E141 - BioCON: Biodiversity, CO2, and Nitrogen

Introduction

CO2, Nitrogen, and Biodiversity Many ecosystems around the world are experiencing simultaneous increases in atmospheric CO2 levels and N deposition, and decreases in biodiversity. The potential importance of these aspects of global environmental change, coupled with a lack of understanding of their interactions (Vitousek 1994), led us to develop the long-term BioCON experiment, which addresses the direct and interactive effects on grassland ecosystems of elevated CO2, added N, and varying plant diversity, including shifts in both richness and composition (e.g., Reich et al. 2001ac, 2004, 2006ab). The BioCON project addresses basic scientific questions about coupled biogeochemical cycles, biodiversity, and other issues while also providing information relevant to society about the implications of these global change variables.

BioCON focuses on 4 key questions: (1) Do CO2 and N interact at physiological, whole plant, multitrophic, community and/or biogeochemical scales, on short- and long-term time horizons? (2) Do plant species and/or functional group diversity and composition influence responses to CO2 and N? (3) Are there linear or non-linear temporal changes in effects of treatments on individual, community, or ecosystem metrics? (4) What mechanisms (physiological, biotic interaction, biogeochemical, etc.) explain the patterns observed in addressing questions 1-3? In other words, how does the integration of plant, consumer, mutualist, and decomposer interactions at multiple temporal scales lead to the responses observed at tissue to ecosystem scales across various time scales?

The BioCON experiment (E141) directly manipulates plant species numbers (1, 4, 9, or 16 perennial grassland species randomly chosen from a pool of 16 species, planted as seed in 1997), soil N availability (ambient soil vs. ambient soil + 4 g N m-2 yr-1), and atmospheric CO2 concentrations (ambient vs. +180 ppm, beginning in 1998) in a well-replicated split-plot experiment. It includes 296 individual plots, each 2 m x 2 m, in six 20-m diameter rings, three exposed to ambient CO2 and three to elevated CO2 using freeair CO2 enrichment. Additional fully factorial experiments (many plots serve multiple experiments) include tests of species composition (in monoculture) x CO2 x N (n=128 plots, Reich et al. 2001c), functional group composition x CO2 x N (n=176, Reich et al. 2004), species richness x CO2 x N at a standard functional group richness (n=176), and functional group richness x CO2 x N controlling for species richness (n=123).

Global change-related shifts in temperature, precipitation, atmospheric CO2, and N deposition will each likely impact terrestrial ecosystem processes, however the effects of each global change element alone may be influenced by other global change factors, via antagonistic and synergistic impacts and by indirect effects on soil resources and soil biota that modulate subsequent ecosystem responses. Yet, considerable uncertainty exists regarding the direction, magnitude, and ubiquity of such interactions, posing a significant challenge for predicting ecosystem feedbacks to multiple global change drivers. In 2007, we began a 5-year sub-experiment examining interactions of CO2, N and water availability. The water manipulation (ambient and -45% rainfall, achieved via temporary, portable rainout shelters) was added to the ongoing CO2 x N treatments that began in 1998. The objective of a new BioCON research subexperiment is to incorporate experimental warming into the ongoing grassland manipulation of precipitation, CO2, and N to elucidate their interactive effects on long-term ecosystem response. The WWCON experiment is thus designed to determine the direct and interactive effects of warming, water, CO2 and N on the productivity, biogeochemical cycling, and dynamics of plant and soil communities in a perennial grassland ecosystem. WWCON uses 48 plots from BioCON, all originally planted with 9 species in 1998.

Key Results

BioCON is a highly productive experiment that led to 44 peer-reviewed papers during our current award. Its most novel element is the simultaneous, long-term manipulation of and examination of effects of multiple global change drivers on ecosystem processes ranging from plant physiology to plant and soil communities to ecosystem biogeochemistry. BioCON is to our knowledge one of only three studies in the world capable of providing long-term evidence on joint effects of CO2 and N on biodiversity and ecosystem function, and the only experiment involving either CO2 or N and biodiversity. Here we highlight its two most important findings of the past 6 years.

(1) CO2 and N interact non-additively in influencing plant biodiversity (Fig. 7; Reich 2009). Over 10 years, elevated N reduced species richness by 16% at ambient CO2 but by just 8% at elevated CO2.This resulted from multiple effects of CO2 and N on plant traits and soil resources that altered competitive interactions among species. Results of this study have important implications for natural ecosystems under global change, because they demonstrated that altered CO2 and N regimes had significant, interactive, persistent impacts on species diversity resulting from direct, but mostly indirect effects on plant and ecosystem processes. The sensitivity of plant diversity to factors that themselves were sensitive to CO2 and N suggests that predicting responses of biodiversity at local scales may be challenging, as responses to multiple global change drivers may not be generally predictable from the responses to each alone.

(2) Nitrogen limitation of plant growth, which is common worldwide, constrains biomass responses to CO2 over the long-term (Reich et al. 2006ab, Reich and Hobbie In Prep) (Fig. 5). In 2007, the Intergovernmental Panel on Climate Change (IPCC) stated that the largest uncertainty in the global C cycle – and hence a key to predicting future climate change – involves the size of the so-called CO2 fertilization effect. Although photosynthesis and plant productivity generally increase with rising CO2 levels in most plant communities, whether this response will decelerate or "saturate" is not known, and hence we lack the ability to predict the fraction of future global C emissions that terrestrial ecosystems acquire and store. The long-term constraint on the CO2 fertilization effect due to natural N limitation confirms a key criticism (Hungate et al. 2003) of earlier IPCC efforts.

