Observations of environmental variables from satellites and aircraft, or collected in the field represent snapshots of a continuously varying Earth system at specific places and times. Satellite offer a global view but miss important details on local environmental conditions.  The emergence of advanced remote sensors on airplanes and drones offer new opportunities to bridge local and regional data with this global view. We are developing methods using advances in spectral techniques, machine learning, and Bayesian statistics to integrate these observations to help us better resolve and understand underlying ecological processes. 

 

We are creating new scale-oriented approaches to assess recovery of forests in the western U.S. from combined disturbances by integrating data from satellites, drones and the National Ecological Observatory Network (NEON) with advances in machine learning. We are gaining new insights by:

  • applying deep learning to identify plant species from NEON airborne hyperspectral and lidar data;
  • unlocking the 40+ year Landsat satellite archive using spectral methods linked to field and drone data to resolve changes in vegetation state; and
  • Applying A  hierarchical Bayesian model to test hypotheses related to disturbance co-occurrence and forest recovery.

 

We are also researching how we can utilize small remote sensing drones and multispectral commercial satellite imagery for ecological research by:

  • building a community of practice to reduce technical and knowledge barriers for use of,  and provide guidance for collecting and processing drone data;
  • Exploring how we can obtain repeatable and reliable vegetation structure and plant species information from drones; 
  • Investigating use of high resolution satellite imagery to compare vegetation cover measurements to those from USGS and NASA satellites, 
  • Investigating use of high resolution satellite imagery to derive block-level urban vegetation cover and impervious surface material composition.

Projects

Project Lead

Anna Spiers

Earth Lab Ecology & Evolutionary Biology

Drones are revolutionizing the way natural scientists measure their study systems. We are researching how measurements from small remote sensing drones, aka uncrewed aerial systems (UAS), can complement existing data to answer environmental questions in new ways.

Project Lead

Barbara Buttenfield

Department of Geography, University of Colorado Boulder

We develop methods to search across temporal, spatial and spectral information for relationships that can be exploited to inform analysis and interpretation of patterns, trends and outliers.

Project Lead

Jennifer K. Balch

Earth Lab

(303)735-8447

If you’re like us and you love studying fire ecology, you know that there are many satellite-derived datasets to choose from, but the collection is not easily combined and a steep learning curve exists. That’s where FIRED (Fire Events Delineation) comes in.

Project Lead

At Earth Lab, we often use data from diverse sources to facilitate inquiry, from the more conventional remote sensing datasets such as multispectral satellite imagery and radar backscatter, airborne lidar data, and high-resolution UAV imagery, to the less traditional datasets such as social media feeds, housing layers, and event databases.

Project Lead

From centimeter-scale imagery collected from UAVs, to airborne hyperspectral imagery at the meter-scale, to the 10’s of meter scale from satellite multispectral imaging systems, the diversity of data representing the Earth’s surface at different scales enables us to ask questions from the hyperlocal to continental and global scale. We combine these data to better understand processes and change occurring on the Earth.

Project Lead

Jennifer K. Balch

Earth Lab

(303)735-8447

We convened the first community organized NEON Science Summit to build community and better utilize NEON data for ecological research and education.

Team Members

Joe McGlinchy

Remote Sensing Scientist

Cutting-Edge Earth Analytics, Earth Data Across Scales

Michael Koontz

Research Scientist

Earth Data Across Scales

Adam Mahood

Graduate Student

Earth Data Across Scales

Victoria Scholl

Graduate Student

Earth Data Across Scales

Anna Spiers

Graduate Student

Earth Data Across Scales

Stefan Leyk

Professor

Earth Data Across Scales, Adaptation Science, Extremes & Natural Hazards

Barbara Buttenfield

Professor

Earth Data Across Scales

Nayani Ilangakoon

Postdoctoral Associate

Earth Data Across Scales, Extremes & Natural Hazards

Natasha Stavros

Analytics Hub Director

Cutting-Edge Earth Analytics, Earth Data Across Scales, Extremes & Natural Hazards

Max Cook

Graduate Student

Earth Data Across Scales, Extremes & Natural Hazards