How Data Helps Us Observe—and Preserve—the World
The onslaught of bad news around climate change and its effects on communities around the world can be dispiriting. No individual data scientist, after all, will singlehandedly solve issues that require systemic action and global coordination.
Data professionals nonetheless have important roles to play in the movement towards a sustainable future. Their skills and creativity put them in a unique position to assess current challenges, measure the efficacy of proposed solutions, and draw our attention to human impacts on the world we inhabit.
This week, we highlight recent articles that help us understand the environment in concrete, actionable, decidedly not doom-and-gloom ways. Let’s get started.
- Getting your hands on the right data is essential. While climate data is abundant, María Bräuner rightly points out that “some freely available data might not be as easily accessible as one might think.” This clear and patient tutorial walks us through the process of obtaining and processing satellite-generated NetCDF files in R, a crucial first step for any climate-centered project.
- See the (literal) forest for the trees. Still on the theme of satellite geospatial data, Sixing Huang devoted a recent article to current methods for monitoring the health of forests over time—which can be helpful for a number of use cases, like measuring deforestation and predicting wildfire risk.
- How to leverage open data for sustainability. It’s hard to promote renewable energy sources like solar power without a clear understanding of their ability to meet demand. Ang Li-Lian’s nuanced walkthrough shows how we can use open-source data repositories (and a few lines of Python) for the complex task of estimating solar panels’ output.
- Another approach for gauging renewables’ potential. Having explored solar power, why not take a close look at wind energy, too? Abiodun Olaoye does just that in his new article, which covers an open-source module for predicting wind-turbine power production.
- When you need a reminder for what is at stake. The pressures of climate change rarely stay abstract when we’re out in nature. Not sure how to start or where to go? Evan Diewald’s latest project might inspire you to grab a bottle of water and put on some comfortable shoes: it’s an open-source hike-planning app and route-optimization engine, and Evan provides all the code he wrote to create it, so you can tinker with and customize the app on your end.
If you’re looking for a few more standout articles to read this week (on the way back from an energizing hike, perhaps?), you can’t go wrong with any of the following:
- We chatted with researcher and entrepreneur barrysmyth about shifts in the academia-vs-industry debate, and how data scientists should position themselves for success.
- Peggy Chang’s new deep dive is a must-read for anyone who’d like to understand how similarity search works on a massive scale. (How massive? billion-vector-datasets massive!)
- Computer vision finds a promising use case in Mihir Garimella’s first TDS post, which explains how a MediaPipe- and Keras-based workflow can detect sign-language characters in real time.
- Sophia Yang helps readers make sense of current approaches to model explainability with a high-level overview of eight popular techniques and tools.
- For data scientists who want to contribute to their companies’ bottom line, we’ve collected some of our favorite resources on data’s business impact in our latest Monthly Edition.
If you’d like to support our authors’ work, consider becoming a Medium member—you’ll also gain access to our entire archive along the way.
Until the next Variable,
TDS Editors
