Oecologia DOI 10.1007/s00442-014-3191-4

CONCEPTS, REVIEWS AND SYNTHESES

Reproductive allocation in plants as affected by elevated carbon dioxide and other environmental changes: a synthesis using meta‑analysis and graphical vector analysis Xianzhong Wang · Daniel R. Taub · Leanne M. Jablonski 

Received: 29 May 2014 / Accepted: 12 December 2014 © Springer-Verlag Berlin Heidelberg 2014

Abstract  Reproduction is an important life history trait that strongly affects dynamics of plant populations. Although it has been well documented that elevated carbon dioxide (CO2) in the atmosphere greatly enhances biomass production in plants, the overall effect of elevated CO2 on reproductive allocation (RA), i.e., the proportion of biomass allocated to reproductive structures, is little understood. We combined meta-analysis with graphical vector analysis to examine the overall effect of elevated CO2 on RA and how other environmental factors, such as low nutrients, drought and elevated atmospheric ozone (O3), interacted with elevated CO2 in affecting RA in herbaceous plants. Averaged across all species of different functional groups and environmental conditions, elevated CO2 had little effect on RA (−0.9 %). RA in plants of different reproductive strategies and functional groups, however, differed in response to elevated CO2. For example, RA in iteroparous wild species decreased by 8 %, while RA in Communicated by Ylo Niinemets. Electronic supplementary material  The online version of this article (doi:10.1007/s00442-014-3191-4) contains supplementary material, which is available to authorized users. X. Wang (*)  Department of Biology, Indiana University-Purdue University Indianapolis, 723 West Michigan Street, Indianapolis, IN 46202, USA e-mail: [email protected] D. R. Taub  Biology Department, Southwestern University, 1001 East University Avenue, Georgetown, TX 78626, USA L. M. Jablonski  Marianist Environmental Education Center, 4435 East Patterson Road, Dayton, OH 45430, USA

iteroparous crops increased significantly (+14 %) at elevated CO2. RA was unaffected by CO2 in plants grown with no stress or in low-nutrient soils. RA decreased at elevated CO2 and elevated O3, but increased in response to elevated CO2 in drought-stressed plants, suggesting that elevated CO2 could ameliorate the adverse effect of drought on crop production to some extent. Our results demonstrate that elevated CO2 and other global environmental changes have the potential to greatly alter plant community composition through differential effects on RA of different plant species and thus affect the dynamics of natural and agricultural ecosystems in the future. Keywords  Drought · Elevated ozone · Nutrient stress · Iteroparity · Semelparity

Introduction Reproduction is a key process in the completion of the life cycle in a plant and the regeneration of new plant populations (Bazzaz and Ackerly 1992). The amount of resources allocated to reproduction in a plant is an important measure of how successful a plant is in converting resources into propagules and thus a measure of plant fitness (Harper and Ogden 1970; Reekie and Bazzaz 1987b). The total amount of resources available for distribution in a plant, however, is not unlimited, and the allocation among competing functions is considered mutually exclusive (Bazzaz and Ackerly 1992; Harper and Ogden 1970). For example, allocation of more resources to the root system for nutrient uptake reduces the amount of resources available for allocation to leaves for carbon (C) assimilation from the atmosphere (Bazzaz and Ackerly 1992). Plants balance allocation of limited resources among competing demands for growth,

13

Oecologia

reproduction, defense and other functions under a particular set of environmental conditions (Karlsson and Mendez 2005; Reekie and Bazzaz 2005). Although various currencies have been proposed to quantify resource allocation in plants, biomass is perhaps the most widely applicable and most commonly measured currency (Bazzaz et al. 1987; Harper and Ogden 1970; Hickman and Pitelka 1975; Reekie and Bazzaz 1987a). In this meta-analytical review, we focused on biomass as the currency of interest, examining allocation of biomass to reproduction for plants of different reproductive strategies and functional groups in the context of elevated carbon dioxide (CO2) in the atmosphere. We concurrently examined how other global environmental changes, particularly low nutrient, drought and elevated ozone (O3) level in the troposphere, interacted with elevated CO2 in affecting pattern of reproductive allocation (RA) in plants. Different functions within a plant compete for a limited supply of biomass, thus any environmental factor that affects growth of a plant may alter biomass allocation. It is well documented that elevated CO2 in the atmosphere greatly enhances biomass production in terrestrial plants (Bazzaz 1990; Curtis and Wang 1998; de Graaff et al. 2006; Norby et al. 2005; Nowak et al. 2004; Wang 2007). Because biomass production in plants grown at elevated CO2 increases, the total amount of biomass available to be allocated to various organs, including reproductive structures, will increase in a CO2-enriched environment of the future. Over the past decades, there have been a large number of empirical studies that examined biomass production as well as reproduction in plants grown at elevated CO2 (Jablonski et al. 2002; Wang and Taub 2010; Wang 2007). Total reproductive biomass (RB) typically increases at elevated CO2 in plants (Hikosaka et al. 2011; LaDeau and Clark 2001), but the effect of elevated CO2 on RA, namely the proportion of biomass allocated to reproductive structures, is much less studied and not well understood. Here, we extracted data from peer-reviewed articles on RB published between 1977 and 2011 and used meta-analysis to examine RA in plants as affected by elevated atmospheric CO2, the best-documented anthropogenic global environmental change (IPCC 2013). We focused on herbaceous species since there have been few studies that grew woody species in elevated-CO2 environments for their entire growth period until reproductive maturity. Because biomass production and allocation are highly dynamic, we examined “standing RA”, i.e., the instantaneous measurement of RB instead of “lifetime RA”, i.e., the measurement of RB at the completion of a plant’s life cycle (Bazzaz and Ackerly 1992; Harper and Ogden 1970; Karlsson and Mendez 2005). Because of potential impacts of RA on future plant community composition and ecosystem dynamics, we were particularly interested in whether RA in plants of different reproductive strategies (iteroparous vs. semelparous species) and functional groups

