4 Theory and methods in seed trait ecology
4.1 Identifying conservation and knowledge gaps in cultivated Brassicaceae: characterization of functional traits in crop wild relatives
Castillo-Lorenzo, E.1, Viruel, J.1, Breman, E.1
1Royal Botanic Gardens, Kew, UK
Wild species hold a key source of diversity to improve and help adapt our traditional crops to deal with future climate scenarios. Those wild species with a close genetic relationship to a crop, named crop wild relatives (CWRs), are likely to possess useful traits due to greater genetic variability and adaptability compared to crops, making them of paramount importance for research and conservation. Previous studies have focused on major crops such as pulses, cereals and forages, and their respective CWRs, but relatively little work has been done on oil crops, vegetables or fruits, such as in the family Brassicaceae. Brassicaceae CWRs have a wide diversity in morphology and genetics, and host desirable traits such as drought and salt tolerance. In this study, we investigate potential CWRs of crops in Brassicaceae, collating information published from different databases (e.g., Genesys, USDA-GRIN), with the objective of understanding the quality and diversity of the accessions available and identifying conservation priorities and knowledge gaps. We will investigate 12 crops and more than 500 wild relatives of Brassicaceae, of which 260 species are found in only one location. Conservation assessments are urgently needed, since around 50% of the CWRs are Data Deficient or not assessed for their global threatened status, and more than 50% have little or no information available about their traits nor DNA data available. A target list of key CWRs will be obtained to prioritize seed conservation and explore potential traits.
4.2 Towards a tropical seed and germination trait database
Silveira, F.A.O.1, Ordóñez-Parra, C.A.1
1Universidade Federal de Minas Gerais, Brazil
A broad understanding of the ecology and evolution of seed traits depends critically on the availability of trait data from tropical regions, given their unique biodiversity and evolutionary history. Here, we introduce two recent initiatives to compile tropical seed functional trait databases: “MelastomaTRAITS” and “Rock n’ Seeds”. MelastomaTRAITS is a database of 67 functional traits, including 27 seed traits, for 2,569 species distributed in 161 genera from all 21 Tribes of the pantropical Melastomataceae – the 9th largest plant family with more than 5,800 species. Melastomataceae species occur in a large diversity of habitats, growth-forms, functional groups, geographic distribution, reproductive, pollination, and seed dispersal systems. Therefore, the family provides a promising starting point for testing a wide variety of ecological and evolutionary questions. In turn, “Rock n’ Seeds” is a database of seed functional traits and germination experiments from the Brazilian rock outcrop vegetation, recognized as outstanding centers of diversity, endemism and refugia. “Rock n’ Seeds” includes data for 16 seed functional traits containing the major axes of the seed ecological spectrum for 383 species from 149 genera, 50 families and 25 Orders. Raw data for 48 germination experiments for a total of 9,779 records for 280 species is also provided. Notably, 8,250 of these records include daily germination counts. Given the widespread occurrence rocky outcrop vegetation, “Rock n’ Seeds” paves the way for comparative seed functional ecology at the global scale. As a result, both of our databases can contribute to improving our understanding of trait-based seed ecology across levels of organizations, plant-animal interactions, regeneration ecology, and support conservation and restoration programs.
4.3 Looking for the orthogonality of germination indexes: minimalism meets seeds functional ecology
Ordóñez-Parra, C.A.1, Negreiros, D.1, Silveira, F.A.O.1
1Universidade Federal de Minas Gerais, Brazil
carlos.ordonez.parra@gmail.com
The advancement of seed functional ecology relies on the proper characterization of the axis that composes the seed ecological spectrum. Since germination is one of these components, seed ecologists must seek to describe it in the most complete and concise fashion. Different methods have been developed to describe the germination axis, including many seed germination indexes. While it is true that a single index is insufficient to describe the germination behavior completely, there is still no consensus on a minimum set of indices that should be used to provide a complete description of this complex phenomenon. A further complication arises from a pervasive lack of consistency in the terminology to describe germination, the apparent redundancy of indices available, and the development of software that allows the calculation of several indexes with ease. In this research, we assessed the covariation of the 44 germination indices available at the germinationmetrics R package using a database of germination experiments from Brazilian rock outcrop vegetation in order to produce an orthogonal set of indices that characterize the different dimensions of germination behavior. The indices representing potentially different germination behavior dimensions, namely germinability, time, rate, synchrony, and uniformity, were calculated for a set of germination experiments under optimal, suboptimal, and supraoptimal temperatures. When tested for linear correlations, most indexes were found to be highly correlated between them. After systematically selecting from indexes different dimensions with little correlation between each other, we were left with only four indices: germination percentage, t50, coefficient of variation of germination time and germination uncertainty. These four indices arise as a promising group for characterizing the multiple dimensions of seed germination behavior across lineages and biogeographic regions.
