Environmental Science Division (EVS)a Division of Argonne National Laboratory
Predictive environmental understanding

Benchmarking Argonne’s UAS capability for environmental studies

December 20, 2019

Use of unmanned aerial systems (UASs) has become a widely adopted method of scientific data collection. The low cost of instruments, ease of operation, and flexibility of data collection makes them useful tools across a range of scientific disciplines — from wildlife and crop monitoring to assessment of fire damage — in large, remote, or inaccessible areas. A team from the Environmental Science (EVS) division at Argonne recently benchmarked the Laboratory’s UAS capabilities for environmental research.

Yuki Hamada, a biophysical remote-sensing scientist, led the study. Hamada wanted to enable Argonne researchers to use UASs to study environmental and cultural resources. “To use UAS effectively, we needed to understand our capabilities and determine how to best characterize and monitor environments using Argonne’s current UAS infrastructure and skill sets,” she explained.

In using UASs, a researcher first needs to know its capabilities and limitations. For example, can the aircraft carry an appropriate sensor? How long can it fly? How much data can it collect? The answers will help researchers to decide whether the UAS is a suitable tool for their research. Researchers also need to know how to convert the images collected using UAS into meaningful information.

The goal of the project was to gain firsthand experience and share it with Argonne researchers to eliminate fundamental mistakes and manage uncertainty when using UASs for their projects. The team focused on assessing and monitoring three areas of application: cultural resources (presence of subsurface artifacts), natural resources (grasslands and woodlands changes), and infrastructure integrity (road damage). The team collected images using two types of UASs: one equipped with an RGB (red-green-blue) camera, which captures images similar to those seen with the human eye, and the other with a thermal infrared (TIR) camera, which senses the temperature gradient of land surfaces, which may be influenced by underground conditions such as the presence of artifacts beneath the earth.

In a collaborative effort with the Strategic Security Services (SSS) and Decision and Infrastructure Sciences (DIS) divisions, the team endeavored to demonstrate how UASs could be used to collect environmental observations. Eric Swanson and Tommy Sunderhaus (DIS) developed flight plans and flew UASs over a portion of the Argonne site to collect thousands of images. Robert Herrera (EVS) used the collected data to create a three-dimensional model, known as a point cloud, then mosaicked the thousands of images to create a single composite, “analysis-ready” image that captured the entire study area. Hamada further processed the seamless mosaicked image to extract information that researchers could use for their studies such as indication of the foundations of old buildings beneath the ground; types and condition of vegetation, including decline and regrowth of plants; and damage to roads.

The study showed examples of what environmental data UASs can collect; the type, detail, and quality of the images; as well as the benefits and limitations of existing UASs in the Laboratory. This information will allow researchers to make sound decisions when designing new research and projects.

“This was a necessary first step, and we learned so much,” Hamada said. “We can use this experience and lessons learned for designing future research knowing the effectiveness and limitations of UASs, as well as plans for mitigating their shortcomings.”

Knowing how to deploy the right resource at the right time — either a UAS or a human — is an important factor that determines the quality and quantity of data for environmental science research and projects.

Variation of vegetation in a part of the Argonne site. Combination of plant properties such as type, condition, height, and age generates a range of colors in a processed images collected using UAS.
Variation of vegetation in a part of the Argonne site. Combination of plant properties such as type, condition, height, and age generates a range of colors in a processed images collected using UAS. [Source: Argonne National Laboratory]
portrait of Yuki Hamada
portrait of Bobby Herrera