NHDPlus V2 Mean Monthly and Low Flow Estimates

undefined
1
Mean Monthly and Low Flow
Estimates in NHDPlus Version 2
Tim Bondelid- NHDPlus Team
Consulting Engineer
2014 AWRA Spring Conference on GIS and
Water Resources VIII
May 12 to May 14, 2014
Snowbird, Utah
This Presentation
This Presentation
NHDPlus V1 and NHDPlus V2 include
mean annual (MA) flow and velocity
estimates
All 2.7 M networked stream segments
We have now released mean monthly
(MM) flow and velocity estimates
We have a first-round test of 7Q10 flows
I will talk about how we did these
MM and 7Q10 flow estimates
2
3
NHD
Plus
 V02 Enhanced Runoff
Method (EROM)
Contributors:
The NHD
Plus
 Team
EROM:
Dave Wolock, USGS
Kernell Ries, USGS
Greg Schwarz, USGS
4
Importance of
Stream Flow Estimates
Support the water element goals
expressed in the EPA Strategic Plan
Support the water mission of the USGS
Water-quality modeling (
SPARROW
)
Regional and national stream flow assessments
goal of National Stream Flow Information Program
National Water Census
Watershed planning and management
Response to emergencies (ICWATER)
NHD
Plus
 V02
EROM Flow Estimation Steps
1.
Runoff based on water balance model (RO)
2.
“Excess ET” component that takes into
account excess evapo-transpiration in the
stream channel area (EET)
3.
A regression of Step 2 flows on Gage flows
using Reference gages (RGR)
4.
A capability for users to add, remove and
transfer flows (PlusFlowAR)
5.
Adjustments to observed gage flows
6.
A flow QA module to evaluate accuracy of
the flow estimates
6
U.S.: Wolock and McCabe; CN and MX: Canadian Forest Service
1971 - 2000
7
                                     
QA Statistics
Gage       Runoff            EET             Ref Reg.       PlusFlow        Gage Seq.
Mean    Mean   SEE    Mean   SEE    Mean   SEE   Mean    SEE         SEE
2.36     2.17   58.8
   2.17   58.8   
2.37    32.6
   2.37   32.6        23.0
Step 3: Reference Gage
            Regression (SR)
EROM MM Flows
EROM MM Flows
EROM built for MA and MM flows
QA issues with MM flows:
RO results in winter months in some
areas produced very poor results
Sometimes EET helped, sometimes it
hurt
RGR helped most of the time, but
sometimes made the QA worse
These QA issues led to the decision
to not release MM flows
8
Poor RO Results
Poor RO Results
Occurs in winter months in the Great
Plains
The runoff model is not able to
model the VERY low baseflows in the
persistently cold months.
This issue occurs in 17 VPU/months.
Approach: Replace the runoff grids
with a regression equation.
Need to model unit runoff (cfs/km
2
)
EROM uses a flow accumulation
approach
The Water Balance Model
(Slide courtesy of Dave Wolock, USGS)
Unit Runoff Regression
Unit Runoff Regression
Tested several statistical model forms
Used Reference Gages
What works well:
Log-log regression using Precip, Temp, and
Baseflow Index (BFI):
URO (cfs/Km
2
) = 10
a
 * Pcm
c
 * TDegC
d
 * BFI
e
 *
BCF
a, c, d, e = regression coefficients
BCF = Bias Correction Factor
Required CA3T allocations for Temp
and Precip for each of the 17
VPUs/months
undefined
Runoff QA With RO Reg
Runoff QA With RO Reg
Setting “Switches” for EET and
Setting “Switches” for EET and
RGR
RGR
Implementing Mean Monthly Flows
Implementing Mean Monthly Flows
Made a new production version of
EROM (Version 10)
Limits the RGR to the regression
bounds
Minimum of 10 years of gage record
A new input file named
“EROMOPTS.DBF” is used:
One record for each VPU/Timeperiod
21 VPUs * 13 flows = 273 records
“Switches” for running/not running EET
and RGR and RO Regression coefficients
What Is Available
What Is Available
EROMExtension:
MA and MM flow and velocity estimates
MA and MM QA files
A QA spreadsheet; please use it
VPUAttributeExtension:
Catchment-level MA and MM Runoff,
Temperature and Precipitation
Cumulative MA Temperature and
Precipitation
15
7Q10 Flow Estimates
Estimate 7Q10 flows on ungaged
flowlines (stream segments) in
NHDPlusV2
Use networked flowlines
Removes many canals and ditches
Note: less than 1% of the flowlines have
gages
Leverage the advanced capabilities in
NHDPlusV2 and the Enhanced Runoff
Method (EROM) to test a cost-
effective method for estimating
interpretive flow statistics
16
Approach
Calculate 7Q10 at gages
Use the DFLOW program in BASINS
USE NHDPlusV2 watershed characteristics:
drainage area (DA), mean precipitation (MP)
and temperature (MT), mean watershed
elevation (MWE), watershed slope (WS), Base
Flow Index (BFI)
EROM MA and MM Flows
Develop regionalized regression equations
by “Vector Processing Units” (VPUs)
7Q10 = Fn(EROM Flow, DA, MP, MT, MWE, WS,
BFI)
Use a log-log regression
17
Approach (Cont.)
Develop regionalized regression equations
by “Vector Processing Units” (VPUs)
7Q10 = Fn(EROM Flow, DA, MP, MT, MWE, WS,
BFI)
Additional Steps
Intermittent flowlines: 7Q10 = 0
“Large River” Adjustments
18
Challenges
Low flows are affected by factors
that can be difficult to quantify on a
broad scale:
Groundwater/baseflow
Geology
Physiography
Reservoirs, other flow alterations
The scope for this project is MUCH
larger than other such studies
2.7 Million Stream Segments across
CONUS
The VPUs are MUCH larger than the
geographic areas normally used.
19
Regression Accuracy Measures
Regression Measures: R
2
, Model
Error Variance [log
10
] (MEV),
Standard Error of the Regression
(SER) in %.
Q_E: EROM Flow
Q_E * BFI works best overall
VPU 02 Statistics:
20
 
