Exploring New Algae Biomass Estimation Approach in Multispectral Imaging

This research presents a novel approach for assessing algae growth in aquatic systems through dilution. Marine micro-organisms like Chlamydomonas, Spirulina and Chlorella, commonly found in abundance in muddy waters, ponds, rivers, and lakes, are of economic significance due to their high lipid content, needed for zero-emission biofuels. These organisms' growth and characteristics have been assessed through satellite on rivers, lakes and ponds. But, several challenges hinder the success of these methods. This study employs an unmanned aerial vehicle (UAV) equipped with a multispectral camera to detect and monitor the presence and growth of the algal communities. Image detection was conducted using a radiometric-calibrated UAV to identify the presence and characteristics of micro-organisms. The primary objective is determining the optimal dilution sampling condition and taking UAV images. The project aims to predict biomass based on image data and eliminate sampling. The dilution involves transferring the first 25% of the tank contents into another container for biomass harvest, while the remaining 75% is iteratively augmented with 25% tap water for eleven consecutive times. Samples were collected, and drone flights were conducted iteratively for the eleven flights. Biomass was determined in the laboratory using circular dry membranes placed in foil, while the optical density (OD) was determined using a spectrophotometer. The resulting data were compared with drone imagery for comprehensive analysis under certain conditions involving two variables. Among the four conditions utilized, only one demonstrated superior performance with R
²
>90%, both for Optical Density (OD), Biomass and NDVI. This study established a correlation model using a vegetation index (NDVI), absorption index and a viable method to estimate algae biomass through UAV-based multispectral imaging. The accuracy of the estimated model was validated by a strong positive correlation between estimated and manually measured biomass.