See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/261328764 Development of a Pollination Service Measurement (PSM) method using potted plant phytometry Article in Environmental Monitoring and Assessment · April 2014 DOI: 10.1007/s10661-014-3758-x · Source: PubMed CITATIONS READS 4 198 5 authors, including: Thomas Stuart Woodcock Peter Kevan Rare Charitable Research Reserve University of Guelph 30 PUBLICATIONS 194 CITATIONS 338 PUBLICATIONS 6,886 CITATIONS SEE PROFILE SEE PROFILE Some of the authors of this publication are also working on these related projects: Dioecy in the High Arctic flora View project All in-text references underlined in blue are linked to publications on ResearchGate, Available from: Peter Kevan letting you access and read them immediately. Retrieved on: 17 October 2016 Environ Monit Assess DOI 10.1007/s10661-014-3758-x Development of a Pollination Service Measurement (PSM) method using potted plant phytometry Thomas S. Woodcock & Laura J. Pekkola & Cara Dawson & Fawziah L. Gadallah & Peter G. Kevan Received: 10 September 2013 /Accepted: 21 March 2014 # Springer International Publishing Switzerland 2014 Abstract The value of pollination to human society is less certain and more labour-intensive methods of polli- not limited to agricultural production, but also in the nator collection and identification, and shows promise for sustainability of ecosystems and the services that they implementation in pollination monitoring and bioassess- provide. Seed set can be used as a comparative measure ment practices. of pollination effectiveness, with minimum variability expected when other resources are not limiting. Six spe- Keywords Sustainability. Ecosystem service . cies of self-incompatible fall asters (Symphyotrichum) Pollinator conservation . Biomonitoring . were used to evaluate pollination service at 12 sites across Symphyotrichum a spectrum of expected levels of pollination. Seed set per inflorescence was generally lower at sites with lower pollinator numbers and diversity, although as expected Introduction pollinator assemblage characteristics were highly variable within and between sites. However, rankings of sites Conservation of native, wild pollinators is critical to showed consistency of response across phytometer spe- ensuring the continued reproductive success and biodi- cies and between years; the summed ranks across multi- versity of the plants on which ecosystem structure and ple species appears to have as the greatest value in function depend (Fontaine et al. 2006; Ollerton et al. Pollination Service Measurement (PSM). Abundance, 2011; Frund et al. 2013). Flowering plants form the richness, and Shannon diversity of pollinator assem- trophic basis of productivity in most terrestrial ecosys- blages were highly autocorrelated and showed variable tems, and the maintenance and sustainability of plant relationships with seed set depending on plant species populations, independent of human intervention, is cru- and temporal scale of pollinator assemblage assessment. cial for the sustainability of the ecosystems themselves. Use of seed set to directly measure pollination service at a Animal pollinators, including but not limited to bees, site was consistent and cost effective when compared to flies, butterflies, beetles, bats, and birds, play a vital role : : : in mediating the sexual reproduction of approximately T. S. Woodcock (*) L. J. Pekkola C. Dawson 85 % of the world's flowering plants, and 78 % in P. G. Kevan temperate regions such as southern Canada (Ollerton Canadian Pollination Initiative (NSERC-CANPOLIN), School of Environmental Sciences, University of Guelph, et al. 2011). While pollination in agricultural systems Guelph, ON N1G 2 W1, Canada is routinely improved through the use of honey bees or e-mail: [email protected] other managed pollinators, similar approaches to polli- nation management in natural ecosystems are neither F. L. Gadallah Information and Indicators Division, Environment Canada, economically nor logistically feasible (Mader et al. 10 Wellington St, Gatineau, QC K1A 0H9, Canada 2010; Kjohl et al. 2011). Whether pollination service is Environ Monit Assess delivered by wild or managed pollinators, or a combi- technical expertise (and therefore incurs lower costs) nation thereof, there is a need for assessment and mon- than surveys of pollinator assemblages. For a plant to itoring of pollination success in both agricultural and be useful as a phytometer for PSM, its seed set must non-agricultural landscape elements. indicate pollination by animal pollinators and no other Historically, bioassessment and biomonitoring means. The plant must therefore be non-apomictic (un- methods have measured some aspect of community able to set