Trends in Precipitation and Streamﬂow and Changes in Stream … · 2002. 6. 17. · tion trends that could affect trends in streamﬂow. Also, because of the isolated nature of

U.S. GEOLOGICAL SURVEY

Water-Resources Investigations Report 00–4130

Trends in Precipitation and Streamflowand Changes in Stream Morphologyin the Fountain Creek Watershed,Colorado, 1939–99

By Robert W. Stogner, Sr.

Denver, Colorado2000

Prepared in cooperation with theTURKEY CREEK SOIL CONSERVATION DISTRICT,EL PASO COUNTY SOIL CONSERVATION DISTRICT,CENTRAL COLORADO SOIL CONSERVATION DISTRICT, andPUEBLO COUNTY

U.S. DEPARTMENT OF THE INTERIORBRUCE BABBITT, Secretary

U.S. GEOLOGICAL SURVEY

Charles G. Groat, Director

The use of firm, trade, and brand names in this report is for identification purposes only and doesnot constitute endorsement by the U.S. Geological Survey.

For additional information write to: Copies of this report can be purchased

U.S. Geological SurveyInformation ServicesBox 25286Federal CenterDenver, CO 80225

from:

District ChiefU.S. Geological SurveyBox 25046, Mail Stop 415Denver Federal CenterDenver, CO 80225–0046

CONTENTS III

CONTENTS

Abstract.................................................................................................................................................................................. 1Introduction............................................................................................................................................................................ 2

Purpose and Scope ....................................................................................................................................................... 2Approach...................................................................................................................................................................... 2

Description of Study Area and General Land Use ................................................................................................................ 6Precipitation ........................................................................................................................................................................... 6

General Precipitation Characteristics........................................................................................................................... 6Temporal Trends in Precipitation................................................................................................................................. 8Synthesis of Precipitation Analysis ............................................................................................................................. 15

Streamflow ............................................................................................................................................................................. 15Temporal Trends .......................................................................................................................................................... 17

Trends in High Streamflow Statistics................................................................................................................. 17Instantaneous Peak Streamflow ............................................................................................................... 17High Streamflow Percentiles ................................................................................................................... 19Streamflow Duration................................................................................................................................ 21

Trends in Low Streamflow Statistics ................................................................................................................. 24Low Streamflow Percentiles .................................................................................................................... 24Streamflow Duration................................................................................................................................ 25

Spatial Trends .............................................................................................................................................................. 28High Streamflow ................................................................................................................................................ 28Low Streamflow................................................................................................................................................. 30

Relation between Precipitation and Streamflow.......................................................................................................... 32Synthesis of Streamflow Analysis ............................................................................................................................... 34

Stream Morphology ............................................................................................................................................................... 36Generalized Stream Channel Characteristics............................................................................................................... 36Sediment Transport Capacity....................................................................................................................................... 37Descriptive Assessment of Changes in Channel Morphology..................................................................................... 37

Summary and Conclusions .................................................................................................................................................... 41Selected References ............................................................................................................................................................... 43

FIGURES

1. Map showing location of Fountain Creek watershed, streamflow and precipitation monitoring stations,and river reaches......................................................................................................................................................... 3

2–14. Graphs showing:2. Population of El Paso, Pueblo, and Teller Counties, 1890–1990, with projected year 2000 population........... 73. Annual precipitation, departure from annual mean precipitation, and cumulative departures from

annual mean precipitation at Ruxton Park and Colorado Springs ..................................................................... 94. Annual precipitation, departure from annual mean precipitation, and cumulative departures from

annual mean precipitation at Fountain and Pueblo ............................................................................................ 105. Distribution of precipitation at stations in and near the Fountain Creek watershed for period of

record, pre-1977, and post-1976 time periods ................................................................................................... 126. Total annual precipitation with estimated trend line for the period prior to 1977, 1977 to 1999,

and period of record ........................................................................................................................................... 137. Annual hydrograph of average mean-daily streamflow at Pinon, 1973 through 1999....................................... 168. Annual instantaneous peak streamflow at gaging stations in the Fountain Creek watershed,

1940–99.............................................................................................................................................................. 189. Magnitude of 7-, 14-, and 30-day high flows at gaging stations in the Fountain Creek watershed,

1940–99.............................................................................................................................................................. 22

IV CONTENTS

10. Magnitude of 7-, 14-, and 30-day low flows at gaging stations in the Fountain Creek watershed,1940–99.............................................................................................................................................................. 26

11. Normalized high flow for selected reaches of Fountain and Monument Creeks, 1977 through 1999............... 29

12. Normalized low flow for selected reaches of Fountain and Monument Creeks, 1977 through 1999................ 31

13. Relation between cumulative average precipitation at four precipitation gages in the Fountain Creekwatershed and cumulative daily-mean streamflow at Pueblo, October 1959 through August 1997 ................. 33

14. Relation between cumulative average precipitation at four precipitation gages in the Fountain Creekwatershed and cumulative daily-mean streamflow at Pueblo, November through March and Aprilthrough October, 1959–97 ................................................................................................................................. 35

15. Photograph showing view looking south at the Overton Road bridge and edge of high terrace onApril 26, 1999. ........................................................................................................................................................... 39

16. Aerial photograph of Fountain Creek at the Overton Road bridge in September 1991 and periodicchanges in the general location of the high terrace from 1955 and 1970 through 1999 after theApril 1999 flood ......................................................................................................................................................... 40

17. Photograph showing streambank erosion and sediment deposition on the flood plain of Fountain Creek................ 41

TABLES

1. Precipitation stations in and near the Fountain Creek watershed and period of record .......................................... 4

2. Streamflow-gaging stations with 23 or more years of continuous record, period of record for each station,and drainage area..................................................................................................................................................... 4

3. Range in magnitude of daily precipitation at precipitation-monitoring stations in and near the FountainCreek watershed ...................................................................................................................................................... 8

4. Seasonal distribution of daily precipitation of indicated magnitude at precipitation-monitoring stations inand near the Fountain Creek watershed .................................................................................................................. 8

5. Frequency of receiving precipitation of 0.01 inch or greater during a 24-hour period at locations in theFountain Creek watershed for water year 1960 through the end of record, 1960 through water year 1976and 1977 through the end of record ........................................................................................................................ 11

6. Kendall trend analysis of annual precipitation for the period of record, and pre-1977 and post-1976 timeperiods ..................................................................................................................................................................... 14

7. Summary of Kendall trend analysis of spring precipitation.................................................................................... 14

8. Wilcoxon rank sum test of differences between mean annual precipitation during the pre-1977 andpost-1976 time periods............................................................................................................................................ 15

9. Kendall trend analysis of annual instantaneous peak streamflow for the indicated period of record,and pre-1976 and post-1975 time periods............................................................................................................... 19

10. Summary of five largest streamflow events on Fountain Creek at Pueblo, Colorado, and magnitude andgeneral location of precipitation.............................................................................................................................. 19

11. Kendall trend analysis of 70th, 90th, and 100th percentiles of streamflow for the respective periods ofrecord, pre-1977, and post-1976 time periods ........................................................................................................ 20

12. Kendall trend analysis of 7-, 14-, and 30-day high daily-mean streamflow duration ............................................. 23

13. Kendall trend analysis for 0th, 10th, and 30th percentiles of streamflow for the respective periods ofrecord, pre-1977 and post-1976 time periods ......................................................................................................... 25

14. Kendall trend analysis of 7-, 14-, and 30-day low streamflow duration ................................................................. 27

15. Kendall trend analysis for 70th, 90th, and 100th percentiles of normalized streamflow for the post-1976time period .............................................................................................................................................................. 30

16. Kendall trend analyses for the 0th, 10th, and 30th percentiles of normalized streamflow for the post-1976time period .............................................................................................................................................................. 32

CONTENTS V

CONVERSION FACTORS AND VERTICAL DATUM

Sea level: In this report, “sea level” refers to the National Geodetic Vertical Datum of 1929 (NGVD of 1929)—a geodetic datum derivedfrom a general adjustment of the first-order level nets of both the United States and Canada, formerly called Sea Level Datum of 1929.

Multiply By To obtain

cubic foot per second (ft3/s) 0.02832 cubic meter per secondfoot (ft) 0.3048 meter

inch 25.4 millimetersinch per year (in/yr) 25.4 millimeters per year

mile 1.609 kilometersquare mile (mi2) 2.590 square kilometer

Abstract 1

Trends in Precipitation and Streamflow andChanges in Stream Morphology in the FountainCreek Watershed, Colorado, 1939–99

By Robert W. Stogner, Sr.

Abstract

The Fountain Creek watershed, located inand along the eastern slope of the Front Rangesection of the southern Rocky Mountains, drainsapproximately 930 square miles of parts of Teller,El Paso, and Pueblo Counties in eastern Colorado.Streamflow in the watershed is dominated byspring snowmelt runoff and storm runoff duringthe summer monsoon season. Flooding duringthe 1990’s has resulted in increased streambankerosion. Property loss and damage associated withflooding and bank erosion has cost area residents,businesses, utilities, municipalities, and State andFederal agencies millions of dollars. Precipitation(4 stations) and streamflow (6 stations) data,aerial photographs, and channel reconnaissancewere used to evaluate trends in precipitation andstreamflow and changes in channel morphology.Trends were evaluated for pre-1977, post-1976,and period-of-record time periods.

