SISTER: PRISMA L2B Vegetative Biochemical Traits 30 m V001

Generated by the Space-based Imaging Spectroscopy and Thermal pathfindER (SISTER) activity in support of the NASA Earth System Observatory's Surface Biology and Geology (SBG) mission, this dataset utilizes existing airborne and spaceborne sources to generate prototype data products spanning terrestrial ecosystems, inland and coastal aquatic ecosystems, hydrology, and geology. The objective of SISTER is to mature many of the workflows, algorithms, and data products envisioned for SBG, lay the groundwork to develop a robust cal/val network, and build a vigorous and expansive user community ahead of launch. This dataset contains experimental Level 2B vegetative biochemical traits (nitrogen concentration, leaf mass area, and chlorophyll content) at 30-m spatial resolution derived from data measured by the PRecursore IperSpettrale della Missione Applicativa (PRISMA) satellite (https://www.asi.it/en/earth-science/prisma/). For the purposes of SISTER, only a handful of scenes have been selected from this mission, with a temporal range between 2020-02-16 and 2022-05-08 and a spatial coverage that is global in scale. PRISMA measures reflected radiance at better than 12-nanometer (nm) intervals in the visible to shortwave infrared spectral range between 400 and 2500 nm. The raw PRISMA data is first processed to L1B orthocorrected calibrated radiance by the Italian Space Agency (ASI) and then ingested into the SISTER platform for further downstream analysis. This collection was derived from corrected surface reflectance and fractional cover datasets using partial least squares regression (PLSR) models. Permuted PLSR models for estimating chlorophyll content, nitrogen concentration, and leaf mass per area were developed using coincident NEON AOP canopy spectra, downsampled to 10 nm, and field data collected by Wang et al. (2020). Biochemical trait estimates were only calculated for pixels with greater than 50% vegetation cover. In addition to mean biochemical trait estimates, per-pixel uncertainties were calculated along with a quality assurance mask which flags pixels with trait estimates outside of the range of data used to build the model. The output products are provided in GeoTIFF format in the following units: chlorophyll content (micrograms/cm2), nitrogen concentration (milligrams/gram), and leaf mass per area (grams/m2).

There are 486 files with this dataset which includes three files in cloud optimized GeoTIFF format (.tif) for each of 27 flightlines. The GeoTIFF files provide vegetative biochemical traits (nitrogen concentration, leaf mass area, and chlorophyll content). Additional files for each flightline include a quicklook image as well as processing product generation information for experimental reproducibility and a Product Generation Executable (PGE) log file.

This dataset contains experimental Level 2B vegetative biochemical traits (nitrogen concentration, leaf mass area, and chlorophyll content) at 30-m spatial resolution derived from data measured by the  PRecursore IperSpettrale della Missione Applicativa (PRISMA) satellite (https://www.asi.it/en/earth-science/prisma/). For the purposes of SISTER, only a handful of scenes have been selected from this mission, with a temporal range between 2020-02-16 and 2022-05-08 and a spatial coverage that is global in scale.

Project:  SISTER 

The Space-based Imaging Spectroscopy and Thermal pathfindER (SISTER) is a NASA project aimed at prototyping workflows and generating SBG-like data products in efforts to sustain and build the community to increase prospects for major scientific discovery post launch.

These data are associated with experimental products run by the SISTER Science Team as pre-launch modeling tools and data for algorithm development are investigated.The SISTER files for the Composite Release ID (CRID) 001 experimental run contain 19 separate collections (Table 1) that include four instruments and five data products (with the exception of the DESIS instrument). The output range for all sensors except DESIS is 400-2500 nm, while the DESIS output range is 400-990 nm. 

Table 1. Summary of SISTER Sensors, Products, and Coding for all CRID 001 outputs available as separate datasets.

