2. Data Validation

2.1 Data Time Stamps

All continuous and non-continuous data in the province are to be collected and reported in standard time throughout the year. In most of the province Eastern Standard Time (EST) is to be used, however, in north-western Ontario (parts west of longitude 90° west), Central Standard Time (CST) is to be used. For continuous data, this applies not only to the ‘raw’ real-time data but also to the means calculated from this data. The ministry recommends that computer clocks used with real-time data acquisition systems be checked at least monthly against a reliable time standard such as the computer system’s server clock. The 24:00 hour system is to be used: the first hour of the day is to be shown as 0:00 hr, such that a 24-hour monitoring period is from 0:00 to 23:00 hrs. An hour is defined as the 'hour beginning', for example, the hourly mean for 14:00 hours is computed from readings collected from 14:00 EST to 14:59 EST hours. This reporting convention is applicable to shorter-time averaging periods, for example, half hour, five minute and other time resolutions.

2.2 Method Detection Limits

The Method Detection Limit (MDL) will be the limit used to qualify data in order to ensure that a simplified and standardised approach is used across the province in dealing with measurements that are reported as being below the limit of the sampling and analytical methods employed. The ministry may also accept low-level data using data qualifiers, such as W to determine the limit of measurement and T as the limit of reliability. The ministry’s recommended approach for calculating the MDL is found in Appendix 6.

As previously stated, the ministry requires an MDL which is at least a factor of ten (one order of magnitude) lower than the applicable O. Reg. 419/05 schedule 1, 2 and 3 standards, POI guidelines and Ambient Air Quality Criteria (AAQC). Site operators or emitter’s environmental representatives should contact their ministry regional office at the program planning stage if it is suspected that the analytical method(s) contemplated for the proposed air monitoring program may not be able to meet this requirement.

The analytical lab must provide the MDL for each substance analyzed in the final engineering units reported to the ministry (e.g., µg/m3). In addition, the lab will report a result with a value lower than the MDL as ‘< 0.00x’ for an MDL of 0.00x.

For the purpose of performing statistical analyses (e.g., means, etc.) and in keeping with a commonly accepted practice, a value of half the MDL must be substituted for concentrations less than the MDL. These values will be counted as non-detects. The percentage of the data with values above the MDL should also be reported.

The goal of this approach is to provide a level of consistency for emitters reporting non-continuous data to the ministry. For data generated by continuous monitors, the analyzers have MDLs typically much lower than the applicable limits. In this instance, DASs report values less than MDLs as zeros. It is common practice to report these as zeroes in the database and to treat these as zeroes in calculating short and longer term means.

In the event that stated MDLs are comparable to or greater than applicable standards, especially for substances whose standards are based on human health impacts or substances with Upper Risk Thresholds (URT) (see schedule 6 of O. Reg. 419/05), a more refined statistical approach may be required to deal with MDLs if the data are to be used to estimate risks. This would occur when there is a need for risk assessors to impute a series of values that represent concentrations below the stated MDL, and when the non-detects comprise a significant portion of the data set. This is to be handled on a case-by-case basis in consultation with ministry experts.

2.3 Meteorological Data

The submission of meteorological data to the ministry as part of air quality monitoring activities does not constitute approval of the data for regulatory purposes. If there is any intention to use meteorological data for modelling under O. Reg. 419/05, the meteorological instrumentation must meet the requirements of O. Reg. 419/05 and be included in the audit program. Approval for the use of site specific meteorological data for modelling must be obtained from the ministry as per the requirements of O. Reg. 419/05. The approval form is entitled Request for use of site specific meteorological data and is available through the Air dispersion modelling resources and guides. Notwithstanding the regulatory use of meteorological data, all meteorological instrumentation generating data as part of air quality monitoring will be subject to audit.

The standard equations for calculating wind speed and direction values are shown in Appendix 5.

2.4 Data Validation/Editing

A data validation process to filter out erroneous data is critical to maximize data integrity. Validation can be done using automated or manual procedures. Regardless of the process followed, judgment to accept or reject suspicious or unusual data is required. Many factors need to be considered in this process, which requires regular inspection of all data by experienced staff that have an understanding of local pollutant and climatic conditions as well as knowledge of air pollution principles and analyzer behavior.

