelsor

2001). This timing of measurement was chosen because although root competition was not apparent after 7 days (P > 0.70), it was apparent soon thereafter. Additionally, the drought resulted in early senescence of the existing vegetation and mortality of all transplants less than 1 month later. Neighbourhood-scale diversity, composition, and root biomass (i.e. crowding) were measured immediately adjacent to one + competition seedling in each subplot (100 total). Neighbour roots were extracted from a soil core (1.9 cm diam 15 cm deep), separated from the soil, dried, and weighed. All vascular plant species rooted within a 20 50 cm area were recorded, from which species richness and composition were calculated. Species composition was represented as the axis loadings from a non-metric multidimensional scaling (NMS) (McCune & Mefford 1999) with a single axis (final stress = 48.39; R2 = 0.273). Species richness ranged from 3 to 12 and root biomass varied by over one order of magnitude among plots. The effects of competition on mortality were determined using a Generalized Linear Mixed Model (GLMM) (SAS 2001). All plants in each plot were included in the analysis, along with block and subplot (nested within block) as random effects and competition a fixed effect. GLMMs were also conducted for each species separately. The effects of diversity, composition, and root crowding on survival were determined in forward stepwise logistic regressions (SPSS 2002). Only the single target plant immediately adjacent to the diversity plot in each subplot was used (+ competition), and neither subplot or competition treatment were included in the model. Plant status (alive/dead) was the response variable, with diversity, species composition, and root biomass as independent variables. As the data generally do not support for the diversity-resistance hypothesis, I chose statistical methods prone to Type I errors, but decrease the probability of Type II errors. First, analyses were conducted on both the whole data set and each species separately. As there are only ten replicates per species, the aggregated analysis allows determination of any overall diversity effect with a larger sample size and power. This analysis is justified as the question being tested is whether there is an inherently negative relationship between neighbourhood diversity and competition. Second, as sample sizes in the species specific analyses are lower, I used = 0.10, rather than the conventional = 0.05.

Results
In the data pooled across species, root competition increased mortality (33.5% vs. 80% mortality; F1,299 = 120.2; P < 0.0001), and neither diversity, species composition, nor root crowding were associated with mortality (Table 1). When forced into the model none of these variables had slopes different from zero (P > 0.10), and the model fit barely changed (2 LL: stepwise = 103.26; all forced = 102.79). There were no correlations among the independent variables themselves (P > 0.35). In the species specific analyses, competition increased mortality for most species (Fig. 1; Table 1). Diversity influenced mortality for two species, with the predicted increase in competition for one species, and a decrease in competition for the other (Table 1). Even when ignoring the significance of the diversity effect on mortality, there was no evidence that mortality was generally higher within increased diversity (Table 1). Species composition was associated with mortality for two species, and root crowding was unrelated to mortality (Table 1). There were no

apparent relationships between any of the species characteristics recorded and the effects of diversity, composition, or root biomass on mortality (Table 1). Figure 1. Proportion of transplanted individuals which were dead at the end of the study separated by species and competition treatment. Grey shading represents the + competition treatment and the black shading represents the competition treatment. Competition increased mortality in seven of the 10 species (Table 1).

Discussion
Despite strong effects of competition on mortality (Fig. 1), there was no evidence to support the ideas that diversity is positively associated with (root) crowding (Kennedy et al. 2002), that competition varies consistently with community composition (Naeem et al. 2000), or that there is an inherently negative relationship between competition and diversity (Levine et al. 2002). What accounts for the lack of support for the main mechanisms and assumptions underlying the diversityresistance hypothesis? (1) The lack of a relationship between composition and competition (Table 1) indicates the competitive environment is constant with respect to species identity. This makes sense from the seedling point of view whose roots will primarily be shallow and experiencing intense competition with neighbours. Spatial separation of root systems (e.g. McKane et al. 2002) would seem more likely after the seedling stage. (2) Although other data show that shoot biomass is positively associated with diversity in this field (F1,38 = 14.18; P < 0.001), we found no relationship between root biomass (crowding) and diversity (F1,98 = 0.36; P = 0.55). Additionally, root biomass was high (mean > 1000 g m2), and any observed variation was unlikely of a magnitude sufficient to significantly alter the strength of root competition (Cahill & Casper 2000). (3) Naturally occurring levels of diversity could already be above the threshold level at which diversity matters (Naeem et al. 2000). Low diversity areas of self-assembled communities are not necessarily functionally equivalent to low diversity artificially assembled communities (Diaz et al. 2003). The process of self-assembly here apparently results in homogeneous root distributions and competitive intensity at least with respect to diversity and composition. (4) Although it has been posited that native species may act differently than non-native species (Levine & D'Antonio 1999), there was no evidence to suggest diversity effects varied as a function of species occurrence within the field site (Table 1). The lack of response found here does not appear to be simply because of species selection. (5) Diversity may have been important if the study lasted longer. Despite the complete mortality of all transplants during the drought, there was a window at which mortality was not uniform among transplants. At this time, competition was important, while neighbour diversity, composition, and root crowding were not (Table 1). It is certainly possible that had the drought broken, transplant growth would have been affected by these neighbourhood traits, although this question is beyond the scope of the current study.

