Competition for Light Causes Plant Biodiversity Loss After Eutrophication

See allHide authors and affiliations

Science  01 May 2009:
Vol. 324, Issue 5927, pp. 636-638
DOI: 10.1126/science.1169640


Human activities have increased the availability of nutrients in terrestrial and aquatic ecosystems. In grasslands, this eutrophication causes loss of plant species diversity, but the mechanism of this loss has been difficult to determine. Using experimental grassland plant communities, we found that addition of light to the grassland understory prevented the loss of biodiversity caused by eutrophication. There was no detectable role for competition for soil resources in diversity loss. Thus, competition for light is a major mechanism of plant diversity loss after eutrophication and explains the particular threat of eutrophication to plant diversity. Our conclusions have implications for grassland management and conservation policy and underscore the need to control nutrient enrichment if plant diversity is to be preserved.

Fertilization experiments (14) and studies of nutrient deposition in terrestrial ecosystems (5) show that increases in the availability of nitrogen (5, 6), phosphorus (7), and other nutrients—both alone and in combination (1, 4)—usually increase primary productivity and decrease plant diversity. Given that anthropogenic activity has doubled global phosphorus liberation and plant-available nitrogen during the past 50 years (8, 9), and that nutrient inputs are predicted to be one of the three major drivers of biodiversity loss this century (10), understanding the mechanisms responsible for diversity loss after eutrophication will be important for the development of effective conservation policies (11).

Most of the hypotheses proposed to explain the reduction in plant diversity after eutrophication focus on changes in competition (1215). Fertilization may increase the strength of competition generally—that is, both above and below ground (15)—or it could increase the strength of aboveground competition for light only: an asymmetric process due to the directional supply of this resource (13, 14). The hypothesis of increased competition for light (14) predicts that as productivity increases, availability of light to plants in the understory is reduced, leading to their exclusion by faster-growing or taller species that preempt this directionally supplied resource (16, 17). Surprisingly, 35 years after these alternative hypotheses were suggested, there is no consensus on the role of competition as a mechanism of plant diversity loss after eutrophication (18, 19).

To test whether diversity loss after eutrophication is due to increased competition for light, we added light to the understory of fertilized grassland communities—a manipulation inspired by competition experiments with algae (20, 21). A key advance of our approach relative to earlier work (22) is that it restores light to the species in the lower canopy that are thought to decrease in diversity as a result of deeper shading after the increase in aboveground productivity caused by eutrophication. We conducted a glasshouse experiment that combined addition of fertilizer and supplementary light in a fully factorial design. The 32 experimental plant communities were pregrown in the field for 4 years (23) before they were extracted with intact soil blocks and moved to the glasshouse. For generality, the communities comprised four different sets of six species (23) that had similar levels of diversity and, as we show, responded similarly to the experimental treatments.

Light was added to the understory of each treated community using a system of three fluorescent tubes that were raised as the canopy grew (Fig. 1). Reflectors were placed above the fluorescent tubes to direct light into the understory and to prevent it from shining up onto the underside of the leaves of the taller species. To keep conditions other than light and fertilization as similar as possible, we installed the same system of fluorescent tubes in communities without supplementary light, but in this case reflectors were placed above and below the tubes to form a closed chamber from which the light could not escape. With this system, we were able to experimentally manipulate light in the understory while holding other conditions (such as temperature) constant. Aboveground biomass was harvested twice a year during 2006 and 2007 to coincide with the cutting regimes typical of European meadows, and other key variables (including belowground biomass production, canopy height, availability of light in the understory, soil pH, and plant diversity) were regularly monitored (24).

Fig. 1

Schematic representation of the experimental understory light addition. To save space, two open lights and one closed light are shown in the same experimental unit. The four treatment combinations were “control” (unfertilized, closed lights), “fertilization” (fertilized, closed lights), “light” (unfertilized, open lights), and “fertilization + light” (fertilized, open lights). For generality these four treatments were applied to four different plant communities, with each combination replicated twice (n = 4 × 4 × 2 = 32).

After 2 years of treatment, fertilization had increased net aboveground biomass production and decreased diversity (24). During the second year, fertilization significantly increased production from an average of 356 ± 39 g m−2 (mean ± SEM) per harvest in the control communities to 450 ± 39 g m−2 in the fertilized treatment (Fig. 2A and table S1). The percentage of photosynthetically active radiation in the understory of the fertilized plots (5 ± 4%) was significantly lower than for the controls (13 ± 4%) (Fig. 2B). Notably, when increased production was accompanied by decreased light in the understory, fertilization significantly reduced species richness (Fig. 2C): On average, 2.6 species were lost in the fertilization treatment relative to the control, around one-third of the original species richness. This loss of diversity after eutrophication is consistent with longer-term field studies (1, 5).

Fig. 2

Effects of fertilization and supplementary understory light on grassland diversity and functioning. (A) Average aboveground plant biomass per harvest in 2007. Addition of fertilizer and fertilizer-plus-light significantly increased aboveground biomass. (B) Light in 2007 measured as PAR (photosynthetically active radiation). Increased aboveground biomass significantly reduced light availability in the understory unless compensated by experimental illumination to levels comparable to control plots. (C) Species richness between 2006 and 2007. Fertilization significantly reduced species richness unless prevented by the addition of supplementary light to the understory. Points denote treatment means, and the intervals show least significant differences (treatments with nonoverlapping intervals are significantly different at P = 0.05).

