Defining Remediation Extents: Why an Exceedance Does Not Automatically Define an Excavation
A laboratory result exceeds an investigation criterion.
How much soil needs to be excavated?
It sounds like a straightforward question, but the analytical result alone cannot answer it.
One of the more important distinctions in contaminated land assessment is the difference between identifying contamination that warrants further consideration and defining the extent of soil that actually requires remediation.
An exceedance may identify a potential hotspot, indicate that an exposure pathway requires further assessment, or demonstrate that contamination has not yet been adequately characterised.
What it does not automatically provide is an excavation boundary.
Defining an appropriate remediation extent requires consideration of the contaminant distribution, the Conceptual Site Model, proposed land use, source characteristics, exposure pathways, investigation confidence, remediation objectives and the practical requirements of the development.
This distinction matters because poorly defined remediation extents can lead to two very different problems:
Under-remediation, where contamination that should have been addressed remains inadequately characterised or managed.
Or:
Over-remediation, where significantly more soil is excavated and disposed of than is necessary to achieve the remediation objectives.
A well-designed Remediation Action Plan should avoid both.
Investigation Criteria Are Not Automatically Remediation Criteria
This is the starting point.
The National Environment Protection (Assessment of Site Contamination) Measure — the ASC NEPM — is explicit that investigation and screening levels are not clean-up or response levels.
They are screening tools used to determine whether contamination warrants further consideration.
The NEPM also warns that using investigation and screening levels as default remediation criteria can result in unnecessary remediation, increased development costs, unnecessary environmental disturbance and unnecessary consumption of landfill capacity.
That distinction is fundamental.
If a soil sample reports a contaminant concentration above a Health Investigation Level, for example, the correct conclusion is not automatically:
Everything represented by this sample needs to be excavated.
The result instead raises further questions.
What does the result represent?
Is it an isolated hotspot or part of a broader contaminant distribution?
What is the likely source?
At what depth does contamination occur?
Does it extend laterally or vertically?
Is the relevant exposure pathway complete?
What will happen to this area under the proposed development?
Is excavation actually required to manage the risk?
These questions move the assessment from screening into interpretation and remediation design.
An Analytical Result Is a Data Point, Not a Boundary
A soil sample represents material collected from a particular location and depth.
It does not, by itself, tell us exactly where the contamination begins and ends.
Consider a sample collected from shallow fill at one investigation location that reports lead above the relevant assessment criterion.
The contamination might extend:
only a short distance around the sample;
throughout the surrounding fill layer;
along one side of a former structure;
vertically through several fill horizons;
across a broader portion of the site; or
irregularly through material that has historically been moved and redistributed.
Drawing a circle around that sampling location and calling it the remediation extent would therefore be arbitrary unless there is evidence supporting that geometry.
This is why a Detailed Site Investigation is intended to establish the nature, extent and significance of contamination rather than simply identify whether one or more concentrations exceed screening criteria.
The NSW EPA's current sampling design guidance similarly emphasises that contaminated land assessment is risk-based and should use a weight-of-evidence approach, with sampling designed around the Conceptual Site Model and identified data gaps.
Start With the Conceptual Site Model
Before deciding where additional samples should be collected or where remediation should stop, the consultant should return to the Conceptual Site Model (CSM).
The CSM provides the framework for interpreting the analytical result.
The key questions are:
What is the source?
How could the contaminant have been distributed?
Where could it have migrated?
What media could be affected?
Who or what could be exposed?
How will the proposed development alter those pathways?
A concentration in shallow imported fill is a different contamination problem to a concentration associated with a leaking underground petroleum storage system.
The first may be associated with heterogeneous material physically distributed across an area.
The second may involve vertical migration, groundwater transport, preferential pathways and potentially vapour.
The same numerical exceedance could therefore require completely different investigation and remediation strategies.
This is why defining a remediation extent is not simply a mathematical exercise based on the distance between an exceeding sample and a non-exceeding sample.
The spatial interpretation needs to make sense in the context of the source and contaminant behaviour.
Delineation Should Answer a Specific Question
Where an exceedance represents contamination that may require remediation, additional sampling is often undertaken to establish its lateral and vertical extent.
This is commonly referred to as delineation.
But delineation should not mean collecting samples around an exceedance until enough non-exceeding results are obtained to draw a convenient polygon.
Each additional sampling location should answer a specific question.
For example:
Does contamination extend north toward the property boundary?
