Urban water resources and drainage systems are anticipated to face significant pressures and challenges in the future, particularly regarding water quantity. The problem is further exacerbated by the anticipated rise in urbanization. The importance of the problem lies in how to make real data better inform the model, in other words, how to quantify the misrepresentation between the model's expression and the real data. This paper proposes a new framework for quantifying the uncertainty in spatial data in soil data at different scales. This new framework is based on Fuzzy C-Means (FCM) clustering for designing membership functions and calculating fuzzy entropy for different soil type data. By assigning a membership degree value to each soil membership degree, reflecting the extent to which it belongness, this method is suitable for handling complex soil datasets. In the Saanich area case study, the cluster centers obtained from the FCM algorithm are used to integrate the raw data to the shapes of the membership functions for “poor”, “medium” and “well draining” soil categories, and fuzzy entropy is calculated using Shannon’s entropy formula. The results show that this framework can effectively track changes in membership functions and quantify data uncertainty at different scales. This data-driven approach has significant advantages in handling complex datasets, capturing the inherent variability and uncertainty within the data more accurately.