Discovering Hyper-local Pollution

Cleair builds network of IoT devices to measure & analyse air pollutants and provides insights to help reduce pollution

 
 

Mission

Approach

Monitor

Data is collected in real-time every 15 minutes to one hour via miniature, modular sensor units utilizing state of the art sensing technology and uploaded to the cloud. The monitoring framework integrates data from satellites, air quality reference monitors (where available), and compliments these measurements from low-cost sensor networks to improve spatial and temporal coverage.

Measure

Air quality can vary sharply even within the same city block, and pollution exposure can vary even 30% between individuals living in the same residence. Through a deployment of air quality sensors together with our hyper local spatiotemporal model we estimate the underlying temporal varying subspace that provides predicted pollution level at a very local level.

Analyze

Cleair uses advanced machine learning algorithms to analyze spatiotemporal pollution data based on real-time and historic data, both from sensors in the field and open sources. The data which includes hyperlocal air pollution forecast, air pollution hot spot is delivered to users through the Cleair app, as well as customized for users.

Act

By making air pollution visible and measurable, Cleair provides a more comprehensive picture of air pollution across cities. Data insights delivered by Cleair’s analytics can be used to make effective pollution management decisions. Cleair’s insights into highly local personal risks from air pollution allow government, industry, citizens to take bold actions.

Solutions

Cleair develops practical solutions that empower people to make better choices, from taking the clean route, to stepping out at the right time.

Real-time monitoring

An Internet of Things (IoT) based Mobile Air Pollution measurement device.                                                                                            

Built-in Analytics & Diagnostics

Measure the exact concentration of major air pollutants PM2.5, PM10, SO2, NO2, and CO along with Temp, Pressure & Humidity.

Data Enrichment by Crowdsourcing & Satellite Data

Automatically connect to thousands of external data sources in a single platform.                                                                             

Artificial Intelligence-based Modelling

The data collected [Ground Level Monitoring data] to be collated with satellite data and other rich data sources.

Applications


GIS / MAP

  • Real time hyperlocal pollution heatmap
  • Pollution Data Layer
  • Green Route [identify least polluted path between source / destination] - Real time pollution map on road network

Mobility

  • Green Route [identify least polluted path between source / destination] - Real time pollution map on road network
  • Route optimization / Go-green (Electric Vehicle) deployment strategy

Hospitality

  • Differentiation based on hyper-local pollution data
  • Check-in / check-out based on diurnal data analysis - helps in promotion
  • Display pollution data on properties

Travel

  • Listing / filtering based on pollution data
  • Helps in travel plan with seasonal variation / prediction

Dining

  • Listing / filtering based on pollution data
  • Promote al fresco dining

Insurance

  • Making pollution exposure part of actuarial model
  • Product refinement with pollution data rider

Real Estate

  • Real-estate planning and construction [green area]
  • Listing / Advertising for properties

Weather

  • Hyperlocal real time pollution data and forecasting

Government

  • Public Awareness
  • Governance: real time hotspot identification for rapid response                                                                            
  • Governance: real time hotspot identification for rapid response
  • Source Apportionment
  • Policy formulation

Others

  • Public Awarenessl
  • Advertisement revenue from display                                                                                              

Real-time and hyper-local data using low cost sensor devices

Cleair analytics uses sophisticated algorithms to figure out low pollution routes in a city. Using the open-source data and using a combination of average concentrations, distance, ventilation rate for walkers, and bikers, the Cleair route finder calculates alternative routes with the lowest concentration of pollutants along each route for nitrogen dioxide (NO2) and PM10 and PM2.5.

World Bank funded electric vehicle project in Kolkata, India

Fleet electrification is one of the strategies for improving urban air pollution in Kolkata. The project identified high polluting corridors in Kolkata using low cost sensors and these corridors were selected for EV deployment. Data is being collected over two years with a network of sensors and shows the following results:

Notable observations made for the data collected are as follows:

The Northern zone of the city has high levels of particulate pollutants, exceeding the advised levels at all times. Even rain did not have a discernible effect on pollutant levels.

The rest of the city of Kolkata showed high levels of particulate pollutants on most days, with PM levels being acceptable only on days with considerable rain.

The Howrah region showed similar levels of particulate pollutants as with most of Kolkata.

Varanasi Cleair Air-Quality Network

Cleair started the deployments in Varanasi for measuring hyperlocal air quality data in November 2020 by deploying three devices complementing one existing government station. The data will be collected for one month for the initial hyper-local modeling exercise. Downstream the model will continuously improve with the data collated from the network and will be open to individuals for hyper-local pollution planning.

New Town proposed Cleair Network

Cleair will be starting to collect hyper-local air quality data for the New Town area by installing 4 devices at strategic locations. This will be the third city covered by Cleair network.