Sensor	Product	Sensor_Level_Product
AVIRIS Classic	Resampled Surface Reflectance and Uncertainty	AVCL_L2A_RSRFL
	Corrected Surface Reflectance	AVCL_L2A_CORFL
	Fractional Cover	AVCL_L2B_FRCOV
	Vegetative Biochemical Traits	AVCL_L2B_VEGBIOCHEM
	Snow Grain Size	AVCL_L2B_SNOWGRAIN
AVIRIS Next Gen	Resampled Surface Reflectance and Uncertainty	AVNG_L2A_RSRFL
	Corrected Surface Reflectance	AVNG_L2A_CORFL
	Fractional Cover	AVNG_L2B_FRCOV
	Vegetative Biochemical Traits	AVNG_L2B_VEGBIOCHEM
	Snow Grain Size	AVNG_L2B_SNOWGRAIN
DESIS	Resampled Surface Reflectance and Uncertainty	DESIS_L2A_RSRFL
	Corrected Surface Reflectance	DESIS_L2A_CORFL
	Fractional Cover	DESIS_L2B_FRCOV
	Vegetative Biochemical Traits	DESIS_L2B_VEGBIOCHEM
PRISMA	Resampled Surface Reflectance and Uncertainty	PRISMA_L2A_RSRFL
	Corrected Surface Reflectance	PRISMA_L2A_CORFL
	Fractional Cover	PRISMA_L2B_FRCOV
	Vegetative Biochemical Traits	PRISMA_L2B_VEGBIOCHEM
	Snow Grain Size	PRISMA_L2B_SNOWGRAIN

Related Datasets:

See SISTER datasets with the Composite Release ID (CRID) version 001 (filter e.g.: V001)

Townsend, P., M.M. Gierach, P.G. Brodrick, A.M. Chlus, H. Hua, O. Kwoun, M.J. Lucas, N. Malarout, D.F. Moroni, S. Neely, W. Olson-Duvall, J.K. Pon, S. Shah, and D. Yu. 2023. SISTER: Composite Release ID (CRID) Product Generation Files, 2023. ORNL DAAC, Oak Ridge, Tennessee, USA. https://doi.org/10.3334/ORNLDAAC/2231

• SISTER_CRID_001.json - SISTER Composite Release ID (CRID) file contains details of repositories and versions used for this V001 run.

Spatial Coverage: Selected scenes/flight lines across the globe

Spatial Resolution: 30 m

Temporal Resolution: One-time estimates

Temporal Coverage: 2020-02-16 and 2022-05-08

Site Boundaries: Latitude and longitude are given in decimal degrees.

Site	Westernmost Longitude	Easternmost Longitude	Northernmost Latitude	Southernmost Latitude
PRISMA Lines	-157.9832	19.6244	66.1514	-38.8589

Data File Information

There are 487 files with this dataset which includes three files in cloud optimized GeoTIFF format (.tif) for each of 27 flightlines. The GeoTIFF files provide data for chlorophyll content, nitrogen concentration, leaf mass per area. Additional files for each flightline include a quicklook image as well as processing product generation information for experimental reproducibility and a Product Generation Executable (PGE) log file. In all, there are 18 files per flight line.

Data File Naming Conventions

File naming convention: 
SISTER_<instrument>_<processing_level>_<product>_<flight_id>_<ver>_<var>.<ext>, where 

where:

• <instrument> = the spectroscopy instrument that provided input radiance (Table 3)
• <processing_level> = the NASA Earthdata Data Processing Level
• <product> = the SISTER Project data product  (Table 2)
• <flight_id> = flight line identifier, <YYMMDD>T<HHMMSS>, encoding the date and time by year (YY), month (MM), day (DD), and the UTC hour, minute, and second for the start of flight. 
• <ver> = the SISTER processing version, also known as the CRID
• <var> = individual variable or measured parameter (CHL, LMA NIT)
• <ext> = file extension 