Emitters will be responsible for ensuring that the data editor follows the recommended data editing and validation protocols outlined in Section 4.1 for continuous data and in Section 4.2 for non-continuous data. These are minimum requirements. Staff responsible for data editing and validation can also refer to, and use, protocols established by other agencies (e.g., ECCC, US EPA), as long as the requirements outlined in in this manual are met.

2.4.1 Continuous Data

As per the guidance in Section 4.1, raw and edited continuous data must be submitted electronically to the ministry on a quarterly basis. The data is to be submitted in a resolution of half-hour means. The ministry’s regional office may request alternate data resolutions. The resolution required to be submitted is subject to change if the averaging period for the limits change. The emitter is advised to contact their regional ministry office to confirm the data resolution required as well as associated reporting requirements.

Automated (DASs) used by emitters or site operators must meet the requirements for acceptable data editing and validation.

Section 6.2 outlines data editing requirements for continuous data; the following general rules apply:

  • Editing of continuous data, as stated above, will be done on half-hour resolution data. The ministry may periodically request the editing and submission of data collected at a resolution of 5 minutes, for example in the case of an exceedence or a spill. Data collected with resolution times shorter than 5 minutes (e.g., 1 minute means) are to be left as is, as a permanent record. This may be especially important in cases, such as where the data is used to assess compliance with 10 minute odour standards.
  • For ministry submissions, a minimum of 75% of the readings within an averaging period must be valid for the means to be considered valid. For example, a valid 1- minute mean requires at least 45 1-second readings, a valid 5 minute mean requires at least 4 1-minute means and a valid 1 hour mean requires at least 45 1-minute means. This qualification must be applied for the determination of % valid data collected.
  • For the purpose of calculating a valid annual mean, at least 75% of the half-hour or hourly means must be valid, that is, 6,570 valid hours out of a total of 8,760 hours, are required, or a valid monthly mean requires 23 valid daily means, with the exception of February. A valid 24-hour mean requires at least 18 valid hours out of 24 hours. Additionally, for a good representation of quarterly or seasonal means at least 75% of the data in each quarter or season must be valid.
  • The tolerance limits for editing of hourly data are ±10% of the calibration standard.
  • Data should be edited within 30 days of the end of the month in which it was collected.
  • Zero drifts, beyond 2 ppb for sulphur dioxide, ozone, oxides of nitrogen and total reduced sulphur (SO2, O3, NOx and TRS), and 3 µg/m3 for suspended particulate matter (SP), require an off-set adjustment.
  • Zero drifts, beyond 0.2 ppm for carbon monoxide (CO), require an off-set adjustment.

All edits of continuous data must be reported to the ministry in an edit log table such as the example shown in Appendix 4 (Table 12). Reasons for edit actions must be included in the edit log table.

Data Storage/Backup

The ministry requires that all continuous data for the current and previous calendar year be stored in the data acquisition central computer. Older data should be archived in an acceptable format, in a reliable electronic medium, and must be stored by the emitter for a period of at least 7 years after its collection. It is recommended that the continuous data be backed up at least weekly.

2.4.2 Non-continuous Data

The data editing SOPs outlined in section 6.3 are based on those currently used by the ministry and are considered to be minimum requirements. Additional guidance is also provided in field method documents referenced in the SOPs and in the ministry’s Laboratory Methods/Procedures documents. The use of other sample and data validation procedures is acceptable as long as the requirements for acceptable data editing and validation are met.

For air quality monitoring purposes, non-continuous data is usually obtained from the collection of monthly (30 days) samples (e.g., dustfall, fluoridation candles) and daily (24-hour) samples (e.g., TSP, PM10, VOCs, PAHs, etc.). The daily samples are 24-hour samples, and sampling starts at midnight; the start time may be altered with the approval of the ministry. Particulate samples are collected every sixth day on the North American schedule. A different schedule (i.e. every third day) is sometimes used or required to collect more samples in a given year. The sampling frequency for polycyclic aromatic hydrocarbons (PAHs), volatile organic compounds (VOCs), and dioxins/furans (PCDD/PCDF) typically occurs every twelfth day on the North American schedule. These sampling schedules, or other schedules possibly required for special studies or assessments, need to be reviewed by the ministry for concurrence. The sampling schedule for this year’s 1-in-3, 1-in-6 and 1-in-12 day schedules can be downloaded from the US EPA.

