See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/261642976

Biodiversity and Ecosystem Functioning

Chapter · January 2009

CITATIONS READS

13 903

2 authors, including:

Andrew Wilby
Lancaster University
82 PUBLICATIONS 3,130 CITATIONS

SEE PROFILE

All content following this page was uploaded by Andrew Wilby on 27 April 2016.

The user has requested enhancement of the downloaded file.

Winterthurerstr. 190
CH-8057 Zurich
http://www.zora.uzh.ch

Year: 2009

Biodiversity and ecosystem functioning

Hector, A; Wilby, A

Hector, A; Wilby, A (2009). Biodiversity and ecosystem functioning. In: Levin, S A. The Princeton guide to
ecology. Princeton, NJ, US, 367-375.
Postprint available at:
http://www.zora.uzh.ch

Posted at the Zurich Open Repository and Archive, University of Zurich.

Originally published at:

1 Biodiversity and Ecosystem Functioning

2 Andy Hector1 and Andy Wilby2

6 Outline

7 • Background and history

8 • Biodiversity and ecosystem functioning relationships

9 • Mechanisms

10 • Multitrophic systems

11 • Ecosystem multifunctionality

12 • Ecosystem service provision

13 • The next phase of research

14

15 Forecasts of ongoing biodiversity loss prompted ecologists in the early 1990s to question

16 whether this loss of species could have a negative impact on the functioning of

17 ecosystems. Ecosystem functioning is an umbrella term for the processes operating in an

18 ecosystem, that is the biogeochemical flows of energy and matter within and between

19 ecosystems (e.g. primary production and nutrient cycling). This first general phase of

20 research on this topic addressed this question by assembling model communities of

21 varying diversity to measure the effects on ecosystem processes. The results of the meta-

22 analyses of this first wave of studies show that biodiversity generally has a positive but

23 saturating effect on ecosystem processes which is remarkably consistent across trophic

24 groups and ecosystem types. These relationships are driven by a combination of

25 complementarity and selection effects with complementarity effects nearly twice as

26 strong as selection effects overall. However, diverse communities rarely function

1
 Princeton University Press Guide to Ecology

27 significantly better than the best single species, at least in the short term. In the longer

28 term biodiversity can provide an insurance value similar to the risk spreading benefits

29 of diverse portfolios of financial investments. The effects of biodiversity on ecosystem

30 functioning may have been under-estimated by the first phase of research due to the

31 short duration of many studies and due to the focus on single ecosystem processes in

32 isolation rather than considering all important ecosystem functions simultaneously. The

33 next phase of research will in part focus on whether the benefits of biodiversity seen in

34 experiments translate to real-world settings.

35

36 Glossary

37 • Biodiversity: A contraction of biological diversity that encompasses all biological

38 variation from the level of genes, through populations, species and functional groups (and

39 sometimes higher levels such as landscape units).

40 • Ecosystem functioning: An umbrella term for the processes operating in an ecosystem.

41 • Ecosystem processes: The biogeochemical flows of energy and matter within and between

42 ecosystems, e.g. primary production and nutrient cycling.

43 • Ecosystem service: An ecosystem process or property that is beneficial for human beings,

44 e.g. the provision of foods and materials or sequestration of carbon dioxide.

45 • Selection effects: The influence that species have on ecosystem functioning simply due to

46 their species-specific traits and their relative abundance in a community (positive

47 selection effects occur when species with higher-than-average monoculture performance

48 dominate communities).

49 • Complementarity effect: The influence that combinations of species have on ecosystem

50 functioning as a consequence of their interactions (e.g. resource partitioning; facilitation,

51 reduced natural enemy impacts in diverse communities).

52

2
 Princeton University Press Guide to Ecology

53 Background and History

54 Darwin, in the Origin of Species, initially proposed that changes in biodiversity could affect

55 ecosystem functioning if niche space is more fully occupied in more diverse communities

56 than depauperate ones. We use ecosystem functioning as an umbrella term to embrace all the

57 bio-geo-chemical processes that operate within ecosystems, primary production for example.

58 This early work was apparently forgotten until the early 1990s, but the same reasoning was

59 around in the mid-twentieth century where it was proposed that more diverse mixtures of fish

60 species should lead to greater productivity: “Presumably fish production will increase as the

61 number of niches increases…[and]…probably the proportion of occupied niches increases as

62 the number of species of fishes increases”. Indeed, both of these early studies even presented

63 data in support of this relationship (Figure 1).