Analysis revealed the lack of trend intotal annual and seasonal precipitation duringthe pre-1977 time period. In general, the analysisalso revealed the lack of trend in seasonal precipi-tation for all except the spring season during thepost-1976 time period. Trend analysis revealed asignificant upward trend in long-term (period ofrecord) total annual and spring precipitation data,apparently due to a change in total annual precipi-tation throughout the Fountain Creek watershed.During the pre-1977 time period, precipitationwas generally below average; during the post-1976 time period, total annual precipitation

was generally above average. During the post-1976 time period, an upward trend in total annualand spring precipitation was indicated at twostations. Because two of four stations evaluatedhad upward trends for the post-1976 period andstorms that produce the most precipitation areisolated convection storms, it is plausible thatother parts of the watershed had upward precipita-tion trends that could affect trends in streamflow.Also, because of the isolated nature of convectionstorms that hit some areas of the watershedand not others, it is difficult to draw strongconclusions on relations between streamflowand precipitation.

Trends in annual instantaneous peakstreamflow, 70th percentile, 90th percentile,maximum daily-mean streamflow (100th percen-tile), 7-, 14-, and 30-day high daily-mean stream-flow duration, minimum daily-mean streamflow(0th percentile), 10th percentile, 30th percentile,and 7-, 14-, 30-day low daily-mean streamflowduration were evaluated. In general, instantaneouspeak streamflow has not changed significantlyat most of the stations evaluated. Trend analysisrevealed the lack of a significant upward trend instreamflow at all stations for the pre-1977 timeperiod. Trend tests indicated a significant upwardtrend in high and low daily-mean streamflowstatistics for the post-1976 period. Upward trendsin high daily-mean streamflow statistics may bean indication that changes in land use within thewatershed have increased the rate and magnitudeof runoff. Upward trends in low daily-mean

2 Trends in Precipitation and Streamflow and Changes in Stream Morphology in the Fountain Creek Watershed,Colorado, 1939–99

streamflow statistics may be related to changesin water use and management. An analysis ofthe relation between streamflow and precipitationindicated that changes in water management havehad a marked effect on streamflow.

Observable change in channel morphol-ogy and changes in distribution and density ofvegetation varied with magnitude, duration,and frequency of large streamflow events, andincreases in the magnitude and duration of lowstreamflows. Although more subtle, low stream-flows were an important component of day-to-daychannel erosion. Substantial changes in channelmorphology were most often associated withinfrequent large or catastrophic streamflowevents that erode streambed and banks, alterstream course, and deposit large amounts ofsediment in the flood plain.

INTRODUCTION

The Fountain Creek watershed, in and alongthe eastern slope of the Front Range section of thesouthern Rocky Mountains, drains approximately930 mi2 of parts of Teller, El Paso, and PuebloCounties in eastern Colorado (fig. 1). Land-surfaceelevation varies from 14,110 ft at the summit of Pike'sPeak to 4,640 ft at the confluence of Fountain Creekand the Arkansas River.

Over the past several years, landowners,farmers, resource managers, municipal, county,local, and Federal agencies have expressed concernthat possible increases in streamflow have adverselyaffected channel stability, resulting in substantialbank erosion along Fountain Creek. Bank erosionhas resulted in property losses and damage to roads,bridges, and other structures along the creek. Damagesassociated with floods are costing property ownersand local and State governments millions of dollars toreclaim, replace, and (or) repair affected property andstructures (National Oceanographic and AtmosphericAdministration, 1999).

In 1999, the U.S. Geological Survey (USGS), incooperation with the Turkey Creek Soil ConservationDistrict, El Paso County Soil Conservation District,Central Colorado Soil Conservation District, andPueblo County, began a study to evaluate precipita-tion and streamflow trends and changes in streammorphology in the Fountain Creek watershed.

Purpose and Scope

This report describes trends in precipitation andtrends in streamflow in the Fountain Creek watershedand presents a qualitative assessment of changes inchannel morphology of selected reaches of FountainCreek downstream from Colorado Springs, Colorado.Trends in precipitation were evaluated for four stationsin or near the Fountain Creek drainage basin for theperiod of record (fig. 1, table 1). Selection of thesestations was based on length of record and locationin or near the Fountain Creek watershed. Trendsin streamflow were evaluated for the period of recordat six streamflow-gaging stations (fig. 1, table 2).The six streamflow-gaging stations were selectedon the basis of length of streamflow record and loca-tion. The length of continuous record varied from 23to 40 years. In addition to continuous record, severalstations had historical data, pre-dating the period ofcontinuous record, which also were included in theevaluation of trends in streamflow. Changes in channelmorphology were evaluated using aerial photographsof Fountain Creek downstream from Colorado Springstaken between 1947 and 1999 and field reconnaissanceduring 1999.

For the purpose of this study, the existence orlack of trend was determined by the reported signifi-cance level as determined by the statistical test used.Statistical significance was defined as highly signifi-cant, significant, and moderately significant, withcorresponding p-values of less than 0.01, 0.05, and0.1, respectively. A nonsignificant trend was indicatedwhen the p-value was greater than 0.1. The estimatedslope of the trend, or rate of change per unit time(trend slope estimate), was compared to trend slopeestimates of other stations.

Approach

Daily, seasonal, and annual precipitation datafrom four stations in and near the Fountain Creekwatershed (fig. 1) were used to evaluate spatial andtemporal variations and trends in precipitation. Annualprecipitation statistics were computed for each wateryear. A water year extends from October 1 throughSeptember 30 of the following year and is identifiedby the year in which it ends. Complete streamflowrecords at some stations were not available until 1977;therefore, precipitation data were divided into pre-1977 and post-1976 time periods to (1) evaluate

INTRODUCTION 3

COLORADO

Pueblo

Grand JunctionColorado Springs

Denver

U.S. Geological Survey streamflow gage and station identifier

Precipitation gage and station identifier

Fountain Creek watershed boundary

R1 Morphological assessment reach and reach number

EXPLANATION

Base modified by Denver Water Board, 1998From U.S. Geological Survey Digital Line Graphs, 100,000, 1981-1983Roads and cities from Colorado Department of Transportation, 50,000, 1998.Albers Equal-Area projectionStandard parallels 37°30' N and 40°30' NCentral meridian 105°30' W

PUEBLO COUNTYEL PASO COUNTY

TELLER COUNTY

COLORADO SPRINGS

PUEBLO

Fountain Creek

Monum

ent Creek

Fountain Creek

Arkansas River

Pikes Peak

Ruxton Park

Colorado Springs

Fountain

Pueblo

Pikeview

Near CO Spgs

Nevada Street

Security

Pinon

Pueblo

MONUMENT

PALMER LAKE

WOODLAND PARK

R4

R2

R1

R3

Cotto

nwoo

d Cree

k

Fort Carson Military Reservation

MANITOU SPRINGS

38°52'30"

38°30'

105° 104°30'

STUDY AREA

0 5 10 15 MILES

0 5 10 15 KILOMETERS

Figure 1. Location of Fountain Creek watershed, streamflow and precipitation monitoring stations, and riverreaches.


whether changes in precipitation occurred within orbetween these two time periods, and (2) make generalcomparisons between trends in precipitation andtrends in streamflow. Kendall trend and Wilcoxon ranksum tests were used to evaluate temporal trends for theperiod of record and compare precipitation character-istics for pre-1977 and post-1976 periods. The Kendalltrend test is a nonparametric test used to evaluate thesignificance of a monotonic trend in a variable overtime; a monotonic trend exists if variable x generallyincreases or decreases as variable y (time) increases(Helsel and Hirsch, 1995). It was used to evaluatetrends in total annual precipitation for the periodof record and for the pre-1977 and post-1976 timeperiods. The Wilcoxon rank sum test is also a non-parametric test and is used to compare the distribu-tion of two populations (Ott, 1993). In this report, theWilcoxon rank sum test was used to compare totalannual and seasonal precipitation during the pre-1977period to total annual and seasonal precipitation

during the post-1976 period. In addition, precipitationdata were evaluated to determine long-term mean,annual departure from mean, and cumulative departurefrom mean.

Streamflow at gaging stations is computedafter developing the relation of stage to streamflowfor a particular location (Buchanan and Somers, 1968;Carter and Davidian, 1968; Kennedy, 1983, 1984).At a typical USGS streamflow-gaging station, streamstage, or the level of water in the stream, is recordedat periodic intervals. Based on the defined relationbetween stage and streamflow, recorded stream stageis converted to streamflow.

