Relevance of CO₂-based IAQ indicators: Feedback from long-term monitoring of three nearly zero-energy houses

Highlights

• Difference of 12–34% in the average concentration in bedrooms between years.

• 10 CO₂-based IAQ indicators show different levels of stuffiness for the same room.

• The most sensitive indicator to the occupancy scenario is the cumulative exposure.

• Cumulative exposure indicators without an occupancy scenario increase up to 257%.

Abstract

There are a large number of indicators that use CO₂ concentration as parameter to assess air stuffiness and, consequently, to asses IAQ. Their comparison is difficult since they are not usually linked to each other. The aim of this article is to compare the results of 10 CO₂-based IAQ indicators and determine if they classify a house in a similar way during heating seasons. We propose a method to normalize the results based on the reference values of each indicator, and we highlight the sensitivity of the indicators to the choice of one occupancy scenario among several possibilities. The database used contains the CO₂ concentration measured over 2–3 years in the living room and the parental bedroom of three new and occupied nearly-zero energy houses in France (COMEPOS project) with low-cost probes sampling every minute. The results indicate that the IAQ of the same house in the same heating season can be classified differently depending on the indicator and threshold chosen. Moreover, an indicator can show different results for the same room over the years. For example, the IAQ of the bedroom of House 2 is classified poor in 2017 and 2019 but good in 2018 according to the mean CO₂ concentrations with a 1000-ppm threshold. The indicators also present different levels of sensitivity to occupancy scenarios, being the cumulative exposure the most sensitive by increasing up to 257% without an occupancy scenario, which highlight the importance of the systematic implementation of a standard occupancy scenario for the CO₂-based IAQ performance indicators.

Introduction

As people start to spend more time at home, it is important to consider the indoor environmental quality (IEQ) of their dwellings. There are warnings that energy-retrofitted buildings can present risks for the health of inhabitants related to the IEQ [38]. Indeed, as for positive, zero and nearly-zero energy buildings, those buildings tend to be air tighter, reducing air infiltrations. If they are not equipped with efficient and well-maintained ventilation systems, indoor pollution can become high and molds are prone to appear.

The assessment of IEQ is complicated by the lack of consensus regarding measurement protocols, category weighting schemes, and assessment class limitations [20]. Therefore, the performance of the same building can be different depending on the regulations and the evaluator's interpretation.

Analyses of recent results of an experimental campaign in a zero-energy building [11] and in multi-unit residential buildings [1] showed that IEQ mainly depends on the indoor air quality (IAQ). However, the IAQ is a complicated issue. Despite the fact that IAQ parameters such as the concentration of CO₂, particle matter and volatile organic compounds are measureable and have effects on human wellbeing, there are no agreed measures that can quantitatively describe the IAQ and that will facilitate the assessment of measures to improve energy performance [44].

A performance indicator is an assessment and decision support tool. It reports the particular situation or state of something based on certain parameters. Wei et al. [50] found nearly 100 parameters that are used in green building schemes to describe IAQ and the quality of the thermal, acoustic, and visual environment. It is complicated and impractical to measure all the parameters necessary to calculate the performance indicators of the different standards, especially given the difficulty of long-term monitoring of some of them without interfering with the normal activities of the inhabitants (e.g. formaldehyde and particle matter). Therefore, this article focuses on one parameter that has been widely used to this end: CO₂. In fact, several standards such as NF EN 15251, NF EN 15665 and NF EN 16798 [8,36,37] have proposed the measurement of CO₂ concentrations as a parameter for evaluating IAQ.

The CO₂ concentration is correlated with human respiration and air renewal [2]. As the body constantly produces CO₂, the presence of this compound inside buildings is an indicator of occupancy and air renovation due to ventilation. A high rate of CO₂ indoors is commonly accompanied by human bioeffluents [[52], [53]]. According to Zhang et al. [54], the exposure to human bioeffluents and CO₂ concentrations around 1600 ppm at a ventilation rate of 4.10⁻³ m³s⁻¹ per person cause sensory discomfort but it do not cause negative effects on cognitive performance or acute health symptoms. However, other studies suggest that at CO₂ concentrations of 1000 ppm there is a moderate decrease in decision-making performance [24,45].

The concentration of CO₂ indoors also facilitates the calculation of air stuffiness indicators [40,43]. A high level of air stuffiness indicates an air renewal unsuited to the occupancy density of the site and thereby signifies a possible accumulation of other substances emitted inside the building. Thus, the identification of a highly confined space using CO₂ concentrations can potentially indicate the presence of other substances, such as certain gaseous compounds or bio-aerosols, which can degrade the IAQ.

Currently, there are numerous CO₂-based IAQ indicators and their description is often ambiguous, since not all of them have reference values, a concrete period, a time step, and a specific place for taking the measurements. Even rarer is finding indicators that include occupancy scenarios. Standards such as NF EN 15251 and NF EN 16798 [8,36] indicate that CO₂ measurements should be made where it is known that occupants spend most of their time, preferably in winter, but they do not include the sample size or the time step to guarantee the quality of the results.

The measurement periods of recent studies carried out in inhabited residential buildings are quite varied: one week or less [9,29,49], one month [5], one year [10,22,23], more than one year [4,12,13,30], some days in different seasons of the same year [46] and some days at the same season but different year [21,41]. The measuring range and time step of the probes are also different between studies: the minimal measure is 0 or 400 ppm, the maximal measure varies from 2000 to 10 000 ppm, and the time step varies from 1 to 30 min.

With the aim of contributing to a future consensus on the CO₂-based IAQ indicators for characterizing residential buildings, the present paper focuses on knowing if the different indicators classify a house in a similar way during a certain period by:

• Testing several CO₂-based IAQ indicators selected from the literature, standards and regulations, such as the air stuffiness index (ICONE), the mean concentration, and the cumulative exposure,

• Proposing a method to normalize and compare the indicators results, and

• Highlighting the sensitivity of the indicators to the choice of one occupancy scenario among several possibilities.

The database used in this study comes from a long-term measurement campaign (between 2 and 3 years) in three real and occupied, nearly zero-energy houses in France built within the framework of the “Optimized design and construction of positive energy houses” project (COMEPOS project) [7].