BREATHING DATA
HOW ANALYTICS IS FIGHTING INDIA’S AIR QUALITY CRISIS
“Turns out, the air around us has been keeping secrets. Thankfully, data doesn’t.”
Let’s start with a deep breath. Actually, maybe don’t!
Go ahead, take a deep breath.
Seems harmless, doesn’t it? After all, breathing is the one thing we do without giving it a second thought. But what if that single breath carried more than just oxygen? What if it also carried a story, one hidden in millions of air quality readings collected over the years?
According to global air quality reports, 9 of the world’s 10 most polluted cities are in India.
This is where Analytics steps in!
Behind every Air Quality Index (AQI) reading lies valuable information waiting to be explored. Analytics helps us understand where pollution occurs, when it becomes most severe, and how those insights can guide smarter decisions.
THE DATASET: Six years, 26 Cities, Thousands of Clues.
Every good analysis starts with one question: Where do we find reliable answers?
In this case, the answers came from the Central Pollution Control Board (CPCB), India’s official authority for monitoring air quality. The dataset, publicly available on Kaggle, captures six years of environmental observations, offering a detailed snapshot of how the air changed across different regions of the country.
Which cities consistently struggled with poor air quality? Were there seasonal patterns? Which pollutants had the biggest impact? More importantly, how can these insights help authorities make better decisions and improve the air we breathe?
Let’s find out!
CHAPTER 1: THE RANKING NOBODY WANTS TO TOP!
When you have data from 26 cities over six years, the first question is almost impossible to resist: Which city has been breathing the hardest?
The chart makes one thing clear almost instantly: Clean air isn’t distributed equally across India.
Ahmedabad recorded the highest average AQI among all the cities analyzed, followed by Delhi, Patna, Gurugram. These cities have long been associated with poor air quality, and the analysis confirms that the concern isn’t based on perception alone. The numbers tell the same story.
The pattern becomes even more interesting when geography enters the picture. Many of the cleaner cities are either located in the northeast or along the coast, while several of the most polluted cities are densely populated, highly industrialized, or experience heavy traffic throughout the year.
Instead of treating every city the same, environmental agencies can use these insights to identify pollution hotspots, prioritize high-risk regions, and design solutions that address local challenges rather than applying one nationwide approach.
CHAPTER 2: IS THE AIR FINALLY GETTING BETTER?
Once we identified the cities with the highest pollution levels, the next question was obvious:
Are things actually improving, or are we just getting used to bad air?
The trend tells an interesting story.
One city immediately stands out — Ahmedabad! Between 2016 and 2018, its AQI rose sharply, reaching some of the highest values in the dataset.
Then comes 2020. Almost every city in the analysis recorded a noticeable drop in AQI. Delhi, Lucknow, Patna, Ahmedabad, they all showed cleaner air than in previous years. The reason wasn’t a new pollution-control technology or a sudden environmental breakthrough. It was the nationwide COVID-19 lockdown.
This is where analytics becomes especially valuable. By tracking trends over time, we can measure the impact of major events and policies instead of relying on assumptions. If reducing traffic and industrial emissions can produce measurable improvements, similar data-driven insights can help governments design targeted interventions, evaluate clean-air initiatives, and monitor whether those efforts are actually working.
CHAPTER 3: WHAT’S REALLY IN THE AIR?
Get Shruthi G’s stories in your inbox
Join Medium for free to get updates from this writer.
We’ve compared cities. We’ve tracked trends over time. But here’s another question we were asking:
What exactly is making the air unhealthy?
The chart quickly reveals one thing; not all pollutants contribute equally. Leading the list is PM10 (116.7 µg/m³), made up of dust, construction debris, smoke, and other particles. It’s the most abundant pollutant in the dataset, reflecting the impact of rapid urbanization and increasing construction activity across many cities. Close behind is PM2.5 (67.3 µg/m³), and this is where the real concern begins.
