Clustering of environmental functional data

Often environmental scientists face the problem of clustering different sites, areas or stations in a monitoring network in order to identify some common features among data collected at different locations. In a functional data analysis approach, each location can be seen as a specific individual, on which noisy observations from a continuous random function are collected at discrete times. The definition of suitable models for samples of such functional observations, can provide useful insights about the dynamics of the variables of interest. In such a context, a cluster can be defined as a group of individuals (i.e. locations, stations, areas etc.) where the observed trajectories share common salient features. We present some classification results in a water quality network and focus on some open issues.