Quick answer: Under the Australian R&D Tax Incentive, eligibility is assessed activity by activity and must be self-assessed by the claimant. Sizing a wetland, drilling an injection well and routine compliance monitoring that reports against a licence condition generally apply established methods whose outcome is determinable in advance. A core R&D activity may exist where a managed aquifer recharge treatment train has to hold a performance current knowledge cannot predict for that water — a pathogen log credit no guideline supplies, clogging in a specific aquifer matrix, or recovery efficiency in a brackish aquifer — and the answer can only be reached by a systematic progression of work conducted to generate new knowledge. Experimental validation designed to establish such a performance is a different question from monitoring that reports against a condition already set.
24 August 2026 — this article describes the current rules. The 2026-27 Federal Budget announced R&DTI changes; the ATO states the measure is not yet law, and industry.gov.au states the changes will apply to income years starting on or after 1 July 2028.
Greater Adelaide is one of the most heavily instrumented managed aquifer recharge (MAR) regions anywhere. The Department for Environment and Water records over 40 MAR schemes across the metropolitan area, which have injected 48 GL of water for non-potable uses, most built as part of stormwater harvesting schemes since the early 1990s (DEW).
Four decades of local practice means most of a scheme — the gross pollutant trap, the wetland, the injection well, the licensing pathway — is engineering with a determinable outcome. The same maturity makes the questions still open unusually well defined, and they concentrate in one place: whether the train will hold a stated water quality when the influent is stormwater whose composition swings with catchment, season and storm hydrograph.
The Statutory Test, Stated Once and in Full
Under s 355-25(1) of the Income Tax Assessment Act 1997, core R&D activities are experimental activities whose outcome cannot be known or determined in advance on the basis of current knowledge, information or experience, but can only be determined by applying a systematic progression of work that is based on principles of established science and that proceeds from hypothesis to experiment, observation and evaluation, and leads to logical conclusions; and that are conducted for the purpose of generating new knowledge, including new knowledge in the form of new or improved materials, products, devices, processes or services (business.gov.au; ITAA 1997). Eligibility is assessed activity by activity, not project by project, and the claimant self-assesses.
The established-science limb is rarely the difficulty here: hydrogeology, microbial inactivation kinetics and aqueous geochemistry give a recharge investigation a settled body of principle to build on. Many of these activities are assessed primarily against the first limb instead — much MAR work has a determinable outcome precisely because the guidelines and South Australian practice have already determined it. Difficulty, cost and novelty to the proponent are not the test.
Three Places Where the Outcome Is Genuinely Not Determinable in Advance
1. A pathogen log credit that no guideline supplies for this water
The Phase 2 stormwater guidelines record measured retention of faecal indicator bacteria across treatment measures ranging from negative — outflow higher than inflow, generally from bird and animal input on open water — up to about 2 log; limited information relating indicator reductions to design parameters, "which would enable a treatment measure to be designed to achieve a particular retention"; and expected removal of Campylobacter, Cryptosporidium and viruses in wetlands below 0.5 log. Their conclusion recommends a conservative approach: do not assume conventional stormwater treatment measures provide reference-pathogen reduction unless the measure has been validated for the scheme, and validate any measure relied on as a health barrier (stormwater guidelines). They add that no statistically valid relationship has been found between pathogen levels and indicator bacteria such as E. coli. A credit cannot be read off a table, and whether one can be established for a given catchment and train is not known beforehand.
2. Clogging in a specific aquifer matrix
Clogging is a loss of permeability around the entry zone; the MAR guidelines call it one of the most serious operational issues, and note that wells in inappropriately designed and operated schemes may clog within a matter of days or weeks. It arises through filtration of suspended solids, microbial growth on assimilable organic carbon and nutrients, geochemical reactions, and air entrainment — usually more than one at once. On prediction the guidelines are explicit: laboratory column studies on cored material "only offer a guide, and can not always be relied upon to accurately predict field-scale performance", and coupled reactive-transport codes are "highly complex, data-intensive and only suitable for research purposes at present". Definitive Australian source-water quality values for sustainable recharge are "yet to be achieved", and single values used overseas — a Dutch assimilable organic carbon limit below 10 µg/L for sandy aquifers, for instance — are said to be inappropriate for Australia's diversity of source waters and hydrogeology (MAR guidelines). Where viability turns on a sustained injectivity those tools cannot resolve, an experimental activity may exist.
