How One Team Developed Accessible Maps for Colorblind Scientists

Imagine you have to do your job, but you can’t visually process the data right in front of you. About 8% of genetically modified men and 0.5% of genetically modified women have some form of color vision deficiency (CVD), which is a reduced ability to distinguish certain colors. CVD is commonly referred to as color blindness.

Scientists use colors to convey information. Many scientists in the weather radar community have CVD, and using and interpreting color is an important part of their work. Most color maps do not account for CVD; for example, maps show green next to red.

Researchers at the U.S. Department of Energy’s (DOE) Argonne National Laboratory recently published a paper in the Bulletin of the American Meteorological Society detailing their work to create CVD-compatible color maps that highlight important features of clouds and precipitation.

The creation of new maps demonstrates the programming community’s commitment to inclusivity

One type of color map, Ze, shows radar reflectivity or storm intensity. The higher the reflectivity factor, the greater the chance of rain, hail, etc. In general, the higher the Ze, the greater the chance of impact to the public. These color maps are the ones the general public sees most often in local weather forecasts.

Some color maps, such as Ze, make it difficult for people with cardiovascular disease to distinguish between different types of precipitation and convection. Convection is the transport of heat and moisture by the movement of a fluid. Thunderstorms are a form of convection.

Another problem with current color palettes is that they are not perceptually uniform. Perceptual uniformity occurs when changes in color (brightness or color) and data values ​​are weighted equally and do not create artificial structure.

As they interacted with the community, the scientists realized they needed to improve the data display. Their work focused on finding a suitable color representation that was both compatible with CVD and suitable for the radar community.

The researchers used this knowledge and created CVD-friendly velocity and Ze color maps using Python programming tools. They gave each element its own color, from drizzle to rain to hail. The velocity color maps were also tested using an existing oceanographic map.

“The Python programming community aims to be inclusive,” said Scott Collis, an Argonne atmospheric scientist and co-author of the paper. “And users have pushed to make the color maps more accessible.”

Once tested and published, the maps will have a longer-term impact thanks to next-generation meteorologists

The new color maps and existing color maps were tested on different weather events using software that viewed the data through the eyes of a person with cardiovascular disease. Comparisons were then provided to the cardiovascular disease community for feedback. Overall, radar researchers in the focus group agreed that the new color maps were more interpretable than the default color maps currently in use.

This research was supported by the Atmospheric Radiation Measurement (ARM) User Facility, a DOE Office of Science user facility. “ARM makes it easy to do great science, visualizations, and tools that are open to the community,” Collis said. “Without ARM, none of this would be possible, and in the practice of open science, ARM data, tools, and software have more impact.”

The new color maps are now available in a GitHub repository used by a variety of open source radar software packages found in the open radar community.

To further engage the CVD community, the authors presented their color maps at scientific conferences around the world. The community provided feedback, tested the color maps, and contributed to the creation of the library.

“Now that we have better color maps, we know from the community that they are relevant and inclusive,” said Zachary Sherman, co-author and administrator of Argonne Systems Integration. “People have a choice to be colorblind-friendly.”

Some authors have personally seen the impact of their color maps on people with cardiovascular disease. Everyone from students to conference co-chairs expressed gratitude for the color maps they could finally understand. Making these color maps available for cardiovascular disease enables equitable and inclusive scientific visualization.

In the future, the research team plans to develop more color cards and conduct additional outreach efforts. Their goal is to create a community practice around the world.

“Radar meteorology is a unique visual science,” Collis said. “Scientists can see the shape of the storm that could signal strong winds or a tornado. That’s why the color maps are so important to our community.”

It’s not just the radar community that will benefit from CVD-compatible color maps. The landscape for tomorrow’s meteorologists is changing. As aspiring students graduate, they work exclusively with radar meteorology tools that feature CVD-compatible color maps. These types of color maps are the standard for the next class of meteorologists.

Then, as they take on roles on local television stations, they will begin to bring inclusive color maps to the general public through weather reports. So one day in the near future, the general population of CVD will be able to watch the morning weather forecast and see, for the first time, how a color map is supposed to be seen.