A Broadscale Assessment of Sentinel-2 Imagery and the Google Earth Engine for the Nationwide Mapping of Chlorophyll a

Richard A. Johansen (), Molly K. Reif, Christina L. Saltus and Kaytee L. Pokrzywinski
Additional contact information
Richard A. Johansen: Environmental Laboratory, US Army Engineer Research and Development Center, 3909 Halls Ferry Rd., Vicksburg, MS 39180, USA
Molly K. Reif: Joint Airborne Lidar Bathymetry Technical Center of Expertise, 7225 Stennis Airport Rd., Kiln, MS 39556, USA
Christina L. Saltus: Environmental Laboratory, US Army Engineer Research and Development Center, 3909 Halls Ferry Rd., Vicksburg, MS 39180, USA
Kaytee L. Pokrzywinski: National Center for Coastal Ocean Science, National Oceanic and Atmospheric Administration, 101 Pivers Island Rd., Beaufort, NC 28516, USA

Sustainability, 2024, vol. 16, issue 5, 1-17

Abstract: Harmful algal blooms are a global phenomenon that degrade water quality and can result in adverse health impacts to both humans and wildlife. Monitoring algal blooms at scale is extremely difficult due to the lack of coincident data across space and time. Additionally, traditional field collection methods tend to be labor- and cost-prohibitive, resulting in disparate data collection not capable of capturing the physical and biological variations within waterbodies or regions. This research attempts to help alleviate this issue by leveraging large, public, water quality databases coupled with open-access Google Earth Engine-derived Sentinel-2 imagery to evaluate the practical usability of four common chlorophyll a algorithms as a proxy for detecting and mapping algal blooms nationwide. Chlorophyll a data were aggregated from spatially diverse sites across the continental United States between 2019 and 2022. Data were aggregated via a field method and matched to coincident Sentinel-2 imagery using k-folds cross-validation to evaluate the performance of the band ratio algorithms at the nationwide scale. Additionally, the dataset was portioned to evaluate the influence of temporal windows and annual consistency on algorithm performance. The 2BDA and the NDCI algorithms were the most viable for broadscale mapping of chlorophyll a , which performed moderately well (R 2 > 0.5) across the entire continental united states, encompassing highly diverse spatial, temporal, and physical conditions. Algorithms’ performances were consistent across different field methods, temporal windows, and annually. The most compatible field data acquisition method was the chlorophyll a , water , trichromatic method , uncorrected with R 2 values of 0.63, 0.62, and 0.41 and RMSE values of 15.89, 16.2, and 23.30 for 2BDA, NDCI, and MCI, respectively. These results indicate the feasibility of utilizing band ratio algorithms for broadscale detection and mapping of chlorophyll a as a proxy for HABs, which is especially valuable when coincident data are unavailable or limited.

Keywords: water quality; CyanoHABs; Sentinel-2; chlorophyll- a; phycocyanin; spatiotemporal analysis; Google Earth Engine (search for similar items in EconPapers)
JEL-codes: O13 Q Q0 Q2 Q3 Q5 Q56 (search for similar items in EconPapers)
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2071-1050/16/5/2090/pdf (application/pdf)
https://www.mdpi.com/2071-1050/16/5/2090/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jsusta:v:16:y:2024:i:5:p:2090-:d:1350168

Access Statistics for this article

Sustainability is currently edited by Ms. Alexandra Wu

More articles in Sustainability from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().