In order to investigate the genesis and developmental evolution patterns of giant dolomite cave systems, focus on the Chiwuxi Basin in Suiyang, Guizhou, a typical karst cave system development area. Based on 1∶50 000 topographic maps and 12.5 m resolution ALOS DEM data, extracted watershed water system. The Horton’s law, box-counting method, and grid water system method were employed to estimate the fractal dimension of the water system in the study area. Comparing the applicability of three methods in the study area and analyzing the developmental trends of its landforms. The results indicate: 1) Controlled by the complex surface-underground “binary and three-dimensional” hydrogeological structure of karst, Horton’s law exhibits poor adaptability. The branch ratio and river length ratio of higher-order rivers in the study area show significant deviations, failing to comply with the Horton’s law. 2) At similar scales, significant discrepancies exist between water system fractal dimension values estimated from different data sources and methods. The most pronounced difference was observed between topographic map water systems and ALOS water systems values using the Horton’s law method, yielding 2.77 and 2.55 respectively, both exceeding theoretical thresholds. The fractal dimension values obtained from different data sources using the box-counting method and grid water system method were relatively close, ranging between 1.41 and 1.42. 3) Ranking the methods for estimating water system fractal dimension from best to worst adaptability yields the following order: box-counting method, grid water system method, Horton’s law method. The box-counting method is more suitable for estimating water system fractal dimensions in karst regions. Integrating the distribution of water system fractal dimensions estimated by this method with the elevation variation ratio of sub-basins, it is comprehensively determined that the study area’s landform development is transitioning from an early to a mature stage. These findings provide a basis for selecting optimal quantitative research methods for studying landform evolution in karst basins.