Our roadmap to regulatory approval of the Lume — two parallel tracks: Track A begins with a Boulder Creek limited-use ATP for Colorado Regulation 93 / 303(d) compliance monitoring, then expands to Colorado state-wide and nationwide freshwater; Track B targets coastal/marine enterococci monitoring via ASBPA, independently.
The EPA's Alternate Test Procedure (ATP) program, governed by 40 CFR 136.4 and 40 CFR 136.5, allows developers to submit new or modified analytical methods for approval as alternatives to the methods listed in 40 CFR Part 136.[1] These are the required methods for Clean Water Act (CWA) compliance—specifically NPDES permit applications and discharge monitoring reports.
Currently, microbial water quality monitoring relies on culture-based grab sampling methods such as Colilert and membrane filtration. These methods require 18–24 hours for results, trained laboratory personnel, and provide only a single snapshot in time. For recreational water—swim beaches, rivers, and lakes—this means contamination events are detected too late to protect swimmers. An ATP approval for the Lume would establish tryptophan-like fluorescence (TLF) coupled with machine learning as a recognized method for continuous, real-time microbial water quality assessment, starting with recreational water monitoring where the public health impact is most immediate.
Culture-based methods require overnight incubation. Contamination events can pass undetected between grab samples.
The Lume delivers in-situ microbial estimates in 60 seconds, enabling continuous monitoring and real-time alerting.
Autonomous deployment for weeks to months captures short-duration events that weekly or monthly grab samples miss entirely.
An estimated 2 billion visits are made to U.S. beaches each year.[2] And across the country, hundreds of rivers and streams carry active E. coli impairment listings under 303(d) — yet monitoring relies on methods that are decades old, infrequent, and inherently delayed. The result is a monitoring gap that leaves swimmers exposed and makes it nearly impossible for regulators to track impairment in real time.
Only ~3,600 coastal and Great Lakes beaches are systematically monitored under the EPA BEACH Act.[3] Tens of thousands of inland swimming sites at lakes, rivers, and reservoirs lack any federal monitoring program.[4]
EPA’s BEACON database records thousands of advisory and closure actions annually at monitored beaches alone—each representing a potential public health exposure.[5]
Culture-based methods require overnight incubation. By the time results are available, water conditions have often changed—advisories may be posted when water is safe, or not posted when it is unsafe.[6][7]
Epidemiological modeling estimates approximately 90 million recreational waterborne illness cases per year in the U.S.[8]
A family arrives at a river beach on a Saturday morning. A storm two days prior caused a contamination event. The most recent grab sample was taken Monday—five days ago—and showed safe levels. The next sample won’t be collected until the following Monday. Results from that sample won’t be available until Wednesday. The contamination event has come and gone entirely undetected.
Continuous monitoring with the Lume would have flagged elevated microbial indicators within minutes of the event, enabling same-day advisories and protecting the public in real time. EPA’s 2012 Recreational Water Quality Criteria acknowledged the limitations of culture-based methods and recommended moving toward faster monitoring approaches[6]—the ATP pathway provides the regulatory mechanism to make that transition.
The EPA provides two regulatory mechanisms for ATP approval: limited-use (40 CFR 136.5, facility-specific) and nationwide (40 CFR 136.4, via Methods Update Rule). Our strategy runs as two parallel tracks, both progressing from limited-use to nationwide.
Phase 1: Boulder Creek facility-specific ATP under 40 CFR 136.5 — scoped to Colorado Regulation 93 / 303(d) compliance monitoring (E. coli, 126 CFU/100 mL geometric mean threshold), applied for by the City of Boulder Utilities in coordination with CDPHE and EPA Region 8.
A separate, independent track in partnership with the American Shore & Beach Preservation Association (ASBPA) at ocean beach sites in other states. Target analyte is enterococci, reference method is Enterolert. This track runs on its own timeline and is not a prerequisite for Track A approvals.
EPA published finalized protocols in 2018 for evaluating new and modified analytical methods.[9] Because the Lume introduces a fundamentally new determinative technique (fluorescence + machine learning rather than culture-based enumeration), it falls under the Protocol for Review and Validation of New Methods. The microbiological protocol (finalized 2010) also applies given the microbial analyte.[10]
Validation across multiple independent laboratories to demonstrate reproducibility. Each lab must independently deploy and calibrate the Lume sensor and compare results against approved reference methods.
Testing across diverse recreational water types: freshwater rivers and lakes (E. coli / Colilert), coastal and ocean beaches (enterococci / Enterolert), as well as stormwater-impacted and CSO-impacted receiving waters.
Side-by-side comparison against approved Part 136 methods: Colilert for E. coli in freshwater, Enterolert for enterococci in marine water, and membrane filtration (EPA Methods 1600/1603), demonstrating statistical equivalence or superiority in detection performance.
Formal comparability assessment following EPA's Appendix H procedures, including precision, accuracy, method detection limits, and matrix-specific performance metrics.
Complete method write-up in standardized EPA format: procedural steps, sample handling, quality assurance/quality control (QA/QC) requirements, and performance criteria.
Defined quality control procedures including calibration verification, positive and negative controls, field blanks, duplicate analyses, and acceptance criteria for each deployment.
The Lume already has a strong foundation of peer-reviewed validation data, with our most extensive dataset collected on Boulder Creek. Track A (Colorado freshwater) partners with the City of Boulder Utilities and CDPHE, starting with a Boulder Creek limited-use ATP for Regulation 93 / 303(d) compliance monitoring. Track B (coastal) partners with the American Shore & Beach Preservation Association (ASBPA) independently, targeting ocean beaches in other states.