Trends in streamflow were evaluated using theKendall trend test. Streamflow data from six gagingstations in the Fountain Creek watershed (table 1,fig. 1) were evaluated for trends in the followingstreamflow statistics: annual instantaneous peakstreamflow; annual maximum (100th percentile), 90th,70th, 30th, 10th, and annual minimum (0th percentile)

Table 1. Precipitation stations in and near the Fountain Creek watershed and period of record

[AP, airport; WSO, Weather Service Office; mm, month; yy, year]

Colorado ClimateCenter station number

Station nameStation identifier

used in this reportPeriod of recordmm/yy to mm/yy

Elevation abovesea level( in feet)

51778 Colorado Springs WSO Colorado Springs 08/48 to 09/99 6,170

53063 Fountain Fountain 08/48 to 09/97 5,550

56704 Pueblo WSO AP Pueblo 07/54 to 09/99 4,640

57309 Ruxton Park Ruxton Park 09/59 to 09/99 9,050

Table 2. Streamflow-gaging stations with 23 or more years of continuous record, period of record for each station, anddrainage area

[mi2, square miles; mm, month; yy, year]

U.S. Geological Surveystation number

Station nameStation

identifier usedin this report

Period of record(mm/yy to mm/yy)

Drainagearea(mi2)

07104000 Monument Creek at Pikeview, CO Pikeview 10/38 to 09/49,01/76 to 09/99

204

07103700 Fountain Creek near Colorado Springs, CO Near CO Spgs 04/58 to 09/99 103

07105500 Fountain Creek at Nevada Street at ColoradoSprings, CO

Nevada Street 10/21 to 09/24,01/76 to 09/99

392

07105800 Fountain Creek at Security, CO Security 10/64 to 09/99 495

07106300 Fountain Creek at Pinon, CO Pinon 04/73 to 09/99 849

07106500 Fountain Creek at Pueblo, CO Pueblo 01/22 to 09/25,10/40 to 09/65,02/71 to 09/99

926

INTRODUCTION 5

daily-mean streamflow; and annual 7-, 14-, and 30-dayhigh daily-mean streamflow duration and annual 7-,14-, and 30-day low daily-mean streamflow duration.The annual instantaneous peak streamflows werederived from the single highest recorded stream stageduring a given water year. Daily-mean streamflow wascomputed by averaging all the periodic computedinstantaneous streamflows over a day. Mean-dailystreamflow was computed by averaging daily-meanstreamflow for days of consecutive years of data (forexample, October 1, 1978, October 1, 1979, and soon). Annual streamflow statistics were computed bysorting and ranking the daily-mean streamflows. Theannual minimum (0th percentile) equals the minimumcomputed daily-mean streamflow during a year; theannual maximum (100th percentile) equals themaximum computed daily-mean streamflow during ayear. Intermediate percentiles, 10th, 30th, 70th, and90th, are equivalent to the 36th, 109th, 255th, and328th annual daily-mean streamflow values whensorted from smallest to largest. The n-percentile indi-cates that n percent of the annual daily-mean stream-flow is below a given streamflow, or (100 minus n)percent is above it. The 7-, 14-, and 30-day high andlow daily-mean streamflow duration statistics werecomputed by averaging daily mean streamflow for 7,14, and 30 consecutive days. This procedure, termedmoving average (Helsel and Hirsch, 1995), wasused for every day of the water year, meaning that,the average daily-mean streamflow for 7, 14, and30 consecutive days was computed for October 1st,and then for October 2d, 3d, and so on.

Time periods for the streamflow trend analysesvaried because of differences in data requirements andavailability for some of the various streamflow statis-tics evaluated. Trends in annual instantaneous peakstreamflow were evaluated for the period of record ateach station and for pre-1976 and post-1975 timeperiods. This break point was selected because somestations were not operational until January of 1976and annual instantaneous peak streamflow data werenot available until the spring of 1976. Computationof streamflow percentiles and streamflow-durationstatistics required continuous annual record; therefore,trend analyses of streamflow percentiles and flowduration were evaluated for respective periods ofcontinuous record at each station and for pre-1977and post-1976 time periods. Annual streamflowstatistics were computed for each water year.

Spatial trends in selected streamflow statisticswere evaluated for the post-1976 time period. Differ-ences in streamflow for five river reaches were normal-ized by the contributing drainage area. Evaluation oftrends in normalized differences in high daily-meanstreamflow statistics identify reaches within whichhydrologic responses (such as rainfall runoff) may bedifferent, may have changed as a result of changes inprecipitation, or may have been altered as a result ofhuman activities. Human activities that may alter thehydrologic response of a reach include changing landuse from natural, pervious or semi-pervious surfaceto an impervious surface; changing vegetative cover;or concentrating and (or) rerouting surface runoff byconstructing storm drains. Additionally, evaluationof trends in normalized differences in low streamflowstatistics provide information related to changes inwater management such as changes in discharge fromwastewater-treatment plants, changes in irrigationreturn flows, and changes in tributary flow that occurwithin the stream reaches.

The relation between streamflow and precip-itation was evaluated using a double-mass curve(Searcy and Hardison, 1960). In a double-mass curve,the cumulation of one variable plotted against thecumulation of another variable will result in a straightline so long as the data are proportional (Searcy andHardison, 1960). A change in the slope of the double-mass curve indicates that the relation between the twovariables has changed. This analysis compared cumu-lative average daily precipitation from four precipita-tion gages to cumulative daily-mean streamflow at thePueblo station (fig. 1). The average precipitation wasused to smooth out spatial variation in precipitation.

Aerial photography from different time periods,field reconnaissance, and topographic maps were usedto determine geomorphic characteristics and evaluatechanges in channel morphology along selected reachesof Fountain Creek (fig. 1). Periodic aerial photographswere used to characterize flood-plain vegetationpatterns, estimate channel width, and determinechannel location and sinuosity. This information wasused to relate large streamflow events to changes inflood-plain characteristics and channel location as aresult of streambed and bank erosion. Photographsselected for the analysis of changes in channelmorphology were taken in 1947, 1955, 1965, 1970,1980, 1991, and 1999.


Field reconnaissance was used to characterizeerosion patterns, which may be causing the changesobserved in channel location and flood-plain vegeta-tion characteristics. Field reconnaissance wasconducted during April and May of 1999, prior toand during the 1999 flood, and during Septemberand October 1999, when streambanks were visible.

DESCRIPTION OF STUDY AREA ANDGENERAL LAND USE

Land use within the Fountain Creek watershedincludes forests, urban areas, military reservations,agriculture, and rangeland. Forested lands are locatedpredominantly in the northwestern mountainous partof the watershed. The major urban center in the water-shed is the Colorado Springs metropolitan area thatincludes Colorado Springs and several smallercommunities in El Paso County. This metropolitanarea is located in the north-central part of the water-shed. Of the three counties encompassing the water-shed, El Paso County had the greatest economicgrowth and population increase (fig. 2). Between1890 and 1950, the population of Pueblo and El PasoCounties increased at similar rates of about 900 peopleper year. Between 1950 and 1990, the populationof El Paso County increased at a rate of about8,000 people per year. Population projections forthe year 2000 indicate this rate has increased toabout 11,000 people per year during the 1990’s.Between 1960 and 1990, population growth in PuebloCounty was relatively flat. Population projections forthe year 2000 indicate the rate of growth during the1990’s has increased to about 1,600 people per year.From 1900 through 1930, the population of TellerCounty decreased at an annual rate of about 600people per year. The population of Teller Countyincreased during the 1930’s, but declined during the1940’s and 1950’s. Since 1970, the population ofTeller County has increased at a rate of about 300people per year. This information is available athttp://www.ancestry.com/free/censtats/cocens.htm(accessed 07/18/00). A small portion of the watershedat and upstream from the confluence of FountainCreek and the Arkansas River is within the urban areaof the city of Pueblo. Agriculture and rangeland arelocated predominantly south and east of ColoradoSprings. Agriculture is common along the alluvialvalley from Fountain to Pueblo and relies heavilyon water diverted from Fountain Creek. A large

expanse of rangeland is included within the bound-aries of the military reservation at Fort Carson locatedjust south of Colorado Springs, west of Interstate 25and Fountain Creek to just northwest of Pueblo(fig. 1).

PRECIPITATION

Analysis of spatial distribution and temporaltrends in precipitation is constrained by the sparsedistribution of long-term precipitation-monitoringstations in the watershed. Variations in the frequencyof daily precipitation, seasonal distribution, seasonaland annual precipitation at four stations were evalu-ated. Seasonal and annual precipitation data were eval-uated for the entire period of record to evaluate long-term trends and were divided into pre-1977 and post-1976 time periods to be consistent with existingstreamflow record.

General Precipitation Characteristics

Climate within the watershed is broadly charac-terized as semiarid temperate continental. However, itcan vary from alpine arctic to semiarid, depending onelevation and proximity to the Front Range (Hansenand others, 1978). Depending on the precipitationstation, between about 40 and 60 percent of dailyprecipitation that occurs is 0.1 inch or less in magni-tude. Between about 70 and 80 percent of daily precip-itation that occurs in the region is less than or equal to0.25 inch (table 3). Daily precipitation of magnitude0.25 inch or greater occurs most frequently betweenJuly and September (table 4). Of the total number ofdaily precipitation events that occurred during Julythrough September, between 25.2 and 34.1 percentwere greater than 0.25 inch in magnitude. Many of theprecipitation events during the summer are convectionstorms driven by the inflow of monsoon moisture fromthe southwest (Giannasca, 1999). Convection stormsare generally strong, isolated events that occur duringthe late afternoon and early evening.

