
NSF Org: |
DEB Division Of Environmental Biology |
Recipient: |
|
Initial Amendment Date: | August 12, 2020 |
Latest Amendment Date: | August 12, 2020 |
Award Number: | 2017756 |
Award Instrument: | Standard Grant |
Program Manager: |
Matthew Kane
mkane@nsf.gov (703)292-7186 DEB Division Of Environmental Biology BIO Directorate for Biological Sciences |
Start Date: | December 1, 2020 |
End Date: | November 30, 2024 (Estimated) |
Total Intended Award Amount: | $344,914.00 |
Total Awarded Amount to Date: | $344,914.00 |
Funds Obligated to Date: |
|
History of Investigator: |
|
Recipient Sponsored Research Office: |
101 COMMONWEALTH AVE AMHERST MA US 01003-9252 (413)545-0698 |
Sponsor Congressional District: |
|
Primary Place of Performance: |
140 Governors Drive Amherst MA US 01003-9264 |
Primary Place of
Performance Congressional District: |
|
Unique Entity Identifier (UEI): |
|
Parent UEI: |
|
NSF Program(s): | MacroSysBIO & NEON-Enabled Sci |
Primary Program Source: |
|
Program Reference Code(s): | |
Program Element Code(s): |
|
Award Agency Code: | 4900 |
Fund Agency Code: | 4900 |
Assistance Listing Number(s): | 47.074 |
ABSTRACT
The lower atmosphere (i.e., aerosphere) is home to literally billions of organisms, including microbes, insects and birds. Species in the aerosphere often use other airborne organisms for food and are interdependent on one another. In recent decades, populations of many aerosphere organisms, such as birds and butterflies, have been rapidly declining in abundance. This project will examine the ecology of two bird and one bat species, all three of which feed on insects, and how their populations are responding in complicated ways to environmental change. These three species can also all be tracked when they emerge from their roosts by using state-of-the-art computer vision techniques with NEXRAD, the United States weather surveillance radar network. Project researchers will use the vast and ever-growing repository of data from the NEXRAD network to quantify the causes and consequences of ecological change in the aerial feeding and group habits of the two bird species (Purple Martins and Tree Swallows) and Mexican free-tailed Bats. The project will leverage environmental data from the NSF National Ecological Observatory Network (NEON) together with the radar data to identify the drivers of changes in abundance, feeding, reproduction and other seasonal patterns. The massive data sets will be integrated with one another to develop predictions of how these three aerosphere species are changing at regional to continental scale, and in response to environmental changes. These studies will also incorporate training opportunities for a postdoctoral researcher and several graduate students and will include hosting an annual workshop on radar aeroecology for students and researchers (including members of underrepresented groups in science). Project investigators will work with a media team to produce a series of five video presentations on studying the ecology of birds, bats and insects in the aerosphere.
This project has two objectives: (1) understand how global environmental change has impacted seasonal timing and population abundance of aerial insectivores over the past twenty-five years and (2) determine drivers of recent within and between seasonal variation in timing and abundance. Aerial insectivore populations have shown precipitous declines in the last half century ? often at much steeper rates than other aerial taxa. Understanding mechanisms driving these changes would have broad implications for hundreds of species of birds, bats, and insects, and also serve as an indicator of terrestrial and aquatic ecosystem health. However, the data sets needed to understand these mechanisms are currently lacking and urgently needed. While macroscale remote-sensing platforms for animals are rare, NEXRAD has emerged as a comprehensive source of information about flying animals, with large-scale and long-term (>two decades) coverage. The investigators will employ an interdisciplinary approach integrating radar remote sensing, data from NEON, and computer modelling to fill this vital gap and to test questions about population change, phenology, and trophic interactions in response to anthropogenic drivers of macroscale environmental change. The PIs will focus their project on the widespread roosting behaviors of three aerial insectivore species as bellwethers for environmental change and ecosystem health: Purple Martin, Tree Swallow, and Mexican free-tailed Bat. This collaborative and interdisciplinary approach will yield large-scale, quantitative, and predictive insights into changing environments. They will also generate new workflows, methodologies, and insights for the use of NEON data for the study of global change. Through this proposal the investigators will generate the tools and web interface to automatically identify, locate, and disseminate information regarding U.S.-wide roosting phenomena. The status of aerial insectivores is a representation of the seasonal pulse of ecosystem health ? the questions, infrastructural development, and outreach proposed will serve as for monitoring the status of aerial insectivores at the continental scale.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
PUBLICATIONS PRODUCED AS A RESULT OF THIS RESEARCH
Note:
When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external
site maintained by the publisher. Some full text articles may not yet be available without a
charge during the embargo (administrative interval).
Some links on this page may take you to non-federal websites. Their policies may differ from
this site.
PROJECT OUTCOMES REPORT
Disclaimer
This Project Outcomes Report for the General Public is displayed verbatim as submitted by the Principal Investigator (PI) for this award. Any opinions, findings, and conclusions or recommendations expressed in this Report are those of the PI and do not necessarily reflect the views of the National Science Foundation; NSF has not approved or endorsed its content.
The project developed techniques for the rapid and automated extraction of biological information from the vast and ever-growing archive of weather radar data collected by the United States NEXRAD network, with the goal of quantifying ecological change and its drivers in aerial habitats. The work focused on three key species that serve as early indicators of ecosystem health: Purple Martins, Tree Swallows, and Mexican Free-tailed Bats. These species depend on the airspace to forage for insect prey and belong to the group of aerial insectivores, which have experienced dramatic population declines. The mechanisms behind these changes are not well understood, but likely shared across hundreds of insect, bird, and bat species.
These species form communal roosts, where large numbers of individuals gather in the same resting location. The daily and nightly departures of thousands to millions of individuals from roosts can be seen and quantified using weather radar, which is the only reliable approach to measure roosts at large scales. The NEXRAD network includes 159 stations and has operated for nearly 30 years, offering a powerful and unique historical resource about these populations. However, the data resource is too vast for scientists to access without automated methods to locate and measure roost signatures in radar images.
Our work significantly improved the accuracy of automated computer vision models for detecting and tracking roosts by incorporating information about how roosts appear and grow over time and by building standardized datasets for training and evaluation and leveraging recent advances in deep learning. We also developed the infrastructure to run these models across multiple radar stations and years, along with user interfaces that enable rapid screening of the results. This framework reduced the cost of analyzing radar data by a factor of seven --- down to roughly one second per radar image.
We first deployed the system on radar data from 12 stations around the Great Lakes, covering 21 years and more than 600,000 radar images. This analysis revealed new insights into the persistence and size of swallow and martin roosts, as well as long-term patterns of yearly timing of aerial insectivore activity in the region. We then deployed the system on data from all 143 radar stations in the contiguous US, analyzing data from 2013 to 2023 to build a comprehensive dataset of swallow and martin roosts. We also processed data from four Texas radar stations to build a corresponding dataset of bat roosts. To improve efficiency, we introduced a new method that uses statistical sampling techniques to generate accurate population estimates with significantly less human effort needed to review the results of the automated system.
Together, these efforts help us better understand how bird and bat populations are shifting at the continental scale --- insights that can inform conservation strategies and deepen our understanding of changing ecosystems.
Last Modified: 04/04/2025
Modified by: Subhransu Maji
Please report errors in award information by writing to: awardsearch@nsf.gov.