Exploitation of pump-and-treat remediation systems for characterization of hydraulic heterogeneity

JOURNAL OF HYDROLOGY

卷:573

Article

Zha, Yuanyuan¹  Yeh, Tian-Chyi J.^2,3  Illman, Walter A.⁴  Mok, Chin Man W.⁵  Tso, Chak-Hau M.⁶  Carrera, Barbara A.⁵  Wang, Yu-Li³ 

[1] Wuhan Univ, State Key Lab Water Resources & Hydropower Engn S, Wuhan, Hubei, Peoples R China

[2] Tianjin Normal Univ, Tianjin Key Lab Water Resources & Environm, Tianjin, Peoples R China

[3] Univ Arizona, Dept Hydrol & Atmospher Sci, Tucson, AZ USA

[4] Univ Waterloo, Dept Earth & Environm Sci, Waterloo, ON, Canada

[5] GSI Environm Inc, Oakland, CA USA

[6] Univ Lancaster, Lancaster Environm Ctr, Lancaster, England

关键词: Hydraulic tomography;    Inverse modeling;    Heterogeneity;    Pump and treat operation;    Geological borehole log;   

DOI  :  10.1016/j.jhydrol.2019.03.089

来源: Elsevier

【 摘 要 】

Pump-and-treat (P&T) is a widely adopted technology for groundwater containment/remediation. Over the past decades, we have recognized that mapping aquifer heterogeneity, especially the low hydraulic conductivity (K) zones, has become an urgent need in remediation efforts. Meanwhile, many studies have proved that hydraulic tomography (FIT) utilizing groundwater responses from a series of cross-hole pumping tests is a viable technology for detailing hydraulic heterogeneity (high and low K zones) of aquifers. At a P&T field site, operation of pumping and injection wells often vary, creating changes in flow fields analogous to those in FIT surveys. Therefore, collecting groundwater monitoring data and analyzing them is tantamount to a cost-effective largescale FIT survey during P&T operations. To demonstrate this idea, water level data was collected at 41 wells for one year at the U.S. Air Force Plant 44 (AFP44) P&T site in Tucson, Arizona, where the pumping and injection rates of the P&T system were changed many times during the year. This dataset was analyzed with a geostatistical inverse model to estimate the hydraulic property distribution, which can reasonably predict the pressure responses of the site under different conditions. Moreover, we showed that large-scale geological information, built based on 245 borehole logs at the site, could improve the model calibration performance through a two-step calibration procedure. Additionally, we used synthetic experiments to verify that the use of a larger variety of P&T configurations can yield non-redundant FIT information, and enhance site characterization. Overall, we demonstrated that the proper scheduling of pumping and injection operations at a P&T site and collecting corresponding groundwater responses are equivalent to conducting a large-scale HT survey, with minimal additional costs.

【 授权许可】

Free