Imbalances of Water and Solutes in Experimental Watersheds: Spatial or Temporal Origin?
Abstract
Experimental watersheds where originally conceived as a tool to measure water balance in different landscapes and climates and in particular evapotranspiration fluxes. Pioneering experimentalists where paying attention to all possible causes of unmeasured losses, for example by ensuring that the watershed outlet was lying on impervious bedrock and that the hydrological boundary was consistent with the topographic divide. Nowadays, watershed studies encompass a large range of objectives, from hydrological process quantification, to diffuse pollution assessment. In many cases, the above-mentioned experimental precautions are not much considered and the closure of the water balance is rarely achieved, although this fact is even more rarely publicized in the scientific communications. As a consequence, it is very often very difficult to determine whether an observed difference between input and output of water or solutes is due to some hidden deep losses or to variation of storage in internal compartments of the watershed. In this presentation, we will discuss this issue based on long term hydrological and geochemical monitoring of experimental watersheds belonging to the research observatories BVET (http://bvet.omp.obs-mip.fr/ ) in South India and AgrHys (http://www.inra.fr/ore_agrhys/) in Western France. In the South Indian forested watershed of Mule Hole (10 years of monitoring) we demonstrated that transpiration by deep tree roots was a significant component of the water balance, and that the main pathway for hydrological and geochemical fluxes was groundwater underflow. In the French agricultural watersheds of Kerbernez and Kerrien (20 years of monitoring) significant water and solute losses through groundwater underflow was also demonstrated. In both sites, a model with flexible structure was calibrated and validated on the observation and then long-term simulations were produced using available long term weather data series of 50 years. Results demonstrated that the water or solute imbalances at the watershed scale depended, to a large extent, on the duration considered in the analysis. In the Mule Hole watershed, water storage in the unsaturated weathered bedrock was the major cause of water imbalance for short time series (less than 10 years) while deep loses were the only source of imbalance for long term analysis (more than 30 years). On the contrary, in the Kerrien and Kerbernez watersheds, solute imbalance were mainly attributed to underflow for short term analysis (less than 10 years) while variation of solute storage in the weathered bedrock became a major source of imbalance for long term analysis (more than 20 years). Discussion will focus on the consequences of these results on the validity of the hypotheses used in hydrological and hydrochemical modeling studies, and on the interest of long term environmental observatories for understanding water and element cycles.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2012
- Bibcode:
- 2012AGUFM.H11G1270R
- Keywords:
-
- 1030 GEOCHEMISTRY / Geochemical cycles;
- 1836 HYDROLOGY / Hydrological cycles and budgets;
- 1879 HYDROLOGY / Watershed