Annual precipitation generally decreases withdistance from the headwaters of the watershed and aselevation decreases. The Ruxton Park station (fig. 1)receives more precipitation than the Colorado Springsstation. Total annual precipitation at the ColoradoSprings station (fig. 3) generally is equivalent to totalannual precipitation at the Fountain station (fig. 4).

PRECIPITATION 7

550,

000

50,0

00

100,

000

150,

000

200,

000

250,

000

300,

000

350,

000

400,

000

450,

000

500,

000

POPULATION

1880

019

0019

2019

40Y

EA

R20

2019

6019

8020

00Pro

ject

ed y

ear

2000

popu

latio

n

Pro

ject

ed y

ear

2000

popu

latio

n

El P

aso

Cou

nty

Pue

blo

Cou

nty

Telle

r C

ount

y

Fig

ure

2.

Pop

ulat

ion

of E

l Pas

o, P

uebl

o, a

nd T

elle

r C

ount

ies,

189

0–19

90, w

ith p

roje

cted

yea

r 20

00 p

opul

atio

n.


Total annual precipitation at the Colorado Springsand Fountain stations generally are greater than totalannual precipitation at the Pueblo station.

During the respective periods of record,recorded annual precipitation was variable from yearto year at all stations (figs. 3 and 4). The Ruxton Parkstation consistently received the most precipitationannually, reporting a mean annual precipitation of24.5 inches. The reporting station at the Pueblostation received the least rainfall annually, reportinga mean annual precipitation of 11.9 inches.

Temporal Trends in Precipitation

Temporal changes in precipitation characteris-tics were analyzed. The analysis revealed no appre-ciable difference in the percentage of days withprecipitation greater than or equal to 0.01 inch(table 5) during the post-1976 time period comparedto the pre-1977 time period. Differences in thefrequency distribution of daily precipitation betweenthe pre-1977 and post-1976 time periods comparedto the entire period of record were slight (fig. 5).

Table 3. Range in magnitude of daily precipitation at precipitation-monitoring stations in and near the Fountain Creekwatershed

Magnitude

Precipitation monitoring station name

Colorado Springs Fountain Pueblo Ruxton Park

Percentage of daily precipitation values less than or equal to indicated magnitude

0.10 57.2 42.9 57.5 40.6

0.25 78.7 69.0 79.1 70.7

0.50 90.9 86.4 91.9 87.6

0.75 95.3 92.6 95.6 94.1

1.00 97.3 95.7 98.3 96.8

2.00 99.5 99.3 99.8 99.4

3.00 99.9 99.8 100.0 99.9

4.00 100.0 99.9 100.0 100.0

5.00 100.0 100.0 100.0 100.0

6.00 100.0 100.0 100.0 100.0

Table 4. Seasonal distribution of daily precipitation of indicated magnitude at precipitation-monitoring stations in and near theFountain Creek watershed

[≤, less than or equal to; >, greater than]

Magnitude(inches)

October–March April–June July–September

Number ofevents during

period of record

Percentageof events


period of record

Percentageof events


period of record

Percentageof events

Colorado Springs

≤0.25 1,342 29.2 736 16.0 1,535 33.4>0.25 176 3.8 264 5.8 540 11.8

Fountain

≤0.25 688 22.4 658 21.4 819 26.7>0.25 165 5.4 316 10.3 424 13.8

Pueblo

≤0.25 1,005 31.8 709 22.4 809 25.6>0.25 156 4.9 209 6.6 273 8.6

Ruxton Park

≤0.25 1,071 25.7 785 18.9 1,084 26.1>0.25 315 7.6 386 9.3 520 12.5

PRECIPITATION 9

1940

060 2040

ANNUAL PRECIPITATION, IN INCHES

1950

1960

1970

1980

1990

Rux

ton

Par

k

Mea

n 24

.5

2000

1940

2000

1950

1960

1970

1980

1990

-1020 010

DEPARTURE FROM ANNUALMEAN PRECIPITATION, IN INCHES

1940

2000

1950

1960

1970

1980

1990

050 2040

2000

1940

1950

1960

1970

1980

1990

20 -10010 194

020

0019

5019

6019

7019

8019

90

WAT

ER

YE

AR

-6020 -40

-20

Col

orad

o S

prin

gs

Rux

ton

Par

kC

olor

ado

Spr

ings

Col

orad

o S

prin

gs

Mea

n 16

.4

2000

1940

1950

1960

1970

1980

1990

WAT

ER

YE

AR

-6020 -40

-200

CUMULATIVE DEPARTURE FROMANNUAL MEAN PRECIPITATION, IN INCHES

Rux

ton

Par

k

0

Fig

ure

3.

Ann

ual p

reci

pita

tion,

dep

artu

re fr

om a

nnua

l mea

n pr

ecip

itatio

n, a

nd c

umul

ativ

e de

part

ures

from

ann

ual m

ean

prec

ipita

tion

at R

uxto

n P

ark

and

Col

orad

oS

prin

gs.


Fou

ntai

n

Fou

ntai

nP

uebl

o

Fou

ntai

nP

uebl

o

Pue

blo

1940

30 1020ANNUAL PRECIPITATION, IN INCHES

1940

-1015 -510

DEPARTURE FROM ANNUALMEAN PRECIPITATION, IN INCHES

1940

-6020 -40

-20

CUMULATIVE DEPARTURE FROMANNUAL MEAN PRECIPITATION, IN INCHES

019

5019

6019

7019

8019

90

2000

1950

1960

1970

1980

1990

1940

030 1020

2000

2000

1950

1960

1970

1980

1990

WAT

ER

YE

AR

1950

1950

1960

1970

-1015 -50510 195

019

60-6

020 -40

-200

Mea

n 15

.7

2000

1980

1990

2000

1970

1980

1990

WAT

ER

YE

AR

2000

1960

1970

1980

1990

Mea

n 11

.9

005

Fig

ure

4.

Ann

ual p

reci

pita

tion,

dep

artu

re fr

om a

nnua

l mea

n pr

ecip

itatio

n, a

nd c

umul

ativ

e de

part

ures

from

ann

ual m

ean

prec

ipita

tion

at F

ount

ain

and

Pue

blo.

PRECIPITATION 11

Trend estimates for the period of record werecomputed for all stations (fig. 6). Trend tests indicatedstatistically significant (p < 0.05) upward trends intotal annual precipitation over the total period ofrecord at the Colorado Springs and Fountain stations(fig. 6, table 6), and moderately significant upwardtrends in total annual precipitation at the Ruxton Parkand Pueblo stations (fig. 6, table 6).

Trend analysis was also done on the precipita-tion data for the pre-1977 and post-1976 time periods.During the pre-1977 period, all stations generallyrecorded below-average total annual precipitation andcumulative departures from the mean annual precipita-tion generally were increasingly negative (figs. 3 and4). During the post-1976 period, all stations generallyrecorded above-average total annual precipitation(figs. 3 and 4). Kendall trend analysis revealed nosignificant trends in total annual precipitation at anystation for the pre-1977 time period, and the ColoradoSprings and Fountain stations for the post-76 timeperiod. Kendall trend analysis revealed a moderatelysignificant upward trend at the Pueblo station, andsignificant upward trend in total annual precipitationat the Ruxton Park station for the post-1976 period

(fig. 6, table 6). The indication of significant trends atsome stations and not others is probably a result ofstorms that hit some areas but not others, which makesit difficult to draw strong conclusions on relationsbetween precipitation and streamflow at differentstations.

Daily precipitation data were divided intoseasons to evaluate seasonal trends; October toDecember – fall, January to March – winter, April toJune – spring, and July to September – summer. Withthe exception of summer precipitation for the period1949 – 97 at the Fountain station (p = 0.0018), trendtest revealed no significant trend in seasonal precipita-tion during the fall, winter, and summer for the periodof record, the pre-1977 period, or the post-1976period. Kendall trend tests revealed highly significantto significant upward trends in spring precipitation atthe Ruxton Park, Colorado Springs, and Pueblostations for the period of record and moderatelysignificant trends in spring precipitation at the RuxtonPark and Pueblo stations in the post-1976 time period(table 7). Trends in annual precipitation at the RuxtonPark and Pueblo stations (table 6) are a function oftrends in springtime precipitation.

Table 5. Frequency of receiving precipitation of 0.01 inch or greater during a 24-hour period at locations in the FountainCreek watershed for water year 1960 through the end of record, 1960 through water year 1976 and 1977 through the endof record

[


070 102030405060

PERCENT

EX

PLA

NAT

ION

0.10to0.01

0.25to0.11

0.50to0.26

0.75to0.51

1.00to0.76

2.00to1.01

3.00to2.01

4.00to3.01

5.00to4.01

grea

ter

or5.01

DA

ILY

PR

EC

IPIT

ATIO

N, I

N IN

CH

ES

Fou

ntai

nP

uebl

o

070 102030405060

Per

iod

of r

ecor

dpr

e-19

77po

st-1

976

0.10to0.01

0.25to0.11

0.50to0.26

0.75to0.51

1.00to0.76

DA

ILY

PR

EC

IPIT

ATIO

N, I

N IN

CH

ES

2.00to1.01

3.00to2.01

4.00to3.01

5.00to4.01

grea

ter

or5.01

0.10to0.01

0.25to0.11

0.50to0.26

0.75to0.51

1.00to0.76

2.00to1.01

3.00to2.01

4.00to3.01

5.00to4.01

grea

ter

or5.01

070 102030405060

Rux

ton

Par

kC

olor

ado

Spr

ings

0.10to0.01

0.25to0.11

0.50to0.26

0.75to0.51

1.00to0.76

2.00to1.01

3.00to2.01

4.00to3.01

5.00to4.01

grea

ter

or5.01

070 102030405060

PERCENT

Fig

ure

5.

