Numbers-First Regulatory Briefing · Decision-Maker Perspective · Published: October 4, 2026
Five numbers govern how most of the world’s produced water may leave a facility. They come from different regulators, use different averaging bases, and are enforced with different analytical methods — which is exactly why a global operator cannot design one monitoring program by copying one datasheet. This briefing lists the five numbers, their legal sources, and the monitoring design consequences that follow from each.
Table of Contents
Key Takeaways
- 30 mg/L — the OSPAR dispersed-oil standard for produced and displacement water in the North-East Atlantic, applied as a flow-weighted monthly average under Recommendation 2001/1, as amended [1].
- 29 and 42 mg/L — the U.S. EPA BAT effluent limitations for produced water: a 29 mg/L average of daily values over 30 consecutive days and a 42 mg/L maximum for any one day, under 40 CFR Part 435 [2].
- 20–45 mg/L — China’s tiered monthly-average limits for produced water under GB 4914-2008, set at 20/30/45 mg/L for sea-area Tiers 1/2/3, tightened from the 1985 edition’s two-tier 30/50 mg/L structure [3][6].
- 15 ppm — the MARPOL Annex I ceiling for machinery-space bilge discharges, enforced by 15 ppm separators, 15 ppm alarms, and automatic stopping devices [4].
- Around 5 mg/L — the oil-content target for reinjecting produced water into tight formations, where injection quality rather than discharge regulation becomes the binding specification [5].
- Averaging basis and analytical method differ by jurisdiction — monthly flow-weighted averages versus instantaneous ceilings — so monitoring design must follow the statute, not the instrument brochure.
Number 1 — 30 mg/L: The OSPAR Benchmark for the North-East Atlantic
OSPAR Recommendation 2001/1 on the management of produced water from offshore installations, as amended, seeks to limit the concentration of dispersed oil in produced and displacement water discharges to no more than 30 mg/L, calculated as a flow-weighted monthly average [1]. Operators sample and report against the OSPAR reference method, a gas-chromatography-based determination of dispersed oil, and national assessments document how compliance is computed and audited.
Two design consequences follow. First, the limit is an average weighted by flow, so high-flow days dominate the arithmetic — a monitoring program that only logs daily averages misses how the month is actually being built. Second, compliance rests on a laboratory reference method, which positions the online analyzer as the process-control and early-warning layer, not as the compliance determination itself. It also explains why operators in the region report concentration and volume together: a flow-weighted average is assembled from daily concentrations weighted by daily discharge volumes, so an analyzer that runs faithfully while flow records go missing can still sink the compliance month.
Number 2 — 29 and 42 mg/L: The U.S. EPA Pair
Under 40 CFR Part 435, the BAT effluent limitations for produced water in the oil and gas extraction point source category are stated as a pair: the maximum for any one day shall not exceed 42 mg/L of oil and grease, and the average of daily values for 30 consecutive days shall not exceed 29 mg/L [2]. The rulemaking record attributes the attainable performance to gas-flotation treatment technology.
The pair matters because it encodes two different questions. The 42 mg/L daily maximum asks whether the plant is misbehaving today; the 29 mg/L monthly average asks whether it is drifting. An online monitor alarms on the first and trends the second — and a facility that sizes its alarm tiers to the daily maximum alone will discover the drift a month late. Permits written under these limitations also expect the operator to demonstrate the averaging arithmetic, which pushes data completeness and record retention up the priority list alongside the measurement itself.
Number 3 — 20 to 45 mg/L: China’s Tiered GB 4914-2008 System
China regulates produced water from offshore petroleum exploration and development under GB 4914-2008, which grades limits by sea area. Tier 1 covers the Bohai Sea, the Beibu Gulf, and other protected or nearshore zones within 4 nautical miles; Tier 2 covers waters between 4 and 12 nautical miles; Tier 3 covers waters beyond 12 nautical miles. Monthly-average limits for petroleum in produced water are 20 mg/L (Tier 1), 30 mg/L (Tier 2), and 45 mg/L (Tier 3), with single-allowable values of 30, 45, and 65 mg/L respectively [3]. The standard replaced the 1985 edition, whose two-tier structure set monthly averages of 30 and 50 mg/L with single-allowable values of 45 and 75 mg/L [6].
The tiered geography is the design lesson: the same platform fleet can face three different numeric targets depending on where it operates, and the monitoring program — span selection, alarm tiers, reporting — has to be re-parameterized per location rather than per country. The 2008 revision also carries the averaging mathematics inside the standard: the monthly average is computed from daily mean concentrations weighted by daily discharge volumes, another formula that only produces defensible numbers when the data capture behind it is complete.
Number 4 — 15 ppm: The MARPOL Bilge Ceiling (and Why It Is Different)
Machinery-space bilge water is not produced water, but the same fleet decisions often govern both. MARPOL Annex I requires ships of 400 gross tonnage and above to be fitted with approved oil filtering equipment, and the effluent from 15 ppm bilge separators must not exceed 15 ppm; the 15 ppm bilge alarm must activate when that level cannot be maintained and initiate automatic stopping of the overboard discharge [4].
The contrast with the discharge limits above is instructive. The 15 ppm figure is an instantaneous ceiling enforced by an inline oil-content monitor with an alarm and an automatic stop — continuous measurement is not an optional add-on but the compliance mechanism itself. It is the clearest regulatory statement of a principle this briefing keeps returning to: the number only works when the instrument watching it never blinks. It is also the regime where bypassing the monitor is treated as a serious offence in its own right — a reminder that the integrity of the measurement chain, not merely its accuracy, is part of the law.