Is the impact confined to the shallow fill layer?
Does contamination extend beneath the underlying natural soil?
Is the result associated with a former structure or a broader fill unit?
Does the contamination continue beneath an existing building?
Is groundwater potentially affected?
Is there evidence of a separate hotspot nearby?
This approach produces a more defensible understanding of contaminant distribution than simply applying a fixed offset distance around every exceedance.
The NSW EPA's 2022 sampling guidance specifically requires the preliminary CSM to inform sampling design and recognises that areas of concern should be separately considered using appropriate targeted or stratified sampling approaches. It also notes that further sampling may be required for delineation or validation.
Lateral and Vertical Delineation Both Matter
Contamination has three dimensions.
It is easy to concentrate on the plan view — where contamination extends across the site — while giving less attention to depth.
But vertical delineation can materially change the volume of soil requiring remediation.
Consider a 20 m by 20 m area of impacted fill.
If contamination extends only through the upper 0.2 m, the theoretical affected volume is approximately 80 m³.
If it extends to 1.5 m below ground level, the volume increases to approximately 600 m³.
That difference can substantially affect excavation, transport, disposal, backfilling and overall project cost.
Vertical distribution can also tell us something about the source.
A contaminant confined to the upper soil horizon may reflect historical surface application or shallow filling.
A hydrocarbon impact extending progressively deeper toward groundwater may indicate migration from a subsurface source.
An isolated deeper concentration beneath otherwise clean shallow soil might suggest buried material or a historical feature that has not yet been identified.
The investigation therefore needs to characterise contamination in three dimensions sufficiently to support the intended remediation decision.
A Non-Exceeding Result Does Not Always Mean the Boundary Has Been Defined
The opposite problem can also occur.
A consultant obtains one non-exceeding result beside an exceeding result and assumes that contamination has been delineated.
Sometimes that may be reasonable.
Sometimes it is not.
The significance of the non-exceeding result depends on factors including:
the distance between samples;
the likely size of the hotspot;
source geometry;
soil heterogeneity;
contaminant mobility;
depth;
the sampling method;
site history; and
the degree of confidence required for the decision being made.
If two samples are 20 metres apart, an acceptable result at one location does not demonstrate that every metre of soil between the two points is acceptable.
There is always uncertainty between investigation locations.
The role of sampling design is to reduce that uncertainty to a level appropriate for the decision being made.
The NSW EPA Sampling Design Guidelines specifically address hotspot detection and recommend sampling densities based on the size of hotspot that the investigation needs to be capable of detecting. The guidelines make clear that the recommended numbers are minimums and that additional targeted sampling may be required where areas of concern are present.
The appropriate question is therefore not:
“Have we found a clean sample?”
It is:
“Do we now have enough evidence to define the contaminant distribution with sufficient confidence for the decision we need to make?”
Example: A Localised Lead Exceedance in Fill
Consider a commercial development where a DSI identifies elevated lead in shallow fill at one sampling location.
The surrounding investigation locations report substantially lower concentrations.
Does the elevated result automatically mean an excavation should be centred on the sample?
Not necessarily.
The consultant should first consider the source.
Suppose the exceedance occurs adjacent to the former wall of an old workshop where lead-based paint deterioration is suspected.
That provides a plausible mechanism for a localised impact.
Targeted sampling along the former building perimeter and at increasing distances from the affected location may then demonstrate that the contamination is confined to a relatively narrow drip-line area.
Vertical samples might also show that the impact is restricted to the upper 100–200 mm of soil.
That information can support a relatively specific remediation extent.
Now change the scenario.
The same lead result is found in heterogeneous imported fill containing ash, slag, demolition material and other anthropogenic inclusions, and similar fill extends across much of the site.
The source model is completely different.
Simply offsetting four samples around the original exceedance may not adequately characterise the issue because the contamination could be irregularly distributed throughout the fill.
The same laboratory result can therefore lead to a very different investigation strategy.
Example: Hydrocarbon Contamination Around Former Fuel Infrastructure
Petroleum contamination is another useful example of why fixed-radius excavation boundaries can be misleading.
Suppose elevated petroleum hydrocarbons are identified next to a former underground storage tank.
A simple circular excavation around the sample might appear conservative, but it may bear little resemblance to the actual contaminant distribution.
Hydrocarbons may migrate:
downward through permeable soil;
along bedding materials surrounding fuel lines;
through preferential pathways;
toward groundwater;
in the direction of groundwater flow; or
laterally from a leaking tank or pipe connection.
