Ecosystem resilience is affected by the loss of biodiversity

Jeremy J. Heath

27 March 2009

 

            This article describes how biodiversity relates to ecosystem resilience.  It begins with a description of biodiversity and the extrinsic importance to humans in terms of ecosystem services such as clean water and soil formation.  These services alone should be sufficient to justify its protection, but this article goes on to show that the importance of biodiversity in maintaining ecosystem resilience may have even more profound implications.  This is in part because biodiversity is a key factor in maintaining ecosystem services at a desired level.  The article ends with an overview of what we are doing to protect biodiversity and what major changes need to occur to facilitate protection efforts.

 

             Biodiversity is more than the sum of all the unique organisms on earth; it is inherently dependent on scale and perspective.  Biodiversity is often conveyed as the number of unique (i.e., reproductively isolated) species at a regional or ecosystem level; but it can also be expressed at the level of the planet, biome, landscape, ecosystem, community, population, or individual (i.e., genotypes).  Organisms, particularly plants, manufacture a diverse group of secondary metabolites that could be considered as a measure of biodiversity.  Furthermore, the various molecular processes that organisms have or the genes that drive those processes are diverse.  One could even consider the multitude of ways that an ecosystem process such as primary production, decomposition, mineralization, or evapotranspiration (see Chatterjee) might be manifested in a system as biodiversity. 

Changing the scale at which biodiversity is considered can alter ones perspective of the impact of a particular ecosystem or regional change in biodiversity.  These various scale-dependent perspectives on biodiversity are important and ecologists take account of this effect of scale with different measures of biodiversity such as alpha, beta, and gamma diversity (Figure 1, also see Harrison et al. 2004).  For instance, in terms of species richness, the establishment of an invasive species from a remote area will increase a region’s biodiversity (alpha diversity), but will decrease the difference between regions (beta diversity) and have no immediate effect on global biodiversity (gamma).  This idea of scale is considered here to facilitate a projection of the implications of the research described below on human societies.  Keep it in mind, it will become apparent later.

 

Figure 1. Venn diagrams representing alpha, beta, and gamma biodiversity.  Alpha diversity is simply the number of unique units (e.g., species) in a particular region.  Beta diversity is the number of unique units not shared by two areas.  And gamma diversity is the sum of all unique units in all regions.  Biodiversity can be calculated at any level of organization from genotypes to biomes, hence the use of the generic term “units.”

 

 

             Biodiversity is important to humans for a number of obvious reasons, which are well laid out in the Millennium Ecosystem Assessment (MEA 2005).  There the authors list the various ecosystem services provided by biodiversity which include provisioning, regulating, cultural, and supporting services.  Provisioning services include things such as food, water, fiber, and fuel.  Biodiversity has the capacity to regulate the quality and intensity of climate, disease, and water supplies. 

Increasingly ecosystems are being engineered or conserved to remediate or maintain clean water supplies (see Powell).  Biodiversity is important to culture through providing recreational and spiritual services and opportunities.  It provides space and materials for education and a sanctuary for reflection, which is of considerable importance to many cultures and to human welfare in general.  Primary production and soil formation are clearly linked to biodiversity and without these supporting services humans and most of the life on earth could not exist.  Tropical regions are in fact the most diverse and productive regions on earth.  Given that human existence is linked to the diversity of life on earth, one may be surprised that when its actual economic value is calculated it makes up only about 11% of the global economy (Pimentel et al. 1997).  It appears obvious that we are not considering the indirect economic value of biodiversity nor are we internalizing the costs associated with the destruction of biodiversity that accompanies the harvesting natural resources.

 

             Lack of political will, market pressure, general apathy, bureaucratic hurdles, and national security will all contribute to a slowed response to the need to protect this valuable resource; therefore, it is necessary to prioritize which natural areas need immediate and focused attention.  However, deciding which of these regions to focus on is difficult.  Do we give priority to those regions with the highest number of species or those regions that are most vulnerable to human impacts?  Are these two factors actually the same thing or does increasing biodiversity equate to increasing resilience and thus lower vulnerability?  Enter the area of ecological research called ecosystem resilience.  This is the study of the ability of a particular ecosystem to resist change in the face of natural or human perturbations.  It is generally assumed that by protecting those ecosystems that are the most sensitive to perturbations, we will maximize the ecosystem services that we can ultimately preserve.  I am not arguing that more diverse regions are less important to protect, in fact we will see that even the most diverse systems are still vulnerable to change.

