India, currently is operating over 300+ stations over 150+ cities across the country. The Central Pollution Control Board (CPCB) facilitates in daily assessment of Air Quality Index (AQI) based on the guildiness set by the Goverment of India. In this reagrds, as on today (03-01-2022) on a daily basis AQI of 100+ cities are reported. In the year 2015, The country has approxiametly 35 stations covering over 20 cities and ranking could be carried out manually on the basis of daily AQI bulletin released. Considering the upgradiation in CAAQMS network over the larger geographical extent of the country, performing the conventional techniques is challenging as it is laborous and time consuming with gretaer risk of human error. To resolve this issue, a python based code was developed. ...
The Central Pollution COntrol Board (CPCB), Government of India releases daily AQI bulletin dispalying the AQI status for all the cities with operational Continnous Ambient Air Quality Monitoring Station (CAAQMS). The bulletin is revised on daily basis in the CPCB website after 4.00 PM. The archeives AQI bulletin may be accessed here. The AQI information provided in the bulletin are analysed on daily basis to asecertain the performance of the cities in India. This may sound simple and actually is. The downloaded daily AQI bulletin is converted to dataframe and are sorted in descending order on the basis of AQI value and thereby allocating highest to lowest ranks. The actuall catch is, if two or more cities are identified with same AQI value, then these cities are assigned with the same rank.
The competent authroity started releasing the daily AQI bulletin back from 1st May 2015. However, the formats in which the bulletin's were released changed a few times making it difficult to analyse using single set of code. Tweak in the exisitng codes will do the thing. But here, just to avoid complexity I have managed to bring out 2 set of codes for varying date range. The first set code is to analyse the information pertaing to dates on or after 29th January 2018; the second set of code is to analyse between 15th August 2015 to 28th January 2018. I suggest eliminating analysis for dates prior to 15th August 2015 as there seems to have non-uniform bulletin formats every now and then. Make sure you use the right set of codes based on your required date of interest.
- Create a new conda environment
conda create -n enter_env-name python=3.9 - Activate the environment
conda activate enter_env-name - Change the directory
cd City-ranking-based-on-AQI - Clone the repository
git clone https://github.com/moorthynair/City-ranking-based-on-AQI.git - install the requirements
pip install -r requirements.txt
- If your date of interest is on or after 29th January 2018. The set-1 codes can be
Ranking_based_on_AQI_code_set1.py. - If your date of interest is between 15th October 2015 to 28th January 2018. The set-2 codes can be
Ranking_based_on_AQI_code_set2.py.
The entire process from downloading the bulletin to ranking the cities as per the AQI levels are automated. There exist 03 basic user input, a) Date (YYYYMMDD format); b) provide the path to download the bulleting pdf file; c)List of specific city of interest (The final analysis provides the rank of all the cities mentioned in the daily AQI bulletin. However in case of exracting the ranks of specific city of interest, the list of those may be provided).
To perform anaysis on the latest AQI bulltein, make sure the bulletin is released by the federal agency. Access here to check the status of latest AQI bulletin available. The latest bulletin is released in the CPCB website after 4.00PM daily. 5.00PM woluld be an ideal time to run the code for analysing the latest report. In case of unavaibality of AQI bulletin for the set date, a warning message shall pop up terminating the entire run
The output parent dataframe with file name 'final_file' is generated. This contains city specific Air Quality status (ie., Good, Satisfactory, Moderate, Poor, Very Poor, Severe); Index value (ie., AQI), crticial pollutants identified, No of monitoring stations and the national level city rank based on AQI for the user defined date. For user defined city/cities information, the output dataframe named 'city_ranking' is generated containing all the above mentioned information.
The work is licensed under MIT LICENSE