Remediation of subsurface systems, including groundwater, soil and soil gas, contaminated with light non-aqueous phase liquids (LNAPLs) is challenging. Field-scale pilot trials of multi-phase remediation were undertaken at a site to determine the effectiveness of recovery options. Sequential LNAPL skimming and vacuum-enhanced skimming, with and without water table drawdown were trialled over 78 days; in total extracting over 5 m³ of LNAPL. For the first time, a multi-component simulation framework (including the multi-phase multi-component code TMVOC-MP and processing codes) was developed and applied to simulate the broad range of multi-phase remediation and recovery methods used in the field trials. This framework was validated against the sequential pilot trials by comparing predicted and measured LNAPL mass removal rates and compositional changes. The framework was tested on both a Cray supercomputer and a cluster. Simulations mimicked trends in LNAPL recovery rates (from 0.14 to 3 mL/s) across all remediation techniques each operating over periods of 4–14 days over the 78 day trial. The code also approximated order of magnitude compositional changes of hazardous chemical concentrations in extracted gas during vacuum-enhanced recovery. The verified framework enables longer term prediction of the effectiveness of remediation approaches allowing better determination of remediation endpoints and long-term risks.

中文翻译：

---

现场规模的多阶段LNAPL补救措施：针对连续的现场试验性试验验证了新的计算框架

修复被轻质非水相液体（LNAPL）污染的地下水系统（包括地下水，土壤和土壤气体）具有挑战性。在现场进行了多阶段修复的现场规模的试验性试验，以确定回收方案的有效性。在连续的LNAPL撇除和真空增强撇除（不论是否有地下水位下降）中进行了78天的试验；总开采量超过5 m ³LNAPL。首次开发了多组分模拟框架（包括多相多组分代码TMVOC-MP和处理代码），并将其应用于模拟在现场试验中使用的多种多相修复和回收方法。通过比较预测和测得的LNAPL物质去除率和成分变化，针对连续的先导试验验证了该框架的有效性。该框架已在Cray超级计算机和群集上进行了测试。模拟模仿了LNAPL回收率的趋势（从0.14到3  mL / s）的所有补救技术，在78天的试用期内，每种修复技术都需要4到14天的时间进行操作。该法规还估算了在真空增强回收过程中提取的气体中有害化学浓度的成分变化的数量级。经过验证的框架可以对补救方法的有效性进行长期预测，从而可以更好地确定补救终点和长期风险。