 

            This review focuses on the study of ecosystem resilience with respect to the effects of biodiversity on this important process.  Although ecosystem resilience; that is, the propensity of a system to continue to provide a certain level of service despite a dynamic and changing environment, can be eroded by many factors such as pollution (see Powell), climate change (see Council, Jenkins), land-use change (see Inclan, Jenkins), or other natural and anthropocentric stressors, biodiversity is the focal factor in this review.  This is because ultimately it is the effect these stressors have on biodiversity that results in ecosystem service degradation.  I begin by discussing the study of ecosystem resilience and then move to three major case studies that document the effect of biodiversity on some of these ecosystem services.  The third section reviews the efforts and problems associated with protecting biodiversity.  It also illustrates how focusing on ecosystem resilience and particularly the value of biodiversity in this regard can be used as further justification for protecting biodiversity.

 

Ecosystem resilience

 

            Ecosystem resilience is detected in systems by monitoring ecosystem processes such as primary productivity, decomposition, or conceivably any quantifiable ecosystem service.  Various hypotheses have been proposed to describe the effect of biodiversity on ecosystem resilience.  These hypotheses are referred to as (1) redundancy, (2) rivet-popping, (3) the portfolio effect, (4) idiosyncratic, (5) non-linear, (6) compensating, (7) keystone, and (8) hump-shaped.  Some workers have also considered a null model, which predicts no effect (i.e., a line with zero slope) or variability associated with changing biodiversity and is used to detect significant effects.  The predictions of some of these models are illustrated in Figure 2.

 

Figure 2. Various hypothetical models predicting the effect that changing biodiversity will have on ecosystem function.  The predicted resiliency is evident by the degree of change in ecosystem function.  Taken directly from Naeem 1998.

 

 

             The rivet-popping hypothesis was developed by Ehrlich and Ehrlich (1981) and suggests that ecosystem function will increase in a stepwise fashion with increasing biodiversity.  In practice this would result in a linear relationship unless enough data is available to resolve the steps.  Imagining this process occurring as species are removed suggests that there is a threshold number of species below which the ecosystem process is negatively affected.  It is analogous to removing rivets from an airplane wing.  Removing a few rivets would result in little effect on the structural integrity of the craft, but removing some threshold number would cause structural modules to collapse.

 

            The idiosyncratic hypothesis (Naeem et al. 1995) predicts that ecosystem function will vary drastically and be ultimately dependent on the specific characteristics of the species added or removed from the system.  This idiosyncratic behavior is characteristic of organisms like earthworms that contribute to ecosystem function indirectly.  For instances, by moving plants seeds in or out of the germination zone earthworms can change the composition of the plant community thereby affecting ecosystem processes (Naeem et al. 1994).  These organisms have unpredictable effects on ecosystem function and are given the name ecosystem engineers.  Another example is Rhopalomyia solidaginis, a small galling insect that causes large bunch galls on the stems of goldenrod.  This tiny fly creates habitat for a number of other creatures and has been called an ecosystem engineer (Crawford et al. 2007).  Beavers are the most obvious ecosystem engineers and many invasive species alter or create habitat that may result in further invasions ultimately homogenizing biodiversity.

 

            The redundancy (Walker 1992) and compensating (Sala et al. 1996) hypotheses are similar in that they both have several species sharing the same functional role, but in the compensating model one or a few species may dominate per function.  For instance, a forest may have many tree species (primary producers), but be dominated by one particular species.  Subordinate species are kept at a residual level by their inability to compete with the dominant species.  During a perturbation the dominant species may be driven to a low level.  This provides the subordinate species with the opportunity to rise in abundance and fill the void and maintain ecosystem function.  Although both the redundancy hypothesis and the compensatory hypothesis share the same overall profile, they differ in the abundance of each species at the onset of a perturbation.  The redundancy hypothesis proposes that species are equally abundant and contribute equally to ecosystem function; therefore, removal of a species has little effect on the system as a whole.  The result of either process is little change in ecosystem function during the disturbance and the ultimate succession of the disturbed area to the original condition (Walker 1992 and Lawton and Brown 1993).

