From DSI to RAP: When Is There Enough Information to Design Remediation?
At some point, a contaminated land investigation has to stop investigating and start solving the problem.
For straightforward sites, that transition may be relatively obvious. A Detailed Site Investigation identifies a localised area of contamination, defines its extent and provides a clear basis for remediation.
Complex sites are rarely that neat.
There may still be uncertainty around the exact edge of an impacted fill layer. Part of the site may be inaccessible beneath an existing building. The depth of contamination may vary between investigation locations. Additional contamination may reasonably be expected to become visible once earthworks begin.
So when is a Detailed Site Investigation (DSI) complete enough to prepare a Remediation Action Plan (RAP)?
The answer is not that every uncertainty must have been eliminated.
The more useful test is whether the investigation has resolved the uncertainties that are critical to the remediation decision.
A RAP should not be used to compensate for an inadequate investigation. Equally, investigation does not need to continue indefinitely simply because some uncertainty remains.
The transition from DSI to RAP occurs when the site is sufficiently understood to define:
what requires remediation or management;
why remediation is required;
what outcome needs to be achieved;
which remediation approaches are technically feasible;
where those approaches need to be applied;
how remaining uncertainty will be managed; and
how successful remediation will ultimately be demonstrated.
That is the point at which investigation becomes remediation design.
The DSI and RAP Have Different Jobs
A Detailed Site Investigation and a Remediation Action Plan are closely connected, but they perform fundamentally different functions.
The DSI asks:
What contamination is present, where is it, how significant is it and what risks does it create?
The RAP asks:
Given what we now know, what are we going to do about it?
The NSW EPA describes the objective of a DSI as providing more complete and definitive information on the issues identified during preliminary investigation. It expects the investigation to consider the type, extent and level of contamination, migration pathways, contaminant behaviour, risks and the adequacy and completeness of the available information, including uncertainty.
The RAP then takes that understanding and converts it into an actionable remediation framework. NSW EPA guidance requires a RAP to summarise the previous investigations, present the refined Conceptual Site Model, document the identified contamination risks, establish remediation objectives and criteria, define the required extent of remediation, assess remediation options and establish how the works will be implemented and controlled.
The dividing line between the two stages is therefore not simply:
“We have finished sampling.”
It is:
“We now have sufficient information to make and defend the remediation decisions.”
A DSI Does Not Need to Remove Every Uncertainty
Contaminated land investigation is inherently based on incomplete information.
We cannot excavate an entire site simply to understand what lies beneath it. Instead, site conditions are inferred from historical information, field observations, intrusive investigation points, analytical data and an evolving Conceptual Site Model.
Some uncertainty will therefore remain on almost every site.
The Data Quality Objectives process is useful here.
NSW EPA guidance describes Data Quality Objectives as performance and acceptance criteria used to determine whether there is enough data of sufficient quality to support decision-making. It also specifically links DQOs to the identification of critical data gaps within the CSM.
That is an important distinction.
The objective is not:
Collect enough data that nothing about the subsurface remains uncertain.
The objective is:
Collect enough reliable data that the uncertainty remaining does not prevent a defensible decision.
For remediation planning, this means distinguishing between decision-critical uncertainty and uncertainty that can reasonably be managed during the remediation works.
The First Question: Do We Understand the Source?
Before a remediation strategy can be designed, there generally needs to be sufficient understanding of what caused the contamination.
That does not mean every historical activity must be reconstructed perfectly.
It means the consultant should understand the contamination source well enough to predict how the contamination is likely to occur and how it can be addressed.
For example, elevated hydrocarbons associated with a former underground storage tank present a different remediation problem to widespread metals within heterogeneous imported fill.
Likewise, isolated bonded asbestos fragments within shallow soil present a different management problem to asbestos fines or fibrous asbestos distributed through fill.
Understanding the source affects:
where contamination is likely to occur;
its likely vertical and lateral distribution;
whether other environmental media may be affected;
whether contamination may continue to migrate;
the likely remediation method; and
the validation strategy.
If the source remains genuinely uncertain and different plausible sources would require materially different remediation approaches, further investigation is usually warranted before the RAP is finalised.
