/nyc-air-quality-analysis

Capturing the change in New York City Air Quality, I use Geo-Spatial data, community district statistics, and machine learning algorithms for forecasting the levels of Nitrogen Dioxide(NO2).

Primary LanguageHTML

About The Project

While still a work in progress, this work hopes to capture the changes in New York City Air Quality while using machine learning models for possible forecasts of Nitrogen Dixoide(NO2) levels throughout the atmosphere.

There are a multitude of factors that often contribute to the general air quality and pollution levels throughout a city. The current indicators for this project mostly consider traffic volume as a major contributor to air pollution.

General Trends and Map Overview

We can begin by plotting the changes in NO2 levels across NYC from 2008 to 2021. This creates a general plot highlighting discernible cyclical patterns.

averageNO2Levels

Mapping historical NO2 levels by Community Districts and aggregating by the average level in a 12-month period could help to assess any changes in NO2 and emerging patterns within our data.

The following procedure creates a choropleth map of NYC with respect to time and is available here:

Change in NO2 Levels in NYC from 2008-2022

Machine Learning Models

K-Means Clustering

Clustering multi-polygons with varying levels of Nitrogen Dioxide in 2020 produced the following clusters:

kmeans_NYC_2020

However, focusing only on our data from 2010, we can create 10 clusters for grouping Community Districts based on average NO2 levels and an aggregation of Traffic Volume per locale.

This creates the following clusters: Available Here

Screen Shot 2023-11-18 at 12 39 38 AM