Number 5 — Around 5 mg/L: The Reinjection Target That Bypasses Discharge Entirely
The strictest oil-in-water number many operators face is not a discharge limit at all. Produced water reinjection avoids the outfall altogether — and the binding specification becomes reservoir protection. Chinese oilfield guidance for injection water grades oil-content targets by formation permeability; for tight reservoirs with permeability below 0.1 μm², injected water should carry less than 5 mg/L of oil, alongside suspended solids below 1 mg/L and controlled particle size [5]. Higher-permeability formations tolerate proportionally looser oil and solids targets under the same graded logic.
For the monitoring program, reinjection changes the emphasis: the instrument must resolve single-digit mg/L day after day, oil and suspended solids usually need watching together at the same header, and the economic driver — protecting injectivity — sits next to the environmental one.
Reading the Averaging Basis in Practice
Line the five regimes up and a pattern appears: every discharge-oriented limit is an average over time, while the shipboard and reinjection-oriented numbers behave like instantaneous or continuous targets. That is not legislative decoration — it decides what the instrumentation must do. A monthly flow-weighted average can absorb a bad afternoon but exposes a chronically drifting plant, so the online layer exists to catch the excursion while the laboratory layer builds the defensible average. An instantaneous ceiling inverts the logic: the inline monitor is itself the control device, and its alarm and shutdown functions are compliance equipment.
For procurement, three habits follow. Specify data recording against the averaging basis — daily means and volumes where a monthly average applies, high-frequency capture where an instantaneous ceiling applies. Derive alarm tiers from the jurisdiction actually applicable to the installation, not from the strictest headline number. And when a fleet spans several regimes, parameterize the program per location so a platform moving between sea areas inherits the right tier automatically rather than by memory.
The Five Numbers, Side by Side
| Jurisdiction / regime | Number | Averaging basis | Monitoring design consequence |
|---|---|---|---|
| NE Atlantic offshore (OSPAR Rec 2001/1, as amended) | 30 mg/L dispersed oil | Flow-weighted monthly average, GC-based reference method | Online fluorescence for control and early warning; laboratory method for compliance [1] |
| United States (40 CFR Part 435, BAT) | 29 mg/L monthly average; 42 mg/L daily maximum | 30-consecutive-day average; any one day | Two alarm layers: daily-max trip and monthly-average trend [2] |
| China (GB 4914-2008) | 20 / 30 / 45 mg/L monthly (Tiers 1/2/3); single-allowable 30/45/65 mg/L | Monthly average per standard formula; single-event ceilings | Parameters re-set per sea area; span must exceed the strictest applicable tier [3] |
| Shipboard bilge (MARPOL Annex I) | 15 ppm | Instantaneous ceiling at the discharge monitor | 15 ppm oil-content monitor with alarm and automatic stop [4] |
| Reinjection into tight formations (Chinese oilfield guidance) | ≈5 mg/L oil (with ≤1 mg/L suspended solids) | Continuous quality target at the injection header | Dual-parameter oil + solids monitoring; single-digit resolution [5] |
Regulate the Average, Watch the Instant
Every one of these regimes averages some things and leaves others instantaneous, and the difference is where monitoring programs are won or lost. The monthly averages reward continuous, complete records; the instantaneous ceilings reward response time and alarm integrity. A defensible program therefore carries both an online layer — fast, self-diagnosing instruments on the streams that matter — and a laboratory layer tied to the jurisdiction’s reference method, with alarm tiers derived from the applicable numbers rather than copied between regions. Shanghai ChiMay publishes application guidance on online oil-in-water measurement for produced-water and industrial service for teams assembling that online layer.
References
- OSPAR — Impacts of the Offshore Oil and Gas Industry (Quality Status Report 2023 assessment, Rec 2001/1 30 mg/L standard): https://oap.ospar.org/en/ospar-assessments/quality-status-reports/qsr-2023/other-assessments/impacts-offshore-oil-and-gas-industry
- eCFR — 40 CFR Part 435, Oil and Gas Extraction Point Source Category (BAT limitations): https://www.ecfr.gov/on/2024-09-06/title-40/chapter-I/subchapter-N/part-435
- China Classification Society — Rules for Offshore Oil and Gas Processing Systems (citing GB 4914-2008 produced water limits, Tier 1/2/3): https://www.ccs.org.cn/ccswz/file/download?fileid=202310131068227124
- IMO — Resolution MEPC.107(49), Revised Guidelines and Specifications for Pollution Prevention Equipment for Machinery Space Bilges (15 ppm requirement): https://wwwcdn.imo.org/localresources/en/KnowledgeCentre/IndexofIMOResolutions/MEPCDocuments/MEPC.107%2849%29.pdf
- CNPC — Oilfield Injection-Water Quality: Permeability-Graded Targets Including <5 mg/L Oil for Tight Formations (industry guidance summary, in Chinese): https://www.cnpc.com.cn/syzs/ktkf/202502/7f10217987874de7915c04a4cfbe35b9.shtml
- Yunhan Tianhe — Offshore Petroleum Development Industry Oily Wastewater Discharge Standard (GB 4914-85 two-tier limits summary, in Chinese): http://www.yhthzb.com/?p=5143/
About the author: Written by the Shanghai ChiMay Technical Editorial Team — instrumentation specialists in oil-in-water and process water monitoring for oil & gas, petrochemical and power applications.