Do We Understand Which Environmental Media Are Affected?
A soil investigation may establish that soil contamination exists.
That does not necessarily establish that only soil is affected.
This is particularly important for mobile or volatile contaminants.
For example, a former fuel facility may have impacted:
soil;
groundwater;
soil vapour; and potentially
surface water or subsurface infrastructure.
Preparing a soil excavation RAP without understanding a credible groundwater or vapour pathway could result in a remediation strategy that addresses only part of the contamination problem.
The same principle applies to other contaminants and pathways.
Before moving from DSI to RAP, the consultant should be able to answer:
Have all environmental media that could materially affect the remediation decision been adequately considered?
If the CSM identifies a plausible pathway into groundwater, but groundwater has not been assessed and the answer could change the remediation strategy, that is likely to be a decision-critical data gap.
If the available evidence reasonably demonstrates that groundwater is not a relevant pathway, installation of monitoring wells simply because the site is contaminated may add little useful information.
Investigation should remain driven by the CSM rather than by a standard list of media that must always be sampled.
Has the Contamination Been Defined Well Enough to Select a Remedy?
The phrase “define the extent of contamination” can sometimes create the impression that every contaminant boundary must be known precisely before remediation planning starts.
That is not always necessary.
The real question is whether the extent has been characterised sufficiently for the proposed remediation strategy.
Consider two different scenarios.
Scenario 1: Localised excavation
A DSI identifies lead contamination associated with a former paint-affected area. The lateral and vertical extent has been reasonably delineated and the contamination is expected to be removed by excavation.
In this case, the DSI may need a reasonably well-defined contamination extent because the RAP is likely to specify a discrete remediation area.
Scenario 2: Widespread impacted fill
A site contains a relatively consistent layer of historically imported fill. The investigation demonstrates that the fill requires management and the proposed strategy involves excavation of the fill across a defined development area.
It may not be necessary to install investigation locations at extremely close spacing simply to establish the exact centimetre-scale edge of every contaminant concentration.
The relevant remediation unit may be the fill layer itself.
The investigation effort required therefore depends on the decision being made.
This is why the ASC NEPM and NSW EPA framework emphasise conceptual site models, data quality objectives and fit-for-purpose investigation rather than simply maximising sample numbers. The ASC NEPM identifies the CSM and DQO process as central components of site characterisation and recognises delineation of lateral and vertical contamination as part of that process.
Would More Investigation Change the Remediation Strategy?
This is one of the most useful practical questions a consultant can ask.
Suppose the next round of investigation finds:
Outcome A: contamination extends another five metres.
Or:
Outcome B: contamination stops where currently inferred.
Would the remediation method change?
Would the remediation criteria change?
Would the project require different controls?
Would another environmental medium need to be addressed?
Would the estimated remediation volume change enough to materially affect feasibility or cost?
If the answer is yes, the data gap may need to be resolved before the RAP is prepared.
If the answer is no, further investigation may provide additional information without changing the decision.
That does not mean the information has no value. It means its value should be weighed against the decision the investigation needs to support.
This is consistent with the NSW EPA's approach to supplementary investigation. Where a DSI identifies remaining data gaps, EPA guidance states that supplementary investigation should focus on the critical data gaps, with effort proportionate to the uncertainties identified and with a clearly defined purpose.
That is a much more useful approach than automatically recommending another round of sampling whenever any uncertainty remains.
Can the Remediation Objectives Be Clearly Defined?
A RAP needs an endpoint.
Before remediation planning progresses, the consultant should be able to state what the remediation is actually intended to achieve.
Depending on the site, the objective might be to:
remove a defined contaminant source;
remove impacted fill from accessible soil areas;
eliminate an unacceptable direct-contact pathway;
remove a vapour source beneath a proposed building;
contain residual contamination beneath an engineered barrier;
manage asbestos-impacted soil;
treat contaminated groundwater;
reduce contaminant concentrations to an established remediation criterion; or
make the land suitable for the proposed use subject to defined management controls.
If the DSI has not provided enough information to establish meaningful remediation objectives, the project may not yet be ready for a RAP.
The NSW EPA requires RAPs to set remediation objectives that will result in the site being suitable for its current or proposed use without unacceptable risk to human health or the environment.