Dis

trib

utio

n of

pre

cipi

tatio

n at

sta

tions

in a

nd n

ear

the

Fou

ntai

n C

reek

wat

ersh

ed fo

r pe

riod

of r

ecor

d, p

re-1

977,

and

pos

t-19

76 ti

me

perio

ds.

PRECIPITATION 13

1940

2000

1950

1960

1970

1980

1990

540 101520253035

TOTAL ANNUAL PRECIPITATION, IN INCHES

Rux

ton

Par

k19

60-9

9, m

ean

= 2

4.5,

sig

nific

ance

0.0

919

60-7

6, m

ean

= 2

3.5,

sig

nific

ance

1.0

0

1977

-99,

mea

n =

25.

3, s

igni

fican

ce 0

.01

Ann

ual t

otal

1940

2000

1950

1960

1970

1980

1990

540 101520253035

Col

orad

o S

prin

gs19

49-9

9, m

ean

= 1

6.4,

sig

nific

ance

0.0

0919

49-7

6, m

ean

= 1

4.8,

sig

nific

ance

0.6

019

77-9

9, m

ean

= 1

8.3,

sig

nific

ance

0.4

2A

nnua

l tot

al

1940

2000

1950

1960

1970

1980

1990

WAT

ER

YE

AR

540 101520253035

TOTAL ANNUAL PRECIPITATION, IN INCHES

Fou

ntai

n19

49-9

7, m

ean

= 1

5.7,

sig

nific

ance

0.0

0519

49-7

6, m

ean

= 1

3.7,

sig

nific

ance

0.9

319

77-9

7, m

ean

= 1

9.1,

sig

nific

ance

1.0

0A

nnua

l tot

al

1940

2000

1950

1960

1970

1980

1990

WAT

ER

YE

AR

540 101520253035

Pue

blo

1955

-99,

mea

n =

11.

9, s

igni

fican

ce 0

.06

1955

-76,

mea

n =

11.

2, s

igni

fican

ce 0

.31

1977

-99,

mea

n =

12.

7, s

igni

fican

ce 0

.08

Ann

ual t

otal

Fig

ure

6.

Tot

al a

nnua

l pre

cipi

tatio

n w

ith e

stim

ated

tren

d lin

e fo

r th

e pe

riod

prio

r to

197

7, 1

977

to 1

999,

and

per

iod

of r

ecor

d.


The Wilcoxon rank-sum test was used to eval-uate whether the mean precipitation during the periodsprior to 1977 and after 1976 were different. TheWilcoxon rank-sum test revealed moderately signifi-cant differences in mean precipitation between the twoperiods at the Pueblo station and highly significantdifferences between the two periods at the ColoradoSprings and Fountain stations (table 8). This test alsoindicated no significant differences in mean precipita-tion between the two periods at the Ruxton Parkstation. Although tests revealed differences in themean precipitation for the periods prior to and since

the mid-1970’s, the distribution in the magnitude ofprecipitation events during these time periods wassimilar (fig. 5). Therefore, differences between the twoperiods are not a result of more rain during any givenday but more days with “typical” rainfall.

The Wilcoxon rank-sum test was also used toevaluate differences in seasonal precipitation duringthe pre-1977 and post-1976 periods. The Wilcoxonrank-sum test revealed no significant differencesbetween mean precipitation during the fall, winter, andsummer seasons, except for summer precipitation atthe Fountain station. The Wilcoxon rank-sum test

Table 6. Kendall trend analysis of annual precipitation for the period of record, and pre-1977 and post-1976 time periods

[Shaded p-value indicates period with significant trend; see table 1, fig. 1 for station location]

Station identifier Time periodTime period(water years)

Kendall tauTwo-sidep-value

Trend slope(inches/year)

Colorado Springs Pre-1977 1950–76 0.07 0.6022 0.0855

Post-1976 1977–99 0.13 0.4298 0.1429

Period of record 1950–99 0.26 0.0086 0.1346

Fountain Pre-1977 1949–76 −0.01 0.9335 −0.0058Post-1976 1977–97 −0.01 1.0000 −0.0143Period of record 1949–97 0.30 0.0049 0.1415

Pueblo Pre-1977 1955–76 −0.16 0.3100 −0.1450Post-1976 1977–99 0.27 0.0804 0.1120

Period of record 1955–99 0.20 0.0613 0.0714

Ruxton Park Pre-1977 1960–76 0.00 1.0000 0.0041

Post-1976 1977–99 0.39 0.0144 0.4012

Period of record 1960–99 0.20 0.0916 0.1468

Table 7. Summary of Kendall trend analysis of spring precipitation

[Shaded p-value indicates period with significant trend; see table 1, fig. 1 for location]

Station identifier Time periodTime period(water year)


Trend slope(inches/year)

Colorado Springs Pre-1977 1950–76 0.07 0.6168 0.0257

Post-1976 1977–99 0.04 0.8215 0.0277

Period of record 1950–99 0.22 0.0279 0.0596

Fountain Pre-1977 1949–76 −0.08 0.5878 −0.0336Post-1976 1977–97 0.06 0.7619 0.0700

Period of record 1949–97 0.16 0.1175 0.0312

Pueblo Pre-1977 1955–76 −0.14 0.3819 −0.0500Post-1976 1977–99 0.28 0.0711 0.0693

Period of record 1955–99 0.22 0.0354 0.0376

Ruxton Park Pre-1977 1960–76 0.24 0.2016 0.1478

Post-1976 1977–99 0.27 0.0907 0.1329

Period of record 1960–99 0.32 0.0047 0.1000

STREAMFLOW 15

revealed highly significant differences in summerprecipitation at the Fountain station (p = 0.0048)between the pre-1977 and post-1976 periods. TheWilcoxon rank-sum test also revealed significantdifferences in spring precipitation between thepre-1977 and post-1976 periods at the Fountain(p = 0.0454) and Ruxton Park (p = 0.0459) stations,and moderately significant differences at the ColoradoSprings (p = 0.0622) and Pueblo stations (p = 0.0747).

Synthesis of Precipitation Analysis

Precipitation data from four stations were usedto characterize and evaluate changes and trends inprecipitation in the Fountain Creek watershed. Trendanalysis revealed significant increases in annualprecipitation at the Colorado Springs and Fountainstations since the late 1940’s and moderate increases atRuxton Park and Pueblo stations since the mid- to late1950’s. An analysis of the pre-1977 period indicatesthat annual precipitation was generally below averageand no trends were detected in annual precipitation.An analysis of the post-1976 period indicates thatannual precipitation was generally above average andupward trends were detected at the Ruxton Park andPueblo stations. In addition, the post-1976 annualprecipitation at the Colorado Springs, Fountain,and Pueblo stations were significantly greater thanthe pre-1977 annual precipitation. However, there hasbeen no change in the distribution or magnitude ofrainfall occurring during a 24-hour period. Therefore,the differences between the two periods result frommore days with rainfall rather than more rain during

any given day. Also, because two of the four stationsevaluated had upward trends for the post-1976 periodand storms that produce most of the precipitation areisolated convection storms, it is plausible that otherparts of the watershed had increasing precipitationthat could affect trends in streamflow.

An analysis of seasonal trends indicated thatsignificant increases in spring precipitation haveoccurred at the Ruxton Park, Colorado Springs, andPueblo stations since the 1940’s and 1950’s. An anal-ysis of the pre-1977 period indicated no significanttrend during any season. An analysis of the post-1976period indicated moderately significant upward trendsin spring precipitation at the Ruxton Park and Pueblostations. The upward trends in annual precipitation forthe post-1976 period at the Ruxton Park and Pueblostations are likely a result of the upward trends inspring precipitation at these stations.

STREAMFLOW

Streamflow in the Fountain Creek watershedvaries seasonally and has three distinct flow regimes:base flow, spring snowmelt and storm runoff thatoccurs during the summer monsoon season (fig. 7).The base-flow period begins in late September or earlyOctober and extends until the following April. Duringthe base-flow period, streamflow is fairly uniform.Depending on temperature and winter snowfallamounts, the snowmelt period begins about mid-Apriland extends until about mid-June. Early in the snow-melt period, streamflow increases substantially from

Table 8. Wilcoxon rank sum test of differences between mean annual precipitation during the pre-1977 and post-1976 timeperiods

[Shaded values indicate significant differences in mean precipitation between the pre-1977 and post-1976 time periods; N, number of complete years ofrecord in period]


N Mean score p-value

Colorado Springs Pre-1977 1950–76 28 20.7 0.0083

Post-1976 1977–99 22 31.7

Fountain Pre-1977 1949–76 27 16.9 0.0006

Post-1976 1977–97 16 30.6

Pueblo Pre-1977 1955–76 22 18.9 0.0620

Post-1976 1977–99 22 26.1

Ruxton Park Pre-1977 1960–76 16 17.5 0.4713

Post-1976 1977–99 21 20.1


Oct

Nov

Dec

Jan

Feb

Mar

Apr

May

June

July

Aug

Sep

t0

600

50100

150

200

250

300

350

400

450

500

550

MEAN-DAILY STREAMFLOW, IN CUBIC FEET PER SECONDB

ase

flow

Sno

wm

elt

Sum

mer

Fig

ure

7.