Although these results refute one model because of the lack of a diversity competition relationship (Levine et al. 2002), and another because of there being strong competition even at low productivity (Wardle 2001), they confirm a third model which argues invasibility increases with resource supply, and will be low when competition is strong regardless of productivity or diversity (Davis et al. 2000; Davis & Pelsor 2001). Here, resource supply was very low (drought) and competition was strong. After 21 days, there was no diversity effect, and at the end of the season the drought resulted in complete transplant mortality (invasibility = 0). The previous year, root competition had no effect on survival of many of these same species in the same field (Cahill in press). This inter-annual variability in transplant success supports Davis ideas that invasibility is not a community trait and is linked to variation in resource supply. This study builds on Davis model by suggesting that because of fundamental differences in above - and belowground competition, the predominant form of competition within a system in a particular year may also influence invasibility. It may not solely be an issue of resource supply, but may also depend upon whether competition for light is present or absent. At a minimum, we should not assume there exists a simple negative relationship between diversity and competition within plant neighbourhoods (Levine et al. 2002), as the nature of plant competition is more complicated than has generally been appreciated in the invasion-resistance debate. More explicit studies separating root and shoot competition at varying levels of naturally and experimentally generated diversity are needed to further the development of a mechanistic understanding of the factors that limit establishment in grasslands.

Acknowledgements
I thank D. Calvert, M. Coupe and J. Haag for assistance in the field and laboratory. Comments from S. Brewer, M. Brown, M. Coupe, J. Haag, S. Kembel, E. Lamb improved the quality of the manuscript. Funding was provided by a research grant from the Natural Sciences and Engineering Research Council of Canada.

References

Cahill, J.F.J., Jr (in press) Below ground competition does not cause a shift in relative allocation to roots or depend on root system size for 10 grassland species. J. Ecol. Cahill, J.F., Jr & Casper, B.B. (2000). Investigating the relationship between neighbor root biomass and belowground competition: field evidence for symmetric competition belowground. Oikos, 90, 311320. Direct Link:

o o o

Abstract PDF(181K) Web of Science Times Cited: 43 Casper, B.B. & Jackson, R.B. (1997). Plant competition underground. Ann. Rev. Ecol. Syst., 28, 545 570.

o o

CrossRef, Web of Science Times Cited: 271

Davis, M.A. & Pelsor, M. (2001). Experimental support for a resource-based mechanistic model of invasibility. Ecol. Lett., 4, 421428. Direct Link:

o o o o o

Abstract Full Article (HTML) PDF(274K) References Web of Science Times Cited: 121 Davis, M.A., Grime, J.P. & Thompson, K. (2000). Fluctuating resources in plant communities: a general theory of invasibility. J. Ecol.,88, 528534. Direct Link:

o o o o o

Abstract Full Article (HTML) PDF(330K) References Web of Science Times Cited: 663 Diaz, S., Symstad, A.J., Chapin, F.S., Wardle, D.A. & Huenneke, L.F. (2003). Functional diversity revealed by removal experiments.Trends Ecol. Evol., 18, 140146.

o o

CrossRef, Web of Science Times Cited: 99 Elton, C.S. (1958). The Ecology of Invasions by Animals and Plants . Methuen, London. Goldberg, D.E., Rajaniemi, T., Gurevitch, J. & Stewart-Oaten, A. (1999). Empirical approaches to quantifying interaction intensity: competition and facilitation along productivity gradients. Ecology, 80, 11181131.

o o

CrossRef, Web of Science Times Cited: 173 Hector, A., Dobson, K., Minns, A., Bazeley, W.E. & Lawton, J.H. (2001). Community diversity and invasion resistance: an experimental test in a grassland ecosystem and a review of comparable studies. Ecol. Res., 16, 819831. Direct Link:

o o o o o

Abstract Full Article (HTML) PDF(267K) References Web of Science Times Cited: 76 Kennedy, T.A., Naeem, S., Howe, K.M., Knops, J.M.H., Tilman, D. & Reich, P. (2002). Biodiversity as a barrier to ecological invasion.Nature, 417, 636639.