64 In the early 1990s, general concern about the impact of anthropogenic biodiversity

65 was voiced at the Rio Earth Summit in 1992. At this the Convention on Biological Diversity

66 (CBD) was launched with the signatures of 150 heads of government. This international treaty

67 designed to promote sustainable development and the protection of biodiversity was evidence

68 of political acceptance that anthropogenic biodiversity loss may have serious detrimental

69 effects on humankind. Concerns highlighted at Rio and in the Convention also led to renewed

70 scientific interest and a concerted effort by ecologists to understand the effects of changes in

71 biodiversity on ecosystem functioning and the likely significance of such changes for

72 humankind. More than a decade’s worth of research has now been published, accompanied by

73 a debate which focused in large part on the mechanisms underlying the relationship between

74 biodiversity and functioning. Synthesis of the first decade of results through meta-analysis is

75 helping to reveal both pattern and mechanism.

76 Biodiversity and ecosystem functioning relationships

77 The main approach that has been used to investigate the relationship between biodiversity and

78 ecosystem functioning is the direct manipulation of biodiversity by the assembly of

3
 Princeton University Press Guide to Ecology

79 synthesized model communities in the laboratory or field. An alternative approach is to

80 remove species from natural communities. A third non-manipulative approach is to infer the

81 relationship between biodiversity and ecosystem functioning by seeing how they are

82 correlated across habitats. All three approaches have strengths and weaknesses. In this piece

83 we focus on the assembly of model communities of varying diversity.

84 Meta-analysis of the first decade’s research clearly shows a positive relationship

85 between biodiversity and ecosystem functioning (e.g. Figure 2); a pattern which is remarkably

86 consistent across trophic groups (producers, herbivores, detrivores and predators) and present

87 in both terrestrial and marine ecosystems. However, the relationship between biodiversity and

88 ecosystem functioning is generally saturating suggesting that the effect of random

89 biodiversity loss on ecosystem functioning will be initially weak but accelerating.

90 The first phase of research on biodiversity and ecosystem functioning was focused on

91 identifying general patterns (whether biodiversity change can affect ecosystem functioning or

92 not) and species were therefore removed at random to generate experimental diversity

93 gradients. Another key result of these studies is that there is considerable variation among

94 species or species assemblages in their impact on functioning. This suggests that the actual

95 effect of biodiversity loss on ecosystem functioning seen in real-world situations will depend

96 strongly on which species are lost. Moving from random to more realistic real-world

97 situations is a key goal for the next phase or research.

98 Mechanisms

99 The early studies mentioned in the introductory background section only identify one way in

100 which biodiversity changes can affect ecosystem functioning, namely by affecting the degree

101 of species complementarity (basically by affecting the number of under-utilized or vacant

102 niches). That is, more diverse communities utilise a greater proportion of available niche

103 space. However, as mentioned above, biodiversity changes can also affect ecosystem

104 functioning by the simple presence or loss of particular species with strong intrinsic effects on

4
 Princeton University Press Guide to Ecology

105 ecosystem processes (so-called sampling or selection effects); more diverse communities are

106 more likely to contain those species or assemblages which strongly affect functioning. There

107 has been widespread debate over the last decade about whether the positive relationships

108 reviewed above were explained by complementarity or selection effects.

109 Additive partitioning methods are one approach which allows separation of the overall

110 net effect of biodiversity on ecosystem functioning into complementarity effects that arise

111 from species interactions and selection effects that are species-specific. Meta-analysis reveals

112 that almost all studies are driven by a combination of these effects but that overall

113 complementarity effects were nearly twice as strong as selection effects (Figure 2). However,

114 even though complementarity effects have a greater effect than selection effects they are not

115 strong enough to cause mixtures to do significantly better than monocultures in most cases

116 (Figure 3). In summary, while the relationship between biodiversity and ecosystem

117 functioning is positive, and complementarity effects contribute approximately twice as much

118 as selection effects in generating these relationships, diverse communities do not generally

119 perform better than the best individual species. However, this result is influenced by the short

120 duration of many of the experiments performed to date since the relationship between

121 biodiversity and ecosystem functioning grows stronger over time (Figure 3) due to increasing

122 complementarity. Nevertheless, it appears that diverse communities are rarely able to do

123 substantially better than a monoculture of the best-performing species that they contain. This

124 appears to be due in part to the fact that communities are often not dominated by the most

125 productive species but by species with a lower performance. In fact, in over 40% of the

126 reviewed studies communities were dominated by a species with a lower-than-average

127 monoculture biomass leading to a negative selection effect with a negative influence on the

128 performance of the ecosystem as a whole. An important implication of this meta-analysis for

129 future research is that studies must be longer-term if they are to reveal the full effects of

5
 Princeton University Press Guide to Ecology

130 biodiversity on ecosystem functioning; experiments to date have, if anything, underestimated

131 the effects of random loss of species on ecosystem functioning.

