Blue Template for Slide Presentations
Urban StreamStats: The development of data and an Arc Hydro tool for implementing stormwater networks in StreamStats for St. Louis County and the City of St. Louis Principle Cooperator: Metropolitan St. Louis Sewer District Pete Steeves, Rodney Southard and Tana Haluska (all USGS)
U.S. Department of the Interior U.S. Geological Survey
StreamStats Implementation • Now due to be implemented in June, 2018 • It will be the first StreamStats study-area that utilizes stormwater networks (in this case, inclusive of combined sanitary sewer networks) as part of the flow network to delineate contributing areas
• Traditional regional regression equations based on basin characteristics are not applicable to storm drain flow. We will be implementing new urban flows for St. Louis with warnings that point this out.
Location of Study within Missouri
St. Charles County
Kansas City Missouri River
St. Louis
Study Area with 3-meter DEM Grid
St. Louis County
City of St. Louis
Metrics of Data Used in Study • Study Area: St. Louis County (523 sq. miles) and City of St. Louis (66 sq. miles) • Elevation data source – 3-meter DEM data created from Lidar data and retrieved from https://nationalmap.gov/ • High resolution watershed boundaries – Generated using 12-digit HUC’s as a guide, as well as the DEM and the infrastructure all verified by imagery • Surface Flow – Generated from 3-meter DEM using a Python script written by Phil Thiem (USGS) and verified by imagery; there was limited NHD stream network in these urban areas
Resolving Data Issues • In the original DEM’s derivative grids, surface flow patterns were not spatially correct in many locations due to watercollecting curbsides not having been hydro-enforced, and small channels not well defined
• Flow direction and connectivity of the combined sanitary sewer network had to be addressed • Integration of the combined sanitary sewer network into the final drainage enforced DEM-derived open channel stream network had to be addressed • Man-made alterations of watershed boundaries and channels due to road structures, culverts, levees, and channelization of streams had to be addressed
DEM Generated Flowlines - Issues Arc Hydro Method
Road Embankments with Culverts not Hydro-Enforced Arc Hydro Method
Culvert
Arc Hydro Method compared to Python Script Arc Hydro Method
Python Script
DEM Resolution and Channel Size Arc Hydro Method
Python Script
Integration of CSS Network
Arc Hydro Stormwater Processing System requirements:
• Must have ArcGIS version 10.3.1 or better with Spatial Analyst Extension
• Must have the Arc Hydro version below: • for ArcGIS 10.3.1, Arc Hydro version 10.3.0.151 • for ArcGIS 10.4.0.32, Arc Hydro version 10.4.x • for ArcGIS 10.5.x, Arc Hydro 10.5.0.22 in cooperation with
Arc Hydro Stormwater Processing Workflow:
•
Started with basic geometric network
•
Iterative process of cleaning up and testing connectivity of the geometric network
•
Followed ESRI documentation to create a new AHStormwater geometric network: http://downloads.esri.com/archydro/archydro/doc Select "Arc Hydro Stormwater v1 10032017.pdf" in cooperation with
Arc Hydro Stormwater Processing Cleaning the geometric network consisted of:
• Snapping inlets to the network • Removing any disconnected pipes and inlets (i.e. catch • •
basins) outside the snapping tolerance Goal was to perform a geometric network trace from the most downstream point on the network and have all upstream features selected Over 70,000 inlets and 87,000 pipe segments
in cooperation with
Arc Hydro Stormwater Processing Characteristics of basic geometric network: • • •
•
No inlet, no flow
Surface water drainage flows through the stream layer and inlets
C B
Inlets connect surface water to pipes, so there must be an inlet anywhere the surface water enters the pipe. (A)
A Pipes can flow directly into streams without an inlet (B)
Must have an inlet at upstream end of pipe network. (C) If no inlet, no flow through pipe
Stream Pipe Inlet Sink in cooperation with
Arc Hydro Stormwater Processing Overview of processing steps:
• Very similar to the standard Arc Hydro processing steps
• There are 17 steps • There are 2 linear network features, streams and pipes • Only the stream layer is used in the DEM hydro enforcement in cooperation with
Direct Surface Contribution Only option
ArcHydro Stormwater Processing
ArcHydro Stormwater Processing
intersection
Arc Hydro Stormwater Processing Summary:
•
Connectivity of the pipe network is based on a "common sense" evaluation of the current data provided to the USGS
•
A CSS pipe network is literally changing daily
•
Spatial location of inlets will dictate where the surface flow enters the CSS network based on the DEM
•
The accuracy of the pipe drainage system (catchment delineation) is dependent on the resolution of the DEM and spatial accuracy of the CSS network in cooperation with
Catchments in CSS Network
Watershed Boundary - Interstate
Watershed Boundary - Levee
Potential for future program!
