FLOOD02665

• County: Larimer
• County: Weld
• Community: Larimer and Weld Counties
• Stream Name: Cache La Poudre River Basin
• Basin: South Platte
• Title: Special Study Cache La Poudre River Basin Volume II - Hydrology
• Date: 10/1/1981
• Prepared By: US Army Corps of Engineers
• Floodplain - Doc Type: Floodplain Report/Masterplan

.

County published in 1976 by the Colorado State University (CSU)

Experiment Station in cooperation with the SCS indicates that most of

the soils in the mountainous portion of the Cache la Poudre River basin

belong to Hydrologic Soil Group B. Technical Release No. 55 published

by the SCS was used to select curve numbers based on soil group and

land use. Selected curve numbers were 60 for forested areas and 66 for

pasture and rangeland. Curve number 66 also describes wood or forest

land with a thin stand, poor cover, and no mulch. A value of 1.0 inch

for detention storage was chosen on a trial and error basis in order to

calibrate the model.

CALIBRATION

The WRC frequency analysis of the gaged data at the mouth of the

canyon was used to calibrate the MITCAT model. Due to a limitation of

a total of 200 catchments and streams in MITCAT, separate models were

developed for the North Fork Cache la Poudre River and the Cache la

Poudre River upstream from the North Fork. Rainfall was applied to the

modeled 988 square miles of the contributing area upstream from the

gage. SCS curve numbers were selected as described in the preceding

paragraph and were held constant. Overland flow lengths were chosen as

either 1,200 feet or 1,600 feet, depending upon the topography of each

catchment. Detention storage was the only parameter that was varied.

By a trial and error analysis, a value of 1.0 inch for detention stor-

age produced a modeled discharge of 16,600 cubic feet per second at the

gage for the laO-year flood event. The calibration studies indicated

that the MlTCAT model was sensitive to detention storage. After

adjusting for an average annual June base flow of 2,000 cubic feet per

second, the resulting lOa-year flood model discharge was 18,600 cubic

feet per second. The same procedure was used to determine the 10-,

50-, and 500-year flood events. The discharge-frequency curve derived

from this analysis is shown in figure 2. It crosses the WRC-developed

curve at about the 30-year level. The departure from the WRC curve at

this point appears reasonable since the record is biased by a mix of

.

13