A CHANNEL CROSS-SECTIONANALYZER

O.S. Department of the Interior

„Bureau of Land ManagementU.S. Department of Agriculture

Forest Service

XSPRO:

A CHANNEL CROSS-SECTIONANALYZER

Technical Note 387August 1992

by

Gordon E. Grant,

Joseph E. Duval

Greg J. Koerper

James L. Fogg

Additional copies available from:Bureau of Land Management Service Center

Printed Materials Distribution SectionBuilding 41, Denver Federal Center

P.O. Box 25047Denver, Colorado 80225-0047

BLM/SC/PT-92/001+7200

ABSTRACT

XSPRO is an interactive, menu-driven soft-ware package designed to assist watershed spe-cialists in analyzing stream channel cross-sectiondata. It has been specifically developed to handlechannel geometry and hydraulic conditions forsingle transects in steep (gradient > 0.01) streams.Several resistance equations are supported,including those specifically designed for largeroughness channels. Analysis options include

developing stage-to-discharge relationships andevaluating changes in channel cross-sectional area.Both graphical and tabular output can be generated.XSPRO can assist resource specialists in analyz-ing instream flow needs, performing hydraulicreconstructions, designing effective channel andriparian structures, and monitoring channelchanges.

ACKNOWLEDGEMENTS AND

DISCLAIMER

The development of this software wassupported by the USDA Forest Service PacificNorthwest Experiment Station (PNW), Corvallis,Oregon, and the Ecology, Range, and WatershedManagement Staff of Region 6, Portland, Oregon.Further software development and testing,editorial assistance, and publication were sup-ported by the Division of Resource Services andthe Technology Transfer Staff, Bureau of LandManagement Service Center, Denver, Colorado.

The authors wish to acknowledge Mr. SteveSwanson of the Bureau of Land Management andDr. Bill Jackson of the National Park Service fortheir review of the draft documentation and testingof the software. We are especially grateful to Mr.Bill Carey of the U.S. Geological Survey for histhorough testing of the software, without whichXSPRO would surely contain many "glitches"that have now been removed. Finally, we wishto thank Linda Hill and Janine Landberg of the

Bureau of Land Management Technology TransferStaff for their tireless efforts to complete thisproject and produce a usable program documenta-tion reference. •

This software is in the public domain, and therecipient may not assert any proprietary rightsthereto nor represent it to anyone as other than aGovernment-produced program. XSPRO is pro-vided "as-is" without warranty of any kind,including, but not limited to, the implied warran-ties of merchantability and fitness for a particularpurpose. The user assumes all responsibility forthe accuracy and suitability of this program for aspecific application. In no event will PNW, theUSDA Forest Service, or the Bureau of LandManagement be liable for any damages, includinglost profits, lost savings, or other incidental orconsequential damages arising from the use of orthe inability to use this program.

TABLE OF CONTENTS

Page

Abstract

Acknowledgements and Disclaimer iii

Table of Contents

List of Figures ix

List of Tables ix

Program Theory and Techniques 1

Introduction 3Applications of Cross-Section Data Analysis 3Features of XSPRO 3

Technical Background 5General Assumptions and Limitations 5Resistance Equations 7

Manning's Equation 7Resistance Equations Suggested by Thorne and Zevenbergen 12Jarrett's Equation for Manning's Roughness Coefficient 13

Subdividing Cross Sections 14

Field Procedures and Techniques 15Reach Selection 15Field Procedures 15

Survey of Cross Section and Water-Surface Slope 16Bed-Material Particle-Size Distribution 17Discharge Measurement 17

Program Operations 19

Introduction to XSPRO 21Getting Started 21

Requirements 21Program Installation 21

Running XSPRO: An Overview 22Input 22Running the Analysis 23Printing Results 23

Page

XSPRO Basic Procedures 25Special Keys 25Help 25Menus 25Error Correction 26

Error in Stage/Slope/Interval Data 26Error in Section Boundary Data 26Error in Manning's n Data 26

XSPRO Input 27Data Entry 27

Input From the Keyboard 27Input From a File 27

Controlling Parameters 27Parameter Editing 28Input Parameters 28Output Parameters 30Analysis Parameters 30

Running an Analysis 33Cross-Section Specifics 33

Stage Data 34Section Boundaries 35Manning's n Values 36

Tabular Hydraulic Output 36Regression Analysis 36

XSPRO Output 39Output File 39Printing 39

Data From the Output File 39Printing Graphics Screens 40

Putting XSPRO to Work 41Examples Using Sample Data 41

Quick Run 41Changing the Controlling Parameters 43

Specific Uses 44Analyzing Geometry Changes From Year to Year 44Entering Data With the Keyboard 46Saving Keyed In Data From the Main Menu 47

Literature Cited 49

Page

Appendix A - File Formats 51Input Files 51

Position-Elevation Free Form 51Elevation-Position Free Form 51User Defined 51

Output Files 51

Appendix B - Importing XSPRO Output into Lotus 1-2-3 53

LIST OF FIGURES

Page

Figure 1. Definition diagram for hydraulic parameters 5

Figure 2. Some typical channel configurations that disrupt uniform flow 6

Figure 3. Diagram of longitudinal profile and plan view of a pool-riffle sequence 16