The following general rules apply to editing non-continuous data:

  • A monthly sample is considered to be valid if the exposure period is within ±3 days of the 30 day period (calendar month); it is also desirable to have the 'on' and 'off' dates as close as possible to the start and end of a calendar month in order to minimize uncertainty in the determination of the exposure month.
  • The tolerance limit for editing daily parameters is ±10% of the air flow calibration standard.
  • A daily sample is considered to be valid if the sampling period is within ±10% of the required 24 hours, that is, from 21.6 to 26.4 hours. Additionally, for SP and PM10 high volume (Hi-Vol) samples, the air volume sampled over this time period must be within ±10% of the required theoretical total air volume of 1631 m3, that is, from 1468 m3 to 1794 m3.
  • For PM10 or PM2.5 samples collected with a discrete sampler (e.g. PQ100 and PQ200) designed to operate with a flow rate of 16.7 liters per minute (L/min), the air volume sampled over a 24 hour period must also be within ±10% of the required theoretical total air volume of 24 m3, that is, from 21.6 to 26.4 m3.
  • A valid annual arithmetic or geometric mean requires at least 75% of the total number of possible samples under the relevant sampling frequency to be valid. Hence, for monthly sampling, at least nine months of valid data are required to calculate an annual mean, whereas for daily sampling the following number of valid 24-hour sample results are required:
    Sampling schedule Number of valid samples required
    Every 3rd day 90
    Every 6th day 45
    Every 12th day 23
  • For calculating quarterly or seasonal means, at least 75% of valid data for each quarter or season of the year must be available.

All edits of non-continuous data must be reported to the ministry in an edit log table such as the example shown in Appendix 4 (Table 12). Reasons for edit actions must be included in the edit log table.

3. Data Reporting

This section discusses the various reporting requirements. Depending on the purpose of the monitoring, reporting requirements may be different. Requirements are usually identified during the development of the monitoring plan and through communications with the ministry. This guidance outlines the ministry’s minimum reporting requirementsfootnote 1.

3.1 Exceedance Notification

For all monitoring data, the emitter shall notify the ministry, as soon as practicable or as otherwise required, of measured exceedances of the limits specified in O. Reg. 419/05, in Environmental Compliance Approvals (ECAs), in orders, or of measured exceedances of other statutory or regulatory limits. Section 28 of O. Reg. 419/05 requires the emitter to notify a provincial officer as soon as practicable in writing of modelled or measured exceedances of the standards in the Regulation, or of discharges that may cause an adverse effect.

Some contaminants are not listed in Schedules 2 and 3 of the Regulation, but are found in the document entitled Summary of Standards and Guidelines to Support Ontario Regulation 419/05 –Air Pollution – Local Air Quality. As well, if the monitoring is being done to support an Environmental Assessment (EA) or a general air quality study in a community, proponents should refer to the contaminants that are listed in the document entitled Ontario’s Ambient Air Quality Criteria.

An exceedance of a guideline value may cause adverse effects, and as such, would trigger the requirement to notify a provincial officer if it is the result of an emission from a particular source. As such, information related to both exceedances of standards and guideline values are captured in the Notification of Exceedance Form under Notifications of exceedances.

Some air monitoring may be conducted to assess community air quality rather than evaluate the performance of a particular source or industry. The resulting measured concentrations should be compared to Ambient Air Quality Criteria (AAQC). In such cases, concentrations above do not necessarily trigger mandatory notification to a provincial officer and the subsequent follow-up required under Section 29 of the Regulation. Nonetheless, the ministry should be made aware of such incidents so that staff may respond appropriately. Thus, operators are requested to communicate these episodes as soon as practicable.

With respect to Upper Risk Thresholds (URT), Subsection 30 (3) of O. Reg. 419/05 requires that a person immediately notify the ministry in writing if there is any reason to believe, based on any relevant information (e.g., unrefined modelling, refined modelling, monitoring, etc.), that discharges of a contaminant may result in the concentration of the contaminant exceeding a URT listed in Schedule 6.

