After exploring the CMIP6 dataset, our group decided we wanted to create a visualization that shows how global warming has intensified in recent years. To do this, we explored different ways to convert the grid data we were given into an interactive map of warming. One of our prototypes involved binning each grid into a region and using projected warming scalars to create the visualization. However, we decided to stay more true to the dataset we were given and continued with our grid data. To encode the warming, we used a blue-to-red continuous color scale, which enables the viewer to easily see how certain areas are more affected by warming than others, as viewers intuitively associate blue with cooler temperatures and red with warmer temperatures. Furthermore, we added a slider at the bottom to allow the user to see the impact of warming year by year, and also implemented a play button to easily see the trends of global warming. Additionally, we added a tooltip for certain major cities like San Diego, London, and Tokyo, among others, to provide local context without cluttering the map. This tooltip includes the average temperature and anomaly of the given year, as well as the coordinates of the city and a small time-series graph of anomalies from 1850 to 2014. Overall, we hope that this visualization provides the viewer with insight into how global warming is affecting different regions at different rates.
For this project, we found the project checkpoint to be quite useful and guiding in getting us to our final product. Exploring multiple design routes as we approached this creative assignment was essential to landing on a concept that maximized the potential of D3 interaction, and the detailed data at hand in CMIP6’s dataset. This section of our process was split evenly with each member of our group completing 1-2 noninteractive graphs and examining the pros and cons of each other's drafts. Luckily, everyone shared the same opinion that the world map experimental visualization would be the most visually interesting and insightful when paired with user interaction capabilities. We then began experimenting with D3, keeping the same idea from our chosen map visualization, but specifically aiming to incorporate a year slider and an annotation clicker for certain points on the map. For this section, we all got together and worked in our own local environments, while sharing ideas aloud and sending files back and forth. We found this form of face to face collaboration to be extremely productive. Between the four of us, we were able to build our visualization and website in about 6 hours across a couple days. Most of this time was spent either straightening out our dataset or fiddling with D3 features so they were easy to use and fit well with the application. Overall, we are very pleased with our development process and the collaboration experience it provided us.