Hu, Chuanmin: University of South Florida (Project Lead)
Project Funding:
2015 - 2017
NRA: 2013 NASA: Suomi National Polar-orbiting Partnership (NPP) Science Team and Science Investigator-led Processing Systems for Earth System Data Records From Suomi NPP
Funded by NASA
Abstract:
NPP Ocean Data Products to work on: chlorophyll a concentration (Chl, primary), and remote sensing reflectance (Rrs, secondary).
Ultimate goals: Increase valid Chl data coverage by at least 50% (possibly 100%); Reduce Chl and Rrs data product uncertainty; Improve cross-sensor consistency for Chl. Problem statement and background: Despite significant progress, two problems still exist in the current global ocean color data products for all ocean color missions and especially for the most recent NPP VIIRS: 1) data coverage (quantity), and 2) data accuracy (quality). These problems not only create data gaps and potential spatial/temporal aliasing, but also lead to large inconsistencies between sensors. For example, monthly mean chlorophyll-a derived from the default algorithms (ChlOCx) can differ by 10-20% between VIIRS and MODIS/Aqua in some major ocean basins. Such differences can overwhelm realistic changes of the ocean (often < 10%) in response to climate variability.
A novel algorithm concept, namely the Ocean Color Index (OCI) algorithm, has shown great potentials in achieving multi-sensor consistency and recovering low-quality (i.e., flagged) data to increase coverage. This is because that the OCI algorithm is nearly immune to most errors in the input remote-sensing reflectance (Rrs) data for clear waters (Chl < 0.25 mg m-3, representing ~80% of the global ocean). The ATBD of this algorithm has been published in 2012, and the algorithm has been implemented by the NASA OBPG for all ocean color missions where the products have been generated and distributed for community evaluations. The algorithm has also been recommended by the Ocean Team Leader of the MODIS Science Team (Franz) to replace the current standard MODIS Chl algorithm.

Specific objectives: The project is designed to fully realize the potentials of the OCI algorithm concept to improve both data quantity and data quality, to refine VIIRS data products to continue the MODIS EOS observations, and to bridge to future ocean color missions such as PACE. Specifically, the project has the following objectives:
1) Increase valid data coverage by at least >50% for most ocean areas by recovering some of the low-quality data. These low-quality data are currently discarded in the global data composites;
2) Refine the OCI algorithm to minimize the impact of colored dissolved organic matter (CDOM) on Chl retrievals, with cross-sensor differences reduced from >10% to <5% and cross-basin biases reduced by more than 50%;
3) Extend the same algorithm concept to higher-Chl waters;
4) Reduce uncertainties in the spectral Rrs data product;
5) Make recommendations to future sensor design and data processing.
Approach: The project relies entirely on optical modeling and analyses of both satellite data (through NASA data archive) and in situ data (NASA’s SeaBASS archive). Specifically,
1) Quality control flags will be examined against Chl_OCI data quality to determine which flags can be relaxed (and to what degree) in order to increase data quantity without compromising quality. Validation will be performed using both in situ data and spatial/temporal consistency checks, with uncertainty estimated;
2) Algorithm coefficients will be fine-tuned using ratios of Rrs(412)/Rrs(443) (or other combinations of the blue bands) and SeaBASS data will be partitioned to major ocean basins in order to minimize the influence of CDOM on Chl retrieval accuracy and to minimize cross-basin biases;
3) The new algorithm concept is applicable to ~80% of the ocean (clear waters). To extend to higher-Chl waters, Rrs spectral slope (or other spectral combinations) will be incorporated;
4) The OCI algorithm also provides a constraint on the Rrs spectral curvature, thus will be used to refine Rrs data products to reduce uncertainties;
5) Based on the results above, recommendations on future sensor design (e.g., signal-to-noise) and data processing (e.g., flag settings) will be made.
Publications:
Barnes, B. B., Cannizzaro, J. P., English, D. C., Hu, C. 2019. Validation of VIIRS and MODIS reflectance data in coastal and oceanic waters: An assessment of methods. Remote Sensing of Environment. 220, 110-123. DOI: 10.1016/j.rse.2018.10.034
Barnes, B. B., Hu, C. 2016. Dependence of satellite ocean color data products on viewing angles: A comparison between SeaWiFS, MODIS, and VIIRS. Remote Sensing of Environment. 175, 120-129. DOI: 10.1016/j.rse.2015.12.048
Hu, C., Barnes, B. B., Feng, L., Wang, M., Jiang, L. 2020. On the Interplay Between Ocean Color Data Quality and Data Quantity: Impacts of Quality Control Flags. IEEE Geoscience and Remote Sensing Letters. 17(5), 745-749. DOI: 10.1109/LGRS.2019.2936220
Hu, C., Feng, L., Guan, Q. 2021. A Machine Learning Approach to Estimate Surface Chlorophyll a Concentrations in Global Oceans From Satellite Measurements. IEEE Transactions on Geoscience and Remote Sensing. 59(6), 4590-4607. DOI: 10.1109/TGRS.2020.3016473
Hu, C., Feng, L., Lee, Z., Franz, B. A., Bailey, S. W., Werdell, P. J., Proctor, C. W. 2019. Improving Satellite Global Chlorophyll
a
Data Products Through Algorithm Refinement and Data Recovery. Journal of Geophysical Research: Oceans. 124(3), 1524-1543. DOI: 10.1029/2019JC014941
Marechal, J., Hellio, C., Hu, C. 2017. A simple, fast, and reliable method to predict Sargassum washing ashore in the Lesser Antilles. Remote Sensing Applications: Society and Environment. 5, 54-63. DOI: 10.1016/j.rsase.2017.01.001
Wang, M., Hu, C. 2018. On the continuity of quantifying floating algae of the Central West Atlantic between MODIS and VIIRS. International Journal of Remote Sensing. 39(12), 3852-3869. DOI: 10.1080/01431161.2018.1447161
Zhang, M., Hu, C., Cannizzaro, J., English, D., Barnes, B. B., Carlson, P., Yarbro, L. 2018. Comparison of two atmospheric correction approaches applied to MODIS measurements over North American waters. Remote Sensing of Environment. 216, 442-455. DOI: 10.1016/j.rse.2018.07.012
More details may be found in the following project profile(s):
