WSP09246

• File Number: 5000.475
• Description: Flood Protection Section - Wetzel Creek Flood - 1986
• State: CO
• Basin: South Platte
• Water Division: 1
• Date: 8/2/1986
• Author: John Henz-Ron Kelly
• Title: Wetzel Creek Colorado Flash Flood - August 2 1986 - A Reconstitution of Thunderstorm Rainfall - 1545-1845 MDT
• Water Supply Pro - Doc Type: Report/Study

000822

Host of the storms on this day moved as the Platteville
storm did with, or to the right, of the upper level winds
from northwest to southeast. In contrast the Wetzel Creek
storm was influenced by a low level circulation along the
cool air boundary and remained focused over the Wetzel Creek
basin for a period of over 60 minutes. Additionally the
atmospheric structure supported the redevelopment of
additional storms over the basin which further aggravated
the rainfall and resulting runoff.

Henz Kelly & Associates has run a flash flood prediction
program(F2P2) for the Urban Drainage & Flood Control
District of Denver since 1979. It is typical that 2-5 days
occur every summer from early Hay until early September
capable of producing flash flood producing thunderstorms of
the Wetzel Creek or Platteville type. So while the storms
experienced on August 2nd were quite intense, they are by no
means unusual and in many ways typical of early August
monsoonal flow conditions,

METHODOLOGY AND DATA SOURCES

The re-constitution of the Wetzel Creek thunderstorm
rainfall relied on the following data sources:

a. Offical rainfall and upper air observations taken
or released by the National Weather Service Office
in Denver.

b. Offical radar observations taken at the National
Weather Service office in Limon, Colorado.

c. Weather observations and forecasts issued by the
National Weather Service or the National
Heteorological Center on the Domestic Data and
DIFAX weather data lines.

d. Weather forecasts and observations 'made by the
Henz Kelly & Associates' meteorologist for the F2P2
in Denver on August 2, 1986.

The methodology followed in re-constituting the
rainfall relied on the combination of "ground-truthing" the
radar observations with obse"ved rainfall rates and
"atmosphere-truthing" the radar observations with model
runs and forecasts from the Henz Kelly Excessive Convective
Rainfall(ECR) model. Once these calibrations of the radar
signal strength and rainfall were correlated and combined
an estimated 5-minutes rainfall rate could be established
for the storm. Additionally rainfall rates were weighted
relative to the storm cell's development cycle.