21
Large River Adjustments
Large River Adjustments
22
Perennial Streams
Perennial Streams
23
Major Issues and Caveats
Major Issues and Caveats
DFLOW:
7Q10 flows are “unsupervised”
Need to modify DFLOW for production
Poor results in the Great Plains:
Subdivide large VPUs
Winter low flows?
Poor Results in Southwest
Very few perennial streams
Some “minor” adjustments
Streamline the process
24
A Low Flow Production System
25
DFLOW
DBMS
Statistics
Regression
Coefficient  Db
Low Flow
Computations
NHDPlus Low
Flow Extension
Evaluation
Regression
Areas, Other
Options
NHDPlus,
EROM, Other
Data
Period of
Record,Flow
Statistic(s)
26
Down the Road
PlusFlowAR:
Incorporate more detailed information
from NHD Stewards, WBD, others!
Seasonal values?
EET Step: Model Coefficients, issues
in some areas
Continue 7Q10, other low flows
Use a 1981-2010 time frame
PRISM has released “normals” for this
period
An issue: international drainage
Read All About It!
Go to NHDPlus.com
Go to NHDPlus.com
This redirects you to the 
This redirects you to the 
ftp.horizon-
ftp.horizon-
systems.com
systems.com
 page
 page
Click on a hot-link in the text to get the
Click on a hot-link in the text to get the
User Guide
User Guide
In the User Guide:
In the User Guide:
Pages 61-69: EROM and Attribute
Pages 61-69: EROM and Attribute
Extension Table Layouts
Extension Table Layouts
Pages 125-147: How EROM works
Pages 125-147: How EROM works
Note caveats for using velocities!!!!
Note caveats for using velocities!!!!
27
28
Questions?
Tim Bondelid - Independent Consultant to EPA
timothy@trbondelid.com
NHDPlus V02:
http://www.horizon-systems.com/NHDPlus
Slide Note
Embed
Share

NHDPlus Version 2 includes mean annual flow and velocity estimates for all stream segments. The release now features mean monthly flow and velocity estimates as well as initial testing of 7Q10 flows. Learn about the methodology behind these estimates and the Enhanced Runoff Method (EROM) contributors. Discover the importance of stream flow estimates in supporting various water-related goals and initiatives such as EPA Strategic Plan, USGS Water-quality modeling, and National Stream Flow Information Program. Gain insights into the steps involved in NHDPlus V2 EROM flow estimation and Reference Gage Regression. Explore the collaborative efforts of key contributors like Dave Wolock, Kernell Ries, and Greg Schwarz in developing the EROM. Follow the presentation to understand the significance of accurate stream flow estimates in water management, emergency response, and watershed planning.

