The beginning of modern in situ subsurface cleanup arguably began in the early 1970s with the application of hydrogen peroxide to shallow aquifers to drive gasoline spill remediation, in what became known as the Raymond method (Committee on In Situ Bioremediation 1993). Since then, this idea has sparked remarkable remediation innovations where processes that occur naturally in the subsurface, including biodegradation, sorption, diffusion, abiotic reactions, and volatilization are enhanced by actions specifically designed to boost their role in lowering contaminant concentrations in groundwater.

By the 1990s, the age of “Natural Attenuation” (NA) was upon us, together with vigorous ethical and technical discussions aimed at understanding and explaining why the technology was not simply a euphemism for “do nothing.” Natural attenuation proponents held that natural processes tend to be protective of human and environmental health, are more cost- and resource effective than active remediation, and less risky by minimizing the aggressive movement of fluids in the subsurface. On the other hand, critics contended that NA amounted to a delay tactic aimed primarily at cost savings, with little or no risk reduction over reasonable time periods. Critically, since the end of the twentieth century, advancements in site characterization and monitoring tools have refined conceptual site models and established multiple lines of evidence that NA is associated with (1) well-constrained plume development and (2) diminished risk over space and time to an acceptable level without active remediation.

As NA gained acceptance, its adoption at sites became primarily dependent on evidence of biologically driven contaminant transformation as the primary attenuation mechanism. In the process, other important and effective processes, including the collective physical, geochemical, and hydrogeologic influences that can govern or complement contaminant mitigation were often overshadowed or completely overlooked. We know that chemical and physical mechanisms can play primary roles in reducing contaminant concentrations. This includes reactions with iron or manganese oxides, hydrolysis, sorption, diffusion, and dispersion, volatilization, and other mechanisms that lower a specified mass of contaminant, or contaminant mixture, in the subsurface. These processes can be influenced by, and feedback to, the biological activity in aquifers; the interconnectedness of biology, chemistry, and flow are better appreciated now than ever in the past. It is therefore intellectually, technologically, and economically wasteful to disregard any part of this compilation of processes when dealing with aquifer restoration. To improve site management and closure efforts, we need process-based conceptual site models that account for all the processes occurring in a groundwater system. Herein, we contend this can be realized by extending the current NA framework and giving proper recognition to a well-known but under-utilized concept: assimilative capacity (AC).

The idea of AC is decades old (e.g., Falk, 1963). Depending on the literature consulted, the usage of the term varies somewhat, but generally alludes to the ability (by any mechanism) of an environmental system to transform or sequester various materials without deleterious effects to the system itself. Most often, AC has been associated with surface water resources (e.g., rivers and wetlands) and atmospheric studies (e.g., greenhouse gases). AC has also been integral in our approach to disposal of wastes (e.g., wastewater treatment effluents, septic systems and land application of biosolids). About 25 years ago the term was discussed in the context of sustainability and the need to avoid exceeding the AC of natural systems. By doing so, our growing and technologically advanced society could coexist with ecosystem services indefinitely (Cairns Jr 1999).

In the specific context of groundwater systems literature, AC has been used since at least the 1970s (e.g., Willis 1976), in arguably vague terms, with little hydrogeological context. Furthermore, following the rationale for NA, these discussions of AC appear to have emphasized intrinsic processes within a flow system rather than specific quantities, as suggested by the term “capacity.” This tendency has blurred the lines between NA and AC. Going forward, it may be useful to make a clear distinction between AC and NA by defining AC as the mass of contaminant(s) that can be transformed or sequestered by a volume of subsurface over a specified period of time, via the chemical and physical mechanisms mentioned previously. This definition frees hydrogeologists and engineers to design aquifer restoration programs and risk reduction projects that take full advantage of all nature's mechanisms for improving groundwater quality. Moreover, this definition applies to all hydrogeologic settings; it is relevant to cases involving small receptors and large plumes in quasi-steady state with no apparent migration. It is relevant to aquifers and aquitards. In the latter case, the concept of AC may be especially relevant because aquitards may be diffusion-controlled environments where time scales are extended and reactivity enhanced. High-resolution aquitard characterization techniques can provide accurate assessment of ‘assimilation’ rates simply through the analysis of concentration profiles (diffusion halos).