<var>: vegetation biological traits variable description

• CHL (Chlorophyll):
  • Band 1:chl_mean; chlorophyll content mean (micrograms per centimeter squared)
  • Band 2: chl_std_dev; chlorophyll content standard deviation (micrograms per centimeter squared)
  • Band 3: chl_qa_mask; quality assurance mask
• LMA (Leaf Mass per Area)
  • Band 1: lma_mean; leaf mass per area mean (grams per meter squared)
  • Band 2: lma_std_dev; leaf mass per area standard deviation (grams per meter squared)
  • Band 3: lma_qa_mask; quality assurance mask
• NIT (Nitrogen Concentration)
  • Band 1: nit_mean; nitrogen concentration mean (milligrams per gram)
  • Band 2: nit_std_dev; leaf mass per area standard deviation (milligrams per gram)
  • Band 3: nit_qa_mask; quality assurance mask

Example file name:

SISTER_PRISMA_L2B_VEGBIOCHEM_20110513T175417_001_CHL.tif

Table 2. The outputs of the L2B vegetative biological traits PGE use the following naming convention and produce the following data products. The naming convention for these files follow the same pattern as described above.

Product description	Example filename
Variable level 3-band geoTIFF file with variable, standard deviation, and quality assurance mask.	SISTER_PRISMA_L2B_VEGBIOCHEM_20210326T193449_001_<var>.tif
Variable level metadata including sensor, start and end time, description, bounding box, product, and processing level.	SISTER_PRISMA_L2B_VEGBIOCHEM_20210326T193449_001_<var>.met.json
Variable level dataset file version (e.g. {"version": "v1.0"}) used internally by SISTER to track product versions.  Note that this is not the same as the CRID	SISTER_PRISMA_L2B_VEGBIOCHEM_20210326T193449_001_<var>.dataset.json
Variable level PGE context: supports data product traceability by providing provenance details, arguments, and runtime settings used during the run of the Product Generation Executable (PGE) on the SISTER platform	SISTER_PRISMA_L2B_VEGBIOCHEM_20210326T193449_001_<var>.context.json
Metadata including sensor, start and end time, description, bounding box, product, and processing level	SISTER_PRISMA_L2B_VEGBIOCHEM_20210326T193449_001.met.json
Dataset file version (e.g. {"version": "v1.0"}) used internally by SISTER to track product versions.  Note that this is not the same as the CRID	SISTER_PRISMA_L2B_VEGBIOCHEM_20210326T193449_001.dataset.json
PGE context: supports data product traceability by providing provenance details, arguments, and runtime settings used during the run of the Product Generation Executable (PGE) on the SISTER platform.	SISTER_PRISMA_L2B_VEGBIOCHEM_20210326T193449_001.context.json
Quicklook image	SISTER_PRISMA_L2B_VEGBIOCHEM_20210326T193449_001.png
PGE runconfig: defines inputs for the dataset's runs (one file per flight line)	SISTER_PRISMA_L2B_VEGBIOCHEM_20210326T193449_001.runconfig.json
PGE log (one file per flight line)	SISTER_PRISMA_L2B_VEGBIOCHEM_20210326T193449_001.log

geoTIFF NoData values = -9999

Table 3. Possible SISTER Project Instruments available in the v001 dataset

Instrument	Instrument fullname
AVCL	AVIRIS Classic
AVNG	AVIRIS Next Generation
DESIS	DLR Earth Sensing Imaging Spectrometer
PRISMA	PRecursore IperSpettrale della Missione Applicativa

The 2018 National Academies’ Decadal Survey entitled, “Thriving on Our Changing Planet: A Decadal Strategy for Earth Observation from Space.” identified Surface Biology and Geology (SBG) as a Designated Observable (DO) with the following observing priorities:

* Terrestrial vegetation physiology, functional traits, and health
* Inland and coastal aquatic ecosystems physiology, functional traits, and health
* Snow and ice accumulation, melting, and albedo
* Active surface changes (eruptions, landslides, evolving landscapes, hazard risks)
* Effects of changing land use on surface energy, water, momentum, and C fluxes
* Managing agriculture, natural habitats, water use/quality, and urban development

To accomplish these priorities, the DO requires the combined use of visible to shortwave infrared (VSWIR) imaging spectroscopy and multispectral or hyperspectral thermal infrared (TIR) imagery acquired globally with sub-monthly temporal revisits over terrestrial, freshwater, and coastal marine habitats. 