132 Multi-trophic systems

133 Alongside the work on biodiversity and ecosystem functioning there has also been significant

134 interest in the functional importance of biodiversity in the context of multi-trophic

135 interactions. Here the focus has been more on the impact of diversity at one trophic level on

136 the population density at the trophic level below. Most commonly this has involved studies of

137 predator species diversity and impact on prey populations. Recently, attempts have been made

138 to link the considerable bodies of work on biodiversity and ecosystem functioning and

139 predator–prey interactions.

140 One striking difference between the predator–prey and biodiversity-ecosystem

141 functioning perspectives is the relative importance ascribed to interspecific interactions

142 among target species. Interactions among species are not explicitly considered in biodiversity-

143 ecosystem functioning studies, whereas predator–prey theory has a long history of

144 investigating direct and indirect interactions among predator species and how these affect the

145 population size of the prey species. For example, intra-guild predation where one predatory

146 species preys on another is a common interaction in nature which has the capacity to reduce

147 the joint impact of the predator species on the original prey species. The opposite outcome

148 can occur when facilitative interactions occur among predator species. One commonly

149 reported example of this is when the avoidance behaviour of the prey to one predator makes

150 them more susceptible to predation by a second. Aphids for example commonly drop from the

151 plant when approached by a foliar predator, but this can leave them susceptible to ground

152 foraging predators so that the functioning of ground and foliar predators together is greater

153 than the sum of the functioning of each alone.

154 Facilitative and negative interactions among constituent species can occur in basal

155 trophic levels, such as primary producers or detritivores. Plants are known to take part in

6
 Princeton University Press Guide to Ecology

156 allelopathic interactions in which they impact each other negatively via the production of

157 toxic chemicals. There is also strong evidence that plant species may facilitate each other by

158 enriching the soil by nitrogen fixation or by moderating harsh abiotic environments for

159 example. A key question is whether the predictive power of biodiversity–ecosystem

160 functioning theory would be improved by the incorporation of such species interactions.

161 Generally, meta-analyses of biodiversity–ecosystem functioning reveal consistent positive

162 effects of diversity on functioning, but results from terrestrial predator–prey systems are more

163 equivocal, with almost half of the studies reporting negative or neutral effects of increasing

164 species diversity on prey suppression. Where significant species interactions occur, it may be

165 useful to think of observed relationships between biodiversity and ecosystem functioning as

166 the net effect of co-occurring positive mechanisms (resource-use differentiation and

167 facilitation) and negative mechanisms (intraguild predation, interference). Experimental

168 evidence from predator–prey systems suggests that at least in some cases negative interactions

169 among species outweigh the positive mechanisms causing reduced functionality in more

170 diverse communities.

171 Diversity and stability

172 Ecological stability refers commonly to one of three general properties of ecosystems: the

173 temporal variation in a property of the ecosystem (e.g. primary production) or the response

174 (resistance) or recovery (resilience) of these properties following perturbation. One possible

175 value of biodiversity to humans is its potential to increase stability by buffering ecosystem

176 processes like production against environmental variation and in making them more resistant

177 and resilient to perturbations. This insurance value of biodiversity has most often been

178 considered in the context of fluctuations over time, where it has been likened to the risk-

179 spreading benefits of diverse portfolios of investments in financial markets, but could also

180 apply to spatial environmental variation. For this insurance effect to occur requires only that

181 fluctuations in the abundances of a guild of species are not perfectly synchronised, because

7
 Princeton University Press Guide to Ecology

182 under perfect synchrony an entire guild or trophic level would effectively behave as one

183 species. When species responses are not perfectly positively correlated changes in some

184 species can be compensated by others and the averaging of their asynchronous fluctuations

185 smoothes the collective productivity of the whole community (Figure 4).

