What Is Normal Air Quality? How Background and Baseline Concentrations Are Chosen

Do we need to know what is normal air quality? When an industrial facility proposes a new source of emissions, one of the first questions in an air quality assessment is deceptively simple:

What is already in the air?

Before we can determine whether a new project may cause an air quality standard to be exceeded, we need to understand the existing concentration of the substance being assessed.

This is sometimes called background air quality, baseline concentration or simply existing ambient concentration. The terminology and methods vary somewhat between jurisdictions, but the underlying problem is the same:

You cannot assess the impact of a new emission source without first understanding the air quality it is being added to. A dispersion model can estimate the concentration contributed by a proposed stack, flare, facility or other source. But that is only part of the picture.

In many assessments, the practical calculation is:

Existing air quality + Project contribution
= Predicted total concentration

The difficult part is often more than running the dispersion model. It is deciding what number reasonably represents what is normal air quality at the project location.

That requires data, regulatory guidance and professional judgment.

What is Normal Air Quality, Really?

What's in your air?

There is no single concentration that represents "normal air quality" everywhere.

Air quality changes with location, season, weather, nearby industry, transportation, agriculture, forest fires, natural sources and many other factors.

A rural location surrounded by agricultural land will not necessarily have the same background concentrations as an industrial area. A forested region may have different particulate concentrations than a city. A monitoring station beside a major highway may not represent a proposed facility located far from traffic.

So, in air quality dispersion modelling, what is normal air quality does not necessarily mean:

• the cleanest air ever measured;
• the average concentration everywhere in a province;
• the highest concentration ever recorded at the nearest monitor; or
• simply the concentration from the closest monitoring station.

Instead, the goal is generally to select a value that is representative of the conditions expected at the project site and appropriate for the purpose of the assessment.

That sounds straightforward. In practice, it can require considerably more thought than simply downloading a number from a monitoring database.

The Number You Need Before You Judge the Project

Imagine that monitoring data show what is normal air quality, an existing concentration of a contaminant in the area. Now suppose an air dispersion model predicts an additional concentration from a proposed facility. Neither number necessarily tells the whole story by itself.

What are your air pollution risks?

The monitoring data may represent existing emissions from industry, transportation, natural sources or other activities. The model estimates the contribution from the project being assessed.

Together, they can provide an estimate of the future total concentration. This is why selecting an appropriate background or baseline concentration matters.

If the background concentration is unrealistically low, the assessment may underestimate the total air quality impact. And if it is unrealistically high or represents conditions that do not apply at the project site, the assessment may become unnecessarily conservative or misleading.

Rather than to find the biggest available number the objective is to select a value that is technically defensible and appropriate for the assessment.

How We Usually Select Background Air Quality Data in Alberta

In Alberta, our experience has generally been to begin with the available ambient monitoring network.

Where suitable data are available, we typically look for:

• The nearest appropriate monitoring location;
• Continuous hourly data for the substance being modelled;
• A recent three-year period of data; and
• A monitoring location in a setting similar to the project site.

That last point is particularly important. The nearest monitoring station is not automatically the best monitoring station.

A station should, as far as reasonably possible, represent what is normal air quality for the location being assessed. For example, we consider whether the project and monitoring site are located in a similar:

• urban setting;
• rural setting;
• industrial area;
• forested area;
• agricultural region; or
• other relevant environmental setting.

The surrounding sources also matter.

A monitoring station located beside a major highway may accurately represent air quality at that location but may not necessarily represent a rural industrial facility located many kilometres from significant traffic.

Similarly, an industrial monitoring station may be appropriate for one project and completely inappropriate for another, depending on the emissions and surroundings.

The process is therefore more than: "Which station is closest?"

A better question is: "Which available monitoring data are most likely to represent the existing air quality conditions at this project?"

That distinction is where experience becomes important.

Background, Baseline and Existing Air Quality: Are They the Same Thing?

The terms are sometimes used interchangeably, but they can have different meanings depending on the jurisdiction and assessment.

In general, background concentration often refers to the concentration already present from sources other than the project being modelled. Those sources may include:

• other industrial facilities;
• transportation;
• residential and commercial activity;
• agriculture;
• natural emissions;
• regional transport of pollutants; and
• other local or regional sources.

Some guidance distinguishes between regional background concentrations and a more localized baseline concentration.

The distinction can be useful. A regional background concentration may describe pollution that is relatively widespread over a larger area. A local baseline may also reflect nearby activities and sources that influence air quality at a particular site. 

The terminology matters less than the fundamental question: What existing concentration should reasonably be combined with the modelled contribution from the project? What is normal air quality here?

That value should be clearly explained and justified in the assessment.

How Do You Choose the Right Monitoring Station?

This is one of the areas where air quality modelling cannot be reduced to pushing a button.

natural sources of air pollution

Suppose three monitoring stations are available.

One is closest to the project but located in a heavily industrialized area.

Another is farther away but located in a rural environment similar to the proposed facility.

A third has the desired contaminant data but only intermittent measurements.

Which one should be used? There may not be a universal answer. A reasonable selection process can include considering:

Distance - The monitoring site should generally be close enough to have some relevance to the project location. But proximity alone is not sufficient.