seed without pollination), an obligate structure, such as richness, diversity, or abundance outcrosser (dioecious or self-incompatible, unable to (Allan et al. 1997; Townsend et al. 1997; Woodcock pollinate itself), and not wind-pollinated (its pollen must et al. 2008;Lissetal.2013), that acts as a proxy for be transported exclusively by an animal vector). In ecosystem services or processes, or an (often poorly addition, seed set must reflect levels of pollination lim- defined) concept of ecosystem “health” or “integrity.” itation, rather than resource limitation or innate limits to Using the community structure of organisms to infer the numbers of seeds produced. For example, some species rate or quality of the ecosystem processes that they of Asclepias are reliant on insects for pollination but will perform is common, but can have mixed or unpredict- produce very few seeds or fruit per inflorescence regard- able results (Karr 1981; Callicott et al. 1999; Schwartz less of initial pollination success (Wyatt 1976;Neyland et al. 2000; Costanza 2012; Liss et al. 2013). High et al. 1999). labour requirements for field sampling and sample pro- In this study, six species of fall asters cessing, and the requirement for expensive taxonomic (Symphyotrichum) which met the above requirements expertise, are significant drawbacks, and high variability were used as the test species. Plants are grown in pots in in the resulting data makes interpretation difficult and a controlled environment with abundant, standard re- development of appropriate responses difficult. In recent sources. Phytometers are experimental units of plants years, the development of approaches that directly mea- (single species or groups of several species) which are sure ecosystem function has been encouraged, although used to measure characteristics of an ecosystem in an none have been developed explicitly for pollination in a area of interest for a variety of experimental purposes biomonitoring context. Evaluation of plant reproductive (Steffan-Dewenter et al. 2002; Albrecht et al. 2007; success using ambient vegetation or potted plant Sahli and Conner 2007; McKinney and Goodell 2010; phytometers has been used to address a variety of eco- Sperling and Lortie 2010). Phytometers also allow for logical questions related to pollination. For example, replicability in both the environmental growing condi- seed set in crops or ambient vegetation has been used tions and in the plant assemblage used to determine to examine landscape-level pollination service and com- pollination success, allowing results obtained from dif- petition among plants for pollinators (Greenleaf and ferent sites to be directly compared. This study will Kremen 2006;Dauberetal.2010; Trant et al. 2010; assess the utility of the PSM approach by examining Hennig and Ghazoul 2011; Liss et al. 2013). Potted seed set of the six Symphyotrichum species at multiple plants have been used to measure pollen limitation sites in southern Ontario, Canada. Multi-year sampling (Campbell 1985; McKinney and Goodell 2010), effects of pollinator assemblages (of varying intensity) at these of neighbouring blooms on plant reproductive success sites indicates a broad gradient of abundance and diver- (Kunin 1997; Bosch and Waser 2001; Schulke and sity that allows evaluation of a priori expectations using Waser 2001; Spigler and Chang 2009; Lazaro and PSM. Totland 2010), and pollination responses to agricultural practices (Brittain et al. 2010a, b), and other habitat conditions (Steffan-Dewenter et al. 2002;Artzand Materials and methods Waddington 2006; Sperling and Lortie 2010). The Pollination Service Measurement (PSM) system Field sites described here directly measures pollination service at a site by evaluating plant reproductive success (seed set) In 2011, test plants were deployed at five sites, at which in a standard array of potted plants. This approach is pollinator sampling was ongoing as part of other studies. expected to have the advantage of directly measuring The sites covered a wide range of expected pollination the target ecosystem service rather than inferring it from service, based on the history of the sites and information pollinator assemblage data, and requires less time and gleaned from ongoing pollinator sampling (Table 1). Environ Monit Assess Ta b l e 1 Descriptions and expected pollination service (PS; based organic farm sites (CVF-1, 2) was insufficient to make a prediction on long-term observations and/or sampling of the bee and syrphid of pollination service. All pollinator sampling occurred between assemblages) at the 12 study sites. Sampling and knowledge of the May 1 and August 31 Site code Site description and sampling history No. of sampling Expected