Questions about Remote sensing HABS

[grimiore]

I'm currently conducting a study about remote sensing harmful algal blooms (HABS) using satellite imagery. However, my main problem is the consistent cloud cover in our area that affects the data when analyzed in SeaDAS because it is located in the Pacific and is always cloudy and/or rainy, especially during red tide blooms which is key to the analysis. Also, my institution does not have a field spectroradiometer that is used to provide in situ data for the validation of the algorithm, though we have other data such as phytoplankton composition & abundance and physico-chemical parameters (nitrate, phosphate, chlorophyll-a, etc.). Which is why I'd like to ask (1) if there is a method to increase coverage despite the cloud cover, and (2) if is there any method/algorithm that can be used as an alternative for using in situ reflectance in detecting HABs? Thank you.

[OB General Science - guoqingw]

Hello,

Thank you for reaching out with your questions regarding the challenges of detecting harmful algal blooms (HABs) in persistently cloudy regions. Here are some potential approaches to consider:

1. Increasing Coverage Despite Cloud Cover
Persistent cloud cover is a common challenge in satellite remote sensing, particularly in tropical and coastal regions. Here are a few strategies to mitigate this issue:

• Multi-Sensor Data Fusion: Combining data from different satellite sensors with varying spatial, temporal, and spectral capabilities can improve coverage.

• Temporal Compositing & Cloud Masking: Instead of relying on single-day observations, you can aggregate multiple images over a period (e.g., weekly or biweekly composites) to fill gaps caused by cloud cover.

• Hyperspectral Imaging: PACE (Plankton, Aerosol, Cloud, ocean Ecosystem) provides hyperspectral data that may enhance bloom detection by leveraging more spectral bands to distinguish HABs.

2. Alternatives to In Situ Reflectance for HAB Detection
If in situ reflectance measurements are unavailable, you can still validate and refine your HAB detection approach using alternative methods:

• Empirical Algorithms Using Available Water Quality Data: Since you have phytoplankton composition, abundance, and physico-chemical parameters, you can develop statistical models that correlate these with satellite-derived chlorophyll-a or other bio-optical parameters. Machine learning approaches can also be trained using historical data.

• Regional Bio-Optical Models: If site-specific field spectra are not available, regional bio-optical models based on similar environmental conditions (e.g., case studies from other Pacific regions) can serve as a proxy for validation.

• Fluorescence Line Height (FLH) and Alternative Indices: FLH, derived from MODIS and VIIRS, can be useful for detecting surface chlorophyll concentration even in complex waters.

• Citizen Science & Auxiliary Data: Collaborating with local researchers, fisheries, or even citizen science projects can provide valuable in situ observations (e.g., Secchi disk transparency, bloom reports) that can be used for indirect validation.

Hope these help.

Guoqing