Cartography and Perseverance through Mapping Mars
August of 2005, I was oft found on the floor of my dorm late into the night coding MySpace profile for friends. Meanwhile, back home in Cape Canaveral, NASA was launching the Mars Reconnaissance Orbiter, which would enter orbit a year later, inevitably helping to establish landing sites for following missions. Last year, the latest of those missions, Perseverance launched from Kennedy Space Center and today finally arrived at Mars. Appropriately, I’ve been back at it, up late, hacking together a map to commemorate JPL’s latest mission.
Once you commit to explore, you may as well accept there’s no going back
As last week came to a close, I was reminded of the news of Perseverance’s proximity to Mars and ultimate timeline for landing. Thinking about how I could celebrate, I considered creating a design kit for the mission, focused on the Mars Helicopter Ingenuity, likely something that would make building a game or app prototype approachable.
The task seemed easy enough, but I assessed constraints first and realized there was a bit more work needed. To start, native map support isn’t abundant with prototyping tools, though there are plenty of ancillary options. I found Framer supported Mapbox, pretty early on, so I opted to go in that direction given that I wanted a live map to work with. Google Earth has Mars data, as well as the other one-off maps, but these options weren’t as viable with the tools.
Unfortunately, work done on Martian maps was pretty limited within Mapbox. There was one relic, evidenced by a decade’s-old post, but when I reached out, their team confirmed that the data hadn’t survived. Fortunately, there were instructions included, and some sources, so I opted to take up the task of rebuilding Mars in Mapbox.
So in under a week, I needed to learn map design, collect appropriate datasets and piece it altogether together. The challenges of map design are many, but to summarize, my experience was more akin to guesswork than intention.
When it comes to maps, everyone’s a critic
Where critics of modern maps are many — with commentary on efficacy and accuracy as commonplace — fewer actually praise the masterful work that’s been done by cartographers for ages. Rather, we’ve become so accustomed to having maps for anything and everywhere, and at our fingertips nonetheless, that we’ve heightened our expectations.
When we hear about map critiques, it’s far more generalized, and much less specific. We don’t have a good grasp for the jargon, or even heuristics, so our assessment of maps often is comparative, rather than evaluative. But maps come in many forms with variaties of purposes. They’re supposed to inspire and inform, or let people explore, not just act as a personal GPS. We take for granted the capability of maps to automate navigation or even warn of upcoming traffic, but it’s rather incredible work.
Hillshade, relief, slope, contour, tiling — maps are made of these things, plus more. There are tons of cartographic terms that sound familiar and logical, but out of context means nothing to the layperson.
As someone looking to design maps (enter me), it’s not particularly easy either. Yes, there are great studios and softwares, but their knowledge and resources are largely out of reach for most. If you want to learn, it’s not encouraging or easy. In talking with friends at Mapbox, NatGeo and GIS firms, I found solace in knowing that despite even the best tools, the tech is too technical and map design is difficult either way. Needless to say, it took me an incredible amount of time, many hours and a couple days worth of dead ends before I started to get it.
Install GDAL, they said. It’ll be helpful, they said.
So building this thing — well, it wasn’t pretty. I kicked off to a pretty rough start. I had recently transitioned computers, so npm, Python, node weren’t updated, let alone installed. And while I grew up on DOS, I don’t use the command line enough to even feel confident referencing files. Fortunately I could understand most of the documentation, but leaned heavily on forums to slog through it.
When I started, I attempted using TileMill and GDAL, but realized the half-life of software, or least my own patience for it, would prevent use. Cartographica and MapTiler helped for checking my exports but were limited otherwise for my needs. And as I started to find Martian data, I would realize that file size was something I’d inevitably have to get around as well. Not all maps involve starting from the ground-up, but that’s exactly what was necessary for this Mars project.
I ultimately landed on QGIS to do most of the processing and found that once I started to understand relevant filetypes and tools a bit more, it was most fitting. In reflection, there are a few things that would have been helpful in retrospect:
- To start, there are a whole set of world files that add a layer of georeferencing data, such as GeoTiff, but many filetypes that are used to simply house data, like CSV and XML are also used.
- Most layers can be processed in QGIS from a single image containing elevation data (GEL), using GDAL. Larger files need to be broken down to process or upload, and tiling processes like gdaltranslate or exporting layers in QGIS to mbtiles will do the heavy lifting. Following me?
- For raster files, styling needs to happen before export, where datasets and vectors can be styled within a tool like Mapbox or in the browser. Most of my earlier exports were rasters, which has quality limitations and little ability to style, but this is unavoidable for satellite imagery.
- Also, there are tons of datasets out there. NASA, JPL and many other sources provide ftp access to their libraries, which are vast and incredibly documented. In fact, most of the conversions, nomenclature and datasets I needed from QGIS were found once I started digging in.
Once I was able to start getting files in an acceptable form for upload, I could start seeing the map come together on Mapbox. And with a first version down (shown top of article), I moved to a new map and have been tinkering with datasets and vectors. As the base layers are uploaded, it becomes much easier to see how it comes together visually. The result includes location names, contour vectors, mapping layers but minimal styling.
So about that UI kit…
I didn’t quite get to that prototype or resource kit, but my pivot into developing a map to start styling with at least left me inspired and eager. There’s an abundance of data to be had and really opportune use cases around the maps, so I hope to continue in building on this map, and perhaps others as well. But most successfully, now there’s a place to start in importing a map, regardless of tools. The Mapbox data is available for others to build on, reference or use as-is.
The process was harrowing, left much to be desired. Not because of the tech or the terminology, but more so the variance in exports and inability to know whether I was going in the right direction or not. Data validation in this regard is a bit more manual, as QGIS exports don’t necessarily guarantee usable data. But that’s probably on my end more than QGIS. Funny enough, as much as I work in the unknown and enjoy figuring things out, I found that this in particular was really out of the familiar, but fun once I got started.
So this pilgrimage for Mars Perseverance was a sort of trial-by-fire. Without really knowing about creating maps, or map design, I more or less brute-forced my way through the documentation, and KyngChaos’ and Unix/python/stack overflow/GDAL forums to functionally get something that represented a map. The vast majority of my time spent was trying to find a viable tool given my constraints, and troubleshooting the usability of the tools. Where map design fundamentals played a moderate challenge, the documentation and availability of data is vast: NASA, JPL, et al really saved time and in many cases provided exactly what was needed.
At the end of it, all of this was only accomplished with a willingness to adapt and figure things out, and with the help of many who are passionate about maps and GIS data. Of course, be it maps or design, when exploring the unknown and pursuinig new paths, Perseverance is key.