CrossRef,

o o o o

PubMed, CAS, Web of Science Times Cited: 265, ADS Knops, J.M.H., Tilman, D., Haddad, N.M., Naeem, S., Mitchell, C.E., Haarstad, J. et al. (1999). Effects of plant species richness on invasion dynamics, disease outbreaks, insect abundances and diversity. Ecol. Lett., 2, 286293. Direct Link:

o o o o o

Abstract Full Article (HTML) PDF(306K) References Web of Science Times Cited: 225 Levine, J.M. & D'Antonio, C.M. (1999). Elton revisited: a review of evidence linking diversity and invasibility. Oikos, 87, 1526.

o o

CrossRef, Web of Science Times Cited: 398 Levine, J.M., Kennedy, T. & Naeem, S. (2002). Neighborhood scale effects of species diversity on biological invasions and their relationship to community patterns. In: Biodiversity and Ecosystem Functioning: Synthesis and Perspectives (eds Loreau, M.,Naeem, S. & Inchausti, P.). Oxford University Press, New York.

Loreau, M., Naeem, S., Inchausti, P., Bengtsson, J., Grime, J.P., Hector, A. et al. (2001). Biodiversity and ecosystem functioning: current knowledge and future challenges. Science, 294, 804808.

o o o o o

CrossRef, PubMed, CAS, Web of Science Times Cited: 800, ADS Lyons, K.G. & Schwartz, M.W. (2001). Rare species loss alters ecosystem function invasions resistance. Ecol. Lett., 4, 358365. Direct Link:

o o o o o

Abstract Full Article (HTML) PDF(296K) References Web of Science Times Cited: 84 MacArthur, R.H. (1970). Species-packing and competitive equilibrium for many species. Theor. Popul. Biol., 1, 111.

o o o

CrossRef, PubMed, CAS McCune, B. & Mefford, M.J. (1999). PC-Ord for Windows. MjM Software, Gleneden Beach, OR. McKane, R.B., Johnson, L.C., Shaver, G.R., Nadelhoffer, K., Rastetter, E., Fry, B. et al. (2002). Resource-based niches provide a basis for plant species diversity and dominance in arctic tundra. Nature, 415, 6871.

o o o o o

CrossRef, PubMed, CAS, Web of Science Times Cited: 180, ADS Naeem, S., Knops, J.M.H., Tilman, D., Howe, K.M., Kennedy, T. & Gale, S. (2000). Plant diversity increases resistance to invasion in the absence of covarying extrinsic factors. Oikos, 91, 97108. Direct Link:

o o o o o o

Abstract PDF(331K) Web of Science Times Cited: 215 Newman, E.I. (1973). Competition and diversity in herbaceous vegetation. Nature, 244, 310311. CrossRef, Web of Science Times Cited: 197, ADS Partel, M. & Wilson, S.D. (2002). Root dynamics and spatial pattern in prairie and forest. Ecology, 83, 11991203.

Web of Science Times Cited: 19 Peltzer, D.A., Wilson, S.D. & Gerry, A.K. (1998). Competition intensity along a productivity gradient in a low-diversity grassland. Am. Nat., 15, 465476.

o o o

CrossRef, CAS, Web of Science Times Cited: 65 Prieur, R.A.H., Lavorel, S., Grigulis, K. & Dos, S.A. (2000). Plant community diversity and invasibility by exotics: invasion of Mediterranean old fields by Conyza bonariensis and Conyza canadensis. Ecol. Lett., 3, 412422. Direct Link:

o o o o

Abstract Full Article (HTML) PDF(357K) References

Web of Science Times Cited: 60 SAS (2001) SAS System for Windows. SAS Institute, Cary, NC. SPSS I. (2002) SPSS for Windows. SPSS Inc., Chicago, IL. Schwinning, S. & Weiner, J. (1998). Mechanisms determining the degree of size asymmetry in competition among plants. Oecologia,113, 447455.

o o

CrossRef, Web of Science Times Cited: 223 Tilman, D., Wedin, D. & Knops, J. (1996). Productivity and sustainability influenced by biodiversity in grassland ecosystems. Nature,379, 718720.

o o o o

CrossRef, CAS, Web of Science Times Cited: 753, ADS Wardle, D.A. (2001). Experimental demonstration that plant diversity reduces invasibility evidence of a biological mechanism or a consequence of sampling effect? Oikos, 95, 161170. Direct Link:

o o o

Abstract PDF(161K) Web of Science Times Cited: 77 Get PDF (82K)

More content like this

Find more content: like this article Find more content written by: James F. Cahill
ABOUT US HELP CONTACT US AGENTS ADVERTISERS MEDIA PRIVACY COOKIES TERMS & CONDITIONS SITE MAP

Copyright 19992012 John Wiley & Sons, Inc. All Rights Reserved.