Clearing the Air: Investigating Air Quality on Campus

Within the gates of our campus, the Mumbai chaos fades into a quieter and greener landscape. With canopies of trees sheltering the roads and glimpses of the Powai Lake peeking through the foliage, the campus’ physical identity is inseparable from its natural beauty. The greenery of our campus is immediately striking, and the air we breathe is often perceived to be cleaner too, seemingly far removed from the bustle of the city.

Yet the campus, too, has evolved significantly over the decades. Now in its late sixties, it has witnessed the emergence of a city from barren land, alongside its own transformation. As it became more densely occupied, infrastructure expanded to keep pace with its growing needs. Alongside this growth came activities that affect the air around it, including increased movement of people and vehicles, construction activity, greater energy and resource consumption, and the multitude of daily activities that accompany a growing campus.

The concern around air quality is not unique to IIT Bombay but a part of a broader trend unfolding across India. Air quality is influenced by a range of pollutants, including particulate matter* and gases like nitrogen oxides, sulfur dioxide, and ozone, each leaving its imprint on our environment. Among these, particulate matter, especially fine particles smaller than 2.5 micrometres (PM₂.₅), is of concern. Its small size allows PM₂.₅ to penetrate deep into the respiratory system, making it particularly harmful to human health. The World Bank identifies PM₂.₅ as the air pollutant with the greatest health burden. A 2022 study published in Environmental Sciences: Atmospheres, tracking India’s PM₂.₅ concentrations from 1998 to 2020, found that annual average concentrations steadily increased across the country, while the share of the population living in areas with acceptable PM₂.₅ levels fell sharply, from 60.5% to 28.4%.

With this worrying picture emerging at the national level, Insight looks closer to home: what does air quality look like within our campus? While the campus around us changes in ways we can see, the air may be changing in ways we cannot.

The Cost of Growth

The number of residents on campus has increased considerably over the past two decades, rising from around 10,000 in 2008 to over 25,000 today. Accommodating this growing population has required the institute to expand its infrastructure and make greater use of the land within its boundaries. This growth is visible across the campus, where construction has become a constant feature of everyday life. New buildings, roads, and facilities have come up across the campus, occupying spaces that were previously open. While this activity is necessary to accommodate an increasing campus population, it also comes at the cost of environmental trade-offs.

The movement of construction materials, excavation, exposed soil, and dust generated during construction activities contributes to high concentrations of airborne particles in and around such sites. While it is one thing to generate this pollution, an episodic by-product of such activities, it is another to manage its impact. There appears to be limited effort to systematically deal with the dust, soot, and bio-generated particulate matter that accumulates on the roads. Instead of being collected and disposed of, these particles simply end up being displaced from one place to another.

Additionally, there has been a notable change in the green cover over the years to make way for the expansion of infrastructure. Vegetation can influence local air quality by firstly providing surfaces on which particulate matter can be deposited and secondly altering airflow, which would in turn affect how pollutants are transported and dispersed (Janhäll, 2015). The removal and fragmentation of vegetated areas can therefore change these processes. While these mechanisms alone cannot establish how changes in green cover have affected air quality on campus, they make it worth examining whether the changes in the campus landscape correspond with changes in measured pollution levels.

To explore this, Insight analysed publicly available data from the Central Pollution Control Board (CPCB) for 2019-2025. PM₂.₅ levels at IIT Bombay were compared with those recorded at five monitoring stations across Mumbai: Sion, Kurla, Colaba, Vile Parle West, and Borivali East. These stations were selected to capture different urban settings across the city, allowing IIT Bombay’s readings to be viewed against a range of city environments.

Before comparing the differences between stations, it is important to account for the uncertainty in these measurements. The PM₂.₅ data from CAAQMS stations are measured using Beta Attenuation Monitors (BAMs), which are commonly used for continuous particulate matter monitoring. Research has found that while BAMs provide reliable PM₂.₅ measurements, their readings can vary depending on the instrument and environmental conditions. According to Prof. Abhishek Chakraborty, BAM measurements can have an accuracy of around ±5%, which is used here as a benchmark for interpreting differences between stations.

The 2025 data show that IIT Bombay recorded higher PM₂.₅ levels than Borivali East and Colaba, with the differences exceeding the 5% benchmark. In contrast, the difference between IIT Bombay and Vile Parle West was within 5%, placing the two stations within the range considered comparable. This raises a crucial question: is the campus experiencing pollution levels comparable to, or even exceeding, those recorded in other urban parts of Mumbai?

