While the effect of turbulence on potential vorticity (PV) modification in the upper troposphere and lower stratosphere (UTLS) has long been identified, previous studies mainly focused on the situation of an upper-level jet and the associated frontal zones. In this study, a more complex flow situation is investigated, which is a case with turbulence-induced PV changes in a warm conveyor belt (WCB) outflow. A numerical simulation of the case is performed with the Integrated Forecasting System (IFS) model from the European Centre for Medium-Range Weather Forecasts, with additional outputs allowing a direct examination of the instantaneous PV tendencies generated by individual non-conservative processes parametrised in the model. Arc-shaped tripolar PV tendency bands due to turbulence are simulated in the UTLS region, which align with the WCB outflow boundary. A detailed analysis of vertical cross sections illustrates that the divergent outflow of the WCB is the key to the emergence of these banded structures. The tripolar pattern of the bands is the result of divergent non-advective PV fluxes associated with turbulent mixing, which is explainable by the analytical framework developed in Part 1 of this study, validating its applicability to complex flows. When accumulated over time, systematic material PV changes along air parcel trajectories occur in the UTLS. Turbulence is the most relevant among the non-conservative processes in causing these PV changes, with a major band of PV decrease identified in the WCB outflow region. The examination of individual air parcel trajectories shows a surprisingly steady accumulation of PV changes despite the local flow being perpendicular to the PV tendency bands. A synchronised movement of the PV tendency bands and the air parcels is revealed as the cause, with the WCB outflow being the driver of such organisation. Coherent PV modification by turbulence is achieved by this delicate coupling of the large-scale flow and individual air parcels, extending the more straightforward pathway of PV modification in the upper-level jet-front system as examined in detail in Part 1. Taken together, this two-part study therefore demonstrates that turbulence is capable of modifying PV systematically in the UTLS even in complex flow situations involving different non-conservative processes. It also reveals the importance of weather systems in imparting the necessary spatial and temporal coherence for turbulence to generate a systematic feedback to larger scales in the form of substantial PV modification. The exact consequences of this upscale effect of turbulence on the dynamics in the UTLS and therefore on medium-range predictability remain to be explored in future studies.