Posted on December 19, 2019
A multinational team of data scientists led by Dr. Indrani Pal, NOAA-CREST research assistant professor at The City College of New York, is participating in the 2020 California Water Data Challenge. The program, held each fall, showcases innovative ways that data can be used to address issues affecting water resources in the United States’ most populous state. From more than 50 submissions, only a dozen teams get invited to present their projects.
Dr. Pal and her team intend to develop a data set of drought indicators that stakeholders could use as a prediction and projection tool as well a mobile app that could deliver notifications about locations where river water resources are at risk. The product, which will be open-source, will analyze interaction among climate, water supply and human factors in order to make predictions.
The team includes participants from Brazil and Canada in addition to four states: California, Illinois, Massachusetts and New York. “Almost all participants are from outside of New York City,” she said. “Being project leader and not personally knowing project participants is a challenge.” Because of this, she runs nightly phone-in sessions on Zoom to coordinate the team’s activities.
Understanding the role that human factors play in water supply resources is a key objective of the project. “Human activities shape water resources,” she said. “We want to understand how and where human factors play a role in water supply and incorporate those factors into the model in addition to natural factors.”
The project will incorporate a novel approach to quantify human impact. Dr. Pal plans to analyze land use change data as well as data on water quantities removed from surface water resources. “If we can factor out natural impacts we can measure human impacts,” she said.
Dr. Pal learned about the competition while attending the Bloomberg Data for Good 2019 conference. The Water Data Challenge presents an opportunity to use her expertise in water resources, statistical analysis and data science “to develop something for social good,” she said.
Much of the early-stage work focuses on identifying and cleaning up data sources drawn from satellite and observed data sets that include precipitation, snow water equivalent, land use and temperature. One group of participants is working on obtaining sources of water data and making it usable for developing a prediction model. Another group is doing the same with climate data. As the project gets closer to completion, a third group will address designing an effective risk communication tool.
In addition, a user group focuses on interaction among stakeholders, i.e. how one stakeholder group’s usage impacts other stakeholders. The two most interested stakeholder groups – fire fighters and utilities – have very different water supply and demand issues. “Utilities are more concerned about the future impact of climate change,” she said. “Fire fighters are interested in seasonal fluctuations in water levels for firefighting.”
Provide your email address and we'll notify you of our news and events