Skip to main content
U.S. flag

An official website of the United States government

Galen Gorski

Galen Gorski is a Machine Learning Specialist for the USGS Water Mission Area, based in Reston, VA. 

I am interested in understanding fundamental connections between hydrologic and biogeochemical cycling in natural and managed landscapes. This requires a suite of methods and approaches including data-driven techniques, field campaigns, and laboratory analysis. I am interested in projects that use machine learning and other data-driven techniques to cut across these boundaries and develop understanding of the connections between water quantity and quality and their implications for water resource management and sustainability.    

In my previous work, I investigated controls on nitrogen cycling in managed aquifer recharge settings in agricultural areas of California. I have also worked to develop local and regional recharge suitability maps in California and the Middle East North Africa Region to help guide future project siting. I’m excited about combining my experiences with the incredible expertise at USGS to tackle important water resource challenges. 

Professional Experience

  • ​2021- Present: Machine Learning Specialist, U.S. Geological Survey, Water Mission Area

  • 2021: Postdoctoral Scholar, University of California, Berkeley, Mentor: Laurel Larsen 

  • 2020: Postdoctoral Researcher, University of California, Santa Cruz; Mentors: Andrew Fisher and Margaret Zimmer 

  • 2014-2015: Biological Technician, University of Minnesota/USDA 

  • ​2013-2014: Laboratory Technician

Education and Certifications

  • ​Ph.D. Earth Sciences, 2020, University of California, Santa Cruz; Advisors: Andrew Fisher and Adina Paytan
    Linking hydrologic and biogeochemical cycling across scales: Implications for nutrient and w

  • B.A. Chemistry, 2013, Carleton College
    Probing the primary events of photosynthesis using ultra-fast lasers

Science and Products

Data and model code used to evaluate a process-guided deep learning approach for in-stream dissolved oxygen prediction

This model archive contains data and code used to assess the use of process-informed multi-task deep learning models for predicting in-stream dissolved oxygen concentrations. Three holdout experiments were run to assess model performance, including a temporal holdout experiment, a spatial holdout experiment with similar sites held out, and a spatial holdout experiment with dissimilar sites held ou
By
Water Resources Mission Area

A deep learning model and associated data to support understanding and simulation of salinity dynamics in Delaware Bay

Salinity dynamics in the Delaware Bay estuary are a critical water quality concern as elevated salinity can damage infrastructure and threaten drinking water supplies. Current state-of-the-art modeling approaches use hydrodynamic models, which can produce accurate results but are limited by significant computational costs. We developed a machine learning (ML) model to predict the 250 mg/L Cl- isoc
By
Water Resources Mission Area

Evaluating a process-guided deep learning approach for predicting dissolved oxygen in streams

Dissolved oxygen (DO) is a critical water quality constituent that governs habitat suitability for aquatic biota, biogeochemical reactions and solubility of metals in streams. Recently introduced high-frequency sensors have increased our ability to measure DO, but we still lack the capacity to understand and predict DO concentrations at high spatial resolutions or in unmonitored locations. Machine
Authors
Jeffrey M Sadler, Lauren Elizabeth Koenig, Galen Agnew Gorski, Alice M. Carter, Robert O. Hall
By
Water Resources Mission Area

Deep learning of estuary salinity dynamics is physically accurate at a fraction of hydrodynamic model computational cost

Salinity dynamics in the Delaware Bay estuary are a critical water quality concern as elevated salinity can damage infrastructure and threaten drinking water supplies. Current state-of-the-art modeling approaches use hydrodynamic models, which can produce accurate results but are limited by significant computational costs. We developed a machine learning (ML) model to predict the 250 mg L−1 Cl− is
Authors
Galen Agnew Gorski, Salme Ellen Cook, Amelia Marie Snyder, Alison P. Appling, Theodore Paul Thompson, Jared David Smith, John C. Warner, Simon Nemer Topp
By
Water Resources Mission Area

Non-USGS Publications**

Gorski G., Zimmer M.A. Hydrologic regimes drive nutrient export behavior in human
impacted watersheds. (2021) Hydrology and Earth System Science. 25, 1333-1345
doi: https://doi.org/10.5194/hess-25-1333-2021
Pensky J., Richardson C., Serrano A., Gorski G., Price A.N., Zimmer M.A., (2021)
Disrupt and demystify the unwritten rules of graduate school. Nature Geosciences 14,
538-539. doi: http://dx.doi.org/10.1038/s41561-021-00799-w
Gorski G., Dailey H., Fisher A.T., Schrad N., Saltikov C. (2020) Denitrification during
infiltration for managed aquifer recharge: Infiltration rate controls and microbial response.
Science of the Total Environment, 727, 138642.
doi: https://doi.org/10.1016/j.scitotenv.2020.138642
Gorski G., Fisher A.T., Beganskas S., Weir W., Redford K., Schmidt C., Saltikov C.
(2019) Field and laboratory studies linking hydrologic, geochemical, and microbiological
processes and enhanced denitrification during infiltration for managed recharge. Environ-
mental Science and Technology, 53, 9491-9501.
doi: http://dx.doi.org/10.1021/acs.est.9b01191
Gorski G., Strong C., Good S.P., Bares R., Ehleringer J.R., Bowen G.J. (2015) Vapor
hydrogen and oxygen isotopes reflect water of combustion in the urban atmosphere. Proceedings of the National Academy of Sciences, 112, 3247-3252.
doi: https://doi.org/10.1073/pnas.1424728112