The IIT Bombay campus is a relatively contained area within the city. While the Institute has no control over pollution entering from its surroundings, it does have a say in what happens within its boundaries. Construction, vehicular movement, and other pollution-causing activities are an inevitable part of a growing institute, but how these activities are managed can make a difference.

These environmental considerations are reflected in IIT Bombay’s 2025 Sustainability Policy, which outlines measures intended to make future infrastructure development more sustainable. The policy calls for new infrastructure to include proper ventilation and aims for a four-star GRIHA rating* for new buildings by planning construction to minimise land use and maintain greenery. To mitigate the dust particles that are generated during this process, it also calls for measures such as using vehicle-mounted road sweepers equipped with water-spraying systems.

The significance of air quality extends beyond only the environmental impacts of development. Speaking to Vasudev, a PhD graduate from the Environmental Science and Engineering Department, revealed that research has also found associations between exposure to air pollution and cognitive functions among children and adolescents, apart from its impact on the respiratory system. A systematic review published in Cognitive and Behavioral Neurology in 2020, for instance, reported links between particulate matter exposure and difficulties in functions such as working memory, attention, and processing speed. These findings highlight why air quality deserves attention beyond its immediate respiratory impacts, particularly in environments where developing minds spend substantial amounts of time. This led Insight to examine the measures that are currently in place on campus to address this problem and how effectively they are being implemented.

How Reliable Are the Readings?

The Brihanmumbai Municipal Corporation (BMC) has laid down a 28-point set of guidelines to mitigate air pollution arising from construction and demolition activities. These guidelines prescribe measures such as protective coverings and continuous water sprinkling to reduce particulate matter in suspension, and call for construction sites to deploy sensor-based air pollution monitors and publicly display their readings. Subsequently, a follow-up circular issued in 2025 made these requirements mandatory at ongoing and new construction sites across Mumbai.

One can find over 10 such dashboards all around campus, from various hostels to other construction and residential areas. The readings from these dashboards help identify periods of elevated pollution, assess whether mitigation measures are working, and provide a basis for corrective action when prescribed limits are exceeded. They also create a record of the conditions around construction sites and make these readings publicly visible. If the numbers being displayed do not accurately represent the air around us, they can affect how pollution is understood, how mitigation measures are evaluated, and whether action is taken when it is needed.

Yet, some of the readings displayed across the campus on these dashboards raise questions about their reliability. For instance, the dashboard near the New Academic Block construction site captured PM₂.₅ and PM₁₀ concentrations of 1 µg/m³ each at 2:01 PM on September 2, 2026. A reading this low beside an active construction site is notably unusual, and something that can be seen at multiple locations on campus. These dashboards are meant to flag pollution spikes and show whether mitigation measures are working. If the measurements are this inaccurate, a system designed to make pollution visible could instead be helping to hide it.

Insight also spoke with Prof. Abhishek Chakraborty in the Environmental Science and Engineering Department to learn more about this. He brought to light another striking example:

“On multiple occasions, even I have observed that dashboards near KV School have PM 2.5 values greater than PM 10, which is impossible, indicating that those measurements are absolutely wrong”

A similar instance was captured near the Rahul Bajaj Technology Innovation Centre at the New Academic Block construction:

PM₂.₅ refers to particles with a diameter of 2.5 µm or less, whereas PM₁₀ includes particles up to 10 µm, which implies that PM₂.₅ is a subset of PM₁₀. This means that when both measurements are taken from the same air sample, the PM₂.₅ concentration should not be higher than the PM₁₀ concentration.

Such readings raise questions about the measurement system. The dashboards around campus display readings from low-cost sensors placed in the vicinity. These sensors typically measure PM₂.₅, PM₁₀, humidity, and temperature. They primarily use the principle of light scattering to estimate particulate concentrations. As particles pass through the sensor, they scatter light, and the resulting signal is then used to estimate the concentration of particles in the air.

However, these readings are subject to variation due to other factors, such as humidity, temperature, wind speed and the physical and chemical properties of the particles. Under humid conditions, for example, hygroscopic aerosols* take up water and change their size and optical properties. This can alter the way they scatter light and consequently affect the concentration the sensor estimates.

Variations are temporal in nature too, with particulate matter concentration following a yearly cycle. Readings during the monsoon season are comfortably low, as the rain washes away air pollutants. In winter, lower temperatures and colder winds reduce the mixing length, trapping pollutants close to the ground and worsening air quality.

Additionally, accuracy at a single point in time is only part of the problem. The discussion with Prof. Chakraborty highlighted another factor. As he puts it:

“More than accuracy, it is the drift and the relative deterioration over time that one has to track accurately.”