3. Recovery efficiency in a brackish aquifer
Recovery efficiency is the proportion of recovered water fit for its intended use, as a fraction of the volume injected. Across six Australian sites, recalculated to a common 1,500 mg/L threshold, the guidelines report it varying from below 0.02 to above 1.0, driven by aquifer thickness, transmissivity, heterogeneity, hydraulic gradient and density contrast — variables intrinsic to a site rather than chosen. It also moves with time, because unrecovered water forms a salinity buffer: at the fractured-rock stormwater ASR scheme at Scotch College in Adelaide, average recovered salinity fell from an ambient 2,100 mg/L to 1,400 mg/L over six seasons. That establishes the phenomenon, not the number at a new site.
None of this makes MAR work R&D by default: sizing a wetland to a load-reduction target, drilling a well, running a pumping test and monitoring to a licence condition are generally established activities where the relevant outcomes can often be determined in advance (what does not qualify).
Where Compliance and Commissioning Work Sits
Section 355-25(2) lists activities that are not core R&D activities, and paragraph (f) reaches a lot of water-quality work:
"(f) activities associated with complying with statutory requirements or standards, including one or more of the following: (i) maintaining national standards; (ii) calibrating secondary standards; (iii) routine testing and analysis of materials, components, products, processes, soils, atmospheres and other things"
In South Australia the surrounding regulation is dense: a well construction permit to drill, an EPA licence or DEW water affecting activity permit to inject, an extraction licence where the resource is prescribed, and recycled-water use regulated by the health portfolio (DEW), with the EPA's Code of Practice for Aquifer Storage and Recovery sitting over the barriers.
A supporting R&D activity must be directly related to core R&D activities (s 355-30(1)). Where an activity is one referred to in s 355-25(2), produces goods or services, or is directly related to producing goods or services, s 355-30(2) makes it a supporting R&D activity only if it is undertaken for the dominant purpose of supporting core R&D activities (business.gov.au). That third limb does real work on a recharge scheme, where recovered water is supplied for an end use, activities directly related to producing that service may require the dominant-purpose test.
Paragraph (f) is not triggered merely because a guideline exists — the MAR guidelines require validation of preventive measures and observe that at a difficult site the true situation "would probably not be known reliably until commissioning with validation monitoring". But "associated with" is broad wording, and some activities directed at demonstrating compliance with a statutory requirement or standard may fall within paragraph (f). In practice what separates a validation experiment from commissioning is documentary, and built before the season starts rather than reconstructed afterwards:
• Keep the validation protocol distinct from the commissioning inspection and test plan, with its own hypotheses, pass criteria and stopping rules recorded before the first trial run.
• Mark each sample at collection as experimental or licence-verification, because the same instruments, laboratory and staff serve both.
• Keep a dated decision log of options tried, rejected and why — the observable form of the evaluation-and-logical-conclusions limb.
• Code experimental labour, plant time and laboratory analysis separately from the season's operating cost.
None of that makes an activity eligible. It makes an activity that already meets the tests visible in the record (record keeping).
A Worked Example (Hypothetical and Illustrative Only)
Invented to show where the boundary falls. It is not a real project and says nothing about whether any actual claim would be accepted, or about who could claim: only an R&D entity can be entitled to the offset and, under s 355-35(3), an exempt entity cannot be an R&D entity. A local governing body is generally income tax exempt, so on facts like these the R&D entity would have to be a taxable participant — and which one is not settled by who runs the rig, because s 355-210 sets conditions an R&D activity has to meet, and under s 355-210(2) an activity conducted to a significant extent for one or more other entities not covered by any paragraph of s 355-210(1) is not one the section applies to. That turns on control, financial risk and ownership of results, and is settled before any activity question is worth asking.
