Development of Empirical Threshold Models for Ice Jam Forecasting

1
Development of Empirical Threshold Models for Ice
Jam Forecasting and Modifications to FLDWAV to
Model River Ice
Dr. Steven F. Daly ERDC/CRREL Hanover , NH 03755
2
Overview

• Objective The overall object is to improve the
  National Weather Service River Forecast Centers
  ability to forecast wintertime flooding due to
  river ice.
• Development of Empirical Threshold Models for Ice
  Jam Forecasting
• Modifications to FLDWAV to Model River Ice

3
Empirical Threshold Models for Ice Jam Forecasting

• The goal is to develop empirical threshold
  indicators of river ice formation, ice cover
  breakup, jam formation, and ice jam flooding. The
  indicators will be based on information that is
  readily available to the RFC's or could be easily
  derived from existing data sources. Proposed
  indicators include Accumulated Freezing Degree
  Days (AFDD), rate of change of discharge, maximum
  discharge, and other factors.

4
Modifications to FLDWAV to Model River Ice

• The effects of an ice cover will be incorporated
  into the NWS FLDWAV model which is part of the
  National Weather Service River Forecast System
  (NWSRFS) to model ice jams under unsteady flow
  conditions.

5
(No Transcript)
6

• Platte River 243
• Yukon River 211
• Kuskokwim River 194
• Connecticut River 90
• Milk River 60
• Kankakee River 40
• Allegheny River/Oil Creek 33
• Mohawk River 30
• Missouri River at Williston, ND 29
• Salmon River 25
• Weiser River 20
• TOTAL 1075

7
(No Transcript)
8
(No Transcript)
9
Analysis

• Assigned each ice event to the next d/s flow gage
  assigned each gage to a met station, average
  flow over POR, DA, Average maximum AFDD
• For each ice event
• Base flow (lowest 40 days discharge in the since
  the start of the water year)
• Q, the flow associated with the event
• DelQ, the flow increase
• Tp, the time to peak
• AFDD, the accumulated Freezing degree days

10
Analysis

• For all the ice events assigned to each gage, we
  then found the overall average of
• Base flow (lowest 40 days discharge in the since
  the start of the water year)
• Q, the flow associated with the event
• DelQ, the flow increase
• Tp, the time to peak
• AFDD, the accumulated Freezing degree days

11
(No Transcript)
12
(No Transcript)
13
(No Transcript)
14
(No Transcript)
15
(No Transcript)
16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
Summary

• Thresholds can be developed for each reach based
  on AFDD and DelQ or other
• There will be uncertainty because of large
  variation in data
• High, medium, low probability of ice event
• Salmon River true freeze up case.
• Remains to write up