Monitoring kelp farms biomass and biofouling using low-cost technologies

Cultivated seaweed is the fastest-growing aquaculture sector worldwide and a multibillion-dollar industry. Monitoring kelp farm biomass, environmental conditions (e.g. temperature, salinity, nutrients and irradiance) and biofouling organisms in a seaweed farm is important for making decisions related to growth optimisation and harvesting logistics. In this study, we showed a combination of low-cost technologies – from affordable underwater robots used for kelp biomass estimations to flow-through microscopes and flatbed scanners for early biofouling detection – all accessible for kelp farmers. These, in combination with “deep-learning” and user-friendly segmentation approaches, have the power to provide a fast and reliable estimation of seaweed biomass of good quality. Here, we show results from the MoniTARE project funded by the Research Council of Norway, where state-of-the-art, yet cost-effective, and scalable technologies were aimed at optimising monitoring in a Norwegian kelp farm. Robotic monitoring of kelp farms, including biomass growth, was assessed using a mini, cost-effective, remotely operated vehicle (ROV). For a fast and reliable estimation of kelp biomass, a robust set of images to build a data-centric machine learning platform was collected, where we developed computer vision applications supported by AI algorithms. Strong correlation (R²=0.85) between the ground-truth biomass (manually collected) and the biomass inferred through 2D computer vision techniques from recorded images. For biofouling estimation and early detection, a low-cost flatbed scanner combined with machine learning methods were used to detect early settlement stages (invisible to the naked eye, but noticeable in the scanning images) and to quantify bryozoan coverage from cultivated Saccharina latissima (Phaeophyceae). Using this method, we attempted to understand (at a finer temporal scale) the influence of a combination of major environmental variables collected using low-cost sensors, i.e. temperature, turbidity, phytoplankton biomass (chlorophyll a), and illumination, on the time of settlement and growth of bryozoan colonies on the kelp lamina. We also tested a low-cost, home-built flow through microscope (AFTI-scope) as a potential semi-automated method to quantify abundance estimates, and to monitor larval size changes of biofouling organisms, such as E. pilosa and M. membranacea, detrimental for the quality of kelp for commercial purposes. Larval size appeared to be a strong factor for successful biofouling of undesired organisms in kelp farms. Low-cost flow-through microscopes have the potential for further development, where the size dependent settlement of larvae on kelp could be a proxy for early fouling detection. Automation of kelp farm monitoring has the potential to revolutionize the industry by offering scalability of production and improved yield predictions.