 

            The keystone hypothesis (Sala et al. 1996) postulates that the removal of a key species can cause the ecosystem process or service to collapse.  The non-linear hypothesis shares similarities with both the compensating and redundancy hypotheses, but differs in the range over which resilience is observed and the existence of different thresholds for collapse and regeneration (Carpenter 1996).  After a system like this changes state, returning to the preferred state with regard to ecosystem function requires that biodiversity increase to a level higher than the point of degradation.  The humped-shaped model suggests that there is more than one state of ecosystem function for any give level of biodiversity.  The humped-shaped response to biodiversity proposed by Rosenzweig and Abramsky (1993) reflects the complexity of species interactions and suggests that it will be difficult to develop any general model for the effect of biodiversity on ecosystem processes.

 

            The portfolio effect (see Tilman et al. 1998 for a theoretical treatment) is a concept to explain how redundant species may contribute to ecosystem resilience and incorporates the idea of response diversity and competitive interactions.  Response diversity is critical to understanding how both the compensatory and redundant systems might operate.  Response diversity simply means that different species within a functional group are affected differently by a given stressor that may diminish ecosystem function.  That is, some may be resistant while others are susceptible to the stress.  A system without response diversity is unlikely to be resilient.  When studying biodiversity with respect to ecosystem function and resilience, it is important to distinguish between apparent redundancy and real redundancy.  A system may be extremely diverse with respect to the number of species within a functional group, but still have low resilience if those species lack response diversity (i.e., are only apparently redundant, Folke et al. 2004). Compensatory dynamics are extremely important in understanding the effect of biodiversity on ecosystem resilience and these dynamics are well illustrated in at least two of the case studies to follow.

 

Three case studies support two of the above hypotheses

 

            I will review three main studies conducted to test the effect of biodiversity on ecosystem resilience and they include Lawton’s Ecotron experiments, Tilman’s Minnesota grassland work, and the Little Rock Lake acidification project.  The grassland and lake acidification projects were both long-term and large-scale projects, but differ in the ecosystems and types of stress applied.  The Ecotron experiments were relatively short-term experiments, but benefitted from being conducted in strictly controlled environments.

 

            The Ecotron Experiments.  Lawton’s Ecotron experiments (Lawton et al. 1993) were conducted in 16, 2 x 2 m, environmental chambers with plots sizes of 1 m² (soil depth of about 40 cm) in the center of each chamber.  The chambers were regulated with respect to temperature, diurnal light cycle, relative humidity, and rainfall.  Two different experiments are described.  The first was conducted to test the effect of increasing the diversity of decomposers and herbivores on a plant community composed of three species: Trifolium dubium (a nitrogen fixing clover), Poa annua (a common grass), and Senecio vulgaris (a common Asteraceae).  The decomposers were collembolan and two earthworm species and the herbivore was a snail.  There were four treatments and all treatments contained collembolans and the three plant species mentioned.  The treatments were: (1) snails and worms, (2) just snails, (3) worms, (4) a control which contained neither snails nor worms.  They measured a number of variables (Thompson et al. 1993), but I will highlight only percent Trifolium cover here.

 

            Midway through the approximately 210-day experiment, percent Trifolium cover in the worm-only treatment increased significantly to more than twice the average of the other treatments.  Lawton (1994) attributes this increase to the idiosyncratic effects of earthworms on Trifolium and its competitors.  He identified three main factors associated with earthworms that may help explain this effect and he places emphasis on the fact that none of these factors are directly related to typical ecosystem processes such as decomposition, direct competition, or predation.  First, the presence of earthworms increased nodulation in Trifolium.  Second, Trifolium seeds germinated more readily and were protected in earthworm casts.  Third, Poa and Senecio seeds were both buried by earthworms at a suboptimal depth.  Overall, there was no obvious relationship between biodiversity and percent Trifolium cover and among the hypotheses described above the closest model was the idiosyncratic one.