Prolonged exposure to conditions such as humidity, temperature, and salinity, along with deterioration of internal components, can cause the sensor’s reading to drift from its true value gradually. These factors can accumulate over time, making it important to distinguish changes in the sensor from real changes in the air it is measuring.

To improve the reliability of these measurements, the BMC guidelines specify requirements for the sensors that can be deployed. BMC-approved sensors must meet prescribed performance standards and have certification or performance evaluations from recognised institutions such as the IITs and CSTEP Bengaluru. The guidelines also specify where and how these sensors should be installed, including requirements for unobstructed airflow and a typical installation height of 3–5 metres above ground level.

These protocols address how sensors are selected and deployed, but they do not eliminate changes in sensor performance over time. To identify and correct such changes, low-cost sensors can be evaluated against reference-grade monitoring systems housed in specialised stations known as Continuous Ambient Air Quality Monitoring Stations (CAAQMS). Equipped with specialised gas analysers, particulate monitors and meteorological sensors, these stations continuously collect and transmit air-quality data. Unlike the relatively simple sensors powering many campus dashboards, these instruments are designed to meet stringent monitoring standards and can provide a reliable benchmark for calibrating lower-cost sensors.

The process of calibration can help account for several factors of variation and correct the sensor’s readings against those of a reference-grade monitor. However, calibration is not a one-time solution. Periodic calibration and maintenance are important to ensure that the displayed values continue to represent the actual conditions around the sensor.

IIT Bombay also has a CAAQMS located near Hostel 15. According to Guttikunda et al. (2023), a commonly used rule of thumb considers a CAAQMS representative of the ambient air within a radius of roughly 2 km. For the H15 CAAQMS, this would encompass a substantial portion of the campus and extend into the surrounding areas of Powai.

Insight compared the readings from the dashboard closest to the CAAQMS against the CAAQMS at the same time to identify any discrepancies.

At 4:29 pm on the 8th of September, the dashboard located near the Ananta building recorded a PM₂.₅ concentration of 5 µg/m³. The CAAQMS station located approximately 250m away, near Hostel 15, however, recorded a PM₂.₅ concentration of 16.9 µg/m³ at the same time.

Given the proximity of the dashboard to an active construction site, one might expect it to record a higher, if not comparable, concentration of particulate matter compared to the CAAQMS, given that localised spikes in airborne particles are more likely to occur near the construction site. Yet the dashboard reported less than a third of the concentration recorded by the CAAQMS. While readings between stations barely 250 metres apart need not be identical, a difference of this magnitude raises questions about how reliably the dashboard is capturing the conditions around it.

In addition, the location of CAAQMS is a concern, as it sits amid dense vegetation and is relatively removed from the roads and other activity-heavy areas of campus. This may not necessarily be the environment in which it was established. As Prof. Virendra Sethi from the Environmental Science & Engineering Department explains,

“The CAAQMS was originally placed there because there was no tree cover in the region; over time, green cover around the station has become prominent”

Located away from many of the campus’s local pollution sources, the CAAQMS may capture the ambient background conditions at its location without reflecting short-term spikes elsewhere. While it remains a useful reference for calibration, its readings therefore cannot, by themselves, tell us what the air quality looks like across the campus or the broader Powai area. Periodic maintenance and reassessment of the station’s placement are especially important given the rapidly changing landscape.

The BMC guidelines also lay down requirements for the placement, connectivity, accessibility and maintenance of these monitors. Under these rules, the onus of installing and maintaining them rests with the project proponents, developers and contractors undertaking the construction work, with compliance overseen by the BMC and the Maharashtra Pollution Control Board (MPCB).

Between October 2025 and January 2026, authorities issued 1,981 show-cause notices and 1,047 “stop-work orders” across the city under the ongoing “Clean Air” initiative. By January 2026, 1,952 of Mumbai’s 2,224 active construction sites had installed low-cost air-quality sensors, while 678 projects were ordered to halt work on January 16th alone for failing to comply with sensor installation requirements.

The scale of enforcement evidently indicates the gravity of the situation as well as the nature of these requirements: they are not merely advisory, but rather a significant part of the mechanisms in place to control air pollution. Yet the unusual values observed on dashboards across the campus call into question the reliability of these readings and the confidence with which they can be trusted. While the responsibility of installing and maintaining these dashboards formally rests with the contractors undertaking these projects, IIT Bombay, as the institution within whose boundaries these readings are displayed, ought to take a stronger stance in ensuring greater transparency around how these monitors are installed, calibrated, and maintained. Without such oversight, these dashboards risk creating an illusion of adequate monitoring without necessarily providing an accurate picture of the air on campus.

