Streamflow Estimation from Advanced Imaging (LSPIV) in Pennsylvania

Science Center Objects

Pennsylvania Water Science Center and the Next Generation Water Observing Systems (NGWOS) Program are implementing new technologies for streamflow estimation at sites in Pennsylvania where conventional measurement techniques are inadequate.

Large-Scale Particle Image Velocimetry (LSPIV) is an emerging technique to obtain measurements of surface velocity in streams and rivers. USGS scientists are applying LSPIV in Chartiers Creek in Carnegie, and Frankford Creek at Castor Avenue in Philadelphia, sites where the dynamic flow and channel conditions make conventional measurement difficult. Every 5 minutes, a 60-second downward-looking video of the moving water surface can be processed to estimate the velocity of water in the stream.

Juxtaposed still images from videos of low and high flows in a stream

Still images from videos of surface water flow overlaid with velocity vectors during low (top) and high (bottom) flow conditions in Frankford Creek, Philadelphia, Pennsylvania. (Matthew Gyves, USGS. Public domain.)


Photo of a camera system that will be tested to measure water level and surface velocity

Example camera system. (Photo by USGS staff, Public domain.)





Photograph of LSPIV Camera, Chartiers Creek at Carnegie, PA

LSPIV Camera, Chartiers Creek at Carnegie, Pennsylvania, installed on the gage house. (Lowell Abbadini, USGS. Public domain)

A camera system to measure water level and surface velocity is being tested at the East Branch Brandywine Creek below Downingtown. This technology is part of the USGS Next Generation Water Observing System in the Delaware River Basin.














Photograph of LSPIV electronics in gage house

Test installation of LSPIV electronics in gage house, Chartiers Creek, Pennsylvania (Lowell Abbadini, USGS. Public domain)

Lidar data processed to define the stream channel

The image below shows about 52 million points collected in July 2020 using laser techniques which define the Frankford Creek channel in three dimensions. With this high resolution survey data, collection of video of moving particles in the water from a bridge camera can be used to estimate the streamflow rate every minute. 

Lidar scan and rectifying frame for Frankford Cr

Lidar scan and rectifying frame for Frankford Creek. Data were used to construct a 3-dimensional stream channel that was used with LSPIV to estimate streamflow. (John Young, USGS. Public domain.)