**Disclaimer: The views expressed in Non-USGS publications are those of the author and do not represent the views of the USGS, Department of the Interior, or the U.S. Government.

Science and Products

Data and model code used to evaluate a process-guided deep learning approach for in-stream dissolved oxygen prediction

This model archive contains data and code used to assess the use of process-informed multi-task deep learning models for predicting in-stream dissolved oxygen concentrations. Three holdout experiments were run to assess model performance, including a temporal holdout experiment, a spatial holdout experiment with similar sites held out, and a spatial holdout experiment with dissimilar sites held ou
By
Water Resources Mission Area

A deep learning model and associated data to support understanding and simulation of salinity dynamics in Delaware Bay

Salinity dynamics in the Delaware Bay estuary are a critical water quality concern as elevated salinity can damage infrastructure and threaten drinking water supplies. Current state-of-the-art modeling approaches use hydrodynamic models, which can produce accurate results but are limited by significant computational costs. We developed a machine learning (ML) model to predict the 250 mg/L Cl- isoc
By
Water Resources Mission Area

Evaluating a process-guided deep learning approach for predicting dissolved oxygen in streams

Dissolved oxygen (DO) is a critical water quality constituent that governs habitat suitability for aquatic biota, biogeochemical reactions and solubility of metals in streams. Recently introduced high-frequency sensors have increased our ability to measure DO, but we still lack the capacity to understand and predict DO concentrations at high spatial resolutions or in unmonitored locations. Machine
Authors
Jeffrey M Sadler, Lauren Elizabeth Koenig, Galen Agnew Gorski, Alice M. Carter, Robert O. Hall
By
Water Resources Mission Area

Deep learning of estuary salinity dynamics is physically accurate at a fraction of hydrodynamic model computational cost

Salinity dynamics in the Delaware Bay estuary are a critical water quality concern as elevated salinity can damage infrastructure and threaten drinking water supplies. Current state-of-the-art modeling approaches use hydrodynamic models, which can produce accurate results but are limited by significant computational costs. We developed a machine learning (ML) model to predict the 250 mg L−1 Cl− is
Authors
Galen Agnew Gorski, Salme Ellen Cook, Amelia Marie Snyder, Alison P. Appling, Theodore Paul Thompson, Jared David Smith, John C. Warner, Simon Nemer Topp
By
Water Resources Mission Area

Non-USGS Publications**

Gorski G., Zimmer M.A. Hydrologic regimes drive nutrient export behavior in human
impacted watersheds. (2021) Hydrology and Earth System Science. 25, 1333-1345
doi: https://doi.org/10.5194/hess-25-1333-2021
Pensky J., Richardson C., Serrano A., Gorski G., Price A.N., Zimmer M.A., (2021)
Disrupt and demystify the unwritten rules of graduate school. Nature Geosciences 14,
538-539. doi: http://dx.doi.org/10.1038/s41561-021-00799-w
Gorski G., Dailey H., Fisher A.T., Schrad N., Saltikov C. (2020) Denitrification during
infiltration for managed aquifer recharge: Infiltration rate controls and microbial response.
Science of the Total Environment, 727, 138642.
doi: https://doi.org/10.1016/j.scitotenv.2020.138642
Gorski G., Fisher A.T., Beganskas S., Weir W., Redford K., Schmidt C., Saltikov C.
(2019) Field and laboratory studies linking hydrologic, geochemical, and microbiological
processes and enhanced denitrification during infiltration for managed recharge. Environ-
mental Science and Technology, 53, 9491-9501.
doi: http://dx.doi.org/10.1021/acs.est.9b01191
Gorski G., Strong C., Good S.P., Bares R., Ehleringer J.R., Bowen G.J. (2015) Vapor
hydrogen and oxygen isotopes reflect water of combustion in the urban atmosphere. Proceedings of the National Academy of Sciences, 112, 3247-3252.
doi: https://doi.org/10.1073/pnas.1424728112

**Disclaimer: The views expressed in Non-USGS publications are those of the author and do not represent the views of the USGS, Department of the Interior, or the U.S. Government.

Was this page helpful?