Environmental Data Science Postdoctoral Fellow
|
|
Employer: Lawrence Berkeley National Laboratory
|
Job location: Berkeley United States
|
Apply before: 13 Dec 2019
|
Summary
The Berkeley Lab’s Energy Geosciences Division has an exciting opportunity for an Environmental Data Science Postdoctoral Fellow to investigate the impacts of streamflow disturbances, such as floods or droughts, on water quality in different river corridors of the United States.
Job location
Job application
Find more information about this
job
Job description
In this role, you will be responsible for integrating large temporal and spatial datasets, conducting analysis using statistical, wavelet, pattern recognition techniques, and implementing predictive machine-learning models. The work will also involve building a data-driven framework that will enable reproducible generation and analysis of integrated watershed datasets. The successful candidate will work within a five-year project funded by the Department of Energy’s Environmental Systems Science program to understand and predict watershed resilience to streamflow perturbations caused by extreme climate events, changes in land use, water management, or other disturbances.
The position offers an excellent environment for working with interdisciplinary teams consisting of environmental and computational researchers based at LBNL. LBNL is a renowned center of scientific expertise in many facets of fundamental and applied sciences related to the environment and water resources.
What You Will Do:
Develop and implement algorithms and data processing methods for repeatable integration and QA/QC of diverse temporal and spatial datasets of river discharge, water quality, watershed characteristics obtained from several federal sources including the DOE, USGS, NOAA, USDA, EPA.
Analyze data and extract patterns using statistical, wavelet, classification, or other data mining methods.
Research and develop machine learning approaches for data-driven modeling of water quality.
Contribute to building a Python-based framework for reproducible data analysis and prediction.
Author peer-reviewed conference or journal papers, and contribute to grant proposals.
What is Required:
Ph.D. in Environmental Sciences/Engineering, Computer Science, Data Science, Applied Mathematics, or other related technical disciplines.
Demonstrated experience with traditional and deep machine learning methods such as methods for time series analysis, classification, and prediction (particularly using neural networks).
Theoretical understanding and application of data analysis methods such as statistical techniques, signal processing, pattern recognition, or data-informed mathematical modeling.
Experience integrating large observational data sets from different sources, QA/QC of noisy datasets.
Programming experience with Python (preferred) or R.
What We Desire:
Familiarity with libraries, frameworks, or workflow tools that enable data analytics and machine learning (e.g., NumPy, Pandas, Scikit-learn, Keras, Tensorflow, Jupyter Notebooks).
Experience with Watershed to regional-scale analysis of hydrology/biogeochemistry in surface or groundwater.
Knowledge of problems related to water resources management.
Established record of publications or conference presentations.
Interest in open science, open data, and implementing maintainable and reusable software/data products for broader scientific use.
Requested Application Materials:
Cover Letter/Research statement: Include a cover letter introducing yourself, your application, and describing your interest in the position.
Curriculum Vitae/Resume: Either an academic CV or a resume is acceptable. Be sure to highlight technical skills, interests, and synergistic activities relevant to the position. Include links to software projects or public code repositories.
References: Provide contact information for