A northern-Adelaide council operates a scheme on a 420 ha mixed residential and light-industrial catchment: gross pollutant trap, a 2.1 ha constructed surface-flow wetland, and an injection well screened from 62 to 98 m in a Tertiary confined aquifer at 1,900 mg/L ambient TDS and 21–23 °C. Recovered water irrigates reserves. The council wants to extend the end use to dual-reticulated toilet flushing in a new 600-lot development — higher exposure, and a salinity limit the scheme has never had to hold.
Baseline and target: Baseline: sustained injectivity 0.48 L/s per metre of head, backwash every 14 days, first-cycle recovery efficiency 0.31 at a 1,200 mg/L cut-off, no validated pathogen credit anywhere in the train. The target written into the validation plan before any trial: 5.0 log reduction of the reference virus across the train, at least 2.0 log of it from aquifer residence; recovery efficiency of at least 0.55 by the third cycle at 1,200 mg/L or less; and injectivity held at 0.40 L/s/m or better across an 80-day injection season, with backwashing costing no more than 8% of injected volume.
Held and varied: Held: the same well and screen interval, catchment and wetland, injection rate 18 L/s, temperature range, one laboratory and enumeration method, chambers at fixed depth. Varied: where the disinfection barrier sits (pre-injection or post-recovery), pre-injection filtration cut, storage time before recovery (28, 56, 90 days), and recovery pumping rate.
The approach that failed (wetland pathogen credit): The first hypothesis credited the wetland with 1.5 log of virus reduction, tested by seeding MS2 coliphage at the inlet and enumerating at the outlet. Across eleven storm events measured reduction ranged from −0.3 to 1.1 log (median 0.6), with outlet counts exceeding inlet counts on two events. Reduction tracked antecedent dry days and hydrograph shape, and negative events coincided with resident birds. The trial did not support the proposed fixed credit under the tested operating envelope.
In-situ inactivation and clogging trials: Diffusion chambers measured 1-log MS2 inactivation time at 5.4 days near the well and 14.6 days down-gradient (tracking redox drop from +40 mV to −90 mV). Bromide tracer test showed exposure time was 44 days (against 56 nominal), yielding 3.0 log MS2 inactivation. Meanwhile, a 5 µm pre-injection filter held turbidity < 1.2 NTU, yet injectivity fell 56% in 34 days due to microbial growth (AOC 180 µg C/L) and pyrite oxidation releasing iron. Replacing it with a saturated-zone biofiltration cell brought AOC to 70 µg C/L, yielding a 9-day backwash interval at 6% volume, no iron response, and 0.43 L/s/m injectivity at day 80.
Result reached: All four pass criteria met. Pathogens: 3.0 log from residence, 0 from wetland, 2.0 from post-recovery UV/chlorine (5.0 log total). Injectivity: 0.43 L/s/m. Backwash: 6%. Recovery efficiency: 0.58 at a constrained 12 L/s recovery rate. Two credit assumptions were wrong; rejected options sit in the decision log.
Activity boundary: Documented from hypothesis and experimental approach through seeded runs, chamber deployments, instrumented injection cycles, evaluation and recorded conclusion. Construction, permits, routine monitoring and normal operation are separate activities needing supporting-activity analysis. Candidate core activity, subject to statutory self-assessment.
What Is Generally Unlikely to Be Core on Those Facts
Activity
Why
Sizing a wetland or bioretention system to a load-reduction target with standard catchment modelling
Established method, outcome determinable in advance
Drilling, developing and equipping the injection and recovery wells
Constructing the asset; outcome determinable in advance
A conventional aquifer pumping test to derive transmissivity and storativity
Established method with a documented procedure
Periodic monitoring and reporting under an EPA licence or water affecting activity permit
Activity associated with complying with statutory requirements — paragraph (f); routine testing and analysis under (f)(iii)
Where an RSP Fits
An RSP is a scientific or technical service provider registered in specific fields that a company can engage to conduct R&D activities on its behalf (business.gov.au). This work sits in our registered field of environmental engineering, and the contribution is upstream of commissioning: framing which barrier is actually unproven, designing the seeding, tracer and in-situ inactivation programme at the resolution a credit requires, and specifying instrumentation before the injection season starts. There is also a threshold point for smaller schemes: Qualifying expenditure incurred to a non-associate RSP may still form part of the offset where total notional deductions are below the usual $20,000 threshold, provided the services are within a research field for which the RSP is registered. In mechanism, where an R&D entity's total notional R&D deductions for the year fall below $20,000, s 355-100(2) works the offset out on a substituted base — expenditure to a registered research service provider that is not an associate, for services in a field the provider is registered for, plus CRC contributions. Offset rates and the two tiers are covered in our article on the refundable and non-refundable offset; see also claiming R&D under $20,000 and R&D Tax Incentive in Adelaide.
