Space Shrimps| To Bloom or Not to Bloom
Project Details
The Challenge | To Bloom or Not to Bloom
Algae?
Our aim is to utilise the data and technology we have to discover the causes behind algae bloom and how we can predict the next occurrence.
We chose the “To Bloom Or Not To Bloom” challenge because we believed our team had the appropriate skill set to come together and create a solution. We had no prior knowledge of the biological factors involved with algal blooms and we decided to challenge ourselves. We wanted to work in an unfamiliar field where we knew our technical backgrounds could be applied to tackle this problem.
Our project involves finding and creating a proof of concept AI algorithm which predicts the spread of algal bloom. This is primarily done by focusing on the multiple factors that cause algae bloom, such as temperature, wind/water currents, sunlight density, nutrients like phosphorus and nitrogen and salinity levels. By focusing on these factors, we concentrated our research on how these factors cause algae bloom and whether if it is just one or multiple factors at play. We looked at various factors, with different team members looking at different variables. A wide range of NASA datasets of satellite mapping was used to compare and contrast the knowledge we obtained and to support our proposed hypotheses.
To create the solution, we used satellite imagery to map the algae bloom more accurately. Chlorophyll, the pigment in algae, strongly absorbs visible light for use in photosynthesis. The cell structure of the algae, on the other hand, strongly reflects near-infrared light. From this, we can begin to detect high and low concentrations of algae across surfaces of water using the near-infrared sensor on the Sentinel 2 satellite. By using this method, we can find algal bloom and compare to NASA's satellite mappings of salinity levels, temperature and other factors we believe may cause the algal bloom to see if there is a correlation between them.
We propose using satellite imagery to track and monitor the factors may have on algal bloom. In the future, we would like to use the same technology plus AI and be able to identify the difference between the toxic algae and the non-toxic algae, as the non-harmful algae have known scientific health benefits, biofuel use cases and food products.
Resources used:
https://neo.sci.gsfc.nasa.gov/view.php?datasetId=MY1DMM_CHLORA
https://aquarius.oceansciences.org/cgi/gal_salinity.htm
https://toolbox.google.com/datasetsearch/search?qu...