Ann

ual h

ydro

grap

h of

ave

rage

mea

n-da

ily s

trea

mflo

w a

t Pin

on, 1

973

thro

ugh

1999

.

STREAMFLOW 17

base-flow conditions. Streamflow decreases fairlyquickly after peaking in early to mid-May. Thesummer flow period follows the snowmelt period andgenerally begins about mid-June and extends throughSeptember, sometimes into October. Streamflowduring the summer period is highly variable. Changesin streamflow during the summer are primarily drivenby afternoon and evening storms.

Temporal Trends

Temporal trends in streamflow were evaluatedfor the following streamflow statistics: annual instan-taneous peak streamflow, high daily-mean streamflowpercentiles (70th, 90th, 100th percentile), high daily-mean streamflow duration (7-, 14-, 30-day), low daily-mean streamflow percentiles (0th, 10th, 30th percen-tile), and low daily-mean streamflow duration (7-, 14-,30-day). The instantaneous peak streamflow statisticwas evaluated for the period of record, pre-1976, andpost-1975 time periods; all other statistics were evalu-ated for the period of record, pre-1977, and post-1976time periods. The mismatch in the time periods isdue to the fact that two of the selected gaging stationswere not operational until January 1976 (water year).The result was an incomplete year of record, whichprevented the evaluation of streamflow percentilesand streamflow duration statistics. Instantaneous peakstreamflow data were available for the partial yearof record. Cumulative streamflow data were comparedto cumulative precipitation data to evaluate the relationbetween streamflow and precipitation.

Trends in High Streamflow Statistics

Trends in annual instantaneous peak stream-flow; 70th, 90th, and 100th daily-mean streamflowpercentiles; and 7-, 14-, and 30-day daily-meanstreamflow statistics were evaluated. Annual instanta-neous peak streamflows were derived from the singlehighest recorded stage during a given water year.Streamflow percentiles, 70th, 90th, and 100th, areequivalent to the 255th, 328th and annual maximumdaily-mean streamflow values when sorted fromsmallest to largest. Flow-duration statistics, 7-, 14-,and 30-day, were computed by averaging daily-meanstreamflows for n-days.

Instantaneous Peak Streamflow

Variations in annual instantaneous peak stream-flow for the six gaging stations are shown in figure 8.Kendall trend analysis indicated significant upwardtrend in annual instantaneous peak streamflow at thePikeview station for the post-1975 period and highlysignificant upward trend in annual instantaneous peakstreamflow at Security (table 9) for the period ofrecord (1967–99). However, significant trends inannual instantaneous peak streamflow were notdetected during the pre-1976 period at any stationand post-1975 time period at any station exceptthe Pikeview station.

Evaluation of long-term streamflow data atPueblo (1941–65, 1971–99) indicates instantaneousstreamflow peaks of 10,000 ft3/s or greater magnitudeoccurred more frequently during the 1990’s than anydecade since the 1940’s (fig. 8). Annual instantaneouspeak streamflow during 1994–97 and 1999 rankedin the top 16 of 59 (27 percent) all-time recordedmaximum streamflow events. Although large stream-flow events occurred more frequently during the1990’s than during previous decades since the 1940’s,no significant trend was detected in the magnitude ofpeak streamflow events. The magnitudes of stream-flow events that occurred during the 1990’s were notatypical of historical peaks.

Examination of streamflow data and historicalaccounts of the period (Colorado Climate Center,1999; National Oceanographic and AtmosphericAdministration, 1999; Snipes, 1974) indicates thatthe four largest streamflow events at Pueblo occurredduring the spring snowmelt period, mid-April to mid-June (table 10). Each of these events was caused byseveral inches of rainfall that fell during intense stormsover large areas of the Fountain Creek watershed andsoutheastern Colorado during a short period of time.In some areas, rainfall amounts received during acouple of days approached or exceeded the averagetotal annual rainfall in the Colorado Springs area(table 10) (National Oceanographic and AtmosphericAdministration, 1999; Hansen and others, 1978;Snipes, 1974). Also significant is the fact that themost recent event, the flood of April 30, 1999, wasestimated to be about a 15-year flood for this station.A 15-year flood is a streamflow event of a givenmagnitude with a probability of recurring once every15 years. This estimate is based on available record,not including the 1935 peak streamflow. The peakstreamflow of 1935 is an estimate and considered


1940

1950

1960

1970

0

6,00

0

2,00

0

4,00

0

Pik

evie

w

1950

1960

1970

0

12,0

00

2,00

0

4,00

0

6,00

0

8,00

0

10,0

00

STREAMFLOW, IN CUBIC FEET PER SECOND

Nev

ada

Str

eet

1940

1950

1960

1970

1980

1990

WAT

ER

YE

AR

(O

CTO

BE

R -

SE

PT

EM

BE

R)

0

20,0

00

5,00

0

10,0

00

15,0

00

Pin

on

2000

1980

1990

2000

1940

0

5,00

0 1940

1950

1960

WAT

ER

YE

AR

(O

CTO

BE

R -

SE

PT

EM

BE

R)

0

30,0

00

10,0

00

20,0

00

Pue

blo

2000

1950

1960

1970

1980

1990

20,0

00

10,0

00

15,0

00

Sec

urity

2000

1970

1980

1990

max

imum

str

eam

flow

dur

ing

1965

47,0

00 c

ubic

feet

per

sec

ond

2000

1950

1960

1970

1980

1990

2000

1940

0

1,00

0

3,00

0

2,00

0

Nea

r C

O S

pgs

1980

1990

1940

Fig

ure

8.

Ann

ual i

nsta

ntan

eous

pea

k st

ream

flow

at g

agin

g st

atio

ns in

the

Fou

ntai

n C

reek

wat

ersh

ed, 1

940–

99.

STREAMFLOW 19

questionable. The flood of April 30, 1999, causedmillions of dollars of damage and resulted in Presiden-tial declaration of several counties in the FountainCreek watershed as well as other downstream countiesas a Federal flood disaster area.

High Streamflow Percentiles

The 70th percentile (Q70), 90th percentile (Q90),and 100th percentile (Q100) daily-mean streamflowswere computed for each year. The n-percentile indi-cates that n percent of the annual daily-mean stream-flow is below a given streamflow, or (100 minus n)percent is above it. Results of analysis of trends instreamflow percentiles are summarized in table 11.

Two stations, Near CO Spgs and Security, hadcontinuous streamflow record dating back to the late1950’s and mid 1960’s. Two other stations, Pikeviewand Pueblo, had historical streamflow record datingback to the early 1920’s and early 1940’s; however,this record was not continuous. Significant upwardtrends (p


regimes were larger than the Near CO Spgs station,with trend slopes of about 3.5, 5.2, and 26.7 ft3/s peryear. Analysis of trends for the period of record atPikeview and Pueblo was not conducted due to thehiatus of continuous record prior to 1976.

Streamflow records were divided into pre-1977and post-1976 time periods. Streamflow record fromfour stations—Pikeview, Near CO Spgs, Security, andPueblo, as noted above—were used to evaluate trendsin the pre-1977 period. Significant trends were not

Table 11. Kendall trend analysis of 70th, 90th, and 100th percentiles of streamflow for the respective periods of record,pre-1977, and post-1976 time periods

[Shaded p-values indicate periods with a significant trend; ft3/s/yr, cubic feet per second per year]



Trend slope(ft3/s/yr)

Annual 70th percentile of daily streamflow (Q70)

Pikeview pre-1977 1939–49 −0.04 0.9372 0.00post-1976 1977–99 0.39 0.0096 1.21

Near CO Spgs period of record 1959–99 0.30 0.0052 0.22

pre-1977 1959–76 −0.07 0.7316 −0.02post-1976 1977–99 0.34 0.0262 0.55

Nevada Street post-1976 1977–99 0.36 0.0187 2.40

Security period of record 1965–99 0.60 0.0000 3.53

pre-1977 1965–76 0.02 1.0000 0.17

post-1976 1977–99 0.48 0.0014 5.00

Pinon post-1976 1977–99 0.51 0.0008 6.78

Pueblo pre-1977 1941–65 −0.16 0.2721 −0.94post-1976 1977–99 0.49 0.0011 6.57


Pikeview pre-1977 1939–49 −0.02 1.0000 −0.20post-1976 1977–99 0.29 0.0571 1.55


pre-1977 1959–76 0.00 1.0000 0.00

post-1976 1977–99 0.26 0.0905 1.30



pre-1977 1965–76 −0.29 0.2160 −2.67post-1976 1977–99 0.31 0.0419 7.62

Pinon post-1976 1977–99 0.29 0.0571 7.75

Pueblo pre-1977 1941–65 −0.20 0.1803 −4.77post-1976 1977–99 0.29 0.0538 8.13


Pikeview pre-1977 1939–49 0.13 0.6404 4.33

post-1976 1977–99 0.42 0.0055 13.07


pre-1977 1959–76 0.01 1.0000 0.14

post-1976 1977–99 0.16 0.2908 3.35



pre-1977 1965–76 0.12 0.6312 28.00

post-1976 1977–99 0.32 0.0324 25.18

Pinon post-1976 1977–99 0.33 0.0302 50.88

Pueblo pre-1977 1941–65 −0.07 0.6238 −11.47post-1976 1977–99 0.41 0.0071 54.00

STREAMFLOW 21

detected in the Q70, Q90, and Q100 streamflow regimesat the Pikeview, Near CO Spgs, Security, and Pueblostations during the pre-1977 time period.

