This paper presents a new method of mining temporal data, using Relational Concept Analysis (RCA), that is applied to sequential datasets, dealing with biological and physico-chemical (PhC) parameters sampled in waterbodies. Our aim is to reveal meaningful and hierachical partially ordered patterns (po-patterns) linking the two types of parameters. We propose a comprehensive temporal data mining process starting by using RCA on an ad hoc temporal data model. Then, we continue with the extraction of sets of subsequences summarized as popatterns. Finally, we select relevant po-patterns, using measures based on the distribution of the concept extents. This process is assessed through some quantitative statistics and qualitative interpretations resulting from experiments carried out on real sequential datasets.