Similar surroundings - Is the monitoring station located in an environment similar to the proposed project? A rural station may be more representative of a rural project than a closer urban station.

The contaminant being assessed - The monitoring station needs data for the substance being modelled. A nearby station with excellent ozone data does not solve a background concentration problem for a substance it does not measure.

Continuous data - For refined assessments, continuous hourly data can be particularly valuable because air quality standards often depend on relatively short averaging periods.

Hourly data also provide a much clearer picture of how concentrations vary over time than occasional measurements.

A sufficiently recent period - Air quality conditions can change.

What is normal air quality if it keeps changing? Facilities begin operation, shut down or change their emissions. Transportation patterns change. Wildfire activity varies. New developments can alter local emissions.

Using a recent period of data helps ensure that the selected baseline reflects reasonably current conditions.

Data completeness and quality - Before using monitoring data, the data should be reviewed for completeness and quality.

Missing data, instrument problems, unusual periods or other issues may affect whether the record is appropriate for the intended assessment.

Why Three Years of Data Can Be More Useful Than One

Air quality changes from year to year. One year may include unusual weather, major wildfires, construction activity, an industrial upset or other events that are not representative of longer-term conditions.

Air quality in cities and rural areas

Using a recent three-year period provides a broader sample of meteorological and emission conditions and determine what is normal air quality. It can help answer a more useful question: What concentrations have reasonably occurred at this type of location over recent years?

That does not mean that three years automatically solves every problem. The data still need to be appropriate for the averaging period and the regulatory method being used.

But relying on a longer period generally provides more information than selecting a single year simply because it contains the highest or lowest concentration.

What Happens When There Is No Suitable Monitoring Station?

This is where the problem becomes more difficult. Not every contaminant is monitored everywhere.

Some proposed facilities are located far from permanent monitoring stations. Other sites may have monitoring data, but the data may not represent the environmental conditions around the project. In those situations, possible approaches may include:

• selecting data from a similar location;
• using another appropriate regional monitoring network;
• using regulatory default or published background values;
• estimating concentrations using other available information;
• conducting additional monitoring; or
• discussing an appropriate approach with the regulator.

The important point is to recognize the limitation. A number should not suddenly become highly precise simply because it has been entered into a dispersion model.

If the background concentration is uncertain, that uncertainty should be understood and, where appropriate, discussed.

Alberta: Establishing a Defensible Baseline

Alberta air quality assessments can require background concentrations to be added to modelled project concentrations. In practice, the challenge is often finding monitoring data that are both available and representative. Our starting point is generally practical:

Find the nearest suitable monitoring site with continuous hourly data for the substance being assessed over a recent three-year period.

Then look beyond distance. Does the site represent similar surroundings? Is it urban, rural, industrial, agricultural or forested?

Are there major local sources near the monitoring station that do not exist near the project? Does the monitoring record appear sufficiently complete to answer what is normal air quality?

These questions can matter as much as the distance between the monitor and the project.

British Columbia: Background Concentrations and Assessment Level

British Columbia's air quality dispersion modelling framework provides different approaches depending on the level and purpose of the assessment.

Monitoring data can be used to establish existing concentrations where appropriate data are available. The method used to select a background concentration may depend on factors such as:

• the contaminant;
• the averaging period;
• the available monitoring record;
• the level of assessment; and
• the purpose of the modelling study.

For more detailed assessments, the selection of the background concentration should be supported by an explanation of why the data are representative of the project location.

This is particularly important in British Columbia because conditions can change dramatically over relatively short distances. A valley, coastal location, industrial corridor and interior plateau may experience very different meteorological and air quality conditions.

Saskatchewan: Background and Baseline Concentrations

Saskatchewan guidance distinguishes useful concepts relating to regional background and more localized baseline concentrations.

The key issue remains the same: model results should not be interpreted as though the project is the only source of the contaminant.

Existing concentrations may include contributions from natural and anthropogenic sources. Monitoring data should be reviewed for quality and representativeness before being used.

In urban or industrial areas, a simple regional value may not adequately represent local conditions. Conversely, using a highly localized concentration influenced by an unrelated source may exaggerate conditions expected at the project site.

Again, selecting the right number requires understanding what the monitor is actually measuring.

Manitoba: Using Representative Existing Air Quality Data

Manitoba also recognizes the importance of representative background concentrations when evaluating potential air quality impacts. Where suitable monitoring data are available, the monitoring site should reasonably represent the study area.

If representative data are unavailable, it may be necessary to consider other sources of information or discuss an appropriate approach with the relevant authority.

For larger projects in areas without suitable data, additional monitoring may sometimes be appropriate.

The Hard Part: Conservative Does Not Always Mean Appropriate

One temptation in environmental modelling is to assume that selecting the highest available number is automatically the safest approach.

Sometimes a conservative approach is appropriate. But conservative and representative are not necessarily the same thing.

Imagine a monitoring station recording high concentrations because it is directly affected by a particular industrial source.