Cleair Hyper-local Models

Cleair Hyper-local model provides hyper-local air pollution with the use of optimum no. of the devices completed with Satellite, open-source, and crowdsourced data. Below is the ward wise pollution obtained from the Cleair network consisting of

  • Existing pollution measurement stations
  • Complemented by Cleair devices.
Air Quality Stations – Kolkata Boundaries
Ward Wise Hyper-local Output
Ward Wise Hyper-local Confidence Scores
Cleair Route Analysis Model

Several studies showed a correlation between long-term exposure to air pollution and increased likelihood of death from COVID-19, likely due to poor lung health. Even short-term improvements to air quality may be helping to reduce the number of deaths from other respiratory illnesses. Cities across the world have understood that now is the time to move towards clean mobility and have already begun to remodel their urban space with more emphasis on walking and bicycling. Cleair route finders calculates alternative routes with the lowest concentration of pollutants along each route made available to citizen via the Cleair app.

S12 East-West corridor route shows. The route is mostly polluted throughout as it starts from Howrah Station, central Kolkata, and pollution level eases as we move to Salt Lake Route close to Sealdah Railways station is severely pollution

S9 North-South corridor route shows. The route is more polluted in the northern part as well as in the Rash Bihari Avenue. This route runs along the Hooghly river and it shows a lower level of pollution alongside the river, And towards the south-western end

Geospatial Temporal Models

COVID-19 lockdown caused most public transport, Factories and construction work suspended have brought the world to a standstill.

The scenario has led to a major cut in air pollutants and cities are recording much lower levels of harmful microscopic particulate matter known as PM 2.5, and of nitrogen dioxide, which is released by vehicles and power plants. Amid the changes brought by COVID-19, we have analyzed the change in particulate levels (PM2.5) concerning the lockdown and with respect to levels in the year 2019 vs 2020.

The results showed a drop of around 30-40 in 2020 compared to 2019 in the city of Delhi and Kolkata India. We could also see a drop in PM levels in Delhi due to the lockdown.

Delhi Lockdown Plot
Severity in COVID19 due to Higher Pollution Levels

Pollution exposure quantification for the region is done by counting the number of days for which the daily pollution mean is greater than poor levels. For each of the twenty-one stations, we define a region around the station that defines the pollution state for that region. The best measure of pollution in that region is approx one km radius of the station.

The range of count of the days of bad pollution exposure varied from 35 to 60 for all regions combined and the range is color codes as in bins of five i.e. 35-40 (inclusive) Light Green, 40-45 (inclusive) Green, 45-50 (inclusive) Light Red, 50-55 (inclusive) Dark Red and 55-60 (inclusive) Dark Red

The figure shows a coarse co-relation between COVID19 severity and the bad pollution exposure levels. The figure also shows the COVID19 hotspots in the more polluted regions.

Partners

Team

Sanjoy Chatterjee

Entrepreneur & Technology Evangelist – founder of Ideation Technology Solutions & Ideal Analytics Solutions. Master of Technology from IIT KGP.

Shovon Mukherjee

25 years of experience in Business & IT consulting. Worked as Executive Director of IBM and Partner/Executive Director in PricewaterhouseCoopers (PWC) Consulting Services.

Rakhi Basu

More than 20 years of global experience across Asia, Middle East, Africa, Europe and USA in advising government and industry to tackle their highest priorities in infrastructure and human development.

Mithu Sengupta

Technology advisor specializing in eCommerce and digital transformation with 20+ years' experience in providing strategic direction on global IT, digital and data projects.

Sukumar Chakraborty

Chief Architect. Statistician, Big Data Specialist. Graduate in Electrical Engineering & Post Gradute from Indian Statistical Institute.

Siddharth Nobell

IoT Specialist Embedded Product Development B. Tech in Computer Science

Arnab Majumdar

Data Scientist. Expert in Mathematical & Statistical Modelling. IIT Kanpur Gard & Post Doctorate from Boston University.

Gora Datta

Gora Datta is a US-based serial entrepreneur and an internationally acknowledged subject matter expert on ICT (Information & Communication Technology),

Soumitra Bose

Functional Expert. Graduate in Physics, Post Graduate in Management Technology and Computer Science.

Advisors

Scott Moura

Clare & Hsieh Wen Shen Distinguished Professor in Civil & Environmental Engineering. Energy, Civil Infrastructure and Climate, Systems

Pravin Krishna

Pravin Krishna is the Chung Ju Yung Distinguished Professor of International Economics and Business at Johns Hopkins University.

V. Faye McNeill

Clare & Hsieh Wen Shen Distinguished Professor in Civil & Environmental Engineering. Energy, Civil Infrastructure and Climate, Systems

IoT based mobile Air Pollution Measurement Device with Data Management and Recommendation Engine.

116 SDF Building
Salt Lake, Kolkata -700091

contact@cleair.com

+91 033 2357 6414

Loading
Your message has been sent. Thank you!