 

            In the second Ecotron experiment, Lawton’s group established three hierarchical levels of biodiversity with the inclusion of four trophic levels: secondary consumers (insect parasitoids), primary consumers (arthropod herbivores), primary producers (plants), and decomposers (earthworms and collembolans).  The highest level of biodiversity included 16 plants, 5 herbivores, 2 parasitoids, and 8 decomposers.  Three measures of primary productivity including total percent cover, CO₂ flux, and percent transmittance all indicated an increase in primary productivity with increasing biodiversity (Naeem et al. 1994); that is, support for all of the proposed models.

 

            Lake acidification.  The lake acidification project was conducted on Little Rock Lake in Wisconsin (Watras and Frost 1989) and was designed to test the effect of acidification on the total biomass and community composition of zooplankton.  They divided this lake in half with a curtain and over a 6-year period progressively acidified the treatment half from a pH of 6.1 to a pH of 4.7.  Over this period the reference half remained near a pH of 6.1.  They monitored zooplankton community composition and biomass about twice a week during open-ice periods and about once every 5 weeks during ice cover. 

Their similarity index indicated that community composition (i.e., the relative biomass of individual zooplankton species) began to change almost immediately after lake acidification started.  However, the total biomass remained constant (Frost et al. 1995) for 3.5 years, but began to fall after this time.  These data strongly suggested that total zooplankton biomass was resilient to the acidification perturbation, but that there was a threshold pH at which system resilience broke down. 

The resilience observed was most likely due to a compensating response because there were rare species of zooplankton that were barely detectable in the reference half and these species increased in abundance upon acidification while others declined (Frost et al. 1995).  Furthermore, they concluded after developing a variance compensation model that it was not possible to predict resilience in this system because: (1) rare species, which were important in the compensatory response were hard to detect in unperturbed systems; (2) even if rare species could have been detected, their low abundance contributes little to the variance compensation model; and (3) natural unperturbed systems are unlikely to experience the same types of stressors as human perturbed systems.  On this latter point they emphasize that human-induced perturbations tend to be long-lasting stresses (i.e., more like press experiments) while natural perturbations are generally over the short-term (i.e., more like pulse experiments).  In conclusion, the only way to predict resilience is to conduct these types of press experiments, which are costly and not practical for assigning resilience to a multitude of varying systems (Frost et al. 1995).

 

             Minnesota grassland.  Tilman and colleagues conducted a large-scale, long-term experiment from 1982 to 1992 (Tilman and Downing 1994).  Plots (n = 207) of varying levels of plant biodiversity were set up in 1982 and characterized and measured over this ten-year period.  In 1987-1988 Minnesota received one of the most devastating droughts to occur in the previous 50 years.  This drought provided the opportunity to test the effects of biodiversity on ecosystem resilience (i.e., biomass).  An index of drought resistance was used to compare plots before and after the drought.  A value of zero indicated that biomass did not change after the drought and a negative value indicated that biomass was reduced (Figure 3).  It is clear from Figure 3 that biomass was resistant to change in the highly diverse plot, but declined sharply in low-diversity plots.  However, no level of biodiversity protected the plots from a reduction in biomass.  The same plots were followed until 1992 and another set of measurements was taken.  Again they calculated an index that compared pre-drought biomass to current biomass and found that many of the highly diverse plots returned to pre-drought conditions, but that low-diverse plots still had not recovered (Tilman and Downing 1994, data not shown here).  Upon closer inspection it was discovered that the observed resilience was due to compensatory growth of relatively drought-resistant C4 grasses.

 

Figure 3.  Increased resistance to drought with increasing plant diversity.  Redrawn from data extracted from Tilman and Downing 1994.  More negative values indicate less resistance to drought in terms of biomass production.  Each point represents the mean of several plots of equal species richness (n = 207 plots).

