Echospace

Echospace

Applied Physics Lab & eScience Institute

University of Washington

Ocean Acoustics + Data Science

We are a diverse group of researchers whose work centers around extracting knowledge from large volumes of ocean acoustic data, which contain rich information about animals ranging from zooplankton, fish, to marine mammals. Integrating physics-based models and data-driven methods, our current work focuses on mining water column sonar data and spans a broad spectrum from developing computational methods, building open source software and cloud applications, to joint analysis of acoustic observations and ocean environmental variables. A parallel but closely related focus of our research involves using echolocating bats and toothed whales as biological model systemss for adaptive and distributed ocean sensing.

Research areas:

  • Acoustical oceanography
  • Machine learning in ocean acoustics
  • Fisheries acoustics
  • Animal echolocation / bioacoustics
  • Data systems and workflows
  • Data science in oceanography

What we value

Code of Conduct & What we value

We review and discuss the code of conduct and what we value as a group regularly, as reminder and inspiration for ourselves.

Recent news!

All news»

[04/2024] Wu-Jung and Valentina presented two talks at the WGFAST 2024 meeting in France on pipelines and software tools for echosounder data processing on both ship and cloud.

[04/2024] Wu-Jung presented on Echosounder Data Processing Levels (with contributions from Emilio, Brandyn, and Valentina) at the Global Acoustics INteroperable (GAIN) workshop associated with the WGFAST 2024 meeting.

[03/2024] Wu-Jung and Valentina hosted 2 Capstone teams in the Master of Science in Data Science program for sonar data processing and automatic bat call detection.

[12/2023] We welcome Dr. Brandyn Lucca to join Echospace as a SEED postdoctoral fellow!

[12/2023] Soham gave a talk at 2023 PyData Global – check out his abstract and the video recording!

[10/2023] Valentina and Wu-Jung gave a talk and a poster presentation in the 2023 North Pacific Marine Science Organization (PICES) meeting in Seattle.

[09/2023] YeonJoon was selected as a UW Data Science Postdoctoral Fellow.

[09/2023] Wu-Jung joined the NOAA NCEI Water Column Sonar Data Archive stakeholder workshop and engaged in Echopype Q&As.

Meet the Team

Principal Investigators

Avatar

Wu-Jung Lee

Senior Oceanographer

Avatar

Valentina Staneva

Senior Data Scientist

Avatar

Emilio Mayorga

Senior Oceanographer

Researchers

Avatar

Don Setiawan

Research Software Engineer

Avatar

Caesar Tuguinay

Research Assistant

Avatar

Soham Kishor Butala

Research Software Engineer Intern

Avatar

Aditya Krishna

Undergraduate Research Assistant

Avatar

Liuyixin Shao

Undergraduate Research Assistant

Avatar

Varun Krishnakumar

Undergraduate Research Assistant

Alumni

Brandon Reyes

Now: HPC specialist at CU Boulder

Derya Gumustel

Now: Instructor Associate at General Assembly

Kavin Nguyen

Now: Operations Automation Engineer, SpaceX

Josie Sachen

Recent Posts

A Summer of Refactoring Echoshader!

Echospace hosted a contributor - Dingrui Lei, to refactor echoshader - a package for interactive visualization of echosounder data.

Hello from Dingrui Lei, GSoC contributor of Echoshader!

Echospace hosted a Google Summer of Code (GSoC) contributor to jump start echoshader, a new package for interactive visualization of echosounder data.

Projects

BOAT: Bridge to Ocean Acoustics and Technology

Democratizing the entrance to the ocean acoustics field through interdisciplinary education and outreach

Analyzing the effects of different environmental conditions on bat activity

Analyzing the effects of weather factors on bat activity through ANOVA, generalized linear model, and difference in difference.

UBNA passive acoustic monitoring project

Using passive acoustic techniques to study bat foraging behaviors in an urban natural area

Machine learning in fisheries acoustics

Accelerating information extraction from fisheries acoustic data through a cloud-based machine learning workflow.
Machine learning in fisheries acoustics

Scalable, cloud-native processing of water column sonar data

Accelerating ocean exploration through cloud-native processing of active ocean sonar data.

Echopop: hake biomass estimation

Modernizing the EchoPro workflow for integrating acoustic and biological survey samples for biomass estimation.

ADCP-equipped underwater glider as a distributed biological sensing tool

Enabling distributed, persistent observation of mid-trophic zooplankton and fish using autonomous underwater gliders equipped with acoustic Doppler current profilers (ADCPs).

Pattern discovery from long-term echosounder time series

Developing algorithms to discover prominent spatio-temporal patterns of animal movement and grouping behavior observed in sonar echoes using data from the Ocean Observatories Initiative (OOI).

Modeling directional hearing in toothed whales

Developing physics-based models for understanding the directional hearing in toothed whales.

Modeling target search by echolocating toothed whales

Modeling the echolocation-based target search behavior of toothed whales as an information-seeking process.

Recent & Upcoming Talks

Echostack: An open-source Python software toolbox that democratizes water column sonar dataand processing

Water column sonar data collected by echosounders are essential for marine ecosystem research, allowing the detection, classifi cation, …

A ship-to-cloud machine learning pipeline built on the open-source Python Echostack software tools

Successful application of machine learning (ML) methodology requires iterative development and testing of not only the models but also …

Scalable and configurable echosounder data workflows

Acoustic fisheries surveys and ocean observing systems collect terabytes of echosounder data that require custom processing pipelines …

Investigation of duty cycles in passive acoustic bat monitoring

Investigation of duty cycles in passive acoustic bat monitoring

Understanding echoes

Keynote Lecture at the 2022 Denver Acoustical Society of America meeting.

Software

Echopype

A Python package that enhances the interoperability and scalability in ocean sonar processing.

Echo Statistics

Matlab code to reproduce all figures in an in-depth tutorial on echo statistics.

Recent Publications

Beluga whale (Delphinapterus leucas) acoustic foraging behavior and applications for long term monitoring

Compact representation of temporal processes in echosounder time series via matrix decomposition

We developd a data-driven methodology based on matrix decomposition to build compact representation of long-term echosounder time series using intrinsic features in the data.
Compact representation of temporal processes in echosounder time series via matrix decomposition

Echo statistics associated with discrete scatterers: A tutorial on physics-based methods

From basic foundational concepts to advanced topics in modeling the statistics of echoes from discrete scatterers, inspired by sonar observation of marine organisms.
Echo statistics associated with discrete scatterers: A tutorial on physics-based methods

Macroscopic observations of diel fish movements around a shallow water artificial reef using a mid-frequency horizontal-looking sonar

Mid-frequency sonar provides a first-of-the-kind macroscopic observation of the nightly foraging runs of fish inhabiting a shallow-water artificial reef in northern Gulf of Mexico.
Macroscopic observations of diel fish movements around a shallow water artificial reef using a mid-frequency horizontal-looking sonar

Contact

  • echospace@uw.edu
  • 1013 NE 40th St, Seattle, WA 98105
  • Henderson Hall, University of Washington