Catchment travel times from composite StorAge Selection functions representing the superposition of streamflow generation processes

Catchment travel times integrate the multitude of hydrological flow processes and provide insights into catchment functioning. StorAge Selection (SAS) functions describe how residence times of water in storage are related to travel times of water in catchment outflows. As such, SAS functions are useful to summarize transport processes in catchments and are ideal to simulate catchment outflows and the concentrations of various solutes and tracers. Recent studies suggested that using one probability distribution function (pdf) for SAS functions may not account for all transport processes in various hydrological systems. In this study we introduced a composite streamflow SAS function defined as a weighted sum of several pdfs (called components), using a uniform pdf for the youngest water and two gamma pdfs for the oldest water. The novel parameterization of this SAS function is nonlinear with respect to catchment storage and to a proxy of storage variations. This composite streamflow SAS function was needed to obtain realistic simulations of the complex high‐resolution (subdaily) stream deuterium (δ2H) dynamics measured for 2 years in the Weierbach, a forested headwater catchment in Luxembourg, whereas various SAS functions using only one component (i.e., pdf) failed. The three components of the composite streamflow SAS function confirmed the superposition of streamflow generation mechanisms with contrasting travel times suggested by previous experimental and modeling studies. Our work suggests that in some catchments, composite SAS functions may offer a more realistic perspective on transport processes derived from high‐resolution tracer data than simpler SAS functions.