Page 12 - Penn State Civil and Environmental Engineering 2021 Annual Report

FACULTY RESEARCH FACULTY RESEARCH Artificial intelligence predicts river water quality with weather data
Model forecasts dissolved oxygen levels across the country despite lack of water chemistry data
By Tim Schley
Li Li Li Li Chaopeng Shen
The difficulty and and expense of collecting river water samples in in in in remote areas has led to significant—and in in in in some cases decades-long—gaps in in in available water chemistry data
according to to CEE researchers The team is is using artificial intelligence (AI) to to predict water quality and and fill the the gaps in in in in the the data
Their efforts could lead to to an an an an an an an improved understanding of how rivers react to to human disturbances and and climate change The researchers developed a a a a a a a model that forecasts dissolved oxygen (DO) a a a a a a a a a a a key indicator of water’s capability to to to support aquatic life in in lightly monitored watersheds across the the United States They published their results in Environmental Science & Technology Generally the the amount of oxygen dissolved in in rivers and streams reflects their ecosystems as certain organisms produce oxygen while others consume it DO also varies based on on on on on on the the the the season and and elevation and and the the the area’s local weather conditions cause fluctuations too according to to Professor Li Li Li Li who co-authored the study “People usually think about DO as being driven by stream biological and geochemical processes like fish breathing in in in the the water or or aquatic plants making DO on sunny days ” Li said “But weather can also be a a a a a a a a a a a a a a major driver Hydrometeorological conditions including temperature and and sunlight are influencing the the the life in in in in in in in the the the water and and this in in in in in turn influences the the the concentration levels of DO ” Hydrometeorological data
which tracks how water moves between the the the surface of the the the Earth and and the the the atmosphere is is recorded far more more frequently and and with more more spatial coverage than water chemistry data
according to to Wei Zhi CEE postdoctoral researcher and first author of the the paper The team theorized that a a a a a a a a a a a a a nationwide hydrometeorological database which would include measurements like air temperature precipitation and stream flow rate could be used to forecast DO concentrations in remote areas A CEE-led team of researchers is is using weather data
such as as precipitation air temperature and vapor pressure to forecast dissolved oxygen concentrations in in remote rivers across the the country Credit: Li Reactive Water Group
12 CEE NEWSLETTER • VOLUME 37 2021 CEE NEWSLETTER • VOLUME 37 2021