**Detecting and Tracking Communal Bird Roosts in Weather Radar Data**
    [Zezhou Cheng^1](http://people.cs.umass.edu/~zezhoucheng), [Saadia Gabriel^2](https://homes.cs.washington.edu/~skgabrie/), [Pankaj Bhambhani^1](https://www.linkedin.com/in/pankajb64/), [Daniel Sheldon^1](https://people.cs.umass.edu/~sheldon/), [Subhransu Maji^1](http://people.cs.umass.edu/~smaji/), [Andrew Laughlin^3](https://www.unca.edu/person/andrew-laughlin/) and [David Winkler^4](http://www.eeb.cornell.edu/winkler/wordpress/)
    ^1 _University of Massachusetts - Amherst_
    ^2 _University of Washington_
    ^3 _University of North Carolina Asheville_
    ^4 _Cornell University_


The US weather radar archive holds detailed information about biological phenomena in the atmosphere over the last 20 years. Communally roosting birds congregate in large numbers at nighttime roosting locations, and their morning exodus from the roost is often visible as a distinctive pattern in radar images. This paper describes a machine learning system to detect and track roost signatures in weather radar data. A significant challenge is that labels were collected opportunistically from previous research studies and there are systematic differences in labeling style. We contribute a latent variable model and EM algorithm to learn a detection model together with models of labeling styles for individual annotators. By properly accounting for these variations we learn a significantly more accurate detector. The resulting system detects previously unknown roosting locations and provides comprehensive spatio-temporal data about roosts across the US. This data will provide biologists important information about the poorly understood phenomena of broad-scale habitat use and movements of communally roosting birds during the non-breeding season.


An Example of Roost Departure
==========================================================================================

<center><video width="640" height="360" controls>
  <source src="roost.mp4" type="video/mp4">
Your browser does not support the video tag.
</video></center>
<center>A flock of tree swallows was departing from their nighttime roosting location in the early morning.</center>
<center>The tornado-like departure pattern leaves a distinct ring-like signature on the radar data.</center>
<center style="font-weight: bold;">(Credit: This video was taken by Dave McNamara, WVUE-TV, New Orleans)</center>

Publication
==========================================================================================

**Detecting and Tracking Communal Bird Roosts in Weather Radar Data**<br />
Zezhou Cheng, Saadia Gabriel, Pankaj Bhambhani, Daniel Sheldon, Subhransu Maji, Andrew Laughlin and David Winkler<br />
AAAI, 2020 (oral presentation, AI for Social Impact Track) <br />
[pdf](./radar-roosts-aaai20.pdf), [supplementary](./radar-roosts-supplement-aaai20.pdf)


Acknowledgements 
==========================================================================================
This research was supported in part by NSF grants #1749833, #1749854, and #1661259, and the MassTech Collaborative for funding the UMass GPU cluster.


<!-- Markdeep: --><style class="fallback">body{visibility:hidden;white-space:pre;font-family:monospace}</style><script src="markdeep.min.js"></script><script src="https://casual-effects.com/markdeep/latest/markdeep.min.js"></script><script>window.alreadyProcessedMarkdeep||(document.body.style.visibility="visible")</script>
<style>
h1:before, h2:before { content: none; }
body{max-width: 900px}
</style>