Designing hydraulic structures, such as culverts, bridges, weirs, and check dams, while planning new flood inundation areas, needs correct assessment of design discharges. In gauged catchments with long time series of discharges, statistical methods are commonly used based on fixed theoretical distributions and on empirical distributions. However, in ungauged catchments, this approach is not possible. In addition to more advanced methods, which are used today, e.g., rainfall–runoff models, much more simple approaches are still needed based on regionalization and empirical models. Thus, the objective of this work is to develop a new empirical model for calculating peak discharge expressed as the median of annual peak discharge (QMED). The innovative aspect of this paper is the use of a new parameter, named landscape hydric potential (LHP), as a descriptor of water storage in catchments. LHP has a crucial role as the descriptor of water storage in catchment and, thus, it has an influence on forming discharges. The work was done in the Upper Vistula basin in the Polish Carpathians. This study was carried out in mountain catchments located in the Upper Vistula basin, in the south part of Poland in in the Polish Carpathians. Results show that the proposed model could provide appropriate calculations in changing climate conditions, as well as when land use is changed. The proposed model is simple and effective; for calculating QMED, it needs only two parameters: catchment area and LHP. Since the model has a significant and high correlation coefficient, it could be used for assessing of QMED in ungauged mountain catchments. The determined form of the empirical equation finds application in the entire Upper Vistula basin, for catchments with a surface area from 24 km2 up to 660 km2.