As alluded to previously, a major issue facing the application of the AC concept, extending the scope of NA, lies in our ability to characterize the subsurface for its capacity to “assimilate.” In diffusion-controlled settings, concentration profiles may be sufficient to accomplish this task. But, in many cases, even open and dynamic aquifers can be characterized for AC with confidence, using high-definition characterization tools and in situ flux characterization technologies. With these recent developments in tools, and a growing appreciation of conceptual site model development, we should now be able to shed light on the complete nature of the complex, and interwoven processes affecting contaminant transport and fate over space and time. For example, the familiar transect approach for assessing contaminant losses from groundwater within an aquifer volume is easily adapted using the new tools. In principle, measurement of contaminant discharges (or fluxes) at two bounding transects, upgradient and downgradient of the test volume, provide the data needed for the AC of that volume to be estimated in situ.

Another potentially fruitful application of AC is in documenting contaminant behavior at sites that have undergone NA or active remediation and are now exhibiting low but persistent concentrations of contaminants. Back-diffusion from low-permeability strata is thought to be a common source of such long-term dilute plumes. NA (as biodegradation) may be difficult to prove in such settings, since the aquifers may have returned to an oligotrophic state in which ambient microbial activity is no longer sufficient to drive contaminant concentrations to regulatory limits. A resulting lack of understanding and confidence that risk is under control will likely delay site closure, sometimes unnecessarily. On the other hand, other processes included in the AC concept may act to fill the gap, and be protective of aquifers to concentrations below safe thresholds. This could open the future to specific benefits including improved risk assessment, enhanced future land use, superior site-specific remediation standards, refined monitoring strategies, and successful technology adoption.

In summary, AC may be the concept of greatest relevance in thoughtful site closure decision making. AC is central to extending natural attenuation processes beyond those that have come to dominate NA thinking and be more versatile for distinct hydrogeologic and contaminant types. The recent high-definition characterization technologies and increasing emphasis on process-based site conceptual models are giving site management strategies more potency than ever before. The time has come for our body of groundwater research to seek to standardize the quantification of AC through the use of process-based site conceptual models developed using modern characterization and monitoring tools. As such, regulatory frameworks can be shaped by a more informed understanding of hydrogeologic principles to safeguard human health and the environment.

现代原位地下清理的开始可以说始于 20 世纪 70 年代初，当时将过氧化氢应用于浅层含水层以驱动汽油泄漏修复，即后来被称为雷蒙德法（原位生物修复委员会 1993）。从那时起，这个想法引发了显着的修复创新，通过专门设计的行动来增强地下自然发生的过程，包括生物降解、吸附、扩散、非生物反应和挥发，以增强其在降低地下水污染物浓度方面的作用。

到了 20 世纪 90 年代，“自然衰减”(NA) 时代已经来临，伴随着激烈的伦理和技术讨论，旨在理解和解释为什么该技术不仅仅是“什么都不做”的委婉说法。自然衰减的支持者认为，自然过程往往能够保护人类和环境健康，比主动修复更具成本和资源效率，并且通过最大限度地减少地下流体的侵蚀性运动来降低风险。另一方面，批评者认为，NA 相当于一种拖延策略，主要目的是节省成本，在合理的时间内很少或根本没有降低风险。重要的是，自 20 世纪末以来，场地特征和监测工具的进步已经完善了场地概念模型，并建立了多种证据表明 NA 与 (1) 良好约束的羽流发展和 (2) 空间风险降低和时间达到可接受的水平而无需主动补救。

随着 NA 获得认可，其在现场的采用主要依赖于生物驱动的污染物转化作为主要衰减机制的证据。在此过程中，其他重要且有效的过程，包括可以控制或补充污染物减排的集体物理、地球化学和水文地质影响，常常被掩盖或完全忽视。我们知道化学和物理机制在降低污染物浓度方面可以发挥主要作用。这包括与铁或锰氧化物的反应、水解、吸附、扩散和分散、挥发以及降低地下污染物或污染物混合物特定质量的其他机制。这些过程可能受到含水层生物活动的影响和反馈；生物学、化学和流动之间的相互联系现在比过去任何时候都得到了更好的认识。因此，在处理含水层恢复时忽视该过程汇编的任何部分在智力、技术和经济上都是浪费。为了改进场地管理和关闭工作，我们需要基于过程的概念场地模型，该模型能够解释地下水系统中发生的所有过程。在此，我们认为这可以通过扩展当前的 NA 框架并适当认识一个众所周知但未充分利用的概念：同化能力（AC）来实现。

