Login / Signup

Interpreting Concentrations Sampled in Long-Screened Wells with Borehole Flow: An Inverse Modeling Approach.

Frederick D Day-LewisRob D MackleyJoshua Thompson
Published in: Ground water (2023)
New approaches are needed to assess contaminant mass based on samples from long-screened wells and open boreholes (LSW&OB). The interpretation of concentration samples collected in LSW&OB is complicated in the presence of vertical flow within the well. In the absence of pumping (i.e., ambient conditions), the well provides a conduit for flow to occur between aquifer layers or fractures as a result of head differences. Under pumping conditions, vertical borehole flow may vary with depth depending on far-field heads and hydraulic conductivity; furthermore, if pumping fails to overcome ambient gradients, outflow from the well to the aquifer may occur. Concentration samples thus represent flow-weighted averages of formation concentrations, but the averaging process is commonly unknown or difficult to identify. Recognition of the importance of borehole flow has motivated the use of multi-level wells, packers, and well liners; however, LSW&OB remain common for numerous reasons, including cost, multi-purpose design requirements (e.g., pump-and-treat, water supply), logging, and installation of instrumentation. Here, we present a simple analytical model for flow and transport within a well and interaction with the surrounding aquifer. We formulate an inverse problem to estimate formation concentration based on sampled concentrations and data from flowmeter logs. The approach is demonstrated using synthetic examples. Our results (1) underscore the importance of interpreting sampled concentrations within the context of hydraulic conditions and aquifer/well exchange; (2) demonstrate the value of flowmeter measurements for this purpose; and (3) point to the potential of the new inverse approach to better interpret results from samples collected in LSW&OB.
Keyphrases
  • air pollution
  • particulate matter
  • magnetic resonance imaging
  • computed tomography
  • machine learning
  • electronic health record
  • big data
  • climate change
  • optical coherence tomography
  • contrast enhanced