Abstract : Geophysical Research Abstracts
Vol. 19, EGU2017-14279, 2017
EGU General Assembly 2017
© Author(s) 2017. CC Attribution 3.0 License.
Relative importance of time, land use and lithology on determining
aquifer-scale denitrification
Tamara Kolbe (1), Jean-Raynald de Dreuzy (1), Benjamin W. Abbott (2,3), Jean Marçais (4,1), Tristan Babey (1),
Zahra Thomas (5), Stefan Peiffer (6), Luc Aquilina (1), Thierry Labasque (1), Anniet Laverman (3), Jan
Fleckenstein (7,8), Philippe Boulvais (1), and Gilles Pinay (3)
(1) OSUR, CNRS, UMR 6118, Géosciences Rennes, Université de Rennes 1, Rennes, France , (2) Michigan State University,
Department of Earth and Environmental Sciences, East Lansing, USA, (3) ECOBIO, OSUR, CNRS, Université de Rennes 1,
Rennes, France, (4) Agroparistech, 16 rue Claude Bernard, Paris, France, (5) Agrocampus Ouest, Sol Agro et Hydrosystème
Spatialisation, Rennes, France, (6) Bayreuth Center of Ecology and Environmental Research, Germany, (7) UFZ-Helmholtz
Centre for Environmental Research, Leipzig, Germany, (8) Water and Earth System Science Competence Cluster (WESS),
Tübingen, Germany
Unconfined shallow aquifers are commonly contaminated by nitrate in agricultural regions, because of excess
fertilizer application over the last decades. Watershed studies have indicated that 1) changes in agricultural
practices have caused changes in nitrate input over time, 2) denitrification occurs in localized hotspots within the
aquifer, and 3) heterogeneous groundwater flow circulation has led to strong nitrate gradients in aquifers that are
not yet well understood.
In this study we investigated the respective influence of land use, groundwater transit time distribution, and
hotspot distribution on groundwater denitrification with a particular interest on how a detailed understanding
of transit time distributions could be used to upscale the point denitrification measurements to the aquifer-
scale. We measured CFC-based groundwater age, oxygen, nitrate, and dinitrogen gas excess in 16 agricultural
wells of an unconfined crystalline aquifer in Brittany, France. Groundwater age data was used to calibrate a
mechanistic groundwater flow model of the study site. Historical nitrate inputs were reconstructed by using
measured nitrate concentrations, dinitrogen gas excess and transit time distributions of the wells. Field data
showed large differences in denitrification activity among wells, strongly associated with differences in transit
time distribution. This suggests that knowing groundwater flow dynamics and consequent transit time distri-
butions at the catchment-scale could be used to estimate the overall denitrification capacity of agricultural aquifers.