For the post-1976 time period, streamflowrecords from all six stations were evaluated. Analysesindicated significant upward trends in the Q70 stream-flow regime at all stations. Trend-slope estimatesindicated that changes in the Q70 streamflow regimegenerally increased from upstream to downstream andranged from 0.6 ft3/s per year at the Near CO Spgsstation to about 6.8 ft3/s per year at the Pinon stationand 6.6 ft3/s per year at the Pueblo station. Moderatelysignificant to significant trends were detected in thepost-1976 Q90 streamflow regime at the Pikeview,Near CO Spgs, Security, Pinon, and Pueblo stations. Asignificant trend was not detected in the Q90 stream-flow regime at Nevada Street during the post-1976period. Trend-slope estimates indicate that changes inthe Q90 streamflow regime increased from upstream todownstream stations and ranged from 1.3 ft3/s/yr atNear CO Spgs to 8.1 ft3/s/yr at Pueblo. Moderate tohighly significant trends were detected in the Q100steamflow regimes at the Pikeview, Nevada Street,Security, Pinon, and Pueblo stations. Trend-slope esti-mates indicated that changes in the Q100 streamflowregime generally increased from upstream to down-stream and ranged from about 13.1 ft3/s per year at thePikeview station to 54.0 ft3/s per year at the Pueblostation. A significant trend in the Q100 streamflowregime was not detected in the post-1976 streamflowrecord for the Near CO Spgs station.

The significant upward trend in precipitationat the Ruxton Park station during the post-1976 period(fig. 6) could possibly explain the upward trends inQ70 and Q90 streamflow at the Near CO Spgs station.In the absence of significant changes over time inpost-1976 precipitation at the Colorado Springs andFountain stations (fig. 6), upward trends in the highstreamflow regimes at the Nevada Street, Security,and Pinon stations that were detected in the post-1976period could be explained by changes in land use and(or) water use and resultant rate of runoff in the water-shed. However, it is plausible that precipitation hasincreased over time in other parts of the watershedbut went undetected because of the sparse spatialdistribution of fairly long-term precipitation stationsand the scattered spatial distribution of convectionstorms. Additionally, changes in high-flow regimesat upstream locations generally would be transferredto downstream locations.

Streamflow Duration

The 7-, 14-, and 30-day high daily-mean stream-flow statistics were evaluated. The n-day high stream-flow duration indicates the highest average daily-meanstreamflow for n-consecutive days during a year. Timeseries of annual 7-, 14-, and 30-day high-flow magni-tude are shown in figure 9.

Analysis of 7-, 14-, and 30-day high daily-meanstreamflow magnitudes at the Pikeview station did notindicate trends during the 1939 through 1949 period(table 12). However, moderately significant to signifi-cant upward trends were detected in the 7- and 14-daydaily-mean streamflow magnitudes during the 1977through 1999 period. Although analysis indicatedupward trends in 7- and 14-day high daily-meanstreamflow since 1977, it is important to note that themagnitudes of the high daily-mean streamflow dura-tions since 1977 generally have been similar to, orslightly less than, daily-mean streamflows of the samedurations during the 1939 through 1949 time period(fig. 9).

At the Near CO Spgs station, analysis indicatedmoderately significant upward trend in the high 7-daydaily-mean streamflow and significant upward trendsin the 14- and 30-day daily-mean streamflow statisticsfor the period of record (1959–99). Significant trendswere not indicated in the 7- and 14-day high daily-mean streamflow during the pre-1977 and post-1976time periods; however, moderately significant upwardtrends were indicated in the 30-day high daily-meanstreamflow statistic for the post-1976 period. Figure 9illustrates that the 7-, 14-, and 30-day daily-meanstreamflow in the mid- to late 1990’s have been greaterthan those of most of the historical record. Therefore,the indicated upward significant trend in 14- and30-day high daily-mean streamflow for the periodof record was affected by the high streamflows thatoccurred in the mid- to late 1990’s.

At the Security station, highly significantupward trends were indicated in the 7-, 14-, and30-day high daily-mean streamflows for the periodof record, and moderately significant to significantupward trends were indicated in the 7-, 14-, and30-day high daily-mean streamflow for the post-1976period. Similar to the Near CO Spgs station, no signif-icant trends were indicated for the pre-1977 period atthe Security station. Again, the upward significanttrends in 7-, 14- and 30-day high daily-mean stream-flows for the period of record were affected by thehigh streamflows that occurred in the mid- to late1990’s.


1940

2000

1950

1960

1970

1980

1990

0

1,20

0

200

400

600

800

1,00

0

1940

2000

1950

1960

1970

1980

1990

0

500

100

200

300

400

1940

2000

1950

1960

1970

1980

1990

0

3,50

0

1,00

0

2,00

0

3,00

0


1940

2000

1950

1960

1970

1980

1990

0

5,00

0

1,00

0

2,00

0

3,00

0

4,00

0

1940

2000

1950

1960

1970

1980

1990

WAT

ER

YE

AR

0

5,00

0

1,00

0

2,00

0

3,00

0

4,00

0

7-da

y hi

gh fl

ow

14-d

ay h

igh

flow

30-d

ay h

igh

flow

1940

2000

1950

1960

1970

1980

1990

WAT

ER

YE

AR

0

5,00

0

1,00

0

2,00

0

3,00

0

4,00

0

Pik

evie

wN

ear

CO

Spg

s

Nev

ada

Str

eet

Sec

urity

Pin

onP

uebl

o

Fig

ure

9.

Mag

nitu

de o

f 7-,

14-

, and

30-

day

high

flow

s at

gag

ing

stat

ions

in th

e F

ount

ain

Cre

ek w

ater

shed

, 194

0–99

.

STREAMFLOW 23

High daily-mean streamflow duration analysisat the Pinon station indicated moderately significantupward trend in the magnitude of the 7-day high daily-mean streamflow for the post-1976 period. Significanttrends were not detected in the magnitude of 14- or30-day streamflow statistics.

At the Pueblo station, analysis indicated nosignificant trends in the magnitude of the 7-, 14-, or30-day streamflow statistics for the pre-1977 period.Significant to moderately significant upward trend wasindicated in the 7-, 14-, and 30-day streamflow statis-tics for the post-1976 period. As occurred with many

Table 12. Kendall trend analysis of 7-, 14-, and 30-day high daily-mean streamflow duration

[Shaded p-values indicate periods with a significant trend; ft3/s/yr, cubic feet per second per year; see table 1, figure 1 for location]




7-day

Pikeview pre-1977 1939–49 0.05 0.8763 1.0

post-1976 1977–99 0.30 0.0447 5.4


pre-1977 1959–76 0.12 0.4954 0.7

post-1976 1977–99 0.24 0.1191 2.0



pre-1977 1965–76 0.18 0.4507 13.4

post-1976 1977–99 0.33 0.0303 15.5

Pinon post-1976 1977–99 0.27 0.0725 15.5

Pueblo pre-1977 1941–65 −0.16 0.2825 −6.1post-1976 1977–99 0.30 0.0447 24.7

14-day

Pikeview pre-1977 1939–49 0.09 0.7555 3.4

post-1976 1977–99 0.28 0.0645 3.3


pre-1977 1959–76 0.18 0.3247 0.6

post-1976 1977–99 0.21 0.1696 2.1



pre-1977 1965–76 0.18 0.4507 8.8

post-1976 1977–99 0.30 0.0447 11.8

Pinon post-1976 1977–99 0.24 0.1131 12.5

Pueblo pre-1977 1941–65 −0.17 0.2336 −8.3post-1976 1977–99 0.27 0.0725 18.0

30-day

Pikeview pre-1977 1939–49 0.02 1.0000 0.6

post-1976 1977–99 0.23 0.1256 2.1


pre-1977 1959–76 0.12 0.4954 0.3

post-1976 1977–99 0.25 0.0960 1.7



pre-1977 1965–76 0.12 0.6312 3.4

post-1976 1977–99 0.27 0.0725 10.2

Pinon post-1976 1977–99 0.21 0.1696 10.1

Pueblo pre-1977 1941–65 −0.21 0.1412 −6.11post-1976 1977–99 0.28 0.0683 15.1


of the upstream stations, the streamflow durations inthe mid- to late 1990’s have generally been greaterthan most of the historical record.

