Aquaculture is the world’s fastest-growing food production sector and plays a vital role in global food security. Recirculating aquaculture systems (RAS) are increasingly favoured due to their efficient water use and reduced environmental impact. However, only ~30% of the feed is assimilated by fish, with the remainder released as waste, including toxic ammonia and suspended organic matter from fish excretion. To overcome the high cost and inefficiency of conventional ammonia removal, a biofloc-based recirculating aquaculture system (B-RAS) was developed. In B-RAS, ammonia and carbon are biologically removed through microbial assimilation in a dedicated reactor, and the resulting microbial biomass is harvested and reused as a protein-rich feed amendment.
System performance depends on the efficiency of microbial assimilation and biomass production, typically indicated by Total ammonical nitrogen (TAN) and total suspended solids (TSS). Real-time monitoring of these parameters is therefore essential for effective and timely system management.
This study aims to optimize B-RAS performance by predicting TAN in both the reactor and the fishpond and TSS in the reactor using a low-cost, portable, real-time monitoring system. The system integrates sensors for pH, dissolved oxygen (DO), redox (ORP), temperature, and turbidity, deployed across both the fishpond and biofloc reactors. Data are transmitted to the cloud, enabling continuous access, predictive analytics, and automated alerts.
Preliminary results demonstrated successful prediction of TAN and TSS levels. The system reliably triggered early warnings as TAN approached critical thresholds. Fish survival remained at 100%, with an average daily growth rate of 1.03%, even under high stocking densities of 55 kg/m³.
This integrated approach supports data-driven management of B-RAS and presents a scalable, sustainable solution for intensive aquaculture.