Investigation should therefore consider the infrastructure layout, soil and geology, groundwater conditions, contaminant composition and likely source history.
The eventual site remediation area may be irregular because the contamination itself is irregular.
This becomes particularly important where groundwater or soil vapour may also be affected.
Soil excavation alone may not address the complete source-pathway-receptor relationship.
Hotspots and Widespread Contamination Need Different Thinking
Not all contamination distributions should be treated the same way.
A discrete hotspot may warrant targeted delineation around a specific source.
Widespread low-level contamination may require a broader assessment of the affected soil unit and its overall significance.
This distinction can have major implications for contaminated soil remediation.
For example, imagine shallow fill across a large industrial property containing relatively consistent metal concentrations.
Trying to excavate small areas around every analytical exceedance may produce a fragmented and practically meaningless remediation strategy.
It may be more appropriate to assess the fill as a broader material unit, understand its statistical and spatial characteristics, evaluate relevant exposure pathways and develop an appropriate site-wide management strategy.
Conversely, averaging a highly contaminated hotspot into a large dataset to reduce the overall mean would not necessarily provide an appropriate assessment either.
The contaminant distribution and CSM need to determine how the data are grouped and interpreted.
NSW EPA sampling guidance recognises both systematic sampling and targeted investigation of specific areas of concern, while requiring sampling design to be appropriate to the contamination model and decision being made.
Remediation Extent Is Not Always the Same as Contamination Extent
This distinction is particularly important.
The contamination extent describes where contamination is present.
The remediation extent describes where physical remediation or another management action is required.
Those boundaries may be related, but they are not necessarily identical.
Consider contaminated fill extending beneath both a proposed building and a landscaped area.
The investigation may establish that essentially the same fill unit occurs across both areas.
However, the exposure conditions under the finished development may be different.
Subject to the risk assessment and remediation objectives, material beneath an engineered building slab may potentially be capable of remaining in place, while accessible contaminated soil within landscaped areas requires removal, clean cover or another form of management.
Alternatively, the development excavation may require all of the fill beneath the building footprint to be removed regardless of its contamination status.
The practical excavation extent may therefore be larger than the contamination extent in one area and smaller in another.
This is why remediation planning should be integrated with the proposed development.
For further discussion of this issue, see Designing Soil Remediation Around the Proposed Development — Confluence Environmental.
The Remediation Objective Should Drive the Boundary
A defensible remediation extent starts with a clear objective.
For example:
remove the identified localised source;
reduce contaminant concentrations to established site-specific remediation criteria;
remove contaminated fill from future accessible-soil areas;
eliminate a vapour source beneath a proposed building;
prevent direct contact with residual contamination;
remove friable asbestos-impacted material;
contain affected fill beneath an engineered barrier; or
achieve a defined land-use suitability outcome.
Once the objective is clear, the consultant can determine what physical works are actually necessary to achieve it.
That is fundamentally different from beginning with:
Sample TP07 exceeded, therefore excavate around TP07.
The NSW EPA reporting guidelines require RAPs to establish remediation objectives, remediation criteria, the selected remediation strategy and the validation requirements that will demonstrate successful implementation.
The remediation boundary should therefore follow the remediation objective — not the other way around.
Remediation Criteria Need to Be Established Before Excavation Starts
Where excavation is proposed, the RAP should clearly define the criteria that determine when the remediation objective has been achieved.
Depending on the project, this may involve:
numerical validation criteria;
removal of a defined soil or fill unit;
removal of visually identifiable material;
removal of a specific source;
excavation to a defined physical boundary;
construction of an engineered containment system;
treatment performance criteria; or
a combination of measures.
The appropriate endpoint depends on the remediation strategy.
This is another reason that simply excavating until field screening or laboratory results “look clean” is generally poor remediation planning.
The endpoint should be established before remediation starts wherever practicable.
Otherwise, the project can become an open-ended process in which excavation continues without a clearly defined technical basis.
Practical Excavation Boundaries May Extend Beyond the Analytical Boundary
An important counterpoint is that remediation does not always follow contamination boundaries with centimetre-level precision.
Excavation is a physical construction activity.
Practical considerations can include:
safe batter slopes;
excavation stability;
access for plant;
existing foundations;
underground services;
property boundaries;
material segregation;
dewatering;
stockpile management; and
the dimensions of excavation equipment.
