For this mapping exercise, I used geocoded disasters data from NASA's Socioeconomic Data and Applications Center
I have no way of knowing how complete this dataset is— especially for Arunachal Pradesh and Jammu and Kashmir. For my analyses, I used a global dataset and sliced it just to keep natural disasters specific to India. At a glance, floods are the most common disaster, followed by droughts, which is consistent with common knowledge.
The dataset includes natural disasters from 1960 to 2018. For future mapping assignments, for a dataset like this, I'd like to be able to show the magnitude of destruction from these disasters too.
The biggest challenge with this map was ensuring I'm able to show all Indian states. Mapbox by default doesn't show disputed territory. Or, rather, shows them from a US-viewpoint. For India, that means that the state of Jammu and Kashmir and Arunachal Pradesh aren't fully shown within India's borders. Google does something similar — showing different borders on its maps to users from different regions.
I didn't want that. I wanted to create a map that's the same for users from everywhere — showing both Arunachal Pradesh and Jammu and Kashmir as part of India.
To do so, I used the example code provided in Mapbox's documentation.
The example code creates a menu of sorts with options to select worldviews from its pre-configured set of options — from India, US, China and Japan. Here's what that looks like:
Notice how the boundary around Arunachal Pradesh changes dynamically. Mapbox's example code makes US the default worldview when a user lands on the page. I configured mine to make it India's and then removed the option to toggle.
The other slight stylistic challenge was— This feature, to toggle worldviews isn't available on all Mapbox styles. So, after much trial and error, I gave in and used the style that was used in their example.