This paper reports about a software package which has been developed to automatically partition hydrological networks (catchments) into clusters of similar size. Such clustering is useful for parallel simulation of catchments on distributed computing systems and is typically based on heuristic graph algorithms.There have been a few approaches to automatically partition catchments, but literature research indicates that there seems to be no systematic investigation of the usefulness of different graph algorithms for catchment partitioning over a reasonable number of real world data sets. Our study aims at making a step in this direction.The paper describes the software package, which has been implemented in Java, its pluggable architecture, and initial experiments using the European catchment dataset ECRINS. The paper presents work in progress.