In recent days, research in underwater fish farming has been seen as a sustainable method to land-based fish farming due to the increasing global population and limitations in land-based farming areas worldwide.The fish farming cage installations are usually operated in the depth area of seawater and are ideal circumstances for fish farming. Monitoring sensor-based data collection from fish farming is highly efficient and more innovative. Due to the unique properties of seawater, the substantial challenge of an Underwater Wireless Sensor Network (UWSN) is guaranteeing reliable data communication between the underwater sensor nodes and surface nodes. To ensure reliable communication, the mathematical approach Stochastic Network Calculus (SNC) is considered for modeling the reliable data flow control in the logical link control layer to satisfy the requirements of the underwater fish farming application. This paper presents the SNC mathematical model construction of stochastic bounds for frame flow, window size, service element analysis, feedback loop analysis, delay, and backlog bounds derived in the proposed model. The proposed model produces the better frame delivery and throughput in an incremental order with respect to the simulation model. There is significant difference (4 to 12 percentage) between the stochastic and simulation model. Thus, the reliable frame flow can be adopted with the dynamic changes in the feedback loop to avoid the overflow by repeated requests.