13

(crop vs. wild species and leguminous vs. non-leguminous species) responded differently to rising CO2 in the atmosphere. We were equally interested in whether other concurring global environmental changes interacted with elevated CO2 in affecting RA in plants. The three non-CO2 environmental conditions that were examined are low soil nutrients, drought and elevated O3 in the atmosphere. Soil nutrient and water levels have long been known to interact with elevated CO2 in affecting biomass production (Ainsworth and Long 2005; Grant et al. 1995; Leakey et al. 2009; Wall et al. 2001; Wechsung et al. 2000). Atmospheric O3 at the ground level is known to reduce plant growth at current concentrations even in relatively unpolluted areas (~40 p.p.b.) and its adverse effect on plant growth, particularly on root growth, becomes more pronounced as O3 concentration increases (Feng et al. 2008; Wittig et al. 2009). A comprehensive meta-analysis that synthesizes results from studies conducted under conditions of elevated CO2 and other environmental changes will therefore greatly improve our understanding of RA under various global change scenarios. In addition to utilizing the synthetic power of metaanalysis, we used graphical vector analysis (GVA) in this meta-analytical review. GVA is a multivariate analysis that we modified and adopted from Haase and Rose (1995) to simultaneously examine RB, above-ground biomass (AGB) and RA in a single graph. GVA shows changes in RA in relation to plant size, which is normalized to 100 %. Normalization of plant size can help distinguish the effects of elevated CO2 and plant size on RA, as plant size itself is an important determinant of RA in plants (Bazzaz et al. 2001; Harper and Ogden 1970). GVA has been successfully used in studying plants to test hypotheses about mechanisms behind changes caused by different experimental treatments (Imo and Timmer 1997; Koricheva 1999). The use of GVA technique complements the conventional metaanalytical approach by quantifying the change of a ratio variable, RA, in response to treatment (elevated CO2). It reveals the magnitude of responsiveness to elevated CO2 in RB and AGB, two variables that determine the magnitude of change in RA, as well as in RA itself. This enables the simultaneous examination of RA and its determinants as affected by elevated CO2 on the same graph, thus improving our understanding of how RA will be affected in plants of different reproductive strategies and functional groups under various scenarios of global environmental changes.

Materials and methods Database construction The database for this meta-analytical synthesis was constructed by extracting results from previously published

Oecologia

journal articles that report impact of elevated CO2 on biomass production and reproduction. One of us (X. Z. W.) has maintained a personal database and library of publications on the effects of elevated CO2 on plants (including any aspect of plant physiological, anatomical, morphological and growth responses). This database has been frequently updated through searches of the ISI Web of Knowledge on the keywords: “elevated CO2,” “increased CO2,” “CO2 enrichment,” and “atmospheric CO2.” We started screening over 8,000 articles on elevated CO2 in this database in 2009 looking through every data table and figure to find articles that report both AGB and RB of plants. The screening continued into 2013 to keep the database and extracted results up to date, although we did not find any 2012 and 2013 publications that met our criteria for inclusion. Compared to common meta-analysis techniques using specialized keyword searches (for example for the current study use of terms such as “reproduction and elevated CO2”), this manual search is likely to have some important advantages. The first advantage is that we will have identified a larger number of studies with relevant data, since in many of these studies the variables that we are interested in were of only secondary interest to the authors of the empirical studies and would not have been noted in the keywords. An additional advantage is that since the data we are interested in were not typically of central importance in the original papers (which, as is typical in elevated CO2 plant studies often report numerous distinct measurements on the plants), publication bias is likely to be much less of an issue than with studies that focus attention on the particular set of variables used in a meta-analysis. Any article reporting results from experiments that met all the following requirements was included: (1) ambient CO2 treatments

Reproductive allocation in plants as affected by elevated carbon dioxide and other environmental changes: a synthesis using meta-analysis and graphical vector analysis.

Reproduction is an important life history trait that strongly affects dynamics of plant populations. Although it has been well documented that elevate...
564KB Sizes 0 Downloads 4 Views