 

 

            Overall there appears to be a consistent increase in ecosystem function with increasing biodiversity and the compensatory model is most widely supported.  In both the grassland and the acidification study compensating factors drove the observed resilience.  And a clear increase in ecosystem process with increasing biodiversity was evident in the second Ecotron experiment, but further work will be necessary in this system to confirm the compensatory model.  The idiosyncratic model was also supported in the first Ecotron experiment with earthworms acting as ecosystem engineers.

 

            I have only described three main studies here, but other studies have shown similar results and new studies are continuously emerging that support similar effects of biodiversity on ecosystem processes.  For instance, Crutsinger et al. (2006) recently demonstrated that net primary production increases with increasing genotypic diversity within one species of goldenrod.  If biodiversity is important at the genotypic level for increasing net primary production, surely it is even more important at the species level and possibly higher.  Indeed functional links between ecosystems have been established in a number of large-scale systems (e.g., Mumby et al. 2004), suggesting that decreasing the diversity of these linked ecosystems will degrade large-scale resilience. It may be difficult to conduct biome level studies of ecosystem resilience, but if one projects the effects I have described above to higher scales (e.g., communities, ecosystems, landscapes, and biomes), the potential outcome is disconcerting.  Consequences of disturbances could be devastating if idiosyncratic of keystone models were true at the biome (e.g., tropical rainforests) or ecosystems (e.g., mangroves) level.

 

What are we doing to preserve biodiversity?

 

            The Millennium Ecosystem Assessment (MEA 2005) outlines what we are doing to protect biodiversity and those items are summarized here.  Three main features summarize our efforts to protect biodiversity to date and these include conservation efforts, sustainable use, and their integration.  Protected areas, species protection programs (e.g., US Endangered Species Act), habitat restoration programs, and seed, tissue, and gene banks are all examples of conservation initiatives aimed at protecting against biodiversity loss.  The development of markets centered on the extrinsic values of biodiversity is an example of sustainable use.  These include markets centered on ecotourism or sustainable harvesting of natural products such as guano from island archipelagos.   Ultimately sustainable use and conservation must be integrated to effectively protect biodiversity.  To achieve this will require extensive public education, better monitoring of the effects of ecosystem change on human well-being, and “increased coordination among multilateral environmental agreements and between environmental agreements and other international economic and social institutions (MEA 2005).”

 

            Clearly lacking from the above list are initiatives aimed at funding more research into understanding ecosystem resilience.  In fact, ecosystem resilience is rarely listed as an overt ecosystem service.  Despite the overarching importance of ecosystem resilience in maintaining desired levels of obvious ecosystem services, it is still unclear which systems are most vulnerable and because of the importance of rare species in compensatory dynamics it is still very difficult to predict ecosystem resilience (see Frost et al. 1995 for a more detailed discussion).  Furthermore, the perspective of scale with regard to ecosystem resilience is particularly worrisome and more research is needed at larger organizational scales to determine if it is appropriate to project findings at lower scales (e.g., genotypes and species) to higher scales.  For instances, are ecosystems with a more diverse set of communities more resilient to perturbations than less diverse ecosystems?  It appears that the effect of biodiversity on ecosystem resilience at the species level is transferrable to the genotypic level (e.g., Crutsinger et al. 2006), but does it project upwards as well?  If it does, ecologists would have an even greater justification for the immediate need for preservation of biodiversity at all scales.

 

What more do we need to do?

 

            Conservation efforts and a move toward more sustainable use are important initiatives in protecting biodiversity, but these efforts are severely hindered by more powerful counter-current forces.  Many of these forces are systemic in the developed world and include subsidies that drive the exploitation of natural resources, unsustainable agriculture, climate change, unsustainable western materialism, externalization of environmental costs, lack of relevant stakeholders in policy decision-making, and the inequitable distribution of scientific knowledge (MEA 2005).  Reversal of these forces will do much to facilitate the development of more conservation efforts and sustainable use initiatives.  Clearly, reversing these forces will be politically and socially challenging, but nevertheless I give a number of examples of how this might occur (Table 1).

 

Table 1.  Systemic problems and how to move toward reducing the destruction biodiversity that is arguably our most valuable natural resource.

