Scaling photosynthetic function and CO2 dynamics from leaf to canopy level for maize – dataset combining diurnal and seasonal measurements of vegetation fluorescence, reflectance and vegetation indices with canopy gross ecosystem productivity

Recent advances in leaf fluorescence measurements and canopy proximal remote sensing currently enable the non-destructive collection of rich diurnal and seasonal time series, which are required for monitoring vegetation function at the temporal and spatial scales relevant to the natural dynamics of photosynthesis. Remote sensing assessments of vegetation function have traditionally used actively excited foliar chlorophyll fluorescence measurements, canopy optical reflectance data and vegetation indices (VIs), and only recently passive solar induced chlorophyll fluorescence (SIF) measurements. In general, reflectance data are more sensitive to the seasonal variations in canopy chlorophyll content and foliar biomass, while fluorescence observations more closely relate to the dynamic changes in plant photosynthetic function. With this dataset we link leaf level actively excited chlorophyll fluorescence, canopy proximal reflectance and SIF, with eddy covariance measurements of gross ecosystem productivity (GEP). The dataset was collected during the 2017 growing season on maize, using three automated systems (i.e., Monitoring Pulse-Amplitude-Modulation fluorimeter, Moni-PAM; Fluorescence Box, FloX; and from eddy covariance tower). The data were quality checked, filtered and collated to a common 30 minutes timestep. We derived vegetation indices related to canopy functioning (e.g., Photochemical Reflectance Index, PRI; Normalized Difference Vegetation Index, NDVI; Chlorophyll Red-edge, Clre) to investigate how SIF and VIs can be coupled for monitoring vegetation photosynthesis. The raw datasets and the filtered and collated data are provided to enable new processing and analyses.

The processed and curated dataset is provided using the local Eastern Daylight Time (EDT). Description of data collection All data were collected under natural growing conditions in the field (i.e., in situ ) at the OPE3 site on maize during the 2017 season.
Canopy eddy covariance observations were collected from a 10 m instrumented flux tower and reported at 30 minutes timestep.
Leaf level Moni-PAM data were collected, using 5 emitter detector probes attached to the 4th plant leaf, at 10 minutes intervals and resampled/interpolated to 30 minutes intervals. As new leaves appeared, the probes were relocated to maintain their position on the 4th leaf from the top.
Optical FloX measurements were collected from 1 m above the canopy, at the frequency defined by the system optimization program and were resampled/interpolated to the 30 minutes tower timestamps. The hight of the optics above the ground was adjusted to maintain the 1 m distance to the canopy during the growing season. Data  The use of trade, firm, or corporation names in this article is for the information and convenience of the reader. Such use does not constitute official endorsement or approval by the US Department of Agriculture nor the Agricultural Research Service of any product or service to the exclusion of others that may be suitable.

Value of the Data
• The data are useful and important for advancing the understanding of the links between the diurnal and seasonal dynamics of vegetation function, leaf chlorophyll fluorescence and canopy reflectance, vegetation indices (VIs), solar-induced fluorescence (SIF) and gross ecosystem productivity (GEP). • The data can benefit ecologists, plant physiologists, foresters, and agriculturalists, by providing inputs for models and linking leaf and canopy processes to demonstrate the connections between reflectance and fluorescence properties for monitoring vegetation photosynthetic function and detecting stress. • The data can be used by remote sensing professionals for calibration and validation of canopy VIs, reflectance and SIF signals currently measured by satellite instruments (e.g., DESIS/ISS, PRISMA/ASI, GOME-2, GOSAT, OCO-2, TROPOMI). • The dataset is useful for simulations and generation of product prototypes characterizing photosynthetic function, as anticipated with space-based spectrometers, such as the existing DESIS (DLR, Germany) and PRISMA (Italy), and the forthcoming Surface Biology, Geology (SBG, NASA), EnMAP (DLR, Germany) and the European Space Agency (ESA) Fluorescence EXplorer (FLEX) mission, which will obtain globally canopy reflectance and SIF and will assess seasonal photosynthetic activity and canopy function. • The fluorescence data might be used for improving the ability to scale and relate the commonly measured leaf-level plant physiology parameters (i.e., active chlorophyll fluorescence metrics such as electron transport rate, ETR and yield to photosystem II, PSII) to the canopy level solar induced fluorescence (SIF) measurements, which are targeted by the future NASA GeoCarb geostationary mission and ESA FLEX mission.

