Epsilon Archive for Student Projects

Abstract

By introducing high-density crop production in controlled environments, vertical farming (VF) offers a sustainable solution to problems with urban food security. Incorporating plant science is necessary to improve an already functioning VF system after technological improvements and breakthroughs like robotics. This is where the following pilot study steps in: examining the output of the hydroponic vertical farming system at SweGreen AB, Stockholm, where we planted basil (Ocimum basilicum), romaine lettuce (Lactuca sativa var. romana), and oakleaf lettuce (Lactuca sativa L.). One objective is to compare plant performance, post-transplant recovery support, and growth forecast across ebb and flow (EF) and the nutrient film technique (NFT) irrigation system. Further, image analysis and chlorophyll contents were used to see whether digital images can be used as a substitute for direct measurements of chlorophyll. Leaf temperature was monitored as a proxy for plant growth that is fueled by photosynthesis. NFT was found to outperform EF for the lettuce species for plant performance and prediction accuracy. Image analysis algorithms for leaf color in RGB color channels were shown to need improvement, e.g. by machine learning, to make robust statements on the correlation to chlorophyll. For leaf temperature, it was found that in NFT leaf temperature has less influence on plant growth but additional studies are needed to fully understand the mechanisms behind it. With this study, I intend to contribute to the autonomy and sustainability of vertical farms to supply people with nutritious, fresh food in the future.