The effect of disturbances on bacterial community characteristics and assembly patterns: insights from microcosms and land-based aquaculture systems

Abstract

Bacterial communities play a vital role in human health, ecosystem stability and various industrial applications, including land-based aquaculture. These communities change over time through predictable deterministic processes such as selection and random stochastic processes such as ecological drift. Our understanding of how disturbances affect community assembly is limited. This knowledge gap hinders accurate predictions of how communities respond to disturbances. This thesis aimed to address this knowledge gap by investigating the effects of various disturbances on community characteristics and assembly processes. Quantifying community assembly is challenging due to the statistical complexities associated with stochastic processes and their lack of predictable patterns. This thesis first explored the limitations of the null models NTI and βNTI that are used to quantify community assembly. These null models have several assumptions that make them unsuitable for implementation in replicated experimental systems without much environmental variability, such as laboratory systems. Therefore, we propose a novel framework, the ‘replicate similarity rate of change’, to quantify selection and drift in replicate experimental microcosms with dispersal limitation. In the context of aquaculture, it has been shown that rearing tank water with a selective pressure for opportunistic bacterial growth reduces the survival of vulnerable fish larvae. However, how the rearing environment affect the characteristics and assembly of the bacterial community in the rearing tank water and in the fish is poorly understood. This thesis investigated this knowledge gap. The results showed that the flow-through, microbially matured and recirculating aquaculture systems influenced the bacterial composition and assembly patterns in the rearing tank water. A key finding was that stochastic community assembly processes in the bacterial communities of the rearing tank water and fish larvae increased over time in a correlated relationship. Furthermore, different rearing environments created distinct stable state attractors, strongly suggesting that microbial management of rearing tank water is feasible. Importantly, we found that small differences in carrying capacity between incoming water and rearing tank water favoured a stable bacterial community with less opportunistic growth and increased larval viability. Ecosystems such as land-based aquaculture rearing tanks are often disturbed. However, there are many uncontrollable variables in the systems that make it difficult to identify the disturbance mechanisms. Therefore, in the third part of this thesis, we aimed to understand how community characteristics and assembly patterns were affected by the four disturbances periodic dilutions, unsuccessful invasions, antibiotic treatment and phage treatment. In these experiments, bacterial communities from the Trondheimsfjord or Jonsvatnet lake were divided into replicated laboratory microcosms with low environmental variability and operated as semi-continuous systems. Using the ‘replicate similarity rate of change’ framework, we found that dilution increased the contribution of selection, while unsuccessful invasions and antibiotics increased drift. The changes in assembly were proposed to be a result of increased resource availability following the disturbance events that initiated opportunistic bacterial growth. Phage treatment had no effect on community assembly, and we found no evidence that the treatment disturbed the communities. In conclusion, this research shed light on how bacterial community characteristics and assembly processes are affected by disturbances. Furthermore, this thesis provides insights into the management of bacterial communities in aquaculture. Overall, the results pave the way for more accurate predictions of bacterial community responses to disturbances. Such prediction is crucial to ensure the stability of bacterial communities in industrial applications, as well as being relevant to human and animal health.