Data Analysis of Energy and Greenhouse Gas Emissions

Posted on May 28, 2022

New York City has already seen a preview of the effects of climate change, from superstorms to increasing number of days with extreme heat. However, the city is also uniquely poised to lead the fight against climate change in the US. NYC has pledged to become carbon neutral by 2050, meaning that its greenhouse gas (GHG) emissions will be equal to the amount of GHGs it can negate. A data-based approach to emissions reductions could help identify a strategy to meet that goal.

More than half of the city’s GHG emissions are produced by the electricity and fossil fuels used to heat and power buildings. This sector will need to cut emissions drastically in order to reach net zero. This exploratory data analysis seeks to analyze current energy use in New York City’s buildings and search for trends and patterns to help determine a path forward.


Local laws require that all buildings over 25,000 square feet--and city-owned buildings over 10,000 square feet--report each year on their energy and water use. This data is publicly available on NYC Open Data. The datasets used in this analysis consist of building data for the calendar years 2020 and 2019. Approximately 29,000 buildings are included in each dataset. While this makes up a fraction of NYC’s approximately one million buildings these are the largest buildings in the city by square footage. As they contribute about 61% of all emissions from buildings in NYC, they represent a good place to start.


I began my analysis with the 2020 dataset, since it was the most recent available dataset. However, as in many industries, 2020 was an unusual year for building operations. Compared to the 2019 dataset, the total GHG emissions were approximately 19.7% lower in 2020 than 2019. The following graph shows the breakdown in GHG emissions by borough.

Figure 1: Comparing Greenhouse Gas Emissions from buildings by borough and by year

 Manhattan saw the greatest decrease, with 23% lower emissions in 2020. The Bronx and Brooklyn saw the smallest decrease, with about 13.5% lower emissions in 2020. It’s possible that NYC saw true decreases in GHG emissions during the initial wave of the Covid-19 pandemic. It’s also possible that some of the GHG emissions were shifted from larger buildings included in the dataset to smaller residential buildings not included in the dataset. It will be interesting to see how emissions trended in 2021 as the city revved back up.

I checked for correlation between factors such as year the building was built and Energy Use Intensity, measured in kBtu/square foot. No correlation was found, suggesting that there is not a significant difference in energy efficiency between buildings built in different years.

The chart below shows the property types contributing the most to greenhouse gas emissions in NYC. Multifamily housing and office buildings were the biggest contributors by far, suggesting they will be good targets for emissions reductions regulations and incentives.

Figure 2: Top greenhouse gas emitting property types in NYC

Green Power Generation Data

Of all the buildings reporting data, only 86 buildings reported onsite green power generation. In this dataset, green power generation means power generated from solar panels or wind  turbines. The majority of the green power in the city is being generated by K-12 schools, as shown in the graph below. They generated nearly 140,000,000 kWh in 2020, which is enough to approximately 19,000 homes' annual electricity use according to the EPA's equivalency calculator.

Figure 3: Property types currently producing renewable energy onsite

New York City’s rooftop solar energy potential may be more limited than less densely built-up areas, and only 86 out of the  29,000 buildings in the dataset have solar panels. Roof space is limited, but it still seems NYC is not maximizing its solar energy potential. I decided to estimate how much green power could be generated if buildings of the same property type generated electricity at the same rate as the current green power producers.

Only five property use types had at least three data points, so I only looked at those five types (shown in the graph above). I estimated that approximately 4.26 terawatt-hours of electricity could be produced annually, which is close to 8% of NYC’s annual electricity use! It should be noted that rooftop square footage was not included in the dataset. Total building square footage is a useful but not totally accurate metric to use instead, since many NYC buildings are high rises with relatively low rooftop to gross square footage. The chart below shows the power generation potential of the five building types. 

Figure 4: Predicted renewable energy generation possible for five property types

Comparison to Energy Star Benchmarks Data

Each entry in the dataset also includes the building’s Site EUI, or Energy Use Intensity, measured in kBtu/square foot. It is the amount of energy used per square foot for a building, and is useful to be able to compare the energy efficiency of buildings of different sizes. It is also useful to be able to benchmark against other similar types of buildings. Energy Star has calculated a list of median EUIs for different property types.

I used the 2019 dataset for this analysis since the 2020 data was so affected by the pandemic. Overall, the buildings in the dataset were well above the Energy Star national median, suggesting that the top emitting property types also have a lot of room for improvement in terms of energy efficiency. In a way, this means the path is straightforward: Tackle the highest emitting property types first to make large dents in citywide energy use and greenhouse gas emissions.