Overall BioCON provides a platform for examining the myriad processes that contribute to interactions such as described above. For instance, the long-term interacting effect of CO2 and N on biomass and biodiversity occur despite a lack of CO2 x N interaction on photosynthesis (Lee et al. 2011) and likely result from complex effects of species identity and diversity, along with CO2 and N, on belowground communities and processes (e.g., Dijkstra et al. 2006a, 2007, West et al. 2006, He et al. 2010, 2012, Chung et al. 2007, 2009, Adair et al. 2009, 2011, Reich 2009, Antoninka et al. 2011, Schnitzer et al. 2011, Reid et al. 2012, Deng et al. 2012; Fig. 21). Collectively, these findings have important implications globally. For instance, because of N limitation and biodiversity losses, global estimates of potential C sequestration in the face of rising CO2 may be currently considerably over-estimated. If this is true, atmospheric CO2 concentrations (and associated global temperatures) may increase more quickly than anticipated. BioCON data have been important in broader analyses, syntheses, and meta-analyses of biodiversity, N, and CO2 effects (Reich et al. 2006b, Isbell et al. 2011, Schnitzer et al. 2011) and development of global databases (Kattge et al. 2011). In addition, a team of non-CDR researchers used results of CDR decomposition studies (data downloaded from our website) in analyzing the effects of plant biodiversity on decomposition rates (Srivastava et al. 2009).

(3) Results to date of the water manipulation include a near total elimination of the CO2 effect on productivity when both water and N are limited (Reich et al. Submitted) and complex pathways by which the resource treatments directly and indirectly drive trophic networks belowground (Eisenhauer et al. 2012). These include both direct effects of CO2 and N on the abundance and diversity of soil animals, as well as indirect effects mediated by changes in other abiotic and biotic components of the soil environment (Fig. 22).

Future Research

Because the responses to CO2, N, and diversity have been highly dynamic temporally (Reich et al. 2006a, Reich and Hobbie In Prep), this experiment will continue providing valuable insights into the interactive effects of multiple global change factors. As one of the longest running multiple global change factor open-air experiments in the world, the results will be among the best available for showing the long-term impacts of such drivers on population, community, and ecosystem responses. Thus, we propose continued annual measurement of many plant, soil, community, and ecosystem variables across all 370 plots each year.

In addition, we will expand analyses of soil communities and trophic interactions, including their response to global change, and impacts on ecosystem structure and function, building on early and ongoing work at BioCON (e.g. Chung et al. 2009, He et al. 2010, 2012, Deng et al. 2012, Eisenhauer et al. 2012, Weisenhorn submitted, Weisenhorn et al. In Prep). This will include use of molecular tools to characterize both bulk soils and the rhizosphere microbiome (e.g., functional gene arrays, 454 pyrosequencing, and 16s ribotyping) in collaboration with research groups led by J. Zhou (Oklahoma), J. Dangl (U. North Carolina), and C. Henry (Argonne National Laboratory). It will also include a comprehensive plan to improve the mechanistic understanding of soil multitrophic interactions in shaping the relationship between producer diversity and ecosystem functioning under varying CO2 and N conditions. This collaborative project will be led by N. Eisenhauer (Technische Universität Darmstadt, Germany), and it comprises several complementary subprojects, including experimental tests of the significance of positive and negative soil feedback effects across biodiversity, CO2, and N combinations.

The new warming manipulation (3°C) will be added in 2012, resulting in a CO2 x N x water x temperature factorial experiment. The study is largely funded through a grant from the NSF Ecosystems program, but will be available as a platform for LTER-related investigations, e.g., by undergraduates, graduate students, and post-docs, just as the core BioCON experiment has been. It will test the overarching hypothesis that global change drivers interact, such that responses will not simply be additive. Unlike interactions of CO2, N, and water, for which multiple limitation theory provides a relatively simple conceptual framework, experimental warming will induce different responses during cool vs. warm, or wet vs. dry, times during the growing season, with myriad possible pathways for interactions. For all four of the treatments, interactions will be influenced by the effects of drivers on the availability of soil resources and on soil biota, and by weather in each specific growing season. Warming will be imposed on a subset of species-rich plots in BioCON, to achieve a 2 x 2 x 2 x 2 factorial manipulation of temperature (ambient and +3°C), growing season precipitation (ambient and -45% rainfall, achieved via temporary rainout shelters), CO2 (ambient and +180 ppm, achieved via FACE technology), and N (ambient and +4 g N m-2 y-1).

Simultaneous soil and vegetation warming will be achieved by synchronized deployment of infrared heat lamps that warm aboveground plant structures, and a network of low-profile, buried electric pins that concurrently warm the soil to 0.75 m deep. Integrated microprocessor-based feedback control will maintain a fixed temperature differential (+3°C) between warmed and ambient control plots. We have successfully deployed similar technology in an experiment in the southern boreal forest in northern Minnesota (Reich et al. submitted, http://forestecology.cfans.umn.edu/B4WARMED.html). The new study will determine the interactive effects of these four global change factors on a suite of responses including: plant physiology, NPP, phenology, symbiotic N fixation, soil N availability, soil CO2 flux, soil food webs, as well as the soil microenvironment. This project will provide one of the few empirical datasets describing the interactive effects of multiple important human-caused global change factors on terrestrial ecosystem processes, thereby enhancing mechanistic understanding of the ecological impacts of global change and informing models that aim to predict biotic feedbacks to such change.