The relative ranks of streamflow magnitude ofthe high daily-mean streamflow duration statisticswere compared. The evaluation of the rank of the highdaily-mean streamflow indicates that the magnitude ofstreamflow within all the high daily-mean streamflowduration statistics for 1999 were either ranked first orsecond for the entire period of record at all the stationsevaluated. This indicates that high streamflows thatoccurred during 1999 remained elevated for a longerperiod of time than any other period of record.

Trends in Low Streamflow Statistics

Trends in 0th, 10th, and 30th daily-meanstreamflow percentiles and 7-, 14-, and 30-day lowdaily-mean streamflow statistics were evaluated.Streamflow percentiles, 0th, 10th, and 30th, are equiv-alent to the annual minimum, 36th, and 109th annualdaily-mean streamflow values when sorted fromsmallest to largest. Low streamflow duration statistics,7-, 14-, and 30-day, were computed by averagingdaily-mean streamflows for n-days.

Low Streamflow Percentiles

Specific streamflow statistics were computedfor each year: minimum (Q0), 10th percentile (Q10),and 30th percentile (Q30) daily-mean streamflow.The n-percentile indicates that n percent of the annualdaily-mean streamflow is below a given streamflow,or (100 minus n) percent is above it. Results of anal-ysis of trend in streamflow percentiles are summarizedin table 13.

Significant upward trends were detected in theQ0, Q10, and Q30 streamflow regime at the Near COSpgs station (1959–99). Highly significant upwardtrends were detected in the Q0, Q10, and Q30 stream-flow regimes at the Security station (1965–99). At theNear CO Spgs station, estimated annual changes in theQ0, Q10, and Q30 streamflow regime were small, 0.05,0.06, and 0.07 ft3/s per year, respectively. Downstreamat the Security station, estimated changes in the Q0,Q10, and Q30 were larger, about 1.7, 2.5, and 2.8 ft

3/sper year, respectively.

Similar to the analysis of trends in the magni-tude of high daily-mean streamflow statistics, analysisdid not indicate trends in the Q0, Q10, and Q30 stream-flow regimes during the pre-1977 time period at thePikeview, Near CO Spgs, Security, or Pueblo stations.

For the post-1976 time period, analysisindicated highly significant upward trends in theQ0, Q10, and Q30 streamflow regimes at the Pikeview,Nevada Street, Security, Pinon, and Pueblo stationsand a moderately significant upward trend in the Q30streamflow regime at the Near CO Spgs station. Nosignificant trends were detected for this period in theQ0 and Q10 flow statistics for the Near CO Spgsstation. Trend slope estimates indicate that changesin the Q0 streamflow regime increased from Pikeviewand Nevada, with trend slopes of about 0.5 and0.7 ft3/s per year, to Security, with a trend slope of2.2 ft3/s per year. The trend slope then decreased atPinon and Pueblo to about 0.8 ft3/s per year. Similarchanges were indicated by the analysis of the Q10streamflow regime. Trend slope estimates indicatethat changes in the Q10 streamflow regime increasedfrom Nevada Street (0.9 ft3/s per year) to Security(about 3.4 ft3/s per year) and then decreased atPinon (about 2.7 ft3/s per year). Trend slope estimatesindicate that changes in the Q30 streamflow regimeincreased from Pikeview (about 0.7 ft3/s per year) andNear CO Spgs (about 0.2 ft3/s per year) to NevadaStreet (about 1.1 ft3/s per yr), Security (about 3.9 ft3/sper year), and Pinon (6.1 ft3/s per year) and thendecreased at Pueblo (about 5.3 ft3/s per year).The upward trends in all three of the low stream-flow statistics for the 1977–99 period indicate thatlow streamflow conditions have significantly increasedthroughout most of the watershed over the past23 years.

In the Fountain Creek watershed, streamflowconsists of native and imported transbasin water.Transbasin water is used to meet the growingdomestic, municipal, and industrial water uses in theColorado Springs metropolitan area. Return flowsassociated with lawn watering throughout ColoradoSprings and the municipalities along Monument Creekmay explain the indicated trends in the low streamflowregimes from the Pikeview to Nevada Street stations.Return flows may explain a portion of the indicatedtrends in low streamflow downstream from the NevadaStreet station. Upstream return flows combined withincreases in wastewater effluent and changes in watermanagement downstream from the Nevada Streetstation result in the indicated increases in the low-flowregimes (Q0 – Q30) downstream from the NevadaStreet station.

STREAMFLOW 25

Streamflow Duration

The magnitudes of low streamflow with 7-,14-, and 30-day durations were evaluated. Low n-daystreamflow duration indicates the lowest averagedaily-mean streamflow for n-consecutive days during

a year. Unlike graphical analysis of high daily-meanstreamflows (fig. 9 and table 12), trends in the magni-tude of 7-, 14-, and 30-day low streamflows were quiteobvious for most stations (fig. 10) and are summarizedin table 14.

Table 13. Kendall trend analysis for 0th, 10th, and 30th percentiles of streamflow for the respective periods of record,pre-1977 and post-1976 time periods

[Shaded p-values indicate stations and time periods with a significant trend; ft3/s/yr, cubic feet per second per year; see table 1, figure 1 for location]




Annual minimum (daily mean) streamflow (Q0)

Pikeview pre-1977 1939–49 0.33 0.1844 0.17

post-1976 1977–99 0.51 0.0006 0.47


pre-1977 1959–76 −0.01 0.9697 0.00post-1976 1977–99 0.13 0.4125 0.03



pre-1977 1965–76 0.35 0.1305 0.70

post-1976 1977–99 0.50 0.0010 2.20

Pinon post-1976 1977–99 0.71 0.0000 0.73

Pueblo pre-1977 1941–65 0.19 0.1889 0.01

post-1976 1977–99 0.62 0.0000 0.88Annual 10th percentile of daily streamflow (Q10)

Pikeview pre-1977 1939–49 0.31 0.2101 0.22

post-1976 1977–99 0.58 0.0001 0.57


pre-1977 1959–76 −0.21 0.2377 −0.03post-1976 1977–99 0.14 0.3548 0.08



pre-1977 1965–76 0.30 0.1905 0.54

post-1976 1977–99 0.54 0.0004 3.36

Pinon post-1976 1977–99 0.64 0.0000 2.70

Pueblo pre-1977 1941–65 0.02 0.9248 0.00

post-1976 1977–99 0.61 0.0001 3.00Annual 30th percentile of daily streamflow (Q30)

Pikeview pre-1977 1939–49 0.24 0.3502 0.29

post-1976 1977–99 0.48 0.0014 0.70


pre-1977 1959–76 −0.20 0.2535 −0.06post-1976 1977–99 0.25 0.0952 0.17



pre-1977 1965–76 0.30 0.1905 0.67

post-1976 1977–99 0.60 0.0001 3.85

Pinon post-1976 1977–99 0.58 0.0001 6.06

Pueblo pre-1977 1941–65 −0.15 0.2922 −0.08post-1976 1977–99 0.51 0.0007 5.33


1940

2000

1950

1960

1970

1980

1990

025 5101520

1940

2000

1950

1960

1970

1980

1990

214 4681012

1940

2000

1950

1960

1970

1980

1990

060 2040


1940

2000

1950

1960

1970

1980

1990

0

150

50100

1940

2000

1950

1960

1970

1980

1990

WAT

ER

YE

AR

0

120

20406080100

7-da

y lo

w fl

ow

14-d

ay lo

w fl

ow

30-d

ay lo

w fl

ow

1940

2000

1950

1960

1970

1980

1990

WAT

ER

YE

AR

0

150

50100

Pik

evie

wN

ear

CO

Spg

s

Nev

ada

Str

eet

Sec

urity

Pin

onP

uebl

o

Fig

ure

10.

Mag

nitu

de o

f 7-,

14-

, and

30-

day

low

flow

s at

gag

ing

stat

ions

in th

e F

ount

ain

Cre

ek w

ater

shed

, 194

0–99

.

STREAMFLOW 27

Analysis of 7-, 14-, and 30-day low streamflowsat Pikeview did not indicate trends during the 1939through 1949 time period. Highly significant trendswere detected in the 1977 through 1999 time period.Estimated trend slopes in the 7-, 14-, and 30-day lowstreamflow for the 1977 to 1999 time period wereabout 0.5 and 0.6 ft3/s/yr. Increases in low streamflow

appear to have begun in the early 1980’s, peaked inthe mid-1980’s, and have remained relatively constantsince 1985 (fig. 10).

Analysis of 7-, 14-, and 30-day low streamflowat the Near CO Spgs station indicated significanttrends for the period of record, 1959–99. However,although the analysis indicated that the trends were

Table 14. Kendall trend analysis of 7-, 14-, and 30-day low streamflow duration

[Shaded p-values indicated stations and time periods with significant trend; ft3/s/yr, cubic feet per second per year; see table 1, figure 1 for location]




7-day

Pikeview pre-1977 1939–49 0.27 0.2758 0.21post-1976 19

Documents

Trends in Precipitation and Streamﬂow and Changes in Stream … · 2002. 6. 17. · tion trends that could affect trends in streamﬂow. Also, because of the isolated nature of