Quality assessment of NEON's eddy-covariance flux data products

The National Ecological Observatory Network (NEON) is a continental-scale research platform with a projected lifecycle of 30 years. NEON's purpose is to provide high quality data products that will facilitate discovering and understanding the impacts of climate change, land-use change, and invasive species on ecology. To accomplish this, NEON's data products are designed to enable extrapolating relationships between ecosystem drivers and the ecological response. Measurements by automated instruments are made by the Fundamental Instrument Unit (FIU), a project team within NEON. These data pass a routine quality assurance and quality control procedure (QA/QC, see companion poster), and provide a direct measure of the ecosystem drivers and responses.

One of the FIU key data products is the exchange of heat, water vapor and CO₂ between the ecosystems and the atmosphere. The eddy-covariance (EC) technique will be used to continuously monitor these fluxes at 60 NEON research sites. However, the theoretical assumptions underlying the EC method cannot always be met, in particular for research sites in complex terrain. Here, we present a QA/QC approach which focuses on the reliable and consistent quality rating of EC flux measurements over the range of climates and ecosystems across an entire continent. Our approach advances generally accepted tests and data flows (e.g., AmeriFlux, CarboEurope) to new state-of-the-art functionality. This has become necessary by transitioning EC measurements from principal-investigator-based into observatory-based operations, which also requires placing QA/QC approaches into a production framework.

NEON's scaled flux QA/QC procedure consists of four consecutive work packages (WP), which are combined to a final quality metric. Firstly WP1 "Temporally explicit flux QA/QC" will continuously assess the applicability of the EC method under the flow conditions at the measurement locations. In WP2 "Spatially explicit flux QA/QC" the results from WP1 will be referenced to spatially explicit surface features, enabling the location and interpretation of flow anomalies. Procedures developed in WP3 "Flux un-mixing" will determine whether the source / sink behavior of individual land covers in a heterogeneous ecosystem differs significantly. Where necessary, WP3 will enable to populate the fluxes of individual land covers. The fourth work package WP4 "Complex terrain" aims at understanding complex flow patterns, e.g. in hilly or mountainous terrain, and will combine the findings from WP3 and Large Eddy Simulation.