Development of Chlorophyll-a estimation algorithm from satellite images for seawater monitoring in Vietnam
Chlorophyll-a (Chl-a) content - a parameter related to the level of eutrophication and the growth of phytoplankton, is one of the important indicators reflecting the "health" of the aquatic environment. Monitoring this index is of great significance in assessing water quality, identifying areas at risk of pollution, supporting the management and protection of the marine environment.
Monitoring water quality using traditional monitoring methods is often time-consuming, expensive and difficult to deploy on a large scale. Meanwhile, remote sensing technology has emerged as an effective modern solution, allowing the collection of environmental data remotely via satellites with the ability to continuously monitor and cover a large area. Ocean color remote sensing images have now been widely applied in the world to study parameters such as suspended matter (SPM), Chl-a or colored dissolved organic matter (CDOM).
Grasping this trend, the international cooperation task: "Research and development of Chlorophyll-a algorithm for VNREDSat-1 images and equivalent" (code: QTRU02.13/21-22) has been deployed. The head of the Vietnamese side is MSc. Chu Xuan Huy, Institute of Space Technology - Vietnam Academy of Science and Technology (now part of Vietnam Space Center), in collaboration with the Pacific Institute of Organic Biochemistry (PIBOC), Far Eastern Branch - Russian Academy of Sciences (FEB RAS). The study uses images from VNREDSat-1 satellite, Vietnam's first remote sensing satellite, and Sentinel-2 images provided by the European Space Agency (ESA). These two data sources have many similarities in spectral range, resolution, and especially open access, suitable for building, calibrating and applying Chl-a monitoring algorithms to serve marine environmental management in Vietnam.
Why is a separate algorithm needed for Vietnam's seas?
Responding to this issue, Master Chu Xuan Huy said: Estimating Chl-a concentration from remote sensing images has been widely studied with many algorithms developed in the world such as NASA's OCx (Ocean Chlorophyll-a) chain. However, each algorithm only gives accurate results in a specific area, due to the different chemical and physical characteristics of each water area.
The relationship between the reflectance spectral signal measured from satellites and Chl-a concentrations in water is influenced by specific environmental factors, such as water surface type, suspended matter, or other optical parameters.
For Vietnam's sea areas, especially coastal areas with diverse geographical and hydrological conditions, available algorithms are difficult to achieve high efficiency without being calibrated. Therefore, the research team aims to develop an experimental algorithm that is adjusted to suit actual conditions, while supplementing data for areas that are lacking to enhance the ability to apply remote sensing images in large-scale water quality monitoring.
Achievements and application prospects
Within the framework of the mission, the team organized many field surveys in the coastal lagoon area of Thua Thien - Hue, one of the large and typical lagoon systems in the Central region. In addition, an in-depth survey by the research vessel Academician Oparin (Russia) was also carried out to collect data from offshore areas. In total, the team measured and took samples at 74 points, of which 52 points were used to build the model and 22 points were used to check the accuracy.
Data collection location
Prepare to preserve and collect water samples in the field
The results showed that the OC3 algorithm applied on both types of satellite images achieved a high coefficient of determination (R²), above 0.7 for VNREDSat-1 and above 0.8 for Sentinel-2. Scientists have determined and demonstrated the suitability of the OC3 algorithm to estimate Chl-a content from VNREDSat-1 and Sentinel-2 remote sensing images. Based on the combination of satellite image data and field surveys, the group developed a set of algorithms with experimental correction factors suitable for coastal conditions in Vietnam. The calculation results showed a clear agreement between Chl-a distribution from images and measurement data in Thua Thien - Hue. Chl-a concentrations were low in coastal waters, higher in estuaries, lagoons, especially in aquaculture areas, reflecting the ecological characteristics of the area and confirmed by previous studies.
Chl-a content distribution calculated from remote sensing images
In addition, the field survey dataset in Thua Thien - Hue is a valuable source of information for future marine research. The Chl-a calculation tool software developed by the team on an open source platform helps automate satellite image preprocessing steps, speed up the calculation process and can be extended to other coastal areas with similar conditions. The thematic map system on Chl-a distribution is also an important document for planning, territorial management and monitoring of local water quality.
The study opens up a new, feasible and effective approach to monitoring coastal water quality using remote sensing technology, especially utilizing domestic satellite imagery sources. With high accuracy, automation and potential for wide practical application, the results of the mission not only enhance domestic research capacity but also create a foundation for integrating digital technologies into marine resource management. This is a typical example showing that science and technology can become a real driving force, contributing to improving the effectiveness of planning and conserving marine ecosystems in the process of sustainable development of coastal areas.
Translated by Quoc Khanh
Link to Vietnamese version