Behind The Closed Doors

When discussing air quality, the focus is often primarily on the air we breathe outdoors. Yet an average student here spends much of their day in indoor environments such as hostel rooms, classrooms, labs, among others. The outdoor air that enters these enclosed spaces through doors, windows, and vents can interact with the indoor conditions, changing the quality of the air they breathe. In addition, pollutants generated by the occupants themselves over time further degrade the air quality significantly. If ventilation is inadequate, carbon dioxide released by occupants can accumulate, along with heat and humidity. As a result, the environment can become increasingly uncomfortable and stuffy. Thus, ensuring healthy air quality indoors is, in fact, as important as maintaining it outdoors.

Carbon dioxide concentration is one of the key metrics used to assess ventilation in enclosed spaces. Since students spend the majority of their time in hostels, adhering to ventilation standards is critical, especially when there is little active air circulation except for fans. Cross ventilation, thus, is a highly effective way to flush out pollutants by allowing consistent circulation of air between the indoor and outdoor environments. In older hostels, one can notice several features that ensure proper air circulation. For instance, mesh vents are usually placed directly above the doors, allowing air to pass through even when the windows are closed. Additionally, both sides of the room open to empty spaces, promoting better air circulation. In contrast, the newer hostels are built under the constraint of limited available space on campus, aiming to maximise resident capacity within them. While this focus on efficient use is important, trade-offs in design are made, which may lead to inadequate airflow and ventilation, especially in Hostels 17, 18, and 19.

The typical room layout includes a window opposite the door and no other openings for air to enter, unlike the mesh vents and open-sided layout described above. These hostels are arranged in a rectangular block, with some rooms’ windows facing inward and others facing outward. Cross-ventilation requires an entry and an exit for air movement; with only a single window, this cross-flow is missing. As a result, rooms on the lower floors and those with inward-facing windows often feel suffocating even when the windows are open, due to the limited air circulation.

Additionally, due to inadequate ventilation, carbon dioxide released by the occupants can accumulate. Over longer periods, especially during sleep, CO₂ levels can rise above 1000 ppm, a benchmark commonly used by ASHRAE to assess ventilation adequacy. A field study conducted in student dormitory rooms found that CO₂ levels averaged 2500 ppm with windows closed, compared to 660 ppm with windows open. The lack of airflow also slows the temperature drop in these rooms, thus keeping them warm for longer and resulting in greater difficulty in falling asleep and subsequent poorer sleep quality.

These issues are not limited to hostels. In a conversation with Prof. Krishnendu Sinha from the Department of Aerospace Engineering, he mentioned:

“When we monitored the carbon dioxide concentration in classrooms, we noticed that the acceptable limits were breached within 15 minutes in a partially occupied LT classroom, and 30 minutes in LC rooms, even with air conditioners.”

Classrooms and labs are other campus hotspots where students spend a lot of their time during the day. Most of the classrooms are completely enclosed with no open windows, leaving exhaust fans and air conditioners as the only means of ventilation. Notably, even with air conditioning and exhaust, the carbon dioxide concentration can exceed the recommended levels within a short span. A similar issue arises in the case of labs. In fabrication facilities like the Tinkerer’s Lab and the Makerspace Lab, particulate matter from 3D printers and other machines can accumulate over time due to inadequate exhaust ventilation, resulting in poorer indoor air quality.

Additionally, given the recirculation of air in closed spaces, insufficient ventilation can increase the proportion of air inhaled by one person that is exhaled by another, thus increasing the likelihood of transmission of airborne infectious diseases like seasonal cough, cold and flu. Infection control was one of the key drivers for ventilation during the COVID pandemic. Recently, a multi-disciplinary team of researchers at IIT Bombay have developed an infection control algorithm that automates the exhaust fan operation to minimize infection probability. The algorithm is implemented using IoT devices installed in pilot classrooms.

To address these concerns and examine campus-wide ventilation, a dedicated Ventilation Committee was established in 2021. Convened by Prof. Krishnendu Sinha, the committee’s main aim is to develop policies and solutions tailored to the ventilation needs of campus classrooms and labs.

Since then, several in silico pilot studies have been conducted using Computational Fluid Dynamics (CFD) in an LT classroom. These studies primarily attempted to identify various ventilation strategies and evaluate their effectiveness. These strategies investigated included various combinations of exhaust systems, door openings, fans, and air-conditioning units. CO₂ concentration heatmaps were then generated under different conditions to assess their overall impact on ventilation.

