BOSSE: The Biodiversity Observing System Simulation Experiment for Remote Sensing

Climate change and human activities jeopardize ecosystems’ biodiversity, functions, and services. However, biodiversity monitoring is resource-intensive and unable to provide the coverage and resolution necessary to understand biodiversity responses to these drivers. Remote sensing can contribute to monitoring plant biodiversity status and change by exploiting the variability of the spectral imagery acquired from space platforms. Still, several gaps must yet be solved regarding what approaches, metrics, sensors, and techniques can provide reliable biodiversity maps. One of the main challenges is the generation of field datasets with the spatial coverages and temporal resolutions necessary to determine the best methods.

To overcome this problem from a theoretical point of view, we have developed BOSSE, a biodiversity observing system simulation experiment. BOSSE simulates dynamic scenes in time where vegetation properties change as a function of meteorological conditions and adopt different spatial patterns. High-spatial resolution scenes can be used to quantify plant functional diversity from plant traits. Moreover, BOSSE can simulate hyperspectral reflectance factors, sun-induced chlorophyll fluorescence, and land surface temperature that are coherent with plant traits of meteorology. Spectral imagery can be generated at different spatial and temporal resolutions, allowing us to test different approaches, metrics, and methods to estimate plant functional diversity.

We have used BOSSE to determine the best approaches to characterize plant functional diversity of large areas, which is a fundamental step prior to assessing the links between remote sensing and ecosystem functions. Additional analyses have compared the capability of different spectral signals to capture plant functional diversity, the role of spatial resolution, and the role of seasonality in those estimates. We expect BOSSE to contribute to solving hypotheses and test methods and help determine what field datasets would be necessary to validate remote sensing biodiversity products.