When applied together with fertilization, the additional understory light compensated for the increased shading caused by the greater aboveground biomass production and generated levels of understory light (12 ± 4%) that were indistinguishable from those in the control plots (13 ± 4%) (Fig. 2B and table S1). Supplementing understory light in the fertilization treatment to levels similar to the control plots prevented the loss of species and maintained comparable levels of diversity (Fig. 2C). This result was general across the four different plant communities used in the experiment; the variance component for the different species mixtures accounted for only 10% of the total of the summed variance components and was nonsignificant (likelihood ratio test: log likelihood = 1.05; χ2 = 2.10; P = 0.15). By mitigating the loss of diversity caused by fertilization, this result supports the hypothesis that increased competition for light was the mechanism responsible for the decline in species richness after eutrophication.

Our communities experienced species turnover that resulted from the loss of resident species and the gain of new species from the seed bank. As in several previous studies (2527), the decrease in diversity caused by fertilization was due mainly to a decline in the numbers of species gained (Fig. 3), from 3.2 in the controls to 1.6 in the fertilized plots (table S2). This result was also consistent across the four nonoverlapping communities used in our experiment: The variance component for the different species mixtures only accounted for 2.5% of the total of the summed variance components and was nonsignificant (likelihood ratio test: log likelihood = 0.81; χ2 = 1.61; P = 0.20). There was a marginally significant bias against the establishment of short-statured perennial grasses and forbs, but the overall response was not driven by particular species (24).

Fig. 3

Species turnover. Decreased diversity in fertilized plots was mainly caused by reduced numbers of species gained. Results are shown as in Fig. 2.

Our understory light addition treatment also had consequences for ecosystem functioning. Net aboveground biomass production in the controls was limited by nutrients (although we cannot exclude light limitation of the taller species as well) because it was increased by fertilization (Fig. 2A and table S1). Without fertilization, the productivity of plants in the understory was not light-limited, because supplementary light had no effect when applied to unfertilized communities (Fig. 2A). However, the productivity of plants in the understory of the fertilization treatment was light-limited, because in fertilized communities the additional light increased average net aboveground production per harvest to 575 ± 39 g m−2 (Fig. 2A). These responses suggest colimitation of productivity by light and nutrients, where the taller species are nutrient-limited while understory species in the fertilization treatment are light-limited. More generally, our results suggest that productivity of the upper canopy and understory can be limited by different factors as a result of the directional supply of light.

Species loss could be due to increased competition both above and below ground (15). To address this possibility, in the second year of the glasshouse experiment we added seedlings of two species not originally present to the 32 experimental communities to measure the strength of belowground competition. Transplanted seedlings planted in plastic tubes to reduce belowground competition were compared with seedlings exposed to full root competition. The results were consistent with competition for light as the main mechanism of diversity loss: When grown without root exclusion tubes (that is, with belowground competition), seedling mortality (as a proportion) strongly increased with nutrient addition from 0.29 to 0.87, but was comparable to control plots when fertilization occurred together with understory lighting (Fig. 4 and table S3A). The results provided no support for a role of belowground competition in the loss of biodiversity (table S4): Removing belowground competition from fertilized plots had no detectable impact on seedling mortality (table S3B) or seedling biomass (change in biomass = 0.3 g, 95% confidence interval = –1.0 to 1.4).

Fig. 4

Seedling mortality. Fertilization significantly increased seedling mortality. Removing belowground competition had little impact on seedling mortality, which suggests that competition for soil resources plays no detectable role in diversity loss. Results are shown as in Fig. 2.

Although other processes can also contribute to diversity loss, there was no evidence that they were important in our study. Fertilization can reduce grassland diversity through acidification (2) or through the accumulation of plant litter (25, 26, 28, 29). However, we found no detectable differences in pH after fertilization (fig. S1 and table S5). There was also little buildup of litter during our experiment, which suggests that the negative effects of increased aboveground productivity might have strengthened in the longer term if litter accumulation had occurred.

Together, our results are consistent with increased competition for light as a major mechanism of diversity loss after eutrophication of grassland communities. Fertilization increased productivity and canopy height, and led to reduced light in the understory. In turn, this led to a reduction in diversity, particularly of low-statured perennial grasses and forbs, mainly through reduced recruitment. Other mechanisms also cause loss of plant diversity, but they played no detectable role in our case. Supplementing levels of understory light in fertilized communities reduced competition for light, sustained seedling establishment, and maintained plant diversity despite the additional nutrient inputs.

Some earlier studies (30) have demonstrated the importance of competition for light indirectly by tying back the vegetation. Our results advance a long-running debate in community ecology by providing a direct experimental demonstration of the importance of asymmetric competition for light as a mechanism of plant diversity loss. More generally, our work explains and emphasizes the need to develop conservation policies and management procedures that prevent eutrophication if biodiversity is to be conserved.

Supporting Online Material

Materials and Methods

SOM Text

Figs. S1 to S3

Tables S1 to S7


References and Notes

  1. See supporting material on Science Online.
  2. We thank E.-D. Schulze and C. Körner for discussion of light manipulation, B. Schmid and L. Turnbull for comments on the manuscript, and L. Wacker, E. Vojtech, G. Feichtinger, and T. Zwimpfer for helpful suggestions and field work assistance. Y.H. is funded by Swiss National Foundation grant 3100A0-107572 to A.H. The project was conceived by A.H., designed by A.H., P.A.N., and Y.H., conducted by Y.H., analyzed by Y.H. and A.H., and written up by Y.H. and A.H. with input from P.A.N.
View Abstract

Stay Connected to Science

Editor's Blog

Navigate This Article