Data Description
There is a critical need for temporally dense time series of remote sensing data, collected at the 'right' time of day, frequency and season for monitoring the dymanics in vegetation function. To advance the ability to monitor the parameters governing vegetation function, time series capturing diurnal responses and seasonal changes in plant photosynthesis at leaf and canopy scales are needed. Leaf in situ and canopy proximal remote sensing can now provide such datasets for monitoring vegetation function at the temporal and spatial scales relevant to the dynamics of plant photosynthesis.
The dataset provided with this manuscript is comprised of tables with contemporaneous diurnal measurements collected continuously by three instruments ( Table 1 ), measuring multiple maize leaves and the canopy during the 2017 growing season. Both raw instrument readings and the filtered and collated dataset are provided. The data are organized in comma delimited Table 1 Combined dataset, including diurnal and seasonal observations of leaf and canopy fluorescence, canopy reflectance and photosynthetic CO 2 dynamics for maize ( Zea mays L. Moni-PAM Ft, Fo, Fm, Fm', PAR, YII, ETR, NPQ, etc. * All parameters are defined in the 'read_me' spreadsheets in the excel files containing the data. The parameters listed in the categories above are provided in the combined dataset and listed in the appendix Table A1 . text files and excel files and are coupled by date and time of collection. Each excel file contains a 'read_me' sheet with information describing the specific parameters and processing approach. The raw data are provided for established users of the instrumentation and includes the separate records of each instrument, as follows: • Moni-PAM data files: • MoniPAM_OPE3_2017_WinControl3.zip containing raw instrument readings. The files can be processed using the freely available WinControl3 software ( https://www.walz.com/ products/light/ulm-500/wincontrol-3.html , Heinz Walz GmbH). • MoniPAM2017_all.xlsx containins all readings, organised by time and date into one file.
The file contains read_me sheet with definitions and units for all parameters. • FloX optical data files : • FloX_DataOPE3_2017_raw.zip contains all FloX raw instrument readings. The data are organised in folders by date. • FLOX_R_OPE3_2017all.xlsx contains canopy diurnal reflectance measurements in the 400-850 nm region collected across the 2017 season.
The collated dataset containing corresponding measurements from the three instruments ( Table 1 ) is assembled in the excel file 2017OPE3_Moni_FLOX_FLUX_ALL.xlsx. Table 1 lists some of the parameters, which include: leaf active chlorophyll fluorescence parameters, canopy solar induced fluorescence (SIF), canopy reflectance and canopy eddy covariance measurements of photosynthetic function. The excel file 2017OPE3_Moni_FLOX_FLUX_ALL.xlsx contains a 'read_me' sheet with definitions and units for each parameter, which are provided also in Appendix Table A1 .

Study site
The data were collected in 2017 at the Optimizing Production inputs for Economic and Environmental Enhancement (OPE3) site at the Beltsville Agricultural Research Center (BARC) on rain-fed maize ( Zea mays L.). The OPE3 site is one of the Long Term Agro-ecosystem Research (LTAR) network of sites, operated by the US Department of Agriculture (USDA) Agricultural Research Service (ARS). It is a 22 ha production field located in Beltsville, Maryland, USA (lat/lon: 39.030686/-76.84546, 42 m asl). At OPE3 a 10 m tall flux tower is set up in a rainfed maize field, which is planted annually and maintained under optimal nitrogen treatment. The local climate is warm and temperate, with hot, humid summers, long fall and typically mild winters with occasional freezes, which provide a strong variation in seasonal leaf area index and canopy chlorophyll and biomass patterns.

Leaf-level measurements
Leaf-level fluorescence and photosynthetic efficiency measurements were collected continuously using the pulse amplitude modulated (PAM) approach, with an automated MoniPAM Data Acquisition system (MONI-DA, Heinz Walz GmbH, Effeltrich, Germany) outfitted with five emitter-detector probes. Leaf fluorescence, collected using the PAM approach, can provide the means for assessment of photosystem II (PSII) efficiency, which are highly dependent on the ambient light levels and can vary substantially with a relatively small change in photosynthetically active radiation (PAR) [6] .
Five MoniPAM emitter-detector probes were mounted on five representative plants on the fully developed 4th leaf from the top of the plant. Measurement collection started when the corn canopy was well-established (e.g., 80-85% canopy closure) and continued until the end of the growing season. All readings are provied in the compressed archive MoniPAM_OPE3_2017_WinControl3.zip, which can be re-processed using the WinCintrol3 free software, available for download at https://www.walz.com/products/chl _ p700/monitoring-pam/ downloads.html . Three probes were positioned to measure fully sun-lit leaves, while two probes collected measurements on shaded leaves at varying illumination levels within the canopy.
The chlorophyll fluorescence (F) parameters measured directly include light adapted maximim and transient/steady-state flouorescence (Fm' and F), and also the prohosynthetically active (PAR) excitation levels for each probe. From the direct measurements were derived F yield, photochemical efficiency of photosystem II (YII, or yield to PSII under steady state light), and electron transport rate (e.g., ETR, photochemical transport of electrons through PSII), which are used for characterising the photosynthetic activity of the plants.

Canopy-level measurements
Canopy diurnal reflectance and solar induced fluorescence measurements (SIF, mW/m 2 /nm/sr), both SIF B (in the atmospheric O 2 B) and SIF A (in the atmospheric O 2 A) bands, were collected using an automated, field dual-spectrometer system, the FLoX (Dual FLuorescence boX; JB Hyperspectral Devices UG, Dusseldorf, Germany) [1] . The FLoX downwelling optics were mounted at the top of a portable platform at approximately 3 m height. The upwelling optics were positioned at nadir and maintained at 1.5 m above the canopy throughout the growing season (i.e., by lifting periodically the measurement arm as new leaves developed and the canopy grew taller), viewing a 25 °field of view.
The OPE3 site is instrumented with a 10 m eddy-covariance flux tower, which measures canopy level CO 2 assimilation reported at 30 minutes intervals continuously throughout the growing season. The measured net CO 2 flux (i.e., Net Ecosystem Exchange, NEE) is partitioned into gross primary productivity (GPP, the carbon used by photosynthesis) and ecosystem respiration [4 , 5] .