The remaining pollutants O₃, NO₂, SO₂, and CO, appear in lower concentrations, but that doesn’t make them harmless. Together, they contribute to smog formation, respiratory problems, and poor, urban air quality. If we only monitored AQI, we know that pollution exists. By analyzing individual pollutants, we understand what is driving it. That insight allows environmental agencies to focus on the biggest contributors, design, targeted interventions, and measure whether those efforts are actually making a difference over time.
Because in environmental analytics, asking “How polluted is the air?” is only the beginning.
The more useful question is “What’s causing it, and what should we tackle first?”
CHAPTER 4: DOES AIR HAVE A CALENDAR? IT TURNS OUT, YES.
Does air pollution follow a pattern throughout the year?
The answer is a clear yes. The analysis shows that November (242) and January (232) consistently recorded the highest AQI levels, making winter the most challenging season for air quality. In contrast, July and August (112–114) had the lowest average AQI, thanks to the monsoon rains that naturally help clear pollutants from the atmosphere.
One observation stood out immediately, not a single month recorded an average AQI below the safe limit of 100.
So, what makes winter so different?
It’s a combination of factors. Crop residue burning during the post-harvest season, increased emissions from vehicles and heating, and winter weather conditions that trap pollutants close to the ground all contribute to the sharp rise in AQI.
This is where analytics becomes far more than a reporting tool. Once seasonal patterns are identified, they become predictable. Historical data allows analysts to forecast periods of poor air quality well before they occur, giving authorities valuable time to prepare.
This isn’t just a possibility, it’s already happening. Delhi’s Graded Response Action Plan (GRAP) uses AQI forecasts to trigger actions such as construction restrictions, industrial controls, and traffic management as pollution levels cross predefined thresholds.
It doesn’t just explain yesterday’s air — it helps us to prepare for tomorrow’s.
CHAPTER 5: THE BIGGER PICTURE — WHEN PATTERNS COME TOGETHER
We’ve looked at cities individually. We’ve explored yearly trends, and seasonal changes. But what happens when we put everything together?
Then there are cities like Aizawl, Shillong, Coimbatore, and Thiruvananthapuram, where the lighter shades remain remarkably consistent. Their patterns show that cleaner air isn’t an exception, it’s achievable under the right environmental and urban conditions. This is exactly why visual analytics is so powerful. A heat map allows us to spot trends that would be easy to miss in a spreadsheet. Instead of examining thousands of individual readings, we can instantly identify persistent hotspots, seasonal shifts, and cities that are consistently performing better than others.
Sometimes, one visualization can tell a story that thousands of rows of data never could.
THE CONCLUSION: DATA DOESN’T CLEAN AIR. DECISIONS DO!
After exploring six years of air quality data, one thing became clear: the real value of analytics isn’t in creating charts, it’s in creating clarity. This analysis identified pollution hotspots, uncovered seasonal trends, highlighted the pollutants that matter most, and revealed patterns that might otherwise have gone unnoticed.
But the story doesn’t end with insights. It begins there.
Across India, environmental agencies are already using analytics to forecast pollution, monitor emission sources, evaluate environmental policies, and issue timely public advisories. Instead of reacting after air quality becomes hazardous, data allows authorities to anticipate problems, prioritize resources, and take informed action before conditions worsen.
That’s the true power of analytics. It transforms millions of data points into decisions that can improve public health, strengthen environmental policies, and support cleaner, more sustainable cities. India’s air quality crisis didn’t appear overnight. It is the result of years of urban growth, industrial activity, transportation choices, and environmental challenges. Solving it will require equally thoughtful decisions, and those decisions should be guided by evidence, not assumptions.
The charts in this blog don’t claim to solve air pollution. They simply reveal what the air has been trying to tell us all along. Every AQI reading is more than a number. It’s a signal. A warning. An opportunity to act.
The next time you take a deep breath, remember this: somewhere, another sensor is recording that moment, another data set is growing, and another opportunity to make a better decision is being created.
The question is no longer whether we have the data.
The question is whether we’ll act on what it’s telling us.