This approach presents some interesting challenges. Due to the high spatial and spectral resolution, SBG is expected to generate roughly 90 TB of data products per day. In addition, the number of existing community algorithms for processing this data is large and needs to be evaluated, and there is no particular consensus yet on standard file formats for hyperspectral data.

To help address a subset of these challenges the SBG Algorithms Working Group was formed to review and evaluate the algorithms applicable to the SBG DO. Also, the SISTER activity was created to prototype workflows and generate SBG-like data products. 

This collection is one of several that include the first experimental data products produced by SISTER with the objective of generating SBG-like VSWIR imagery with 30-m spatial resolution and 10-nm spectral resolution. The scenes in the various SISTER collections include examples from the terrestrial, aquatic, snow/ice, and geologic domains, as well as from validation sites. The main purpose of these data are to provide the scientific community with prototype data products from the SBG workflow in order to get early feedback on things like useability, file formats, and metadata.

These products are still considered experimental, and no quality assessment procedure was conducted for this collection.

The datasets in this collection were collected by PRISMA, a spaceborne imaging spectrometer operated by the Italian Space Agency (ASI). PRISMA measures radiance at ~10-nanometer (nm) intervals in the visible to shortwave infrared spectral range between 400 and 2500 nm at 30-m spatial resolution. The PRISMA scenes in this collection are globally distributed and include vegetation, aquatic and snow targets as well as locations with in situ validation datasets, namely RadCalnet sites (Bouvet et al., 2019). Image datasets were downloaded from the ASI PRISMA Data Portal (https://prisma.asi.it) and ingested into the SISTER platform for further downstream processing (Figure 2).

Figure 2. Workflow diagram of the SISTER 001 production run (Click on image to view full-resolution version).

First, images were converted to radiance using provided band gains and offsets and per-pixel sensor and solar geometries were generated using acquisition metadata. A digital elevation model (DEM) covering the extent of the image was then generated from the global Copernicus DEM (https://copernicus-dem-30m.s3.amazonaws.com/). Prior to atmospheric correction a series of correction routines were applied to the dataset to improve geometric registration and radiometry. First a small correction was applied by resampling the radiance data using a precalculated wavelength center array. Next a pseudo flat field correction was applied to the radiance data using a precalculated array of radiometric adjustment coefficients. Finally using a Landsat image as a reference, pixel coordinates were adjusted using an image matching algorithm. 

Then, the radiance data were processed to surface reflectance using an optimal estimation atmospheric correction algorithm, ISOFIT (Thompson et al., 2018) with an open-source neural-network-based emulator for modelling radiative transfer (Brodrick et al., 2021). Next, spectral resampling to a 10-nm sampling interval was performed in a two-step calculation. Bands were first aggregated and averaged to the closest resolution to the target interval then a piecewise cubic interpolator was used to interpolate the spectra to the target wavelength spacing.

Following spectral resampling a combination of topographic and glint correction algorithms were applied to each image. Topographic correction was performed using the Sun-Canopy-Sensor+C algorithm (Soenen et al., 2005), and glint correction was performed using the method of Gao and Li (2021). 

Fractional cover maps were generated from the corrected reflectance datasets using a spectral mixture analysis. Spectral unmixing was performed using a generic four endmember dataset to derive fractional cover estimates of soil, vegetation, water and snow/ice. To minimize the impact of intraclass brightness variability on unmixing results a brightness normalization was applied.

Terrestrial vegetation biochemistry maps were generated from the corrected reflectance datasets using partial least squares regression (PLSR) models. Permuted PLSR models for estimating chlorophyll content, nitrogen concentration, and leaf mass per area were developed using coincident NEON AOP canopy spectra, downsampled to 10 nm, and field data collected by Wang et al. (2020). Biochemical trait estimates were only calculated for pixels with greater than 50% vegetation cover. In addition to mean biochemical trait estimates, per-pixel uncertainties were calculated along with a quality assurance mask which flags pixels with trait estimates outside of the range of data used to build the model. 