A vague objective such as “remove contaminated soil” is generally not enough.
Which soil?
Why?
To what endpoint?
And how will completion be verified?
Those questions need to be answerable before a robust soil remediation plan can be developed.
Can Appropriate Remediation Criteria Be Established?
The investigation also needs to provide enough information to determine how remediation success will be assessed.
Remediation criteria may be numerical, physical, risk-based or a combination of these.
Depending on the strategy, they might include:
contaminant concentration criteria;
removal of a defined material layer;
removal of visible asbestos-containing material;
removal of a contaminant source;
achievement of treatment performance criteria;
installation of an engineered barrier;
achievement of groundwater objectives; or
compliance with a site-specific risk-based endpoint.
The NSW EPA RAP checklist specifically requires the selected remediation criteria to be identified and the rationale, assumptions and limitations associated with their selection to be explained.
If the investigation has not established enough information to select defensible remediation criteria, further assessment may still be needed.
Can the Remediation Options Actually Be Compared?
A RAP should not merely identify that remediation is required.
It should provide a defensible basis for selecting how the contamination will be managed.
Depending on the project, the available soil remediation methods may include:
excavation and off-site disposal;
treatment;
stabilisation;
capping or containment;
in situ management; or
a combination of methods.
To compare those options meaningfully, the investigation needs to provide enough information about contamination conditions and site constraints.
For example, treatment feasibility may depend on contaminant type and concentration.
Excavation feasibility may depend on depth and groundwater conditions.
Containment may depend on whether the remaining contamination can be reliably isolated and managed.
Groundwater contamination may require a completely different strategy to shallow soil contamination.
A DSI is therefore sufficiently advanced when it provides enough information to distinguish between realistic remediation options and reject those that are technically inappropriate.
The RAP then documents the selected strategy and its justification.
Exact Excavation Quantities Do Not Always Need to Be Known
One area where projects can become unnecessarily bogged down is trying to calculate the exact remediation volume before the works commence.
That level of certainty is not always achievable or necessary.
Subsurface contamination is not normally distributed in perfect geometric shapes.
Investigation points provide an interpretation of conditions between discrete locations. Once excavation begins, significantly more of the subsurface becomes visible.
A RAP can therefore sometimes be prepared using an estimated remediation extent and quantity, provided that:
the source and contaminant distribution are sufficiently understood;
the remediation method would not materially change;
reasonable allowances have been made;
decision rules are established for extending or stopping excavation; and
validation is used to confirm the final remediation extent.
The critical distinction is between uncertainty in the exact quantity and uncertainty in the nature of the remediation problem.
Not knowing whether a remediation excavation will contain 450 m³ or 520 m³ of similar material may be manageable.
Not knowing whether contamination extends into groundwater may not be.
When Should Supplementary Investigation Occur Before the RAP?
Some data gaps should generally be resolved before the remediation strategy is locked in.
Examples can include situations where:
the primary contamination source has not been identified;
the lateral or vertical extent is so uncertain that different remediation approaches remain plausible;
groundwater may be affected but has not been adequately assessed;
volatile contamination could create an unresolved vapour pathway;
contamination may extend beyond the site boundary;
different contaminant forms require materially different controls;
asbestos distribution or condition is insufficiently understood to establish the remediation method;
remediation volumes could vary by an order of magnitude;
a potential ecological receptor has not been adequately assessed; or
the available data are not of sufficient quality to support the remediation decision.
In these circumstances, a supplementary investigation may be more appropriate than attempting to manage fundamental uncertainty through contingencies in the RAP.
The EPA specifically provides for supplementary site investigations to address critical gaps identified during the DSI and states that their purpose should be clearly defined.
Which Data Gaps Can Sometimes Be Managed During Remediation?
Other uncertainties may reasonably remain when a RAP is prepared.
Examples might include:
the exact edge of an excavation where validation sampling will establish the final boundary;
contamination beneath a structure that cannot be accessed until demolition;
minor variation in fill thickness where all of the relevant fill will already be excavated;
small uncertainties in the volume of material requiring disposal;
localised unexpected contamination that can be addressed under a clearly defined contingency procedure; or
final placement levels associated with a containment system that will be documented during construction.
