H24D-07
Lagrangian simulation of age and environmental tracer concentrations through integrated hydrologic systems

Tuesday, 15 December 2015: 17:30
3024 (Moscone West)
Nicholas B Engdahl, Washington State University, Pullman, WA, United States and Reed M Maxwell, Colorado School of Mines, Hydrologic Science and Engineering Program and Department of Geology and Geological Engineering, Golden, CO, United States
Abstract:
Hydrologic systems are complex, containing many sources, feedbacks, and interfaces, as well as heterogeneities and transience that increase the difficulty associated with simulating age or reconstructing tracer concentrations. Investigation of the system’s components in isolation ignores these feedbacks, which may have a profound impact on the simulated result. This presentation demonstrates how high-resolution, integrated hydrologic models can be combined with Lagrangian particle tracking to provide an unprecedented accounting of the particles’ history over time. The time spent in each process domain (groundwater, vadose zone, and surface water) is delineated, which permits the changes in tracer concentrations to be simulated differently in each domain. The example application uses geologically based characterizations (deterministic and geostatistical) and high performance computing to investigate changes in the distribution of age under different potential recharge scenarios for a high-alpine watershed in Colorado. The vadose zone played the most important role in this example and the saturated groundwater system exhibited minimal changes for the different recharge cases. Tracers that remain in equilibrium with the atmosphere throughout the vadose zone would not capture these effects, despite the large shifts in total residence time.

The approach is not limited to steady-state systems and is easily adapted for transient flow fields. The main limitations to these methods are the structural and parametric uncertainty of the flow and transport models, the heavy data requirements, and the difficulty in calibrating transient integrated models. However, the main strength of these methods is for making relative comparisons based on qualitatively validated flow models. In this context, the tools presented here provide a dynamic, virtual hypothesis-testing laboratory for investigating change in complex hydrologic systems.