Data For Science: Visualization

Airline Traffic Between States

Bruno Gonçalves — Tue, 21 Apr 2026 20:53:01 GMT

Pie charts have a rich history. William Playfair first used them in 1801, and despite decades of criticism from data visualization purists, they remain one of the most instantly recognizable ways to show proportions. Today we'll put them to work in a decidedly non-standard way: placing custom-built pie charts directly on a geographic map to reveal where Americans fly.

Subscribe now

You can find the companion notebook on the Visualizarion for Data Science GitHub repository:

Visualization For Science GitHub

The Bureau of Transportation Statistics provides a wide range of datasets on US and global transportation infrastructure. We have used it previously here, here, and here.

The first step in our analysis is to extract the Top destinations for each state. We sort the outgoing destination of each state and extract the Top n:

Choropleth Maps

Bruno Gonçalves — Wed, 02 Apr 2025 14:49:30 GMT

In this post, we will learn how to quickly and easily plot geographical data using Shapefiles from the Census Bureau, geopandas, and matplotlib.

We hope your enjoy this post and don’t forget to

Subscribe now

so you never miss a post and

to help us grow!

The US Census Bureau provides detailed shapefiles for various regions in the US. We download the shapefile for the 50 States. And load it with GeoPandas:

The Effect of Vaccines

Bruno Gonçalves — Wed, 05 Jul 2023 14:56:40 GMT

Welcome to the latest post of the Viz4Sci Substack! In this post we will learn how to reconstruct one of the most famous visualizations of the effects of vaccination of disease prevalence, originally published by the Wall Street Journal in 2015. We’ll use the original WSJ data and build the entire visualization using nothing but matplotlib.

We hope your enjoy this post and don’t forget to

Subscribe now

so you never miss a post and

to help us grow! We’re looking forward to hearing your thoughts, so go ahead and….

Leave a comment

In a previous post, we explore the inverse correlation between vaccination rates and deaths. Here we explore the effect that vaccination has on disease prevalence, the number of infected individuals per capita.

The original visualization was published in the Wall Street Journal back on Feb 11th, 2015 and covers 7 different disease/vaccine pairs. For simplicity, we’ll focus on the figure for Measles, a childhood diesease that used to kill up to 2.6 million people per year, before the advent of va…

Waterfall Chart

Bruno Gonçalves — Sun, 26 Feb 2023 15:44:58 GMT

The Gross Domestic Product is perhaps the most common measure of the economic health of a specific country or region, and it is composed of the monetary value of all the goods and services produced and sold in a given quarter or year. Naturally, such a complex metric is composed of many individual components each with its contribution.

In this post, we will download GDP data from the Bureau of Economic Analysis (BEA) and use a Waterfall Chart to compare how it changed between the second quarter of 2019 and 2020. Waterfall Charts are 2D diagrams used to highlight how different components contribute to explaining a change in the value of a specific quantity, making them ideal for this specific application. McKinsey, a consulting firm, popularized Waterfall Charts and they’ve since gained popularity in financial applications.

The data we will be using can be found in Table 14 of this spreadsheet which can be downloaded from the BEA website.

The data can be downloaded directly into a pandas …

Photo Color Scheme

Bruno Gonçalves — Wed, 31 Aug 2022 00:33:00 GMT

Welcome to the latest post of the Viz4Sci Substack! This week we explore how to manipulate, extract the main colors, and reduce the number of colors of arbitrary images using just matplotlib and sklearn.

We hope your enjoy this post and don’t forget to

Subscribe now

so you never miss a post and

to help us grow! We’re looking forward to hearing your thoughts, so go ahead and….

Leave a comment

In a previous post, we took our first steps in image manipulation by simulating how a given photo looks for people with different kinds of color blindness. Today we continue exploring how to easily extract the dominant colors from an image.

For the purposes of this post, we’ll being using a colorful photo from a Buddhist temple in Singapore:

Rotating Global Heatmap

Bruno Gonçalves — Sun, 24 Jul 2022 19:57:55 GMT

As record temperatures are being felt across the globe, this week we visualize the temperature anomaly for the month of June 2022 dowloaded from the website of NASAs Goddard Institute for Space Studies. The data compares the average temperature during the past month with the average observed in the period between 2007 and 2016 at a resolution of 250km.

As always, all the code public in the form of a self contained Jupyter notebook.

Subscribe now

and:

we’re looking forward to hearing your thoughts, so go ahead and

Leave a comment

NASA makes its data available in multiple formats, including the fixed width text file we’ll be using.

The Mandelbrot Set

Bruno Gonçalves — Thu, 30 Jun 2022 22:15:00 GMT

This week we explore the Mandelbrot Set, one of the best known fractals, and use them to illustrate how we can use ConnectionPatch to draw lines connecting different subplots within the same figure in a simple and convenient way.

As always, all the code public in the form of a self contained Jupyter notebook.