For all datasets in the SISTER 001 processing version (also known as the CRID), workflow components, versions, and links to source code are summarized in Table 4 as provided in the SISTER_CRID_001.json dataset file.

Table 4. Key SISTER workflow components, versions, and links to source code used for V001 production (Townsend et al., 2023).

software	version	url
maap-api-nasa	2.0	https://gitlab.com/geospec/maap-api-nasa/-/tags/2.0
sister-preprocess	2.0.0	https://github.com/EnSpec/sister-preprocess/releases/tag/2.0.0
sister-isofit	2.0.0	https://gitlab.com/geospec/sister-isofit/-/releases/2.0.0
sister-resample	2.0.1	https://github.com/EnSpec/sister-resample/releases/tag/2.0.1
sister-reflect_correct	2.0.0	https://github.com/EnSpec/sister-reflect_correct/releases/tag/2.0.0
sister-fractional-cover	1.0.0	https://gitlab.com/geospec/sister-fractional-cover/-/releases/1.0.0
sister-algorithm_router	1.0.0	https://github.com/EnSpec/sister-algorithm_router/releases/tag/1.0.0
sister-trait_estimate	1.0.0	https://github.com/EnSpec/sister-trait_estimate/releases/tag/1.0.0
sister-grainsize	1.0.0	https://github.com/EnSpec/sister-grainsize/releases/tag/1.0.0

Brodrick, P.G., D.R. Thompson, J.E. Fahlen, M.L. Eastwood, C.M. Sarture, S.R. Lundeen, W. Olson-Duvall, N. Carmon, and R.O. Green. Generalized radiative transfer emulation for imaging spectroscopy reflectance retrievals, Remote Sensing of Environment, Volume 261, 2021, 112476, ISSN 0034-4257. https://doi.org/10.1016/j.rse.2021.112476

Gao, B.C., and R.R. Li. 2021. Correction of Sunglint Effects in High Spatial Resolution Hyperspectral Imagery Using SWIR or NIR Bands and Taking Account of Spectral Variation of Refractive Index of Water. Advances in Environmental and Engineering Research, 2(3), 1-15. https://doi.org/10.21926/aeer.2103017

Queally, N., Z. Ye, T. Zheng, A. Chlus, F. Schneider, R.P. Pavlick, and P.A. Townsend. 2022. FlexBRDF: A Flexible BRDF Correction for Grouped Processing of Airborne Imaging Spectroscopy Flightlines. Journal of Geophysical Research: Biogeosciences, 127(1), e2021JG006622. https://doi.org/10.1029/2021JG006622

Soenen, S.A., D.R. Peddle, and C.A. Coburn. 2005. SCS+ C: A modified sun-canopy-sensor topographic correction in forested terrain.IEEE Transactions on geoscience and remote sensing, 43(9), 2148-2159. https://doi.org/10.1109/TGRS.2005.852480

Thompson, D.R., V. Natraj, R.O. Green, M.C. Helmlinger,B._C. Gao, and M.L. Eastwood. 2018. Optimal estimation for imaging spectrometer atmospheric correction. Remote Sensing of Environment 216, 355-373.https://doi.org/10.1016/j.rse.2018.07.003

Townsend, P., M.M. Gierach, P.G. Brodrick, A.M. Chlus, H. Hua, O. Kwoun, M.J. Lucas, N. Malarout, D.F. Moroni, S. Neely, W. Olson-Duvall, J.K. Pon, S. Shah, and D. Yu. 2023. SISTER: Composite Release ID (CRID) Product Generation Files, 2023. ORNL DAAC, Oak Ridge, Tennessee, USA. https://doi.org/10.3334/ORNLDAAC/2231

Wang, Z., A. Chlus, R. Geygan, Z. Ye, T. Zheng, A. Singh, J.J. Couture, J. Cavenderâ Bares, E.L.  Kruger, and P.A. Townsend. 2020. Foliar functional traits from imaging spectroscopy across biomes in eastern North America. New Phytologist, 228(2), pp.494-511. https://doi.org/10.1111/nph.16711