In 2016, our project director Stephen Goldsmith wrote about how Washington D.C.’s Urban Forestry Administration (UFA) used lidar—a radar-like technology that generates 3-D representations using laser pulses—to count and asses its city’s many trees, ultimately producing a map that detailed their individual heights and crown widths. After this initial success, the UFA discovered a number of opportunities for the technology. For example, they found that field arborists could use the map to quickly and efficiently determine the legality of tree removals or use elevation data to mitigate storm water with strategic tree planting.
Utah has had a similarly positive experience with lidar. In 2011, the state began using lidar to assess the pavement conditions of its roads, but soon found a wide number of applications for the gathered data, including hydraulic modeling and roadside measurements. In 2017, the state reported impressive savings: Through the use of lidar, Utah cut its spending on billboard inventory and measurements from $144,000 dollars down to $400; spending on a bike corridor inventory went from $15,000 down to $25.
The takeaway from each example is that lidar data is relatively cheap and widely applicable. This makes sense, if you consider how all-encompassing the data is. Even if the original purpose for a lidar project was to count trees or assess pavement conditions, it’s not as if the data captured has only to do with treetops or asphalt. Instead, lidar captures everything—trees and shrubs, hills and valleys, cars and houses, billboards and road signs—and it’s the work of algorithms to sort out what’s relevant after the fact, like a keen eye picking out details in a high-resolution photograph. Once their work is done, the data still remains, ready to be analyzed for some other purpose.
This points to a general guideline for what to do with lidar data: Make it public. Groups and agencies like the National Ecological Observatory Network (NEON) and NOAA have already taken this step with the understanding that the data can be used for many different research projects—e.g., this geomorphology project, which developed an automated model for classifying barrier islands using NOAA lidar data. And in fact, in 2016, the UFA was also working off of borrowed data, namely data gathered by a lidar flight the U.S. Geological Survey ran after Hurricane Sandy to assess storm damage.
Cities have also continued the trend. Columbus, Ohio; Philadelphia, Pennsylvania; San Diego, California; and many others have shared lidar data online. The city of Singapore is soon to release Virtual Singapore, a three-dimensional visualization of the city built on a base of lidar imagery. To maximize the resource’s utility, the city plans to make the model accessible through the Internet, allowing citizens and other interested individuals to make use of the same data that urban planners intend to use to better model traffic and connect IoT sensors across the city.
Currently, city and state governments use lidar devices affixed to planes and fleet vehicles, but a new resource is soon arriving: self-driving cars. Almost all self-driving cars utilize lidar to navigate the world around them, so it’s worth asking whether city and state governments could piggyback off the gathered data to do the sort of work that the Utah Department of Transportation and D.C.’s UFA have already begun. This is not a new idea—researchers and app developers alike have been using GPS data from cars for years now—but lidar data from Autonomous Vehicles (AV) might soon offer a unique and cost-efficient opportunity for cities to map their cities in real-time. Data sharing of this sort is already beginning: Though once hesitant about sharing data, Uber is now working with cities to develop curbside maps using the company’s data on pick-up locations.
As more and more agencies discover uses for lidar technology, the argument for open data strengthens. Since there’s no way of knowing beforehand what data will be useful for what purpose, the number of potential cost-cutting applications for a resource like lidar is effectively immeasurable. Perhaps it’s better, then, for the data to remain in the hands of those who stand to benefit from it: everyone.