Whether this is appropriate depends on the site.
The key test is whether the RAP provides a clear process for resolving the uncertainty without undermining the remediation objective.
This is why contingencies and unexpected finds procedures are an important part of remediation planning. NSW EPA guidance expressly requires RAPs to address contingencies and unexpected finds.
Contingency planning is not a substitute for site characterisation.
It is a mechanism for dealing with uncertainty that could not reasonably or efficiently be eliminated before works commenced.
Example: Former Fuel Facility
Consider a former service station where a DSI identifies petroleum hydrocarbon contamination around a former underground storage tank.
Soil impacts have been identified and partially delineated.
Should a RAP now be prepared?
That depends on what remains uncertain.
If the investigation demonstrates that:
the tank and associated infrastructure are the likely source;
impacted soil is relatively localised;
groundwater is sufficiently deep or otherwise demonstrated not to be a relevant pathway;
vapour risks have been adequately assessed;
off-site migration is not indicated; and
excavation and validation provide a technically appropriate remedy,
the remaining uncertainty around the exact excavation edge may be capable of being managed during remediation.
A RAP could establish the anticipated excavation, field observations, validation requirements and decision rules for extending the excavation if necessary.
Now change one fact.
Groundwater is shallow and petroleum hydrocarbons are present close to the water table, but no groundwater investigation has been completed.
That uncertainty could materially change the remediation strategy.
Excavating contaminated soil may not address dissolved contamination, free product or ongoing migration.
The project may therefore require additional investigation before the RAP can be properly designed.
The difference is not simply that one site has more data.
It is that one site has resolved the decision-critical uncertainty and the other has not.
Example: Contaminated Fill Across a Development Site
Consider a redevelopment site containing heterogeneous imported fill.
A DSI confirms metals, hydrocarbons and localised anthropogenic material within the fill. Sampling indicates that the fill requires management across a substantial portion of the proposed development.
Does every metre of the fill need to be exhaustively delineated before a RAP can be prepared?
Not necessarily.
If the development requires excavation of the affected fill across the relevant area and the remediation strategy will treat the fill as a defined material unit, additional sampling simply to generate a more detailed contamination contour may not change the remediation approach.
The RAP may instead focus on:
defining the affected material;
excavation and segregation;
waste classification;
stockpile management;
controlled on-site management where appropriate;
environmental controls;
unexpected finds; and
validation.
Conversely, if the project intends to leave large portions of that fill in place, selectively excavate only higher-risk areas or reuse material in different parts of the site, a greater level of characterisation may be necessary.
Again, the required investigation effort depends on the decision the RAP needs to make.
Existing Buildings Can Create Legitimate Data Gaps
Another common situation involves contamination beneath existing structures.
A DSI may identify contamination adjacent to a building but be unable to investigate beneath the slab without disrupting ongoing operations.
Does that automatically prevent remediation planning?
Not always.
If the building will be demolished as part of the development, the RAP may establish a staged process whereby:
accessible contamination is addressed;
the building is demolished;
the previously inaccessible area is inspected and investigated;
the CSM is confirmed or revised; and
remediation proceeds under defined decision rules.
That can be a legitimate approach where the likely contamination conditions are sufficiently understood and the RAP provides an appropriate framework for resolving the remaining uncertainty.
However, if contamination beneath the structure could fundamentally alter the remediation method or demonstrate a significant unidentified source, additional investigation may be needed before the remediation strategy is accepted.
For a broader discussion of how remediation interacts with building footprints, earthworks and construction sequencing, see Designing Remediation Around the Proposed Development.
Validation Strategy Is a Useful Test of Whether the DSI Is Sufficient
One of the best ways to test whether the project is ready for a RAP is to ask:
Can we design the validation program now?
If the consultant can define:
what remediation must achieve;
what areas or materials require validation;
what media need to be assessed;
what criteria will apply;
how the remediation endpoint will be determined; and
what evidence will demonstrate successful completion,
the underlying investigation is often sufficiently mature to support remediation planning.
If those questions cannot be answered because the contamination problem remains poorly understood, more investigation may be required.