Talk to Ignition Research before the first injection season, while the trial can still be separated from commissioning. As a Registered Research Service Provider at Lot Fourteen in Adelaide, we design and conduct the experimental programme and produce the technical record while the work is happening. We are not a registered tax agent: your company self-assesses and remains responsible for its own claim. Get in touch.
Frequently Asked Questions
Q: Is designing a stormwater treatment train an eligible R&D activity?
A: Generally not, as a core R&D activity. Selecting and sizing a gross pollutant trap, wetland and filtration stage to meet a load-reduction target applies established methods with an outcome that a competent professional in the relevant field could determine in advance from published design guidance. A core activity may exist in the narrower case where the train has to hold a performance the available guidance cannot predict for that source water — most often a pathogen barrier credit — and the answer comes only from a systematic progression of work conducted to generate new knowledge. Eligibility is assessed activity by activity and must be self-assessed by the claimant.
Q: Why can't a constructed wetland simply be credited with a pathogen log reduction?
A: Because the measured performance is too variable to design against. The Phase 2 stormwater guidelines report indicator-bacteria retention across stormwater treatment measures ranging from negative — outflow higher than inflow, typically from birds on open water — to about 2 log, note limited information linking design parameters to a particular retention, and expect protozoan and virus removal below 0.5 log. Their recommended default is to assume no reference-pathogen reduction unless the measure is validated for the scheme.
Q: Is aquifer clogging testing R&D or commissioning?
A: It is assessed as an activity, not by the phase it happens in. Monitoring injectivity and backwashing on a documented schedule is operations. Determining whether a sustained injection rate is achievable at all in a particular aquifer matrix and source water, where the guidelines' own predictive tools — column studies and coupled reactive-transport codes — are said not to resolve it reliably at field scale, may be a core R&D activity if the work proceeds from a recorded hypothesis through experiment, observation and evaluation to logical conclusions. Distinct work that is not itself core — running the plant to produce the water a trial injects, for instance — is analysed separately as a possible supporting R&D activity, and where s 355-30(2) applies, the dominant-purpose test must be satisfied for activities that produce goods or services or are directly related to producing them.
Q: Is recovery efficiency in a brackish aquifer predictable in advance?
A: Partly. The controlling variables are documented — aquifer thickness, transmissivity, heterogeneity, hydraulic gradient, density contrast, storage time, recovery rate and cycle number — and the improvement over successive cycles from salinity buffering is a known phenomenon. What is not generally determinable in advance is the value at a specific site, which the MAR guidelines show varying from below 0.02 to above 1.0 across Australian schemes. Whether closing that gap is a core R&D activity depends on the facts of the particular activity and on self-assessment against the statutory tests.
Sources & Further Reading
waterquality.gov.au — Australian guidelines for water recycling
environment.sa.gov.au — Managed aquifer recharge (Department for Environment and Water)
epa.sa.gov.au — Code of Practice for Aquifer Storage and Recovery
legislation.gov.au — Income Tax Assessment Act 1997 — Div 355, incl. ss 355-25, 355-30, 355-100 and 355-210
Related: environmental engineering · what does not qualify · what an RSP is · claiming R&D under $20,000 · refundable vs non-refundable offset · R&D Tax Incentive in Adelaide · more Insights
This article is general information from a Registered Research Service Provider about the R&D Tax Incentive. It is not tax, legal or financial advice; eligibility depends on your circumstances and you should self-assess and seek your own advice.
Thinking about a project like this?
If you're weighing up an AI, software or technical improvement project and can't tell yet whether it's implementation or research, start with a quick read on where it sits.