A remediation extent may therefore deliberately extend beyond the interpreted contamination boundary because doing so provides a safer or more constructable outcome.
For example, removing an additional metre of otherwise acceptable soil may be more practical than attempting to leave a narrow strip of material between two remediation areas.
That does not mean the additional soil should automatically be classified or treated as contaminated.
It means the physical excavation footprint and the environmental contamination footprint should be understood as related but distinct concepts.
This distinction becomes particularly important for waste management.
Material leaving the site must be appropriately assessed and classified rather than being assumed to have the same classification merely because it originated within a remediation excavation.
Confluence's waste classification services can form part of this broader remediation and material-management process where excavated soil requires off-site disposal.
Validation Is the Final Test of the Remediation Extent
The investigation defines where remediation is expected to be required.
The RAP establishes how the work will be undertaken.
Validation then demonstrates whether the remediation objective has actually been achieved.
For an excavation, this may involve assessment of the base and walls after contaminated material has been removed.
If validation identifies residual contamination above the established acceptance criteria, further excavation or another management response may be required.
If the results satisfy the validation requirements, the remediation extent can be considered complete for that area.
The NSW EPA requires remedial works to be validated against the objectives established in the RAP and states that the extent of validation depends on factors including the original contamination, remediation process and proposed or current land use.
This is why validation should be designed alongside the remediation strategy rather than after earthworks are finished.
Learn more about Confluence Environmental's remediation and validation services.
What If the Remediation Extent Changes During the Works?
This is common.
Contaminated land investigations are based on discrete observations and samples. Earthworks expose substantially more soil than an investigation ever can.
During remediation works, the consultant may encounter:
unexpected staining;
odours;
additional fill;
buried waste;
asbestos-containing material;
tanks or structures;
contamination extending beyond the anticipated boundary; or
different subsurface conditions to those inferred during investigation.
A good RAP should anticipate this possibility through contingency and unexpected-find procedures.
If field observations or validation results demonstrate that contamination extends beyond the anticipated remediation boundary, the CSM should be reconsidered and the remediation extent adjusted where required.
That is not necessarily evidence that the original investigation failed.
It reflects the inherent uncertainty associated with characterising subsurface conditions from discrete investigation points.
The important issue is whether the remediation framework allows new information to be assessed systematically and incorporated into the works.
A Defensible Remediation Extent Is an Interpretation of Multiple Lines of Evidence
Ultimately, the boundary shown on a remediation plan should represent more than a line drawn around laboratory exceedances.
It should reflect the combined interpretation of:
site history;
source characteristics;
field observations;
analytical results;
lateral and vertical delineation;
contaminant behaviour;
geology and hydrogeology;
the Conceptual Site Model;
proposed land use;
development requirements;
relevant exposure pathways;
remediation objectives;
remediation criteria; and
validation requirements.
That is why contaminated land remediation requires professional judgement.
Laboratory data are critical, but laboratory results do not interpret themselves.
The role of the environmental consultant is to translate those data into a defensible understanding of the site and then develop a remediation strategy proportionate to the actual risk.
The Objective Is Not to Excavate Until Everything Is Below an Investigation Level
Perhaps the simplest way to summarise the issue is this:
An investigation criterion tells us when contamination requires further consideration. It does not automatically tell us where to put an excavator.
The ASC NEPM specifically cautions against treating investigation and screening levels as default clean-up criteria because doing so can produce unnecessary remediation and development costs.
A good remediation strategy instead asks:
What contamination is present?
Why is it there?
How far does it extend?
What risk does it create?
What will the proposed development change?
What remediation or management action is actually necessary?
And how will we demonstrate that the objective has been achieved?
Those questions provide the technical basis for defining a remediation extent.
Contaminated Land Remediation in NSW
Confluence Environmental provides contaminated land assessment, remediation planning, remediation support and validation for development, infrastructure, industrial and government projects across NSW.
Our contaminated land and remediation services include Detailed Site Investigations, Remediation Action Plans, soil and groundwater assessment, asbestos in soil assessment, waste classification, remediation support and validation.
Our approach is to define remediation requirements from the available evidence, Conceptual Site Model, proposed development and project objectives rather than treating individual analytical exceedances as predetermined excavation boundaries.
Where additional delineation is required, the investigation is designed to resolve the specific uncertainty needed to develop a practical and defensible remediation strategy.