Accounting for spatial variability when combining fluxes and proximal sensing techniques.

The continuous observation of ecosystems fluxes between the biosphere and atmosphere provides a much-needed foundation for effective real-world management of our geo-ecological life support systems. Over a thousand flux measurement sites globally use sophisticated eddy-covariance (EC) instruments and are organized in international monitoring networks, e.g. FLUXNET, NEON, ICOS. The integration of automated spectrometers measuring irradiance, reflectance and sun-induced chlorophyll fluorescence (SIF) adds valuable optical proxies for photosynthesis and carbon fixation at top-of-canopy level:  field spectroscopy provides  a powerful tool for understanding measurements of plant carbon uptake, thus providing a link between fluxes and satellite remote sensing and enabling local information to be scaled up to the globe. In the last years many efforts have been made for merging these two types of measurements. Nevertheless, when combining these two sources of information one major limitation is to match the different areas seen by the EC flux measurements and field spectroscopy. Typically the proximal sensing instruments have a fixed field of view (FOV) which limits the measured area to a portion of the underlying canopy. The FOV depends on the set up defined and can vary between a few degrees and 180 degrees, and, alongside the mounting height, determines the size of the monitored area. On the contrary, classical EC flux measurements provide data that typically refer to a larger and variable FOV, also referred to as footprint. The footprint of EC data varies spatially according to meteorological conditions, so – unless a site is perfectly homogenous – the comparison between proximal sensing and EC is always flawed by the footprint mismatch. Additionally, the results of a classical EC time series are difficult to attribute to individual sources and sinks within an upwind source area due to spatial aggregation. Recently the FluxMapper EC approach has been shown to transcribe high-frequency temporal information onto half-hourly Flux Maps around the tower which resolve fluxes spatially through signal disaggregation. Analogous to the proximal sensing techniques, the spatial resolution of the Flux Map depends on the sensor distance from the canopy. In this contribution, for the first time, we provide preliminary results of combining field spectroscopy techniques and Flux Mapper EC, and evaluate spatial heterogeneity effects on the interpretability relative to classical EC. Broader impacts include cost-effective measurement, reporting, and verification of nature-based and technological climate solutions in support of the Glasgow Climate Pact and the Dubai Climate Summit net-zero commitments.