Data inputs
Potential inputs include satellite imagery, sensor readings, industrial process parameters, meteorological data, source profiles, and ambient monitoring records, depending on the project.
Methodology
Research-grounded, clearly scoped environmental evidence.
Zolena Lab combines satellite remote sensing, low-cost sensors, industrial process data, machine learning, and atmospheric science methods where suitable data and project conditions exist.
Analytical methods
Methods may include machine-learning calibration, PEMS modeling, receptor modeling, regional transport analysis, OFP analysis, and structured evidence reporting.
Boundaries
Each method has defined limits. Zolena Lab provides data, analytical evidence, and report-ready technical support; regulated reporting, legal compliance, carbon-credit verification, or formal assurance requires the relevant qualified parties and review processes.
Research visuals
Method evidence should be visible, sourced, and bounded.
.jpg){kind=link}
PEMS method
Predictive monitoring uses site-specific industrial data.
Zolena Lab's PEMS approach uses existing facility process parameters and machine-learning models to predict combustion-related emissions such as NOx. The method is designed for industrial facilities where operating data are available, relevant to emissions behavior, and suitable for model training and validation.
A typical PEMS model may use eight core process parameters: generator output, fuel temperature, exhaust temperature, O2 concentration, fuel flow, exhaust heat balance, turbine speed, and compressor air temperature.
The reference PEMS case reported 28 months of industrial field validation, 99.93% data availability, MAE of 0.5982, r value of 0.9451, and a 0.14% total-emissions difference in the test set. The reference case was evaluated against US EPA PS16, European CEN/TS 17198:2018, and Alberta AER/RATA-related requirements. These values describe a specific reference case; they should not be read as universal performance guarantees for every facility.
Methane QMRV
QMRV is a data and evidence workflow, not a legal compliance service.
For methane monitoring, Zolena Lab uses QMRV to describe a workflow connecting quantification, monitoring records, report-ready datasets, and evidence organization. Depending on project conditions, a methane workflow may combine satellite screening, continuous low-cost sensor networks, machine-learning calibration, hotspot identification, anomaly detection, and trend analysis.
Zolena Lab does not provide statutory compliance determinations, legal verification, regulatory approval, carbon-credit certification, or formal assurance opinions. Our role is to provide monitoring data and analytical evidence that customers, field partners, qualified professionals, verifiers, or regulators may use within their own processes.
Source attribution
Different attribution methods answer different questions.
PMF and CMB should not be treated as a fixed combined model. PMF is a receptor-modeling method that requires suitable monitoring data to infer major sources and relative contributions. CMB depends on reliable source profiles and is appropriate only where representative source signatures are available.
CALIPSO aerosol extinction data and MERRA-2 meteorological reanalysis can support identification of regional particulate transport events and their relationship to local air quality. They should not be described as directly measuring cross-regional particulate mass flux without additional evidence and method support.
Satellite evidence
Satellite data supports screening and documentation.
Satellite-supported applications such as EUDR evidence packages, land environmental risk screening, property ESG evidence reports, and green-space stewardship reports use observable environmental signals where suitable imagery and project context exist.
Satellite evidence can support due diligence, screening, and environmental communication, but it cannot replace all field investigation, subsurface sampling, laboratory testing, legal review, or local professional judgment.
Related