Systemic Problem	How to move toward a solution
Subsidies that drive over exploitation	direct subsidies instead into promoting sustainable agriculture (see Inclan) and to subsidize more sustainable products that do not externalize costs
Climate change (see Council)	protect large areas and/or increase connectedness among protected areas to give plant and animal communities an opportunity to shift with a rapidly changing climate reduce greenhouse gas emissions in part through the development of alternative energies (see Jenkins)
Western-style materialism	numerous studies have shown that economic wealth does not equate to happiness; therefore, more and better public education programs could perturb this desire
Expensive sustainable products	a higher tax on unsustainable products could be used to subsidize more sustainable products, which would make the playing field more realistic
Uninformed consumers	certification programs need more societal and governmental support and they also might benefit from some level of consolidation to promote easier recognition by consumers
Too much market power	strong lobbying pressure consistently translates into muted voices in important policy decisions.  A more diverse group of stakeholders in all government decisions is important to preserving biodiversity
Limited access to scientific knowledge	an informed public will purchase more sustainable products, voice their opinions to government officials, work to neutralize their contribution to biodiversity destruction, and make more rational decisions about how to increase their own well-being

 

Conclusions

 

            I believe the most profound implication to stem from ecosystem resilience research is the possibility that any of the proposed models may operate at larger scales.  If this is true, then humans ought to be extremely worried that a global catastrophe is lurking.  Our complacent harvesting of natural resources and the consequential destruction of biodiversity may be significantly undermining global resilience.  This possibility should drive more research on resilience at higher scales.  We already know that ecosystem services and resilience increase with increasing genotypic and species level diversity.  Is this also true at the community and landscape level?  I would argue that humans have already destroyed enough of certain landscapes and biomes to test this hypothesis explicitly.  Think of the dead zone in the Gulf of Mexico, savannas that have been turned into deserts by overgrazing in sub-Saharan Africa and the transformation of Amazonian Rainforest into unproductive eroded waste land.  Testing this hypothesis experimentally would require significant resources, but would not be impossible.  If the results are positive, ecologists will have even more evidence to warn fervently against the further destruction of biodiversity at all scales.

 

Bibliography (*not cited within, but good additional literature)

 

Carpenter, S. R. 1996. Drivers and dynamics of changes in biodiversity. Pages 311-318 in Global Biodiversity Assessment, editor. United Nations Environmental Programme (UNEP). Cambridge University Press, Cambridge, United Kingdom.

 

Crawford, K. M., G.M. Crutsinger, and N. J. Sanders.  2007.  Host-plant genotypic diversity mediates the distribution of an ecosystem engineer.  Ecology 88(8): 2114-2120.

 

Crutsinger, G.M., M.D. Collins, J.A. Fordyce, Z. Gompert, C.C. Nice, and N.J. Sanders.  Plant genotypic diversity predicts community structure and governs and ecosystem process.  Science 313: 966-968.

 

Ehrlich, P. R., and A. H. Ehrlich.1981. Extinction: the causes and consequences of the disappearance of species. Random House, New York.

 

*Elmqvist, T., C. Folke, M. Nystrom, G. Peterson, J. Bengtsson, B. Walker, and J. Norberg. 2003. Response diversity, ecosystem change, and resilience. Frontiers in Ecology and the Environment 1(9):488-494.

 

Folke, C., S. Carpenter, B. Walker, M. Scheffer, T. Elmqvist, L. Gunderson, and C.S. Holling.2004.  Regime shifts, resilience, and biodiversity in ecosystem management.  Annu. Rev. Ecol. Evol. Syst. 35: 557-581.

 

Frost, T. M., S. R. Carpenter, A. R. Ives, and T. K. Kratz. 1995. Species compensation and complementarity in ecosystem function. Linking Species & Ecosystems :224-239.

 

*Grime, J. P. 1997. Ecology - biodiversity and ecosystem function: The debate deepens. Science 277(5330):1260-1261.

 

*Grime, J. P. 1998. Benefits of plant diversity to ecosystems: Immediate, filter and founder effects. Journal of Ecology 86(6):902-910.

 

Harrison, I., M. Laverty, and E. Sterling.Alpha, Beta, and Gamma Diversity. Connexions. 29 July 2004 <http://cnx.org/content/m12147/latest/> 13 Mar. 2009.

