Project description

I have worked in the Financial Times graphics department for the past 23 years. I have gone from working solely for the newspaper, to creating Flash interactive graphics, and learning HTML, Javascript and CSS.
My main area of interest is cartography which is evident in my selection of pieces from the past year. My goal in the pollution pieces was to raise awareness and challenge traditional thinking about where are the most polluted places in the world are. Using satellite data from Nasa really helped drive home my point is a visually appealing way.

What makes this project innovative?

China and India pollution –– China and India are both known for their huge populations and their choking air pollution – but China’s smoggy skies usually get more media attention. I drew on data files from NASA to show convincingly that India has it worse. This project was unique at the FT in that the story was initiated out of graphical research, and then expanded to include the written text. I created the graphics and then approached South Asia bureau chief Amy Kazmin and Lucy Hornby, deputy bureau chief in Beijing, to provide the interviews and context for the articles. I deployed 3D data animation and digital illustrations to create pieces that appealed to the brain and the heart. This collaborative reporting by the FT highlights the power of smartly designed, easily understood graphics to bring public policy questions to life. Together the graphics showed conclusively that air quality in Northern India has been far worse than in China in recent years, contrary to popular opinion. In India in particular, the graphics generated a public debate over a problem that is still not fully appreciated. The original goal behind “China’s polluted skies” was to show in a compelling visual way how levels of nitrogen dioxide (NO2) – a common air pollutant -- had changed over the past decade or so, and investigate the possible causes. A dark splodge on the southwest corner of the maps sparked our curiosity about the situation in India. When I pulled the additional files from NASA, he was shocked at the dark wall of pollution that smothered the Ganges plain. I created an animation to directly compare India and China and emphatically dispel the long held assumption that China’s pollution is worse than India’s. The heart of both pieces was the use NASA’s rich resources. For the China piece I downloaded monthly average levels of NO2 since Jan 2005; the India piece drew off annual particulate matter (PM2.5) data going back to 1998. The illustration at the top of each article was important, as it showed the human side of a story dominated by maps and charts. Choosing to use the illustrations created a look and feel that wouldn't have been possible with a photograph, and helped spark empathy in the reader. For India, where the air pollution problem is just beginning to be acknowledged, the question I tried to answer was how many people were impacted. The centrepiece of this article was the Marimekko chart of pollution data for over 80 countries. This chart type is the perfect way to show two types of information – the total population and the percentage affected, on a single graphic. I overlaid the Nasa pollution data onto a population density grid from the European Commission to produce the dataset. In the scatterplot chart I went one step further and adjusted the data for population density, to avoid skewing the mean numbers for countries such as China and Russia that have vast unpopulated regions. I used R, a statistical computing application, to perform the analysis. In both countries, coal is a major contributing factor to pollution. The accompanying charts on coal production and consumption, and the numbers of coal-fired power stations, helped give the reader some context as to why these pollution levels are the way they are. The FT has been building on its newspaper heritage by developing graphics and other online content that can drive debate in new media, while remaining firmly grounded in its core reporting on economics and policy. “China’s Polluted Skies” and “Dirty Air” are powerful examples of how innovative data and graphics can go hand-in-hand with our traditional expertise in text and analysis. Struggling to breathe –– The challenge with this project was to show how, even though London was a pioneer in limiting the number of cars, it is almost as polluted as Delhi by some measures. We worked closely with TfL and King's College to produce a series of maps and graphics that highlight problems unique to London. Taking the NO2 data from King's College and categorising it relative to the WHO safe limit highlighted just how bad it is in Central London. Designing a new colour ramp that put across the extreme nature of the top end of the scale worked really well. This was commented on by Joshua Stevens, Data Visualisation and Cartography Lead at Nasa (see additional link). Showing the slowest London bus routes in comparison to the average walking speed really drove home the point at how bad the congestion is in London. These slow moving diesel buses are a major contributor the NO2 pollution in London Prime suspect –– The goal of this project was to highlight not just the eye watering prices of these luxury skyscrapers in an already slowing housing market but to illustrate the impact it would have on the Manhattan skyline. This project began with a list of locations of currently under construction and planned constructions. I painstakingly tracked down the locations of each building using Google maps and drew the new footprints alongside the dataset of building footprints from NYC Open Data Portal. These were then used to create a digital elevation model (DEM) inside QGIS based on the height of the buildings. I then took the DEM into Blender to create the 3d model, colour coding the buildings by date of construction. Running the full extent of the Manhattan skyline along the bottom of the graphic really highlighted how transformational these new buildings would be to the Manhattan skyline. High rise downturn hits Manhattan (social animation) –– The challenge for this was how to take a half page graphic and condense it down into something that will work on Instagram. The only logical solution was to create an animation of the graphic, highlighting some of the key locations. This was created in Blender and completed in After Effects Why water is a growing faultline between Turkey and Iraq –– For this project we were tasked with showing what the impact of Turkey's damming of the Tigris river would have on Iraq and and the ancient town of Hasankeyf. I collaborated with Chris Campbell on this piece who created the large annotated map of Iraq. For the flooding animation and impact on Hasankeyf graphic I used detailed digital elevation models of the region, which were taken into Blender and used in conjunction with raising a flat plane plane to give the effect of what the extent of the reservoir would be once it is filled. Making a comparison to another landmark is always a useful frame of reference. For the close up of Hasankeyf I used google satellite imagery in conjunction with a digital elevation model to show the extent of the flooding of Hasankeyf. This graphic really drove home how an entire town full of ancient ruins of historical significance will be completely submerged. With the existing residents expected to relocate to a new development just to the north.

