Precision aquaculture is founded on a set of disparate, interconnected sensors deployed within the marine environment to monitor, analyse, interpret, and provide decision support for farm operations. This trend parallels developments in agriculture where sensors and other observing technologies lead to enhanced insight into crop production as well as animal health and welfare (Precision Livestock Farming or PLF).  A notable difference for the aquaculture paradigm is that it requires sensing of the ambient environment (e.g., water temperature, oxygen), a consideration that is less important in agriculture where animals can be housed. The fundamental approach has been summarised as a series of steps, namely: observe, interpret, decide, and act (Føre et al., 2018). Traditionally, many of these steps have required human intervention and depended heavily on farmer experience and intuition for correct decisions and actions. As farm size increases, however, and move further offshore, increased automation and data-driven decision-making are imperative to enhance the environmental performance and sustainability of the aquaculture industry. 

Materialisation of precision aquaculture depends on IoT technologies to empower management in a chaotic environment subject to the vagaries of oceans and weather. An obvious impediment is water cover, but other major obstacles exist, including the harsh environment, power and connectivity in offshore locations, large range of spatial scales involved (fish-, cage-, farm- and bay-scale), and the challenges of manual intervention or analysis in the ocean (where access can be regularly impeded or prevented by adverse weather). A fish farm has an imposing array of underwater chains, ropes, moorings, and other infrastructure, so wireless communications are essential. Further, the distributed nature of the industry, composed of many small-scale aquaculture companies and sensor providers, pose challenges related to the integration of diverse – sometimes proprietary – datasets into a unified edge, fog, and cloud ecosystem.

Application of mature monitoring, modelling, prediction, and analysis tools to aquaculture farms has potential to improve operations and alleviate key challenges facing the industry. The behavioural and environmental data obtained by PFF can be used as Operational Welfare Indicators (OWI) that predict the state of the animals when conditions change.Fish feed represents 50-70% of fish farmers’ production costs while the growth rate of fish is intrinsically linked to feed composition and time of supply; precise management can link fish growth with optimal feed schedule and composition, that minimises wastage (and subsequent pollution of surrounding waters) and improves welfare and productivity. Disease and parasite-induced impacts are a major issue for aquaculture farms costing the industry up to $10 billion annually and having severe socio-economic impacts. Further parasite-control treatment in salmon farms constitutes 7.5% of total production costs (Shinn et al., 2015). Farming in the open ocean requires the ability to respond to natural fluctuations that impact operations, such as dissolved oxygen (DO) concentrations or temperatures, both of which act as stressors, impact feeding, and parasitic rates, and even cause mortalities. Today, management of most of these tasks is conducted manually relying on direct human observation or human-centric data acquisition means to observe conditions, combined with decision making based on subjective experience. However, as real-time sensor technologies become more prevalent on farms, the foundation exists to transition the industry from ad-hoc decision-making based on heuristics and intuition, to real-time informed decisions backed by AI insights and IoT connectivity.