Hydrological connectivity drives dissolved organic matter processing in an intermittent stream

Hydrological conditions are key drivers of dissolved organic matter (DOM) processing in intermittent streams. However, there still exist major gaps in knowledge regarding the temporal dynamics of DOM processing during drought periods, as well as the role of the hyporheic zone (HZ). We conducted weekly sampling of surface water and hyporheic pore water during a drying/rewetting cycle and characterized DOM by fluorescence and absorbance properties. Overall, the contribution of allochthonous and humic-like DOM increased during base flow in early summer (pre-drought) and continued increasing throughout the drought period, which covered three phases: contraction, fragmentation and dry. The contribution of autochthonous DOM during this period was restricted to very specific points in time (the transition from contraction to fragmentation phase) and space (the HZ). Hydrological connectivity between the HZ and the surface water was a driver of DOM composition by supplying terrestrial, aromatic DOM to the HZ. The disconnection of the stream from the riparian groundwater enabled us to quantify the DOM retention/release in the HZ. DOM mass balance at the stream-hyporheic interface revealed the occurrence of two time periods with disproportionately high rates for DOM processing (hot moments) during the study period: (1) a short pulse of protein-like, autochthonous DOM net release at the beginning of the disconnection; and (2) a longer time period of increasing net dissolved organic carbon (DOC) retention up to 30% along 25 m of HZ length during fragmentation and dry phase. Remarkably, the net carbon retention was coupled to a decrease of aromatic and high molecular weight compounds, while protein-like, autochthonous DOM was released. This result evidenced that under drought conditions, the HZ becomes a sink for DOM compounds previously assumed to be recalcitrant in aquatic ecosystems and therefore highlights the importance of hydrological drivers on DOM processing. (C) 2017 The Authors. Published by Elsevier GmbH.