What was the impact of your project? How did you measure it?

The two pieces relied on different techniques to answer specific questions about the policy challenges in each country. In the case of China, the government is claiming progress in fighting pollution following a public outcry, and the graphic was designed to show if that is true. The piece generated excitement among environmental campaigners, who have been split over the question of whether policy or economic factors have had the greatest impact. “Dirty Air, How India became the most polluted country on earth” triggered active discussion and raised public awareness in India, where air pollution is a relatively recent public concern. Both pieces also generated tens of thousands of uplikes and hundreds of comments on Reddit’s “data is beautiful” forum, an exceptional performance.   Dirty Air's performance on reddit was exceptional, it generated a fierce debate in the comments section. This generated 9,406 referrals to the story, the 5th most of all time from Reddit. It also generated a huge spike in traffic with an additional 90,000 page views, putting it in the most read top 5 for two days running, one month after the initial publication date. This has been a great driver of traffic to stories that people would not expect the Financial Times to cover and has helped push our subscriber numbers over the 1 million mark in recent days.

Source and methodology

For India and China I collected vast number of geotiffs from Nasa from their Giovanni tool which allows you access to their huge repository of satellite based data. These data files were loaded into a geographic information systems application (called QGIS) and coloured based on the concentration of pollutants. In the case of China I took this data into the third dimension by creating digital elevation models of each month of averaged data and then took these into Blender to create 3d height maps, this helps emphasis the spikes in pollution more than just a simple heat map would. India Socioeconomic Data and Applications Center (sedac) –– Global Annual PM2.5 Grids from MODIS, MISR and SeaWiFS Aerosol Optical Depth (AOD) with GWR, v1 (1998 – 2016), European Commission of Global Human Settlement –– GHS POPULATION GRID, World resources institute, Haver Anaytics, Central Electricity Authority China Nasa/OMI satellite data, Greenpeace, CoalSwarm Global Plant Tracker, China's National Bureau of Statistics, CEIC London pollution TfL, King's College, EEA, Aerosol and Air Quality research, Central Pollution Control Board Manhattan NYC Open Data Portal; Council for Tall Building and Urban Habitat Tigris dam FT research, Nasa digital elevation model

Technologies Used

Primarily the technologies used across all these projects is a combination of QGIS, Blender, Adobe Illustrator, Photoshop and d3 All of the mapping projects are created using QGIS. If they require relief mapping then I obtain the digital elevation model data from Nasa's Earth Explorer and bring it into Blender to create a 3D terrain with an overlay of satellite imagery from Bing that I colour correct in Photoshop. The Manhattan skyline was created by downloading building footprint data from the NYC Open Data Portal and using it to create a digital elevation model inside QGIS based on the height of the buildings. This was then taken into Blender to create the 3d model and animation, which was then finished off in After Effects for use in social media promotion. All of the of the non-cartographic data visualisation is created using a combination of d3 and Adobe Illustrator. We have a vast array of chart types at our disposal thanks to the creation of our own set of d3 templates called the Visual Vocabulary. The illustrations that accompanied the China and India pollution stories were hand drawn using the Apple pencil and iPad Pro in the Procreate app

Project members

I worked alongside Chris Campbell on the Tigris damming project, Leslie Hook on the London pollution project.


Additional links


Click Follow to keep up with the evolution of this project:
you will receive a notification anytime the project leader updates the project page.