Models to determine first-order rate coefficients from single-well push-pull tests

Area of Science:

• Environmental Science
• Geosciences
• Microbiology

Background:

• Push-pull tests (PPTs) are vital for quantifying microbially mediated subsurface processes.
• Existing models often assume complete, instantaneous mixing, treating aquifers as well-mixed reactors.
• This assumption may limit the accuracy of estimated rate coefficients (k).

Purpose of the Study:

• To develop and evaluate alternative models for estimating first-order rate coefficients (k) from PPTs.
• To compare the accuracy of well-mixed, plug-flow, and variably mixed reactor models.
• To assess model robustness in estimating subsurface microbial process rates.

Main Methods:

• Numerical simulations of PPTs were conducted.
• Sensitivity analysis was performed to compare model performance.
• Three models were applied: well-mixed, plug-flow, and variably mixed reactors.
• Published PPT data for nitrate consumption in a petroleum-contaminated aquifer was reanalyzed.

Main Results:

• All models provided reasonably accurate rate coefficient (k) estimates (errors < 13%).
• The variably mixed reactor model achieved the highest accuracy (errors < 1%).
• Well-mixed models overestimated k, while plug-flow models underestimated k, but both bracketed true values.
• Model estimates showed high consistency when applied to real-world data.

Conclusions:

• Alternative mixing models enhance the accuracy of PPT-derived rate coefficients (k).
• The variably mixed reactor model is most accurate but complex.
• The well-mixed and plug-flow reactor models offer a practical approach to bracket true k values.
• All tested models are robust for subsurface microbial process rate estimation from PPT data.