Seed trait variability in species of tropical savannas: testing a new tool to improve seed trait assessment

Abstract

Seed and germination traits can provide essential information on recruitment, seedling establishment, plant function and adaptation, and thereby ecological strategies. However, in fire-prone ecosystems, fire regimes (e.g. fire frequency) affect plant/seed traits and trait variability thus interfering in those essential processes. It is therefore our main aim to specifically capture seed traits associated with the seed and fruit coat biochemical and biophysical properties, as well as molecular traits associated with as fire-associated environmental cues. For that we collected seed from different species at an open Cerrado area in Central Brazil and will evaluate the intra- and interspecific variability of seed traits according to a fire frequency gradient in a tropical savanna. Seeds will be shipped to Royal Holloway University of London (RHUL) in England where we will utilize Multispectral Imaging (MSI) as a novel tool for fire ecological research. At the RHUL lab seeds will be analyzed using the VideometerLab 4TM (VL4), that is a non-destructive MSI device designed for fast and accurate determination of color, texture, and chemical composition of surfaces. Using strobed LED technology VL4 will perform measurements at up to 20 different wavelengths into a single high-resolution spectral image for each species. This image will be used to extract information on variables representing seed traits (e.g., chemical composition), and will be combined with molecular and biochemical analyses. The data acquired with MSI will be added to seed traits previously collected (seed mass and seed shape) and the results will be analyzed. We expect that MSI analysis technique will be a valuable tool to detect seed and fruit coat traits/properties and how they are affected in response to fire and changes in fire regime. Thus, narrowing the barriers in future studies on seed ecology. (AU)