ABSTRACT:

This data contains eddy covariance, soil, and meteorological data collected from the Goose Creek Tower site in central IL from 2016-2024. A 25m tower was installed April 2016 with 15 min measurements, and it was supplemented by a 10m height eddy covariance instrument in August 2021. The "25m" files contain the main tower data (eddy covariance at 25m, soil moisture and temperature, solar radiation and other meteorological variables), and the "10m" is the newer instrument at 30 min resolution. The "raw" files obtained directly from the sensors are provided, with a nomenclature document describing several hundred variable names. We also processed this data using ReddyProc R package (https://cran.r-project.org/web/packages/REddyProc/index.html) for carbon flux partitioning estimates at 30 min resolution, and applied the flux-data-qaqc python package (https://flux-data-qaqc.readthedocs.io/en/latest/) for daily energy balance corrections. We also provide a solar radiation corrected file, due to some sensor errors in incoming and outgoing radiation, and note that some NDVI and PRI (vegetation indices) values contain errors that are not corrected from the raw data. Please see the Readme file for short descriptions of each file type.

ABSTRACT:

This resource contains materials for a 1-day workshop on critical zone time-series data analysis, co-organized by NSF CZ Clusters CINet and Big Data. and CUAHSI. Topics of the workshop include choosing the right tool for your research questions, autocorrelation, correlation, trend analysis, clustering, and causal inference.
Workshop date: Aug 5
Workshop location: University of Illinois at Urbana-Champaign
Session topics include correlation, autocorrelation, trend analysis, clustering, and causal inference based in information theory methods.

ABSTRACT:

15-minute time-series data from flux tower at Goose Creek site, originally part of the NSF-IMLCZO project, currently NSF-CINet. Variables include: wind speed, carbon flux (NEE), pressure, precipitation, air temperature, relative humidity, 5cm depth soil moisture, 5cm depth soil temperature, solar radiation, ground heat flux. These are raw data, date range of 2016-2020, used in Farahani and Goodwell, 2024 (JGR Biogeosciences, DOI: 10.1029/2023JG007815): Causal Drivers of Land-atmosphere fluxes from machine learning models and data. The associated Github repository (https://github.com/allisongoodwell/Farahani_CarbonML2023) contains code for the outlier removal and pre-processing applied to this data, which were used for machine learning models for carbon flux prediction.