  • NHDPlus
  • Flow Estimates
  • EROM
  • Stream Flow
  • Water Resources

Uploaded on Feb 27, 2025 | 1 Views


Download Presentation

Please find below an Image/Link to download the presentation.

The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.

You are allowed to download the files provided on this website for personal or commercial use, subject to the condition that they are used lawfully. All files are the property of their respective owners.

The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.

E N D

Presentation Transcript


  1. Mean Monthly and Low Flow Estimates in NHDPlus Version 2 Tim Bondelid- NHDPlus Team Consulting Engineer 2014 AWRA Spring Conference on GIS and Water Resources VIII May 12 to May 14, 2014 Snowbird, Utah 1

  2. This Presentation NHDPlus V1 and NHDPlus V2 include mean annual (MA) flow and velocity estimates All 2.7 M networked stream segments We have now released mean monthly (MM) flow and velocity estimates We have a first-round test of 7Q10 flows I will talk about how we did these MM and 7Q10 flow estimates 2

  3. NHDPlus V02 Enhanced Runoff Method (EROM) Contributors: The NHDPlus Team EROM: Dave Wolock, USGS Kernell Ries, USGS Greg Schwarz, USGS 3

  4. Importance of Stream Flow Estimates Support the water element goals expressed in the EPA Strategic Plan Support the water mission of the USGS Water-quality modeling (SPARROW) Regional and national stream flow assessments goal of National Stream Flow Information Program National Water Census Watershed planning and management Response to emergencies (ICWATER) 4

  5. NHDPlus V02 EROM Flow Estimation Steps Runoff based on water balance model (RO) Excess ET component that takes into account excess evapo-transpiration in the stream channel area (EET) A regression of Step 2 flows on Gage flows using Reference gages (RGR) A capability for users to add, remove and transfer flows (PlusFlowAR) Adjustments to observed gage flows A flow QA module to evaluate accuracy of the flow estimates 1. 2. 3. 4. 5. 6.

  6. U.S.: Wolock and McCabe; CN and MX: Canadian Forest Service 1971 - 2000 6

  7. Step 3: Reference Gage Regression (SR) 100,000 10,000 EROM Flow (cfs) 1,000 100 10 1 1 10 100 1,000 10,000 100,000 Gage Flow (cfs) QA Statistics Gage Runoff EET Ref Reg. PlusFlow Gage Seq. Mean Mean SEE Mean SEE Mean SEE Mean SEE SEE 2.36 2.17 58.8 2.17 58.8 2.37 32.6 2.37 32.6 23.0 7

  8. EROM MM Flows EROM built for MA and MM flows QA issues with MM flows: RO results in winter months in some areas produced very poor results Sometimes EET helped, sometimes it hurt RGR helped most of the time, but sometimes made the QA worse These QA issues led to the decision to not release MM flows 8

  9. Poor RO Results Occurs in winter months in the Great Plains The runoff model is not able to model the VERY low baseflows in the persistently cold months. This issue occurs in 17 VPU/months. Approach: Replace the runoff grids with a regression equation. Need to model unit runoff (cfs/km2) EROM uses a flow accumulation approach

  10. The Water Balance Model Temperature Potential Evapotranspiration Snow Snow Melt Rain Actual Direct Runoff Snow Storage Evapotranspiration Surplus Runoff Soil-Moisture Storage Capacity Soil-Moisture Storage (Slide courtesy of Dave Wolock, USGS)