Ministry Point of Impingement (POI) standards, guideline values and URTs can be found in Summary of Standards and Guidelines to Support Ontario Regulation 419/05 – Air Pollution – Local Air Quality, as amended from time to time. Ambient Air Quality Criteria can be found in the document Ontario’s Ambient Air Quality Criteria, as amended from time to time. These documents can be found at the ministry’s Rules on air quality and pollution website.

3.2 Exceedance Reporting

Monitoring to determine non-compliance with air standards (or non-conformance with a guideline value) may take the form of samples with durations which match the averaging period specified in the regulation for that contaminant. As well, it may consist of a series of shorter consecutive samples which can be combined to give an average over the required period. If these samples are taken continuously, any consecutive set of them that spans the specified averaging period for that contaminant may be used to assess off-site concentration. Consecutive averages of these measurements are sometimes called running averages or rolling averages. They are also sometimes called a moving mean or rolling mean. A running average is used to analyse measurements by creating a series of different subsets of a larger data set. This document will refer to these as running averages for ambient air monitoring.

For the purpose of reporting to the ministry, when determining the number of exceedances for an averaging period using shorter period measurements, there may be multiple consecutive running averages that lie above a standard or other limit. Since each individual measurement reflects one period of emission and may be used to calculate a series of running averages (both before and after the individual measurement), it is not necessary to report multiple exceedances within a single averaging period. The exceedance reported should be the first value calculated for the averaging period that exceeds the standard. As well, the range of exceedances should be reported for that specific time period. This is clarified in the following two examples; however, for greater detail please refer to Appendix 7.

Example A
A substance may have a 24-hour standard specified in the regulation. At a particular station, a monitor may report concentrations based upon shorter averaging periods (e.g., hourly). Thus, any consecutive 24 one-hour concentrations may be averaged to compare to the standard. An average that is greater than the 24-hour standard constitutes an exceedance, except when an exceedance has occurred within the last 23 hours. For example, if the samples taken between 03:00 EST and 02:59 EST on the following day are used to compute a daily average which is higher than the standard, then an exceedance has occurred.

Example B
In the previous example, should the samples between 05:00 EST on the first day and 04:59 EST on the following day also average to a value which is higher than the standard, this would not be considered a separate reporting requirement as it relies, in part, on the same data as the previously reported exceedance.

For the purposes of comparing monitoring data to O. Reg. 419/05 standards, and URTs, ECAs, order limits or guideline values for any contaminant, the ministry requires the following information to be reported:

  • When the exceedance occurred;
  • The value of the first running average that caused an exceedance; and,
  • The range of running averages for that specific averaging period.

Please note in some cases the ministry may request all data points used in the calculation. This protocol is further explained in Appendix 7 with examples showing how to report the number of exceedances to the ministry.

4. Air Quality Monitoring Data Collected by Emitters

Emitters that operate ambient air quality monitoring stations off property as part of the Source Emissions Monitoring (SEM) program, or as a requirement within an Environmental Compliance Approval (ECA), an order or other legal instrument, have specific reporting requirements which are described within this section. Alternative data reporting arrangements can be made upon approval of the ministry.

4.1 Continuous Monitoring

The ministry may request continuous data to be available in real time in an acceptable format. Additionally, data (continuous and non-continuous) along with associated reports, are to be submitted to the ministry on a quarterly and/or annual basis as described below. Submitted monitoring data will be uploaded into a ministry database.

With the reliability of current technology and the implementation of a strong preventative maintenance program, automated continuous air quality monitoring systems can achieve a high level of valid data collection. Based on many years of ministry experience with the operation of such systems, a minimum target of 90% valid data collection per quarter per parameter (i.e., at least 1966 hours of valid data out of a total maximum of 2184 hours) can be routinely attained. Emitters and site operators should try to better this objective and meet a desirable target of 95% valid data (at least 2075 hours of valid data out of a total maximum of 2184 hours).

Notwithstanding the minimum 90% valid data performance measure, emitters shall notify, as soon as practical, the ministry of any system or equipment failures resulting in missing data of 24-hours or more in length and of the plans and schedule for repairing the failed system or equipment. This is to be followed by a notification to the ministry when the problems have been resolved and the annual report should detail the events and remedial actions taken.