This is important because validation should not be improvised after the remediation works have already occurred.
The RAP should establish the validation requirements in advance so the project team knows how completion will be demonstrated.
Auditor-Reviewed Projects Often Require Greater Certainty
The threshold for moving from DSI to RAP can also depend on the project context.
Where a contaminated land project is subject to independent review under the NSW EPA Site Auditor Scheme, unresolved assumptions are likely to receive closer scrutiny.
An auditor may reasonably expect the consultant to demonstrate that the investigation provides an adequate basis for:
the refined CSM;
risk conclusions;
remediation criteria;
remediation extents;
selection of the remediation strategy; and
validation design.
This does not mean an auditor-reviewed site needs every uncertainty eliminated.
Complex sites often retain uncertainty.
It means that the uncertainty needs to be identified, understood and managed transparently, with the remediation strategy remaining robust if the anticipated conditions vary within a reasonable range.
On more complex sites, early auditor input can therefore help distinguish between data gaps that genuinely require further investigation and those that can appropriately be managed within the RAP.
There Is Also a Cost to Over-Investigation
The risk of insufficient investigation is obvious.
A poorly characterised site can result in:
unexpected contamination;
uncontrolled scope growth;
incorrect waste estimates;
inappropriate remediation methods;
repeated mobilisation;
construction delays;
failed validation; or
an inability to demonstrate site suitability.
But unnecessary investigation also has a cost.
Additional drilling, laboratory analysis, mobilisation and reporting can add significant expenditure without necessarily improving the decision.
It can also delay the point at which the actual contamination problem begins to be addressed.
The objective should therefore not be to maximise the number of investigation locations.
It should be to obtain the information required to support a technically defensible and proportionate contaminated land remediation strategy.
This is the practical purpose of Data Quality Objectives and a well-developed CSM.
A Useful Decision Test Before Progressing to RAP
Before moving from the DSI into remediation planning, the consultant should be able to answer the following questions with reasonable confidence:
Do we understand the contamination sources?
Do we understand the affected environmental media?
Is the contaminant distribution sufficiently characterised for the proposed remedy?
Have the relevant human-health and environmental pathways been assessed?
Are the critical data gaps resolved?
Can remediation objectives and criteria be established?
Can technically feasible remediation options be identified and compared?
Can the required remediation areas or material units be defined sufficiently for implementation?
Can the validation strategy be designed?
Can the remaining uncertainties be safely and defensibly managed through contingencies or staged works?
If the answer to those questions is yes, preparation of a RAP may be appropriate even though some uncertainty remains.
If one of the unanswered questions could fundamentally change the remediation method, risk conclusion or site suitability outcome, supplementary investigation is usually the better next step.
The Transition From Investigation to Remediation Is a Professional Judgement
There is no universal number of samples that marks the point where a DSI becomes “complete”.
Nor is the absence of all uncertainty a realistic endpoint.
The decision to progress from DSI to RAP requires professional judgement about whether the available evidence provides a sufficiently robust basis for the decisions that now need to be made.
The NSW EPA reporting framework reflects this directly.
The DSI is expected to assess the adequacy and completeness of the available information, including uncertainty. Supplementary investigation is then directed toward critical gaps. The RAP uses the resulting understanding to define remediation objectives, remediation extent, criteria, methods, contingencies and validation.
In practical terms:
Investigation should continue until additional information is no longer reasonably expected to change a decision that matters to the remediation strategy.
At that point, the project should stop simply characterising the problem and begin designing the solution.
Detailed Site Investigation and Remediation Action Plans in NSW
Confluence Environmental provides contaminated land assessment and remediation services for development, infrastructure, industrial and government projects across NSW.
Our team prepares Detailed Site Investigations to characterise contamination and resolve the data gaps required for project decision-making, and Remediation Action Plans to convert those findings into practical remediation and validation strategies.
Where an existing DSI has already been prepared, we can review the investigation, identify whether any decision-critical data gaps remain and determine whether the available information provides an adequate basis to proceed to remediation planning.
The objective is not further investigation for its own sake.
It is to establish enough reliable information to design a remediation strategy that can be implemented, validated and defended.