Based on these findings, the proposed interventions were then executed in particular LT classrooms, with adjustments made accordingly. A metal mesh was installed in the lower panel of the door, creating an opening for outdoor air to enter without needing the door to remain open. Three exhaust fans were also installed on the wall opposite the door, with two facing outward to push stale air out of the room, and one facing inward to draw fresh air back in, together creating a continuous cross-ventilation system rather than relying on natural circulation alone. Four low-cost sensors were then placed at various points across the room to track concentration levels in real time, feeding into a centralised controller that automatically turned on the exhaust system when the readings crossed the accepted threshold and switched it off once concentration levels dropped below it, with no manual monitoring required. The same setup was subsequently scaled up for an LC classroom to match its larger size.

Prof. Krishnendu Sinha also mentioned:

“Our immediate plan is to expand these systems in 50-100 different classrooms across the academic area, mimicking the similar solution for LC and LT classrooms. We are now focusing on the design for ESED classrooms.”

Such measures are promising, with practical solutions already being implemented on campus. Yet with current efforts focusing primarily on classrooms, the concerns with hostel rooms remain largely unresolved, with students having to depend purely on the existing architecture and natural airflow. The challenge now is to extend these solutions beyond academic spaces to the residential spaces where students spend a significant part of their time.

Conclusion

While the concerns surrounding campus air quality are becoming increasingly difficult to ignore, efforts to address them are growing as well. From the installation of air-quality monitors to the Ventilation Committee’s ongoing efforts to improve classroom ventilation, there are already several initiatives aimed at understanding and mitigating the problem.

The objective now should be to ensure that these efforts keep pace with the problems they aim to solve. Installing more monitoring sensors as checkpoints across the crowded hotspots in the institute would help in identifying the areas that require more immediate attention and understanding how serious the situation is there. This also means ensuring that the sensors used are regularly and correctly calibrated, ventilation systems are effectively planned and implemented, and the guidelines already in place are consistently followed.

It is essential to ensure that solutions are consistently implemented and easy to adopt widely. We know our campus well through its visible landscape: the buildings, roads, trees, the lake and the constant hum of life within it. Yet surrounding all of this is another landscape, invisible but ever-present, made up of the particles and gases we breathe every day. Perhaps it is time we paid closer attention to this environment that we cannot see.

References

1. Guttikunda, S. and Ka, N., 2022. Evolution of India’s PM2. 5 pollution between 1998 and 2020 using global reanalysis fields coupled with satellite observations and fuel consumption patterns. Environmental Science: Atmospheres, 2(6), pp.1502-1515.

2. Lopuszanska, U. and Samardakiewicz, M., 2020. The relationship between air pollution and cognitive functions in children and adolescents: a systematic review. Cognitive and Behavioural Neurology, 33(3), pp.157-178.

3. Guttikunda, S., Ka, N., Ganguly, T. and Jawahar, P., 2023. Plugging the ambient air monitoring gaps in India’s national clean air programme (NCAP) airsheds. Atmospheric Environment, 301, p.119712.

4. Strøm‐Tejsen, P., Zukowska, D., Wargocki, P. and Wyon, D.P., 2016. The effects of bedroom air quality on sleep and next‐day performance. Indoor air, 26(5), pp.679-686.

5. Mishra, A.K., Van Ruitenbeek, A.M., Loomans, M.G.L.C. and Kort, H.S.M., 2018. Window/door opening‐mediated bedroom ventilation and its impact on sleep quality of healthy, young adults. Indoor Air, 28(2), pp.339-351.

6. Arigela, R., Kannan, S., Jejurkar, P., Sethi, V., Sinha, K. and Balasubramanian, S., 2026. Assessment of Classroom Ventilation Strategies and Infection Risk using Spatiotemporal Modelling of CO2 Concentrations. Building and Environment, p.114393.

7. Janhäll, S., 2015. Review on urban vegetation and particle air pollution–Deposition and dispersion. Atmospheric environment, 105, pp.130-137.

8. Mishra, A.K., Van Ruitenbeek, A.M., Loomans, M.G.L.C. and Kort, H.S.M., 2018. Window/door opening‐mediated bedroom ventilation and its impact on the sleep quality of healthy, young adults. Indoor Air, 28(2), pp.339-351.

9. Strøm‐Tejsen, P., Zukowska, D., Wargocki, P. and Wyon, D.P., 2016. The effects of bedroom air quality on sleep and next‐day performance. Indoor air, 26(5), pp.679-686.