Quarterly Data Reporting

Raw and validated (edited) data are to be submitted (electronically) quarterly to the ministry in time series format, either in Excel or Comma Separated Values (CSV) files, as per the guidance in Appendix 3. Prior to the initial submission of data, the ministry must be contacted to initialize the emitter station in the database and to confirm the acceptability of the data format (Appendix 3). For stations that have 10-minute to hourly limits that apply to them, and determine their exceedances based on rolling averages as described in Section 3.2, the data must be submitted in a 5-minute resolution, unless another resolution has been approved by the ministry. For stations that have 24-hour limits, data must be submitted as hourly averages. In cases where more than one averaging period may apply to a monitored substance, submitted data must be able to address the shortest-term standard applicable to that substance.

The data validation and editing rules discussed in Section 2.4 must be followed for data reports and data submission.

The data must be submitted within 45 days of the end of each quarter, based on the calendar year, that is, first quarter by May 15, second quarter by August 14, third quarter by November 14 and fourth quarter by February 14. Data editing protocols must be followed as per Section 6.2. Data may not be deleted from a data set – it may however be flagged as unsuitable with proper justification and not used in summarising the data; however, the ministry reserves the right to override these decisions. Edits made to the original data must be summarised in an edit log table, as per the example in Appendix 4, and submitted with the edited data.

The units of measurement for various continuously monitored contaminants are listed in Table 1.

Quarterly Report

A quarterly summary report must be submitted along with the data, unless alternate arrangements have been made and approved by the ministry. Do not include a hard copy of the data with the report. The report shall include the following statistics for each measured pollutant parameter:

  • Arithmetic Mean
  • Monthly Arithmetic Mean
  • Maximum for any averaging period for which the data is used for comparison to any limit applied to the emitter; and,
  • Maximum 24-hour, % valid hours, or other averaging period as appropriate.

Please submit only one hard copy to your facility’s ministry District Manager/Director or as directed by the District Manager/Director. Submit an electronic copy of the data to the appropriate address below depending on the region:

Table 1: Units of Measurement Associated with Continuously Monitored Parameters
Continuously monitored parameter Units of measurement
SO2, TRS parts per billion by volume (ppbv)
NOx/NO/NO2 parts per billion by volume (ppbv)
Particulate matter (SP, PM10 and PM2.5) micrograms ⁄ cubic metre (µg/m3)
Wind speed2 Kilometer per hour (km/hr)
Wind direction1,2 Degrees clockwise from north (deg)
Temperature Degrees Celsius (°C)
Flow3 cubic metres per second or cubic feet per minute (m3/s or CFM)
VOCs parts per billion by volume (ppbv)

1 to be taken as the direction it is blowing from, expressed as degrees clockwise from north (i.e., wind blowing from the north is 0° ⁄ 360°, from the east is 90°, etc.

2 to be computed using the equations listed in Appendix 5

3 flows should not be corrected for humidity, pressure, temperature

Additionally, the report must include the number of times that the emitter exceeded a standard in O. Reg. 419/05, a guideline, an AAQC, a limit in an ECA or order, or any other legal limit which applies to the emitter’s facility. The emitter must report all exceedences for each applicable averaging period. This is based on running averages, as discussed in Sections 3.1 and 3.2.

4.2 Non-Continuous Data

A target of 90% valid data collection per quarter for each parameter should be attainable with good routine maintenance and implementation of sound QA/QC practices. The data validation and editing rules discussed in Section 2.4 must be followed for data reports and data submission.

Raw and validated data from non-continuous sampling programs are to be submitted quarterly to the ministry in Excel or CSV format (Appendix 3, Table 11). The units of measurement for various sampled contaminants are listed in Table 2. Data are to be submitted electronically within 45 days of the end of each quarter of the calendar year:

  • 1st quarter due May 15;
  • 2nd quarter due August 14;
  • 3rd quarter due November 14; and,
  • 4th quarter by February 14 of the following year.

If full quarter results are not available from the laboratory, supply all data available, and provide missing results in the following quarterly report. The monitoring schedules are specified in the respective SOPs (see Section 6). If any data correction was performed, an edit log table shall be submitted as per the example in Appendix 4.