  11. Unit Runoff Regression Tested several statistical model forms Used Reference Gages What works well: Log-log regression using Precip, Temp, and Baseflow Index (BFI): URO (cfs/Km2) = 10a * Pcmc * TDegCd * BFIe * BCF a, c, d, e = regression coefficients BCF = Bias Correction Factor Required CA3T allocations for Temp and Precip for each of the 17 VPUs/months

  12. Runoff QA With RO Reg

  13. Setting Switches for EET and RGR

  14. Implementing Mean Monthly Flows Made a new production version of EROM (Version 10) Limits the RGR to the regression bounds Minimum of 10 years of gage record A new input file named EROMOPTS.DBF is used: One record for each VPU/Timeperiod 21 VPUs * 13 flows = 273 records Switches for running/not running EET and RGR and RO Regression coefficients

  15. What Is Available EROMExtension: MA and MM flow and velocity estimates MA and MM QA files A QA spreadsheet; please use it VPUAttributeExtension: Catchment-level MA and MM Runoff, Temperature and Precipitation Cumulative MA Temperature and Precipitation 15

  16. 7Q10 Flow Estimates Estimate 7Q10 flows on ungaged flowlines (stream segments) in NHDPlusV2 Use networked flowlines Removes many canals and ditches Note: less than 1% of the flowlines have gages Leverage the advanced capabilities in NHDPlusV2 and the Enhanced Runoff Method (EROM) to test a cost- effective method for estimating interpretive flow statistics 16

  17. Approach Calculate 7Q10 at gages Use the DFLOW program in BASINS USE NHDPlusV2 watershed characteristics: drainage area (DA), mean precipitation (MP) and temperature (MT), mean watershed elevation (MWE), watershed slope (WS), Base Flow Index (BFI) EROM MA and MM Flows Develop regionalized regression equations by Vector Processing Units (VPUs) 7Q10 = Fn(EROM Flow, DA, MP, MT, MWE, WS, BFI) Use a log-log regression 17

  18. Approach (Cont.) Develop regionalized regression equations by Vector Processing Units (VPUs) 7Q10 = Fn(EROM Flow, DA, MP, MT, MWE, WS, BFI) Additional Steps Intermittent flowlines: 7Q10 = 0 Large River Adjustments 18

  19. Challenges Low flows are affected by factors that can be difficult to quantify on a broad scale: Groundwater/baseflow Geology Physiography Reservoirs, other flow alterations The scope for this project is MUCH larger than other such studies 2.7 Million Stream Segments across CONUS The VPUs are MUCH larger than the 19

  20. Regression Accuracy Measures Regression Measures: R2, Model Error Variance [log10] (MEV), Standard Error of the Regression (SER) in %. Q_E: EROM Flow Q_E * BFI works best overall VPU 02 Statistics: Vars n R2 MEV SER (%) 115% BCF Q_E*BF I, DA 563 0.92 0.16 1.4 20

  21. 21

  22. Large River Adjustments 22

  23. Perennial Streams 23

  24. Major Issues and Caveats DFLOW: 7Q10 flows are unsupervised Need to modify DFLOW for production Poor results in the Great Plains: Subdivide large VPUs Winter low flows? Poor Results in Southwest Very few perennial streams Some minor adjustments Streamline the process 24

  25. A Low Flow Production System Period of Record,Flow Statistic(s) DFLOW NHDPlus, EROM, Other Data DBMS Regression Areas, Other Options Statistics Evaluation Regression Coefficient Db Low Flow Computations NHDPlus Low Flow Extension 25

  26. Down the Road PlusFlowAR: Incorporate more detailed information from NHD Stewards, WBD, others! Seasonal values? EET Step: Model Coefficients, issues in some areas Continue 7Q10, other low flows Use a 1981-2010 time frame PRISM has released normals for this period An issue: international drainage 26

  27. Read All About It! Go to NHDPlus.com This redirects you to the ftp.horizon- systems.com page Click on a hot-link in the text to get the User Guide In the User Guide: Pages 61-69: EROM and Attribute Extension Table Layouts Pages 125-147: How EROM works Note caveats for using velocities!!!! 27

  28. Questions? Tim Bondelid - Independent Consultant to EPA timothy@trbondelid.com NHDPlus V02: http://www.horizon-systems.com/NHDPlus 28

More Related Content

giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#