Subscribe now

and

we’re looking forward to hearing your thoughts, so go ahead and

Leave a comment

The Mandelbrot Set is the set of complex numbers c for which the function

Unknown Pleasures Plot

Bruno Gonçalves — Sun, 22 May 2022 16:38:38 GMT

This week we reconstruct one of the most iconic album covers of the twentieth century using nothing but standard matplotlib and learn more about how matplotlib handles the z-order of the different components of a plot.

As always, all the code public in the form of a self contained Jupyter notebook.

Subscribe now

we’re looking forward to hearing your thoughts, so go ahead and

Leave a comment

In this post we explore how to recreate the cover of the 1979 “Joy Division” album “Unknown Pleasures”. The album cover image has a fascinating story. It is a slightly modified plot from the 1970 PhD thesis by Harold Craft about the first Pulsar ever discovered.

NASA Climate Spiral

Bruno Gonçalves — Wed, 30 Mar 2022 22:59:00 GMT

Welcome to the latest issue of the V4Sci substack. This week we create our own version of the climate spiral animation that NASA published recently: https://climate.nasa.gov/climate_resources/300/video-climate-spiral/. We use the original dataset and add a few modifications of our own!

As usual, we use just matplotlib and make all the code public in the form of a self contained Jupyter notebook.

Subscribe now

We’re looking forward to hearing your thoughts, so go ahead and….

Leave a comment

NASA has recently published a nice animation of the monthly global temperature anomaly (changes from an average) between the years 1880 and 2022. Colder than average temperatures are represented in blue and white, while warmer than average temperatures are shown in red. Unsurprisingly, global temperatures have increased from mainly human activities as time has progressed.

We download the original dataset form here and perform a small amount of preprocessing

3D Surface Plot

Bruno Gonçalves — Sun, 20 Feb 2022 15:57:00 GMT

Welcome to the latest post of the Viz4Sci substack. This week we continue the exploration of matplotlib’s 3D plotting capabilities that we started in a previous post by using plot_surface to generate a 3D view of the population distribution of the 48 lower United States.

We finalize by adding a custom light source and shading to make the resulting figure more appealing.

Subscribe now

We’re looking forward to hearing your thoughts, so go ahead and….

Leave a comment

US population map

For this visualization, we’ll use the population of the 48 contiguous United States from NASA’s Gridded Population of the World project that provides the number of residents within a grid where each cell has a resolution of 0.25°x0.25°.

As always, imshow provides us with a quick heatmap of the data where all the major cities are easily identifiable.

Christmas Tree Animation

Bruno Gonçalves — Sun, 26 Dec 2021 13:49:52 GMT

Welcome to a very special Christmas post of the Viz4Sci Substack! This week we use the matplotlib 3D plotting and animation capabilities to generate a quick rotating Christmas tree GIF.

While relatively simple (and technically not a visualization) this quick example provides us with a minimalistic example of how easy it is to animate plots using matplotlib, even in 3D.

Subscribe now

And we’re looking forward to hearing your thoughts, so go ahead and….

Leave a comment

Let us start by using some basic numpy functions to create a simple 3D plot of two expanding spirals to make a stylized Christmas tree. We even add a Golden Diamond as a topper.

As you recall from our previous 3D post, after we set the 3D projection to the matplotlib Axes object, we can just add a Z component to our points:

Pyramid Plot

Bruno Gonçalves — Sun, 05 Dec 2021 23:23:05 GMT

Welcome to this weeks post of the Viz4Sci Substack! In this post we show you how to use matplotlib and seaborn to create a Pyramid plot by recreating a recent plot by the European Centers for Disease Control comparing the Vaccination rate and number of deaths per million inhabitants in the various EU countries.

Despite its simplicity, it is an informative and visually appealing plot that can be easily modified for other comparisons.

We hope you enjoy it, and of course, don’t forget to:

and

Subscribe now

so you never miss a post!

This week we will show you how to generate an beautiful yet informative visualization recently published by the European Centers for Disease Control (ECDC) that you might have seen on social media.

Florence Nightingale Rose Plot

Bruno Gonçalves — Sun, 21 Nov 2021 17:41:48 GMT

Welcome to this weeks post of the Viz4Sci Substack! In this post we show you how to recreate one of histories most famous plots, the Rose Plot, created by Florence Nightingale over 160 years ago to highlight the toll of infectious disease amount British soldiers during the Crimean war.

Don’t forget to

Subscribe now

and

Florence Nightingale, was a trainer of nurses in Constantinople (now Istanbul) during the 1853-1856 Crimean War. During her work, she collected an amazing amount of statistics on the causes of death of British soldiers. After the war, in 1858, she published the book “Notes on matters affecting the health, efficiency, and hospital administration of the British Army” with a detailed statistical analyses and visualizations that became one of the corner stones of modern day Statistics, Visualization and Public Health.

In her book, Nightingale makes the point that preventable (in particular, infectious) diseases were the greatest killer during the Crimean War. Her data and analyses lead to si…

Geographical Maps with Cartopy

Bruno Gonçalves — Sun, 05 Sep 2021 16:36:55 GMT

Welcome to this weeks post of the Viz4Sci Substack! This week we introduce you to some of amazing functionality of Cartopy, a state of the art package for geospatial processing and visualization. With it, we’ll create a beautiful global visualization of the location of every airport using BTS data.