Minimum Routine Scans for Non-Continuous Parameters

The ministry may require minimum routine analytical scans for contaminants of interest or concern such as: metals and/or anions in particulate matter (SP, PM10 and PM2.5), speciated VOCs, speciated PAHs, and speciated dioxins and furans (PCDD/PCDF). The emitter should contact their regional office of the ministry during the development of the monitoring plan (Section 1.5) to establish the list of parameters for which to analyse and confirm the required Method Detection Limit (MDL). Subsequent changes to the parameters list need to be done in consultation with and with approval by the ministry.

Quarterly Report

A quarterly summary report must be submitted along with the data. These reports should be combined with any continuous data quarterly reports. The report shall include the following statistics for each measured pollutant parameter:

  • No. of valid samples;
  • % valid data;
  • Arithmetic mean;
  • Period geometric mean (SP only);
  • Maximum 24-hour value;
  • Maximum monthly value; and,
  • Sampling dates (start and end) (for example, dustfall, fluorides, etc.).

Additionally, the report shall include the number times that the emitter exceeded a standard in O. Reg. 419/05, a guideline, a AAQC, a limit in an ECA or order, or any other legal limit which applies to the emitter’s facility. The emitter must report all exceedences for each applicable averaging period.

Please submit only one hard copy to your facility’s ministry District Manager or as directed by the District Manager. Submit an electronic copy of the data to the appropriate address below depending on the region:

Table 2: Units of Measurement Associated with Non-Continuous Samples
Sampled Pollutant Unit of Measurement
Dustfall grams ⁄ square metre ⁄ 30 days (g/m2/30d)
Fluoridation rate micrograms of fluoride ⁄ 100 square cm ⁄ 30 days (mg F/100 cm2/30d)
Particulate matter (SP, PM10 and PM2.5) and its constituents (it is called “suspended particulate” in the standard listing) micrograms ⁄ cubic metre (µg/m3)
VOC micrograms ⁄ cubic metre (µg/m3)
PAH nanograms ⁄ cubic metre (ng/m3)
Dioxins/furans (PCDD/PCDF) picograms ⁄ cubic metre - expressed in toxicity equivalents (pg/m3 – TEQ)
Flow1 actual cubic metres per second or cubic feet per minute (m3/s or CFM)

1 flow should not be corrected to temperature, humidity, or pressure with the exception when conducting audits

Annual Reports

The data validation and editing rules discussed in Section 2.4 must be followed for all data reports and data submission.

By May 15 of each year, emitters shall provide an annual summary report to the ministry, with interpretation, of the results obtained in the previous calendar year. This report is a summary of annual operations and data, along with interpretation. Do not include a hard copy of the data within the report. The report shall include:

  • A map showing the location of emitting sources, property boundaries, major structures on site, and monitoring stations; the map must also include a distance scale, a north indicator (arrow), and the location of any significant nearby receptors
  • A summary of overall operations (e.g., summary of parameters monitored and equipment/model numbers, frequency of site visits and calibrations, confirmation of data backups and/or archiving, list of problems) that resulted in significant losses of data, along with remedial actions. Do not include copies of station log entries.
  • A summary of audits and audit outcomes. Do not include copies of the audit reports.
  • Summary statistics, including:
    • Annual Arithmetic Mean
    • Annual Geometric Mean (SP only)
    • Maximum 1-hour (continuous data only)
    • Maximum 24-hour
    • Number of valid hours or sampling periods
    • % valid data
  • A summary of exceedences including the number times that the emitter exceeded a standard in O. Reg. 419/05, a guideline, an AAQC, a limit in an ECA or order, or any other legal limit which applies to the emitter’s facility. The emitter must report all exceedences for each applicable averaging period.
  • All exceedences of criteria, standards or reporting thresholds are to be evaluated by wind speed/direction data for source contribution assessment: This may include “pollution rose” or “wind rose” analysis for continuous data.
  • Comparison to historical data collected by emitters, (preferably using graphics),e.g., comparison of statistics to any previous years' statistics.
  • Evaluation of effects (if any) on monitoring results by abatement actions.

Please submit only one hard copy to your facility’s ministry District Manager or as directed by the District Manager. Submit an electronic copy of the data to the appropriate address below depending on the region:

Footnotes

  • footnote[1] Back to paragraph Note that this guidance outlines the ministry’s minimum expectation with respect to reporting and also informs what may be reported publicly in an Environmental Compliance Report. It does not speak to the number of potential offences which could occur in that averaging period based on the application of running averages.