4.4 An image analysis tool for automated extraction of diaspore morphological traits
Dayrell, R.L.C.1, Ott T.1, Begemann L.1, Horrocks T.2, Poschlod P.1
1University of Regensburg, Germany; 2University of Western Australia, Australia
The field of comparative morphology of diaspores provides basis for the characterization and identification of taxa, and for understanding how traits are linked to functions, processes, and ecological strategies. However, the description of diaspore traits relies mainly on manual measurements of few parameters, such as length and width, and on human classification of structures, both of which can be error prone and time-consuming. Image analysis applications address these shortcomings by offering an alternative approach for an objective characterization of diaspores. Several applications have been developed to extract quantitative traits from images, but these were tailored mainly for crop species, and do not provide a full automated process that works for diaspores in a variety of sizes, surface structures and colors. Here, we present a protocol and software for the measurement of diaspore morpho-colorimetric traits from images captured with flatbed scanners. The tool uses unsupervised segmentation to automate the separation of diaspores from high-contrast backgrounds, without the need for manual fine-tunning or thresholding. Next, the software determines diaspore size and shape descriptors, including length, width, area, perimeter, aspect ratio, circularity, diaspore surface structure, and solidity. Two types of colorimetric measurements are also extracted: values suitable for human recognition purposes and data independent of any particular animal visual system. To determine the application and accuracy of the software, we assessed morpho-colorimetric traits and performed exploratory analyses with a diverse diaspore collection composed of 1,477 taxa belonging to 97 plant families from Central Europe. The protocol is a highly automated solution that allows for rapid and reproducible measurement of diaspore traits of native species. This approach facilitates the acquisition of data that are readily comparable across different taxa, opening up new avenues to explore functional relevance of morphological traits and to advance on tools for diaspore identification.
4.5 Multiple relations between seed traits and germination performance change with temperature
Saatkamp, A.1, Leclerc, L.1, Guerchet, V.1, Vandelook, F.2
1Aix Marseille University IMBE, France, France; 2Meise Botanic Garden, Belgium
Global warming triggers multiple responses in plant populations and communities. Since seed functional traits are essential co-variates to understand germination timing but also survival during the seed bank stage it is crucial to know whether these relationships remain stable across -or vary over- environmental temperatures. Here we present the relationships between seed traits and germination speed at different exposure temperatures for 37 herbaceous species. At low temperatures (10ºC) germination rates increase with shape, lightness, and ratio of embryo to seed. At intermediate temperatures (20ºC), germination rates decrease with mass, but increase with embryo to seed ratio and terminal velocity. At high temperatures (30ºC), germination rates decrease with seed shape. Our work demonstrates the importance of temperature for studying seed trait – performance relations and we suggest to proceed with studying seed-related processes at different temperatures in the seed environment.
4.6 The seed trait handbook
Poschlod, P.1, Chen, S.C.2, Phartyal, S.S.3, Rosbakh, S.1, Saatkamp, A.4, Silveira, F.A.O.5, Dalling, J.6, Dalziell, E.7, Dickie, J.B.2, Fernández-Pascual, E.8, Guja, L.9, Jiménez-Alfaro, B.8, Merritt, D.7, Ooi, M.10, Mašková, T.1, Vandelook, P.11
1University of Regensburg, Germany; 2Royal Botanic Gardens, Kew, UK; 3Nalanda University, India; 4Aix Marseille University IMBE, France; 5Federal University of Minas Gerais, Brazil; 6University of Illinois at Urbana-Champaign, USA; 7Kings Park Science, Australia; 8University of Oviedo, Spain; 9Center for Australian National Biodiversity Research, CSIRO, Australia; 10University of New South Wales, Australia; 11Meise Botanic Garden, Belgium
The need to understand the mechanisms behind any biodiversity pattern, vegetation process or threat have led to a functional approach using plant traits which reflect a species’ response to its environment or its strategy to cope with. The first overview on functional traits by Weiher et al. has now culminated in the TRY database, actually in the fifth version with nearly 12 million trait entries. A deeper look, however, shows that only few seed ecological traits are presented, such as seed morphology, seed size, mass, dispersal mode, seed longevity and seed germination stimulation with by far most entries on seed size and dispersal mode. The limited functional knowledge on seed traits is also reflected by the second edition of the handbook for standardized measurements of plant functional traits by Pérez-Harguindeguy et al. Only six seed traits were included – dispersal mode and potential, dispersule shape and size, seed mass and seedling morphology. It is hence timely to increase knowledge on seed functional traits by providing a collection of standardized measurement methods, to motivate studies along new axes of functional ecology of seeds and to enable more reliable comparative works across different studies. Taking this gap as an opportunity, we organized a workshop in Perth, Australia in September 2016, which first outcome was to define a research agenda for seed-trait functional ecology. We also worked out a list of 60 reproductive traits concerning flowering and fertilization (gametophyte traits), fruit and seed attraction and defense, seed dispersal and dispersal potential, seed persistence, seed dormancy and germination as well seedling emergence and establishment. Each trait was defined and its functionality, source of variability and the methodology (how to measure) were described. An overview of the seed trait handbook will be presented at this talk.