Abiotic and biotic drivers of species diversity across an elevation gradient: an optical trait-based approach for testing ecological theories

Abstract

Predicting the impact of climate change on ecosystems requires understanding how local processes influence diversity across environmental gradients. Explaining diversity patterns has been a central issue in community and ecosystem ecology, and more recently in macroecology. Both biotic and abiotic mechanisms have been shown to explain diversity patterns across gradients, and trait-based ecology has established the use of functional traits to understand diversity patterns at different spatial scales. Still, spatially and temporally explicit information on plant traits is currently lacking, limiting our understanding of how plant communities interact with biotic and abiotic drivers. Remote sensing offers innovative tools for assessing vegetation traits at multiple scales, and next-generation approaches have the potential to revolutionize ecological observation, particularly by using imaging spectrometers. Hyperspectral sensors measure the reflected electromagnetic radiation in hundreds of narrow spectral bands, allowing the reconstruction of spectral signatures and the detection of structural, physiological and even phylogenetic plant traits. Recent studies have shown the linkage between plant species-traits and hyperspectral measurements (spectranomic traits), where reflectance and transmission signals are converted into leaf-trait values. The capabilities of these new remote sensing instruments have led to the concept of "optically distinguishable functional types" ('plant optical types') as a unique way to address the scale dependence of ecosystem processes and vegetation dynamics. Hyperspectral sensors are flown on manned airborne platforms, at a high cost. Unmanned Aerial Vehicles (UAVs or "drones") are a cutting-edge technology undergoing rapid and significant expansion. Its use in spatial ecology is virtually unlimited, but applications for trait-based ecology are few, especially in tropical ecosystems. We propose a complete assessment of the applicability of hyperspectral imaging and UAVs to unravel the mechanisms underlying the patterns of plant community assembly across a tropical elevation gradient at Serra do Cipó, Brazil, based on species- and optical-traits. We will answer the following questions: (i) Are assembly mechanisms similar across elevations, and/or between woody and herbaceous assemblages? (ii) Are trait variation patterns related to competitive interactions at low altitudes and abiotic filtering at high altitudes? (iii) How does leaf spectra change across community assemblages at different elevations? (iv) How does intra-specific optical variability compare to inter-specific variability, at multiple scales? (v) Can plant optical types be linked to plant functional types in tropical ecosystems? The project will be carried out in the Serra do Cipó region, a mountainous environment characterized by an altitudinal gradient ranging from 800 to 1700m. The landscape includes different Cerrado physiognomies, rupestrian fields, and deciduous forests, distributed in patches across the elevation range. We expect to build a species-specific hyperspectral library, based on high quality field spectrometry, which will be associated with a database of functional traits. Scientific and technological results arising from this project will give us an unpaired ability to address current and future environmental issues, taking advantage from the multidisciplinary exchange promoted by the FAPESP/Microsoft Joint Research Center collaboration. We will test relevant modern theories, often cited but rarely tested, about the origins of diversity gradients in the tropics, in terms of variation in functional and optical diversity and community assembly patterns. We expect to reach a new level of understanding on how community traits are distributed across gradients, in the light of niche theory, revealing mechanisms and patterns that can be used to predict the effects of a changing environment, from leaves to ecosystems. (AU)