Data Stratification and Processing
The maize crop conditions and phenology development in 2017 are recorded by the phenocam network (site arsope3ltar, data available at https://phenocam.sr.unh.edu/webcam/sites/ arsope3ltar/ ). By DOY 190 the maize canopy was established, canopy closure was approximately 85% and no additional agricultural treatments were planned [7] . Measurements were collected continuously during the 2017 growing season, periodically re-locating the Moni-PAM probes, as new leaves emerged and lifting higher the FloX optics as the canopy grew taller. To enable analysis of the seasonal variation in the observations depending upon the time of collection and crop growth stage, the data were stratified into four growth stages , as follows: young (Yn, DOY 192-209); mature (M, DOY 220-235); senescent (DOY 236-272) which was subdivided into early senescence (S1, DOY 245-247) and advanced senescence (S2, DOY 255-257). The diurnal measurements were processed to form categorical variables representative of the three distinct periods during the day: morning (AM, 8:45-10:15), noon (11:45-1:15) and afternoon (PM, 15:15-16:45) local time (EDT or UTC-4).
Leaf-level light-adapted PAM fluorescence metrics representative of the canopy were derived by calculating the mean values from the three Moni-PAM emitter-detector probes maintained on fully sunlit leaves, using the simultaneously acquired measurements at 10 minutes interval. Large outlier values from a single probe were removed. Average values for light adapted fluorescence (F, relative units), mean yield of Photosystem II (YII), mean relative electron transport rate (ETR) and PAR were calculated. The night-time measurements were used to derive Fo and Fm, and to calculate non-photochemical quenching (NPQ), which are described in [8] and provided in file MoniPAM2017_all.xlsx.
The FLoX system collects upwelling and downwelling measurements, which were processed to reflectance and solar induced fluorescence (SIF) using two open-source R packages [2 , 3] (i.e., FieldSpectroscopyDP and FieldSpectroscopyCC), which are available for download at the following links: https://github.com/tommasojulitta/FieldSpectroscopyDP and https://github.com/ tommasojulitta/FieldSpectroscopyCC . Quality screening of the measurements was completed using the assigned by the R routines quality flags, which reported information related to the illumination stability during the measurement cycle and the internal noise of the instrument during data acquisition. The dataset was filtered for saturated data points and anomalous readings. Additional screening was done to remove sensor artifacts due to low light levels (i.e., in the early morning and late afternoon, where the solar zenith angle, SZA is > 75 o ). The outputs of the processing include incoming radiance at the surface; top of canopy reflected radiance; apparent reflectance; and SIF estimates, at spectral wavelengths associated with both atmospheric oxygen absorption features centered at 683 (SIF B ) and 760 nm (SIF A ). SIF A and SIF B were retrieved by applying the Fraunhofer Line Discriminator method (version 3, iFLD) and the Spectral Fitting Method (SFM) [9 , 10] . Total SIF A + B was calculated as the sum of SIF A and SIF B [7] . Canopy reflectance from the FLoX was used to calculate indices indicative of vegetation green biomass, chlorophyll content and photosynthetic function (e.g., NDVI, Chl re and PRI), which are described by providing the formulas for their calculation in the 'read_me' sheets of the excel files containing the data.
Incident canopy PAR was measured by the OPE3 flux tower and the FloX instrument. Incident PAR was measured at leaf level by the Moni-PAM emitter-detector probes. Outliers (e.g., values greater than three standard deviations from the mean) from the daily linear trend were removed from all data sets. The PAR absorbed by the canopy (APAR, μmol/m 2 /s) in the FLoX footprint is required for computation of the FLoX fluorescence yields. It was calculated as: APAR = PAR * f APAR, where f APAR, the fraction of incident PAR absorbed by the canopy, is estimated using an equation available for OPE3 from [11] . Canopy SIF yield (YSIF) was calculated as: YSIF = SIF/APAR (for both SIF A and SIF B bands, with the assumption that APAR was equally available to both PSII and PSI. The set of contemporaneous complementary leaf and canopy measurements at OPE3 were collected at different time intervals: 30 minutes (flux tower), 10 minutes (Moni-PAM), and at the time required for system optimization (i.e., approximately 1 minute, FLoX). To compare all three data sets, the most frequently acquired FLoX data set was screened for outliers and smoothed to reduce the random noise using a moving Savitzky-Golay filter function (MATLAB 2019) and linearly interpolated to 10 minutes intervals and resampled at congruent time steps with the Moni-PAM data. The combined Moni-PAM and FLoX data were then linearly interpolated to extract values at the same 30 minutes intervals used by the flux data.

Ethics Statement
The work did not involve the use of human subjects, animal experiments and information collected from social media platforms.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or relationships which have or could be perceived to have influenced the work reported in this article.