Each track uses the appropriate EPA-recommended indicator and reference method for its water type:[6]
Peer-reviewed publications establishing TLF-ML as a viable method for microbial detection.[11][12] Over 94% categorical accuracy demonstrated on Boulder Creek and other sites. Validation across drinking water, freshwater rivers, and coastal environments. Two issued US patents covering the core technology. Existing Boulder Creek data includes R² = 0.67 and 7% MAPE in log-transformed space against Colilert.[11]
Working with CDPHE and EPA Region 8 in Denver to scope the ATP application around Colorado Regulation 93 / 303(d) compliance monitoring. Define analytes, reference methods, statistical thresholds, and study design requirements for the Boulder Creek limited-use approval. The Reg 93 / 303(d) framework — CDPHE’s impaired waters program — provides a concrete, well-defined regulatory context for the initial submission.
Install Lume sensors at City of Boulder’s 6 monitoring locations on Boulder Creek. Collect continuous TLF data paired with Boulder’s existing weekly Colilert E. coli grab samples. Boulder Creek is a CDPHE 303(d)-listed impaired waterbody with an E. coli geometric mean threshold of 126 CFU/100 mL under Reg 93. Data collection spans 12–18 months to cover seasonal variation, spring runoff, and storm events.
The City of Boulder submits a limited-use ATP application to EPA Region 8 under 40 CFR 136.5, scoped to Regulation 93 / 303(d) compliance monitoring on Boulder Creek. If approved, Boulder can use Lume data in fulfillment of its Reg 93 impaired-waterbody monitoring obligations — the first regulatory use of continuous TLF-based microbial monitoring under the Clean Water Act.
After Boulder Creek approval, expand to additional Colorado 303(d)-listed waterbodies through further limited-use ATPs at partner utilities. Pursue CDPHE state-level recognition of the Lume method for all Colorado Reg 93 / 303(d) compliance monitoring. Multiple Colorado sites provide the multi-laboratory evidence base needed for the nationwide freshwater ATP submission.
A separate, parallel track in partnership with the American Shore & Beach Preservation Association (ASBPA) at ocean beach sites in other states. Uses enterococci / Enterolert — not tied to the Colorado Reg 93 work. This track proceeds on its own timeline and leads to a coastal limited-use ATP, eventually feeding a combined nationwide submission with Track A.
Submit complete application package to EPA’s Office of Science and Technology under 40 CFR 136.4. Track A provides freshwater E. coli / Colilert data from Colorado sites; Track B provides marine enterococci / Enterolert data from ASBPA coastal sites. Track A (freshwater) may reach the nationwide threshold before Track B — a freshwater-first submission is viable, with a coastal amendment to follow. Nationwide approval establishes the Lume as a recognized 40 CFR Part 136 method across all CWA compliance monitoring categories.
The Lume’s ATP application builds on a substantial body of validation data, with our most extensive dataset from Boulder Creek—the same watershed where the regulatory validation study will take place.
Site-specific calibrated classification of microbial contamination risk across freshwater environments.[11]
Binary classification at 1 and 10 CFU/100 mL regulatory thresholds with Cohen’s kappa of 0.82–0.84.[13]
Mean absolute percentage error in log-transformed concentration space against Colilert on Boulder Creek—the same watershed as our ATP validation study.[11]
Binary E. coli classification on held-out test data from three sensors deployed along the Seine in Paris.[14]
Published validation studies in Water Research, ES&T Water, Science of the Total Environment, and EarthArXiv.
Validated across groundwater, freshwater rivers, drinking water, coastal/estuarine, and wastewater environments.
For detailed performance data, calibration charts, and the full publication list, see our Research & Performance page.
The EPA ATP program exists because analytical science advances faster than the rulemaking process can incorporate new methods. The Lume represents a fundamental shift from discrete, lab-processed grab sampling to continuous, in-situ monitoring—a transition that regulatory frameworks are increasingly prepared to support.
Approved analytical methods for CWA compliance monitoring.
Procedures for nationwide ATP applications and approval.
Procedures for limited-use (facility-specific) ATP approvals.
Method modification flexibility that labs can exercise without ATP.
Colorado Regulation 93 is CDPHE’s implementation of the federal 303(d) program for impaired waters. Boulder Creek carries an active E. coli impairment listing under Reg 93, with a TMDL-driven geometric mean threshold of 126 CFU/100 mL. The City of Boulder Utilities monitors Boulder Creek weekly against this threshold using IDEXX Colilert-18. A limited-use ATP approval would allow the Lume’s continuous TLF data to supplement or replace the weekly grab sample as the compliance measurement — providing real-time visibility into impairment events rather than a weekly snapshot.
Several convergent trends make this the right moment to pursue ATP approval for continuous TLF-based monitoring:
We are building the regulatory case for continuous microbial water quality monitoring across two tracks. Track A starts with the City of Boulder Utilities on Boulder Creek (Colorado Reg 93 / 303(d) compliance monitoring) and expands to other Colorado utilities and impaired waterways. Track B targets coastal/marine monitoring with ASBPA at ocean beach sites in other states. We are seeking additional Colorado utilities with 303(d)-listed waterbodies, coastal beach operators, state regulators, and research partners interested in deploying the Lume alongside existing Colilert or Enterolert monitoring programs.
Get in Touch