Work Description

Title: Tower observations of solar-induced chlorophyll fluorescence at the University of Michigan Biological Station

Creators: Zachary Butterfield, zbutterf@umich.edu; Daniel Muccio, drmuccio@umich.edu; Gretchen Keppel-Aleks, gkeppela@umich.edu

Funding: NASA Grants 80NSSC17K0116 and 80NSSC21K1070

Keywords: Solar-induced chlorophyll fluorescence, gross primary production, temperate deciduous forest, remote sensing, flux observations, forest productivity

Data Coverage: June 7, 2018 - November 30, 2019

File Inventory: PhotoSpecM1_2018.nc; PhotoSpecM1_2019.nc

Description: Solar-induced chlorophyll fluorescence (SIF) is an emission of photons during photosynthesis that can be used to make inferences about gross primary productivity (GPP) and carbon uptake of vegetation. With a recent proliferation of available satellite-based observations of SIF, there is much interest in assessing how SIF relates to GPP across multiple temporal and spatial scales. Tower-based observations of SIF at high temporal resolution provide a key link between satellite data and local surface-based observations of ecosystem productivity. We collected tower-based observations of solar-induced chlorophyll fluorescence (SIF) and several vegetation indices using a PhotoSpec spectrometer system (Grossmann et al., 2018) deployed on the AmeriFlux tower (US-UMB) at the University of Michigan Biological Station (UMBS). As the data were collected alongside concurrent eddy flux observations of carbon exchange, they provide a unique opportunity to explore how SIF and other vegetation signals relate to GPP in a temperate deciduous forest and better inform the interpretation of satellite observations.

Methodology: This dataset contains tower-based remote sensing observations of SIF and several vegetation indices at the US-UMB AmeriFlux site at UMBS. Observations were collected using a PhotoSpec spectrometer system consisting of two narrowband spectrometers (QEPro, Ocean Optics Inc.) allowing for the retrieval of SIF in the far-red (745-758 nm) and red (680-686 nm) regions of the electromagnetic spectrum, as well as a broadband spectrometer (Flame, Ocean Optics Inc.) allowing for the calculation of canopy reflectances and vegetation indices.

Site Information: The research site at UMBS (lat, lon: 45.5598, -84.7138) is located in the northern Lower Peninsula of Michigan within deciduous broadleaf forest (DBF) composed of aspen, oak, maple, and beech, with some understory pine. Canopy height is approximately 22 m, and stand age is roughly one century following widespread wildfires prior to 1920. Carbon and water flux measurements have been collected at the site as part of the AmeriFlux network since 1999 (Gough et al., 2022).

Additional Methodology: Data were obtained with the PhotoSpec system using a 2D scanning telescope with a field of view of 0.7° which was mounted on the AmeriFlux tower at a height of 45 m. The telescope directed light from the canopy through a fiber optic cable to the three climate-controlled spectrometers. The system was automated to collect observations using azimuthal scans from nadir to 45° below the horizon in three directions: 60° east of south, due south, and 60° west of south. Each of the three azimuthal scans took approximately 30 minutes, thus data are provided as 90-minute averages. For each individual data point, the exposure time of each spectrometer was optimized to maintain consistent detector signal level, and exposures were integrated into 20 s measurements before moving the telescope to the next location. Reference solar spectra were collected at least every 10 scans using an upward-facing diffuser disc. The far-red and red SIF signals were calculated from spectra collected by the QEPro spectrometers using a physical approach based on the infilling of solar Fraunhofer lines following the methodology of Grossmann et al. (2018). Vegetation indices were calculated from canopy reflectance at various wavelengths as observed by the Flame spectrometer.

Following the spectral analysis, observations were filtered to ensure data quality and to only include observations containing significant vegetation signals. Data were discarded if the normalized difference vegetation index (NDVI) was < 0.2, SIF retrieval errors were > 0.1 mW m^-2 sr^-1 nm^-1, or SIF retrievals fell outside the range of -0.1 to 10 mW m^-2 sr^-1 nm^-1 or -2 to 20% of the total light signal, resulting in the removal of roughly 12% of observations. Uncertainty for the red and far-red SIF signal is provided as the standard deviation of included observations for each 90 minute period.

Data for each year (2018 and 2019) are formatted as a netCDF file where each observed variable is provided as a matrix with size (365, 16). The two axes represent day of year and time of day (divided into 90-minute periods), respectively, and are represented in the files as "Time Variables" listed below. Included data are listed below under "SIF Variables" and "Vegetation Indices". Relative SIF values represent the SIF signal as a fraction of the total light signal. Vegetation Indices are calculated based on canopy reflectances (R) at specific wavelengths (in nm), or in the red or near-infrared (NIR) regions of the electromagnetic spectrum.

***Time Variables***
Day_of_Year --> Day of the Year (Day of Year)
Time_of_Day --> Beginning time of a 90-minute interval (Hour of Day)

***SIF Variables***
far_SIF_irr --> far-red SIF irradiance (mW/m^2/nm/sr)
far_SIF_irr_stdev --> standard deviation of far-red SIF irradiance (mW/m^2/nm/sr)
far_SIF_rel --> relative far-red SIF (Unitless)
red_SIF_irr --> red SIF irradiance (mW/m^2/nm/sr)
red_SIF_irr_stdev --> standard deviation of red SIF irradiance (mW/m^2/nm/sr)
red_SIF_rel --> relative red SIF (Unitless)

***Vegetation Indices***
NDVI --> Normalized Difference Vegetation Index (NDVI = (R_NIR - R_Red)/(R_NIR + R_Red); Unitless)
PRI --> Photochemical Reflectance Index (PRI = (R_531 – R_570)/(R_531 – R_570); Unitless)
ChRS --> Chlorophyll_RS (ChRS = (R_850 – R_710)/(R_850 – R_680); Unitless)

To cite data:
Butterfield, Z., Muccio, D., Keppel-Aleks, G. Tower observations of solar-induced chlorophyll fluorescence at the University of Michigan Biological Station [Data set], University of Michigan - Deep Blue Data. https://doi.org/10.7302/sx8c-y281

Licence: Attribution 4.0 International (CC BY 4.0)

Related Publication:
Butterfield, Z., Magney, T., Grossmann, K., Bohrer, G., Vogel, C., Barr, S., & Keppel-Aleks, G. (2023). Accounting for Changes in Radiation Improves the Ability of SIF to Track Water Stress-Induced Losses in Summer GPP in a Temperate Deciduous Forest. Journal of Geophysical Research: Biogeosciences, 128, e2022JG007352. https://doi.org/10.1029/2022JG007352

Other References:
Gough, C. M., Bohrer, G., & Curtis, P. (2022). AmeriFlux BASE US-UMB Univ. of Mich. Biological Station, Ver. 18-5. AmeriFlux AMP, (Dataset). https://doi.org/10.17190/AMF/1246107
Grossmann, K., Frankenberg, C., Magney, T. S., Hurlock, S. C., Seibt, U., & Stutz, J. (2018). Remote Sensing of Environment PhotoSpec : A new instrument to measure spatially distributed red and far- red Solar-Induced Chlorophyll Fluorescence. Remote Sensing of Environment, 216(June), 311–327. https://doi.org/10.1016/j.rse.2018.07.002