 

*Holt, R. D., and J. H. Lawton. 1994. The ecological consequences of shared natural enemies. Annual Review of Ecology and Systematics 25:495-520.

 

Lawton, J. H. 1994. What do species do in ecosystems. Oikos 71(3):367-374.

 

Lawton, J. H., and V. K. Brown. 1993. Redundancy in ecosystems. Pages 255-270 in E. D. Schulze and H. A. Mooney, editors. Biodiversity and ecosystem function. Springer Verlag, New York.

 

Lawton, J. H., S. Naeem, R.M. Woodfin, V.K. Brown, A. Gange, H.J.C. Godfray, P.A. Heads, S. Lawler, D. Magda, C.D. Thomas, L.J. Thompson, and S. Young.  1993.  The Ecotron: A controlled environmental facility for the investigation of population and ecosystem processes.  Philosophical Transactions: Biological Sciences 341(1296): 181-194.

 

Millennium Ecosystem Assessment (MEA). 2005. Ecosystems and Human Well-being: Biodiversity Synthesis. World Resources Institute, Washington, DC.

 

Mumby, P.J., A.J. Edwards, J.E. Arias-Gonzalez, K.C. Lindeman, P.G. Blackwell, A. Gall, M.I. Gorczynska, A.R. Harborne, C.L. Pescod, H. Renken, C.C.C. Wabnitz, and G. Llewellyn.  2004.  Mangroves enhance the biomass of coral reef fish communities in the Caribbean.  Nature 427:533-536.

 

Naeem, S. 1998. Species redundancy and ecosystem reliability. Conservation Biology 12(1):39-45.

 

Naeem, S., L. J. Thompson, S. P. Lawler, J. H. Lawton, and R. M. Woodfin.1994. Declining biodiversity can alter the performance of ecosystems. Nature 368(6473):734-737.

 

Naeem, S., L. J. Thompson, S. P. Lawler, J. H. Lawton, and R. M. Woodfin.1995. Empirical-evidence that declining species-diversity may alter the performance of terrestrial ecosystems. Philosophical Transactions of the Royal Society of London Series B-Biological Sciences 347(1321):249-262.

 

Pimentel, D., C. Wilson, C. McCullum, R. Huang, P. Dwen, J. Flack, Q. Tran, T. Saltman, and B. Cliff. 1997. Economic and environmental benefits of biodiversity. Bioscience 47(11):747-757.

 

Rosenzweig, M. L., and Z. Abramsky. 1993. How are diversity and productivity related? Pages 52-65 in R. E. Ricklefs and D. Schluter, editors. Species diversity in ecological communities: historical and geographical perspectives. University of Chicago Press, Chicago.

 

Sala, O. E., W. K. Lauenroth, S. J. McNaughton, G. Rusch, and X. Zhang. 1996. Biodiversity and ecosystem function in grasslands. Pages 129-145 in H. A. Mooney, J. H. Cushman, E. Medina, O. E. Sala and E. D. Schulze, editors. Functional role of biodiversity: a global perspective. Wiley, New York.

 

Tilman, D., and J.A. Downing.  1994.  Biodiversity and stability in grasslands.  Nature 367: 363-365.

 

Tilman, D., C. L. Lehman, and C. E. Bristow. 1998. Diversity-stability relationships: Statistical inevitability or ecological consequence? American Naturalist 151(3):277-282.

 

Thompson, L., C.D. Thomas, J.M.A. Radley, S. Williamson, and J.H. Lawton.  1993.  The effect of earthworms and snails in a simple plant community.  Oecologia 95: 171-178.

 

*Tonn, W. M. 1985. Density compensation in umbra-perca fish assemblages of northern Wisconsin lakes. Ecology 66(2):415-429.

 

Walker, B. H. 1992. Biodiversity and ecological redundancy. Conservation Biology 6(1):18-23.

 

Watras, C.J. and T.M. Frost.  1989.  Little Rock Lake (Wisconsin): Perspectives on an experimental ecosystem approach to seepage lake acidification.  Archives of Environmental Contamination and Toxicology 18: 157-165.