What is it about?
During flooding events, decision makers and first responders depend on streamflow and flood inundation information for situational awareness, giving them the ability to effectively allocate resources, enable evacuations, and reduce property losses. Risk awareness is largely dependent on the collection, analysis, and communication of accurate hydrologic data from various sources, one of the most important being real-time and historical streamflow data provided by U.S. Geological Survey (USGS) streamgages. Losing near-real time streamgage data can make a difference between life or death, loss of property, or travel time with safety.
Photo by sgcdesignco on Unsplash
Why is it important?
These gages can be damaged by high water and debris, however, and stop transmitting data. Because uninterrupted transmittal of accurate streamflow data is critical for hydrological prediction systems and effective decision-making based on flood forecasts and unfolding real-time events, any missing data due to disruption has significant and wide-ranging implications (NHWC, 2006; Holmes et al., 2012).
Read the Original
This page is a summary of: Streamflow Hydrology Estimate Using Machine Learning (SHEM), JAWRA Journal of the American Water Resources Association, August 2017, Wiley, DOI: 10.1111/1752-1688.12555.
You can read the full text:
Flood Disasters Research - Predicting floods and improving public safety
Streamflow hydrology estimate using machine learning (SHEM) creates a predictive model that can act as a proxy streamflow data when a stream gauge fails. And due to the machine learning capabilities, it can even make estimates of stream levels where there is no actual stream gauge present. SHEM differs from most existing models as it does not rely on distances between stream gauges and their location attributes, but is based solely on machine learning to process from historical patterns of discharge and interpret large volumes of complex hydrology data. This “training” prepares SHEM to predict streamflow information for a given location and time as it is impacted by multivariate attributes.
The following have contributed to this page