AC 的想法已有数十年历史（例如，Falk，1963）。根据查阅的文献，该术语的用法有所不同，但通常指的是环境系统（通过任何机制）转化或隔离各种材料而不会对系统本身产生有害影响的能力。大多数情况下，交流与地表水资源（例如河流和湿地）和大气研究（例如温室气体）相关。 AC 也是我们废物处理方法中不可或缺的一部分（例如废水处理废水、化粪池系统和生物固体的土地应用）。大约 25 年前，该术语在可持续性和避免超过自然系统 AC 的需要的背景下进行了讨论。通过这样做，我们不断发展和技术先进的社会可以与生态系统服务无限期地共存（Cairns Jr  1999）。

在地下水系统文献的特定背景下，AC 至少从 20 世纪 70 年代就开始使用（例如，Willis  1976），其术语可以说是模糊的，几乎没有水文地质背景。此外，根据 NA 的基本原理，这些关于 AC 的讨论似乎强调了流动系统内的内在过程，而不是“容量”一词所暗示的具体数量。这种趋势模糊了 NA 和 AC 之间的界限。展望未来，通过将 AC 定义为可在指定时间内被地下体积通过化学和物理作用转化或隔离的污染物质量，明确区分 AC 和 NA 可能会有所帮助。前面提到的机制。这一定义使水文地质学家和工程师能够自由地设计含水层恢复计划和风险降低项目，充分利用所有自然机制来改善地下水质量。此外，该定义适用于所有水文地质环境；它与涉及小受体和处于准稳态且没有明显迁移的大羽流的情况相关。它与含水层和弱透水层有关。在后一种情况下，AC的概念可能特别相关，因为弱透水层可能是扩散控制的环境，其中时间尺度延长且反应性增强。高分辨率弱水层表征技术只需通过分析浓度分布（扩散晕）即可准确评估“同化”速率。

正如前面提到的，应用 AC 概念、扩展 NA 范围所面临的一个主要问题在于我们描述地下“同化”能力的能力。在扩散控制的设置中，浓度分布可能足以完成此任务。但是，在许多情况下，即使是开放的动态含水层也可以使用高清表征工具和原位通量表征技术充满信心地进行交流表征。随着工具的最新发展，以及对概念场地模型开发的日益认识，我们现在应该能够阐明影响污染物在空间和时间上的迁移和命运的复杂的、交织的过程的完整性质。例如，使用新工具可以轻松调整用于评估含水层内地下水污染物损失的熟悉的断面方法。原则上，在两个边界横断面（测试体积的上坡和下坡）处测量污染物排放（或通量），可以提供现场估计该体积的 AC 所需的数据。

AC 的另一个潜在富有成果的应用是记录经过 NA 或主动修复且现在污染物浓度较低但持续存在的场地的污染物行为。低渗透地层的反向扩散被认为是这种长期稀羽流的常见来源。在这种情况下，NA（生物降解）可能很难证明，因为含水层可能已经恢复到贫营养状态，其中环境微生物活动不再足以将污染物浓度推至监管限值。由此导致的对风险可控的缺乏理解和信心可能会延迟站点关闭，有时甚至是不必要的。另一方面，AC概念中包含的其他过程可能会填补空白，并保护含水层的浓度低于安全阈值。这可以为未来带来具体的好处，包括改进的风险评估、增强未来的土地利用、优越的特定地点修复标准、完善的监测策略以及成功的技术采用。

总之，AC 可能是在深思熟虑的站点关闭决策中最相关的概念。 AC 是将自然衰减过程扩展到那些已经主导 NA 思维的过程的核心，并且对于不同的水文地质和污染物类型具有更广泛的用途。最近的高清表征技术和对基于流程的场地概念模型的日益重视，使场地管理策略比以往任何时候都更加有效。现在是我们地下水研究机构通过使用现代表征和监测工具开发的基于过程的场地概念模型来寻求标准化 AC 量化的时候了。因此，可以通过对水文地质原理的更深入了解来制定监管框架，以保护人类健康和环境。