The Solar Nerd
Resources to help the homeowner go solar.
One of the key strategies to bring users to this website is content development. Performance is also key, both in terms of quickly delivering a lightweight experience to the end user, but also being able to run the site on minimal hardware. I knew from the beginning that static site generation was the direction I wanted to go.
After a little research of the popular options, I decided to do a little experimentation with Hugo. After a couple hours it was obvious that this was the right tool. It is fast to get up and running, has a logical and intuitive approach to content management and templating, and good documentation.
I'll write some more about this in the future, but so far it's been a great backend tool for creating my blog.
In my blog on The Solar Nerd, I'm going to try to do analyses that are relevant to my audience is that you can't find anywhere else. One of the things I've been working on lately is data visualization, especially ones that involve mapping solar-related data in a way that helps people quickly see different dimensions in how solar is spreading.
Whenever possible, I want to try to capture this in a single visualization rather than have to rely on a set of different maps or graphs. I think that if you can figure out a way to distill a complicated concept down to a single image, it becomes a lot more powerful.
The first dataset I wanted to do this with is a list of 1.4 million home solar installations across the United States. The geographic resolution of this data goes down to the zip code. In the zip codes with the highest number of installations, you might have as many as 30,000 solar installations.
This was the main challenge of this map: how to retain both the fine-grained geographic detail of this dataset and the wide range of possible data, which may range from zip codes with a single solar installations to ones with tens of thousands, all in a national US map.
The first approach I tried is a heat map. Tableau is an application that lets you very quickly load a dataset from a CSV and then project that data onto a map, using a color to represent the range of values.
The resulting map looked visually interesting, but there was one big problem. The heat map that Tableau produces requires a large radius of pixels to represent a data point, which eliminates the fine grained geographic resolution of the data.
After a little more searching, I came across CesiumJS (https://cesiumjs.org/) which, in their words, is "An open-source JavaScript library for world-class 3D globes and maps". This sounded like what I needed: by adding a third dimension, I could retain the geographic resolution I wanted, and then use the 3rd dimension to fully represent the dynamic range of the data.
It was exactly what I needed. It look only a few hours to grok the API and produce the map I needed. My blog post, and a link the full map, is here:
https://www.thesolarnerd.com/blog/posts/mapping-million-solar-installs/
CesiumJS was a great choice, and I plan to use it for future data visualizations I have in mind.
Remember that polar vortex last month? That's when I decided to go public with my solar energy-related startup. Yes, I think there's a certain logic to that too!
Comment