The plot below shows the site EUIs per property type of the buildings in the data, with only some of the top GHG emitting property types shown. The red dot indicates the Energy Star median EUI. The chart shows that for most of the top GHG emitting property types in New York, the median EUI is greater than the Energy Star median EUI.

This means those building types are using more energy per square foot than recommended, and are not as energy efficient as they could be. With the exception of Hospital, all property types included in the chart are less energy efficient than the Energy Star benchmark, with Prison/Incarceration EUIs being especially high.

Figure 5: Comparing Median EUI of NYC buildings to the Energy Star Median EUI. Having a higher EUI means a building is less energy efficient.


New York City has a long way to go to tackle its greenhouse gas emissions and fight climate change. However, there are many areas for improvement, including relatively low energy efficiency of some of the largest energy user property types, and a lack of onsite green power generation throughout the city. Targeting these low-hanging fruits first could lead to fast, dramatic cuts in greenhouse gas emissions, while deep decarbonization will take more time.

Expanding the dataset to include smaller buildings would help get a fuller picture of energy use in New York, and see if energy use is shifting away from the larger buildings in the city due to changing work habits during the Covid-19 pandemic.

Future research

It would be interesting to look at the breakdown of fossil fuels and steam heat usage in the different buildings, and determine a path forward for electrification. How much more electricity will be required to support electrification of these buildings?


The skills I demoed here can be learned through taking Data Science with Machine Learning bootcamp with NYC Data Science Academy.

About Author

Caroline Keough

Data Scientist with a background in energy and sustainability. I'm hoping to apply my new skills in Python, R, and Machine Learning to make a positive impact on the planet.
View all posts by Caroline Keough >

Leave a Comment

No comments found.

View Posts by Categories

Our Recent Popular Posts

View Posts by Tags

#python #trainwithnycdsa 2019 2020 Revenue 3-points agriculture air quality airbnb airline alcohol Alex Baransky algorithm alumni Alumni Interview Alumni Reviews Alumni Spotlight alumni story Alumnus ames dataset ames housing dataset apartment rent API Application artist aws bank loans beautiful soup Best Bootcamp Best Data Science 2019 Best Data Science Bootcamp Best Data Science Bootcamp 2020 Best Ranked Big Data Book Launch Book-Signing bootcamp Bootcamp Alumni Bootcamp Prep boston safety Bundles cake recipe California Cancer Research capstone car price Career Career Day citibike classic cars classpass clustering Coding Course Demo Course Report covid 19 credit credit card crime frequency crops D3.js data data analysis Data Analyst data analytics data for tripadvisor reviews data science Data Science Academy Data Science Bootcamp Data science jobs Data Science Reviews Data Scientist Data Scientist Jobs data visualization database Deep Learning Demo Day Discount disney dplyr drug data e-commerce economy employee employee burnout employer networking environment feature engineering Finance Financial Data Science fitness studio Flask flight delay gbm Get Hired ggplot2 googleVis H20 Hadoop hallmark holiday movie happiness healthcare frauds higgs boson Hiring hiring partner events Hiring Partners hotels housing housing data housing predictions housing price hy-vee Income Industry Experts Injuries Instructor Blog Instructor Interview insurance italki Job Job Placement Jobs Jon Krohn JP Morgan Chase Kaggle Kickstarter las vegas airport lasso regression Lead Data Scienctist Lead Data Scientist leaflet league linear regression Logistic Regression machine learning Maps market matplotlib Medical Research Meet the team meetup methal health miami beach movie music Napoli NBA netflix Networking neural network Neural networks New Courses NHL nlp NYC NYC Data Science nyc data science academy NYC Open Data nyc property NYCDSA NYCDSA Alumni Online Online Bootcamp Online Training Open Data painter pandas Part-time performance phoenix pollutants Portfolio Development precision measurement prediction Prework Programming public safety PwC python Python Data Analysis python machine learning python scrapy python web scraping python webscraping Python Workshop R R Data Analysis R language R Programming R Shiny r studio R Visualization R Workshop R-bloggers random forest Ranking recommendation recommendation system regression Remote remote data science bootcamp Scrapy scrapy visualization seaborn seafood type Selenium sentiment analysis sentiment classification Shiny Shiny Dashboard Spark Special Special Summer Sports statistics streaming Student Interview Student Showcase SVM Switchup Tableau teachers team team performance TensorFlow Testimonial tf-idf Top Data Science Bootcamp Top manufacturing companies Transfers tweets twitter videos visualization wallstreet wallstreetbets web scraping Weekend Course What to expect whiskey whiskeyadvocate wildfire word cloud word2vec XGBoost yelp youtube trending ZORI