Thanks for attending! Questions?
U.S. Department of the Interior U.S. Geological Survey
StreamStats Implementation • Now due to be implemented in June, 2018 • It will be the first StreamStats study-area that utilizes stormwater networks (in this case, inclusive of combined sanitary sewer networks) as part of the flow network to delineate contributing areas
• Traditional regional regression equations based on basin characteristics are not applicable to storm drain flow. We will be implementing new urban flows for St. Louis with warnings that point this out.
Location of Study within Missouri
St. Charles County
Kansas City Missouri River
St. Louis
Study Area with 3-meter DEM Grid
St. Louis County
City of St. Louis
Metrics of Data Used in Study • Study Area: St. Louis County (523 sq. miles) and City of St. Louis (66 sq. miles) • Elevation data source – 3-meter DEM data created from Lidar data and retrieved from https://nationalmap.gov/ • High resolution watershed boundaries – Generated using 12-digit HUC’s as a guide, as well as the DEM and the infrastructure all verified by imagery • Surface Flow – Generated from 3-meter DEM using a Python script written by Phil Thiem (USGS) and verified by imagery; there was limited NHD stream network in these urban areas
Resolving Data Issues • In the original DEM’s derivative grids, surface flow patterns were not spatially correct in many locations due to watercollecting curbsides not having been hydro-enforced, and small channels not well defined
• Flow direction and connectivity of the combined sanitary sewer network had to be addressed • Integration of the combined sanitary sewer network into the final drainage enforced DEM-derived open channel stream network had to be addressed • Man-made alterations of watershed boundaries and channels due to road structures, culverts, levees, and channelization of streams had to be addressed
DEM Generated Flowlines - Issues Arc Hydro Method
Road Embankments with Culverts not Hydro-Enforced Arc Hydro Method
Culvert
Arc Hydro Method compared to Python Script Arc Hydro Method
Python Script
DEM Resolution and Channel Size Arc Hydro Method
Python Script
Integration of CSS Network
Arc Hydro Stormwater Processing System requirements:
• Must have ArcGIS version 10.3.1 or better with Spatial Analyst Extension
• Must have the Arc Hydro version below: • for ArcGIS 10.3.1, Arc Hydro version 10.3.0.151 • for ArcGIS 10.4.0.32, Arc Hydro version 10.4.x • for ArcGIS 10.5.x, Arc Hydro 10.5.0.22 in cooperation with
Arc Hydro Stormwater Processing Workflow:
•
Started with basic geometric network
•
Iterative process of cleaning up and testing connectivity of the geometric network
•
Followed ESRI documentation to create a new AHStormwater geometric network: http://downloads.esri.com/archydro/archydro/doc Select "Arc Hydro Stormwater v1 10032017.pdf" in cooperation with
Arc Hydro Stormwater Processing Cleaning the geometric network consisted of:
• Snapping inlets to the network • Removing any disconnected pipes and inlets (i.e. catch • •
basins) outside the snapping tolerance Goal was to perform a geometric network trace from the most downstream point on the network and have all upstream features selected Over 70,000 inlets and 87,000 pipe segments
in cooperation with
Arc Hydro Stormwater Processing Characteristics of basic geometric network: • • •
•
No inlet, no flow
Surface water drainage flows through the stream layer and inlets
C B
Inlets connect surface water to pipes, so there must be an inlet anywhere the surface water enters the pipe. (A)
A Pipes can flow directly into streams without an inlet (B)
Must have an inlet at upstream end of pipe network. (C) If no inlet, no flow through pipe
Stream Pipe Inlet Sink in cooperation with
Arc Hydro Stormwater Processing Overview of processing steps:
• Very similar to the standard Arc Hydro processing steps
• There are 17 steps • There are 2 linear network features, streams and pipes • Only the stream layer is used in the DEM hydro enforcement in cooperation with
Direct Surface Contribution Only option
ArcHydro Stormwater Processing
ArcHydro Stormwater Processing
intersection
Arc Hydro Stormwater Processing Summary:
•
Connectivity of the pipe network is based on a "common sense" evaluation of the current data provided to the USGS
•
A CSS pipe network is literally changing daily
•
Spatial location of inlets will dictate where the surface flow enters the CSS network based on the DEM
•
The accuracy of the pipe drainage system (catchment delineation) is dependent on the resolution of the DEM and spatial accuracy of the CSS network in cooperation with
Catchments in CSS Network
Watershed Boundary - Interstate
Watershed Boundary - Levee
Potential for future program!
Thanks for attending! Questions?