Don’t forget to

Subscribe now

and

Airport Locations

The Bureau of Transportation Statistics collates a huge amount of information about US transportation networks. In the case of airports, we can download detailed information about each airport to which a US airline flies to (or that has direct flights to the US) from the BTS Tables, whimsically named “Master Coordinate” Table that BTS describes as:

This table contains historical (time-based) information on airports used throughout the aviation databases. It provides a list of domestic and foreign airport codes and their associated world area code, country information, state information (if applicable), city name, airport name, city market information, and la…

Candlestick Chart

Bruno Gonçalves — Sun, 29 Aug 2021 18:56:50 GMT

In this weeks post of the Viz4Sci Substack we implement a nice Candlestick Chart from scratch in matplotlib to visualize DJIA stock prices. Candlestick Charts are a powerful way of summarizing a lot of information about the hourly, daily or monthly price movements that dates back to the XVIIIth century Japan and are still used today in the stock market.

We hope you enjoy this post and…

share it so we can continue to grow, and don’t for get to…

Subscribe now

subscribe to make sure you never miss a post. We’re looking forward to hearing your thoughts, so go ahead and….

Leave a comment

Candlestick Charts summarize the four most important values in any trading period, High, Low, Open and Close, in a compact and characteristic shape from where their name is derived:

Matplotlib Heatmap

Bruno Gonçalves — Sat, 07 Aug 2021 21:54:15 GMT

Welcome to the latest post of the Viz4Sci Substack! This week we play around with matploltib’s heatmap functionality and use it to visualize the peak times of the Top 50 countries by number of CoVID-19 cases.

Subscribe now

And we’re looking forward to hearing your thoughts, so go ahead and….

Leave a comment

This week we once again use the JHU CoVID dataset has steadily become the unofficial standard dataset for the number of cases and deaths around the world. Yo can checkout out our previous post using this data repository in the time series state map post.

Today’s application is fairly straightforward. After some normalization and sorting, we simply use plt.imshow (we used it a couple weeks ago to display photos in the Color Blindness post) to display our matrix of values as a heatmap:

GapMinder bubble plot

Bruno Gonçalves — Sun, 01 Aug 2021 18:49:21 GMT

This week we look at how we can combine the all the goodness of Seaborn and Matplotlib to generate Gapminder-like bubble plots.

This kind of plot was developed and popularized by the amazing Hans Rosling, a Swedish physician, public speaker, and science popularizer. You should definitely checkout his TED talk and his book, Factfulness, if you haven’t yet. Truly amazing stuff.

Subscribe now

Bubble plots are ideal ways to encode multiple features of our dataset within a single plot in an intuitive and appealing way. In particular, we’ll generate a plot where we will display 4 different features for each country in our dataset.

Color Blindness

Bruno Gonçalves — Sun, 25 Jul 2021 13:29:54 GMT

Welcome to the latest post of the Viz4Sci Substack! This week we have a look at how we can check out what our figures look like to people with different kinds of color blindness.

Subscribe now

And we’re looking forward to hearing your thoughts, so go ahead and….

Leave a comment

Color blindness is a genetic condition that impacts the way the cones in your retina are able to perceive different light colors (wavelengths) and that affects about 8% of individuals and even well known personalities like Mr. Rogers, Mark Zuckerberg (which is why the Facebook logo is blue) and, reportedly, even Bill Clinton.

3D Plot

Bruno Gonçalves — Sun, 11 Jul 2021 15:10:27 GMT

Welcome to the latest post of the Viz4Sci Substack! This week we have a quick introduction to 3D plotting in matplotlib using a simple Fourier Transform visualization.

Subscribe now

And we’re looking forward to hearing your thoughts, so go ahead and….

Leave a comment

We’re all familiar with matplotlib’s capabilities in all forms of 2-dimensional plots, but did you know that it’s just as easy to generate beautiful 3D plots with everyone’s favorite Python plotting library? This week we explore a little known matplotlib functionality, 3D plots.

To create a 3D Axis object, all you need to do is use add_subplot to add a new subplot to an existing figure and set its projection to “3d”.

Time Series State Map

Bruno Gonçalves — Sun, 04 Jul 2021 14:05:36 GMT

Welcome to the second post of the Viz4Sci Substack! This week we work through a stylized map of the United States where each state is represented by a square with a time series of the number of CoVID-19 cases.

Subscribe now

This week we’ll use the Johns Hopkins University CoVID19 dataset which is the standard dataset for the number of cases and deaths around the world. You can download the continuously updated data from their GitHub repository. In particular, we will use the file (time_series_covid19_confirmed_US.csv) for the United States.

Matplotlib supports a wide range of patches that correspond to simple geometric shapes like Arrows, Circles, Rectangles, etc. For our map, we will use FancyBoxPatch to draw rectangles with rounded corners at specific x, y coordinates.

Patches are added to a pre-existing matplotlib Axes object by creating a PatchCollection that you can instantiate with a list of patches and respective colors. Creating our map is then pretty simple, all we need to do is create a Fanc…