186 One potentially confusing or counter-intuitive aspect of the insurance hypothesis is

187 that diversity has a stabilizing effect on aggregate community or ecosystem properties (like

188 primary productivity) at the same time as the fluctuations of the constituent species may be

189 destabilized due to interactions with greater numbers of species (although destabilization is

190 not inevitable). The key thing to understand is that it is the lack of perfect synchrony of

191 individual species fluctuations that leads to the stabilizing effect of diversity on ecosystem

192 processes. This asynchrony through independent or compensatory species responses can be

193 interpreted as a form of temporal niche differentiation between species.

194 A recent review of the diversity-stability literature emphasizes its breadth and

195 complexity due to the many different types of stability and the range of different variables

196 that stability measures can be calculated for (e.g. stability of population abundance vs total

197 community biomass as introduced above). For experiments where diversity was directly

198 manipulated there are reports of two positive effects of plant species diversity on the stability

199 of biomass production and three positive effects of microbial diversity on the stability of

200 biomass or carbon dioxide production. There are no reports of negative or neutral effects of

201 diversity on temporal stability of ecosystem processes from grassland experiments, but one

202 negative effect of increased multitrophic diversity on the temporal stability of biomass

203 production in seagrass beds, and one neutral and one negative effect of microbial diversity on

204 the stability of microbial biomass production. Observational studies have also looked at

205 stabilising effects of biodiversity on ecosystem processes producing five positive effects of

206 plant diversity on temporal stability and one neutral effect. In summary, evidence from both

207 natural and experimental systems of plants and microbes suggests that insurance effects of

8
 Princeton University Press Guide to Ecology

208 biodiversity on temporal stability may be relatively widespread.

209 Ecosystem multifunctionality

210 As summarised above, meta-analysis of the results of the first generation of experimental

211 research on biodiversity and ecosystem functioning has revealed that individual ecosystem

212 processes generally show a positive but saturating relationship with increasing diversity. The

213 saturating relationship suggests that some species are redundant with respect to a single

214 function. However, nearly all studies to date have been short-term and only address the effect

215 of biodiversity on ecosystem functioning at a given point in time and under a relatively

216 narrow set of conditions. Much of the other work reviewed above suggests that biodiversity

217 can sometimes have an insurance value by buffering ecosystem-level processes in a way

218 analogous to that in which diverse investment portfolios spread financial risk and improve

219 average performance in the longer term. Nevertheless, all of the research to date considers

220 ecosystem processes examined individually, despite the fact that most ecosystems are

221 managed or valued for several ecosystem services or processes: so-called ecosystem

222 multifunctionality. If it is the case that a single species, or group of species, controls

223 ecosystem functioning, then the remaining species are functionally redundant. Although it

224 seems unlikely that a single species could control all ecosystem processes, it is possible that a

225 single group of species may. However, if there is appreciable lack of overlap in the groups of

226 species that influence different ecosystem processes, then higher levels of biodiversity will be

227 required to maintain overall ecosystem functioning than indicated by analyses focusing on

228 individual ecosystem processes in isolation. Only one study of ecosystem multifunctionality

229 exists to date, but this analysis of seven ecosystem processes measured in a network of

230 grassland biodiversity experiments supports the ecosystem multifunctionality hypothesis: the

231 greater the number of ecosystem processes included in the analysis the greater the number of

232 species found to affect overall functioning (Figure 6).

233 Ecosystem service provision

9
 Princeton University Press Guide to Ecology

234 The Millennium Ecosystem Assessment defines ecosystem services as the benefits provided

235 by ecosystems to humans. Ecosystem services include the provision of materials (food,

236 genetic resources, water etc.), cultural and psycho-spiritual well-being, supporting services

237 (nutrient cycling, soil formation etc) and regulating services (pest and disease control,

238 pollination, erosion control, climate regulation etc.). The ecosystem processes that we have

239 covered in this chapter are closely aligned with both supporting and regulating services. The

240 evidence from the meta-analyses discussed above suggests that, in general, we expect the

241 provision of such services to be compromised due to anthropogenic declines in biodiversity.