If that source does not influence the proposed project location, using that concentration without examining its relevance may not accurately describe existing conditions at the project site. The same issue can arise with:

• traffic-influenced monitoring stations;
• unusual industrial events;
• temporary construction activities;
• wildfire periods;
• facilities that have since shut down; or
• changes in regional emissions.

Professional judgment involves understanding the data rather than simply selecting the largest value in a spreadsheet.

A technically defensible assessment should be able to answer: Why was this monitoring station chosen?

and: Why does this value reasonably represent existing air quality at the project location?

If those questions cannot be answered clearly, the baseline selection may need more work.

How Calvin Consulting Approaches Background Air Quality

At Calvin Consulting Group Ltd., we approach baseline air quality as part of the larger modelling problem. We do not simply search for a monitoring station and copy the largest concentration into the model.

Our starting point is to identify appropriate monitoring data for the substance being assessed.

Where suitable Alberta monitoring data are available, we generally look for the nearest appropriate station with continuous hourly data over a recent three-year period and then consider whether its setting reasonably represents the project location. We consider factors such as:

• proximity;
• the substance being measured;
• the availability and completeness of the data;
• whether the data are recent;
• the surrounding land use and environmental setting; and
• whether the monitoring location is representative of the project area.

This approach helps connect the scientific details to the practical question that matters: What air quality conditions is this project actually being added to?

Our work also considers the other major components of an air quality assessment, including emissions, meteorology, terrain, building downwash, existing sources and the regulatory requirements applicable to the project.

The goal is not simply to produce a model run. It is to provide a clear, technically defensible assessment that regulators, engineers and project decision-makers can understand and use.

Need Help Establishing a Background Concentration?

If you are planning a new facility, expanding an existing operation or preparing an air quality assessment, the background concentration can be an important part of the analysis.

Choosing an appropriate value requires more than finding the closest available monitoring station. It requires understanding whether the data actually represent the location and conditions being assessed.

Calvin Consulting Group Ltd. provides air quality dispersion modelling and regulatory support for industrial projects across Western Canada.

​Clean air is our Passion...Regulatory Compliance is our Business.

Air Quality Modelling Is About More Than Running the Software

AERMOD, CALPUFF and other dispersion models can perform extremely sophisticated calculations.

But the quality of the results still depends on the quality of the inputs.

Meteorological data must represent the site.

Emission rates must represent the operating scenario.

Terrain and buildings must be considered appropriately.

And background concentrations must reasonably represent the air quality that already exists.

The computer does not know whether a monitoring station beside a highway represents a rural gas plant.

It does not know whether an industrial facility influencing the monitoring data is located near your proposed project.

Those are questions that require someone to understand the project, the data and the regulatory purpose of the assessment.

That is where good judgment can make the difference between simply producing a modelling result and producing an assessment that is technically defensible.

Common Mistakes When Establishing Background Air Quality

Automatically using the closest monitor - The closest monitor may be located in a completely different environmental setting.  Distance matters, but representativeness matters too.

Using data for the wrong environment - Urban data may not represent a rural location. Industrial data may not represent a forested or agricultural setting.

Ignoring the contaminant - A monitoring station must actually measure the substance required for the assessment.

Using incomplete or poor-quality data - More data are not necessarily better data. Completeness and quality assurance are important.

Treating an unusual event as normal - An extreme concentration may be real, but the reason for the event should be understood before deciding how it should be used.

Forgetting that the emissions environment changes - New facilities begin operation. Others close. Emissions controls change. Historical monitoring data should be considered in the context of what the area looks like today and what it is expected to look like during the life of the proposed project.

Double-counting project emissions - Where an existing facility is being expanded or modified, care may be needed to ensure that emissions already represented in the monitoring data are not inadvertently added twice.

Do you have concerns about air pollution in your area??

Perhaps modelling air pollution will provide the answers to your question.

That is what I do on a full-time basis.  Find out if it is necessary for your project.

Have your Say...

on the StuffintheAir         facebook page

Other topics listed in these guides:

The Stuff-in-the-Air Site Map

And, 

See the newsletter chronicle. 

Thank you to my research and writing assistants, and the author remains responsible for the content.

A Practical Approach to Determining What Is Normal Air Quality

When we establish a background or baseline concentration, the process can be summarized as follows:

Step 1: Identify the substances and averaging periods

Determine which contaminants need background concentrations and the air quality standards being assessed.

Step 2: Find available monitoring data

Identify monitoring stations that measure the required substances.

Step 3: Start with proximity

Look for monitoring locations reasonably close to the project.

Step 4: Check whether the setting is comparable

Consider whether the monitor and project are located in similar urban, rural, industrial, agricultural, forested or other relevant environments.

Step 5: Review the data

Where appropriate, examine:

• the availability of continuous hourly data;
• data completeness;
• quality assurance information;
• the period of record; and
• whether the data are recent enough to represent current conditions.

Step 6: Apply the regulatory method

Select the concentration using the approach appropriate to the jurisdiction, contaminant and averaging period.

Step 7: Document the reasoning

Explain why the monitoring site and resulting baseline concentration were selected.

That final step is important.

A good air quality assessment should allow a reviewer to understand not only what number was used, but why it was used.