242 Direct evidence of impacts of biodiversity loss on ecosystem functioning is accumulating. For

243 example, increased diversity of wild host species has been shown to lead to dilution effects

244 that reduce the probability of human infection by zoonotic diseases. Loss of biodiversity is

245 also implicated in causing reduced carbon sequestration and therefore a net release of carbon

246 into the atmosphere where it contributes to global climate change. However, just as in

247 experimental studies of biodiversity and ecosystem functioning, effects will depend strongly

248 on which species are lost. The provision of ecosystem services has been of particular concern

249 in agricultural systems both because of their spatial extent (they are estimated to cover a

250 quarter of the terrestrial earth surface, rising to almost three-quarters in some developed

251 regions) and the severe losses of biodiversity they endure. Intensification of production in

252 many parts of the globe has resulted in extreme declines in biodiversity in agricultural

253 systems, both in terms of homogenisation of production systems (simplified landscapes and

254 fewer breeds/varieties grown) and declines in the wild species inhabiting agricultural

255 ecosystems. Such simplification requires that compromised services such as pest regulation

256 and maintenance of soil fertility and condition are replaced by synthetic pesticides and

257 fertilisers, which are inherently unsustainable due to their reliance on externally derived

258 energy and materials, and their negative impacts on non-target taxa, including humans.

10
 Princeton University Press Guide to Ecology

259 Enhancement and utilisation of ecosystem services in agriculture is seen as one route to

260 increased sustainability of food production.

261 The next phase of research

262 The first phase of research on biodiversity and ecosystem functioning primarily used

263 experimental communities to investigate the effects of random species loss. Recent meta-

264 analysis suggests that there generally are effects of species loss on ecosystem functioning in

265 these experiments and that these effects are generally positive but saturating. Both

266 complementarity and selection effects play a role in generating these relationships with the

267 effects of complementarity being nearly twice as strong as selection. Nevertheless, diverse

268 mixtures rarely perform better than the best-performing species, at least in the short term

269 (complementarity effects grow stronger over time in these studies).

270 A key goal for the next phase of research is a move away from artificial experimental

271 systems towards more realistic settings and to see if the biodiversity effects seen in the

272 experiments translate to real-world situations. This will also necessitate a move away from the

273 random loss of species used in the first phase of research towards more realistic scenarios of

274 species loss and the incorporation of multiple trophic levels. The move from experimental to

275 real-world settings will also require a move to larger field-scale study systems and, as

276 suggested by the recent meta-analyses, to longer-term research. The first phase of research

277 reviewed here has demonstrated the potential for biodiversity to have positive effects on

278 ecosystem functioning. The question now is whether these experimental results will translate

279 into positive effects of biodiversity on the provision of ecosystem services in the real world.

280 The value of ecosystem services to humans is enormous [cross reference?] and it is now

281 critical to find out what role biodiversity plays in the provision of these services to human

282 societies.

11
 Princeton University Press Guide to Ecology

283 Further reading

284 Balvanera, P., Pfisterer, A.B., Buchmann, N., He, J.-S., Nakashizuka, T., Raffaelli, D., &

285 Schmid, B. (2006) Quantifying the evidence for biodiversity effects on ecosystem

286 functioning and services. Ecology Letters, 9, 1146-1156.

287 Cardinale, B.J., Srivastava, D.S., Emmett Duffy, J., Wright, J.P., Downing, A.L., Sankaran,

288 M., & Jouseau, C. (2006) Effects of biodiversity on the functioning of trophic groups and

289 ecosystems. Nature, 443, 989-982.

290 Cardinale, B.J., Wright, J.P., Cadotte, M.W., Carroll, I.T., Hector, A., Srivastava, D.S.,

291 Loreau, M., & Weis, J.J. (2007) Impacts of plant diversity on biomass production increase

292 through time due to species complementarity: A meta-analysis of 44 experiments.

293 Proceedings of the National Academy of the USA, 104, 18123–18128.

294 Hector, A. & Bagchi, R. (2007) Biodiversity and ecosystem multifunctionality. Nature, 448,

295 188-190.

296 Ives & Carpenter (2007) Stability and Diversity of Ecosystems Science 317, 58-#.

297 Kinzig, A., Tilman, D., & Pacala, S., eds. (2002) The Functional Consequences of

298 Biodiversity: empirical progress and theoretical extensions. Princeton University Press,

299 Princeton.

300 Loreau, M. & Hector, A. (2001) Partitioning selection and complementarity in biodiversity

301 experiments. Nature, 412, 72-76 [erratum: 413 548].

302 Loreau, M., Naeem, S., Inchausti, P., Bengtsson, J., Grime, J.P., Hector, A., Hooper, D.U.,

303 Huston, M.A., Raffaelli, D., Schmid, B., Tilman, D., & Wardle, D.A. (2001) Biodiversity

304 and ecosystem functioning: current knowledge and future challenges. Science, 294, 804-

305 809.

306 Loreau, M., Naeem, S., & Inchausti, P., eds. (2002) Biodiversity and ecosystem functioning:

307 synthesis and perspectives. Oxford University Press, Oxford.

12
 Princeton University Press Guide to Ecology

308 Worm, B., Barbier, E.B., Nicola Beaumont, Duffy, J.E., Folke, C., Halpern, B.S., Jackson,

309 J.B.C., Lotze, H.K., Micheli, F., Palumbi, S.R., Sala, E., Selkoe, K.A., Stachowicz, J.J., &

310 Watson, R. (2006) Impacts of Biodiversity Loss on Ocean Ecosystem Services. Science,

311 314, 787-790.

312

313 Legends

314

315 Figure 1 Early evidence for a link between biodiversity and ecosystem functioning from (top

316 panel) an early nineteenth century large-scale experimental garden at Woburn Abbey, U.K.

317 mentioned by Darwin in The Origin (after Hector & Hooper 2002), and, (lower panel)

318 “Relation between standing crops and numbers of species of fish present in Midwestern

319 reservoirs” (after Carlander 1955).

320

321 Figure 2 Effects of plant species richness on the production of above-ground plant biomass

322 for 11 plant biodiversity experiments. Lines are linear regression slopes as a function of the

323 number of plant species (log2 scale) for the eight BIODEPTH project experiments (black

324 lines; see Hector et al. 1999) the Cedar Creek biodiversity experiment (dashed line; see

325 Tilman et al. 2001) the Jena biodiversity experiment (dotted line; Roscher et al. 2005) and

326 Van Ruijven & Berendse (2005; dotted-and-dashed line).

327

328 Figure 3 Meta-analysis results (Cardinale et al. 2007) for (top left) overyielding, (top right)

329 transgressive overyielding, (lower left) complementarity effects and, (lower right) selection

330 effects. The data for the upper and lower panels are for different (but overlapping) sets of

331 studies. Note the different y axis scales in the two upper panels. Tukey box and whisker plots

332 to the right of each scatter plot summarize the variation averaged over time (notches provide

333 an approximate 95% interval for the medians).

13
 Princeton University Press Guide to Ecology

334

335 Figure 4 Asynchronous population fluctuations buffer total community biomass. In this

336 hypothetical example the total biomass of the three-species community (d) is less variable

337 than the two-species community (b) due to the asynchrony of individual species biomasses (a

338 and c).

339

340 Figure 5 Diversity increases stability (S) of ecosystem primary production (μ = mean and σ =

341 standard deviation of biomass through time; reproduced from van Ruijven & Berendse 2007).

342

343 Figure 6 Positive relationship between the range of ecosystem processes considered and the

344 number of species that affect one or more aspect of ecosystem functioning. The points

345 (jittered for clarity) show numbers of species required for all possible combinations of

346 ecosystem processes. Lines are average predictions based on the mean number of species

347 required for a single process and the average overlap in the sets of species required for each

348 pair of processes (after Hector & Bagchi 2007).

349

350

14
 Princeton University Press Guide to Ecology

Grassland plots
High diversity Low diversity

500 1000 1500

Hay yield

Midwest Reservoirs
1000
Pounds per acre (ln)

500
200
100

1 2 5 10 20

Number of species of fish (ln)

Figure 1

15
 Princeton University Press Guide to Ecology

1200

1000
Aboveground biomass production (g)

800

600

400

200

2^0 2^1 2^2 2^3 2^4 2^5

Plant species richness

Figure 2

16
 Princeton University Press Guide to Ecology

Overyielding Transgression

4
Transgressive overyielding, LRtrans

3
Net diversity effect, LRnet

2
1
0

0
-1
-1

-2
4 5 6 7 8 4 5 6 7 8

Experimental duration, ln(days) Experimental duration, ln(days)

Complementarity Selection
600

600
Compelemntarity effect, CE

400

400
Selection effect, SE
200

200
0

0
-200

-200

4 5 6 7 8 4 5 6 7 8

Experimental duration, ln(days) Experimental duration, ln(days)

Figure 3

17
 Princeton University Press Guide to Ecology

100
50
Species abundance (a) (b)

Total biomass
30

60
10

20
0

0
0 5 10 15 0 5 10 15

100
50

(c) (d)
Species abundance

Total biomass
30

60
10

20
0

0 5 10 15 0 5 10 15

Time Time

Figure 4

18
 Princeton University Press Guide to Ecology

Species richness

Figure 5

19
 Princeton University Press Guide to Ecology

Figure 6

20

View publication stats