Figure 4. Main screen 22

Figure 5. Keyboard entry screen 23

Figure 6. Menu selection fields 28

Figure 7. User defined data format screen 29

Figure 8. Sample input file 30

Figure 9. Sample cross-section graph 33

Figure 10. Stage, slope, increment entry 34

Figure 11. Section boundary selection 35

Figure 12. Sample regression graph 37

Figure 13. Print setup screen 39

Figure 14. Main screen 41

Figure 15. Sample data cross-section graph 42

Figure 16. Menu selection fields 44

Figure 17. Keyboard input 46

LIST OF TABLES

Page

Table 1. Factors that affect roughness of the channel(modified from Aldridge and Garrett, 1973, table 2) 9

PROGRAMTHEORY

ANDTECHNIQUES

INTRODUCTION

Surveys of stream-channel cross sections pro-vide important information for hydrologists, riverengineers, geomorphologists, fishery biologists,and other professionals associated with river man-agement issues. Data from cross-section surveysare useful for analyzing channel form and func-

tion, and the processes that influence a channel'shydrologic performance. Use of survey data toconstruct relationships between streamflow, chan-nel geometry, and various hydraulic characteris-tics provides information that serves a variety ofapplications.

APPLICATIONS OF CROSS-SECTION DATA ANALYSIS

Information on stream-channel geometry andhydraulic characteristics is useful for channel de-sign, restoration of riparian areas, and placementof instream structures. The analysis of cross-section hydraulics, along with an evaluation offlood frequency, is a primary consideration inchannel design. Once a recurrence interval isdefined for bank-full flow, the channel is designedto contain that flow, and higher flows are allowedto spread over the floodplain. Such periodicflooding is extremely important for the formationof channel macrofeatures (e.g., point bars andmeander bends) and for establishment of certainkinds of riparian vegetation. A cross-section analy-sis may also help optimize placement of suchitems as culverts and fish habitat structures.

Additionally, knowledge of the relationshipsbetween discharge and channel geometry and hy-draulics is useful for reconstructing the conditionsassociated with a particular flow situation. Forexample, in many channel-stability analyses, it is

customary to relate movement of substrate mate-rials to some measure of stream power or averagebed shear stress. If the relations betweenstreamflow and certain hydraulic variables (e.g.,mean depth and water-surface slope) are known, itis possible to estimate stream power and averagebed shear at any given level of flow. Thus, achannel cross-section analysis makes it possible toestimate conditions of substrate movement at vari-ous levels of streamflow.

Finally, cross-section analyses provide impor-tant information for instream flow assessments.Various resource values may be altered by changesin hydraulic parameters associated with changesin streamflow. For example, the relation betweenlow-water discharge and channel wetted perim-eter may be an important consideration formacroinvertebrate production or the scenic enjoy-ment of a stream. Similarly, cross-section datamay be used to define the depth-discharge rela-tionship for analysis of fish habitat.

FEATURES OF XSPRO

Computer programs can be useful tools inchannel cross-section analysis. One such programis XSPRO. The XSPRO program is designed foranalyzing channel cross-section data in an interac-tive, user-friendly environment. The program ismenu-driven with easy-to-read input and outputscreens, and an integrated graphics package tofacilitate data entry. XSPRO uses a resistance-equation approach to single cross-section hydrau-

lic analysis (e.g., Manning's equation), and iscapable of analyzing both the geometry and hy-draulics of a given channel cross section. XSPROwas specifically developed for use in high-gradient streams and supports three alternativeresistance equations for handling boundary rough-ness and resistance to flow. The program allowsthe user to subdivide the channel cross section sothat overbank areas, mid-channel islands, and high-

3

water overflow channels may be analyzed sepa- water-surface slopes so that slope may be variedrately. The program also allows input of variable with discharge to reflect natural conditions.

TECHNICAL BACKGROUND

The theoretical background for analyzing chan-nel cross-section data is derived from the basiccontinuity, momentum, and energy equations offluid mechanics. Specifically, streamflow at across section is computed using the simplifiedform of the continuity equation where dischargeequals the product of velocity and cross-sectionalarea of flow. Computation of cross-sectional areais strictly a geometry problem; it is determined byinputting incremental depths of water (stage) to achannel cross section defined by surveyed dis-tance-elevation pairs. In addition to cross-sectional area, the top width, wetted perimeter,mean depth, and hydraulic radius are computedfor each increment of stage (Figure 1).

Once the channel geometry has been com-puted for a given stage, an estimate of mean cross-section velocity is needed to produce an estimateof streamflow. Analysis of the momentum andenergy equations requires that, under certain con-ditions of streamflow, gravitational forces thatcause water to move downhill are balanced byfrictional forces at the channel boundary that tendto resist the downhill flow. Under these conditionsit is possible to estimate resistance to flow and,hence, mean velocity at the channel cross section.Thus, various resistance equations have beendeveloped for estimating mean velocity as a func-tion of cross-section hydraulic parameters.

Figure 1. Definition diagram for hydraulic parameters.

GENERAL ASSUMPTIONS AND LIMITATIONS

As indicated, the mean velocity of streamflowin a cross section may be estimated when certainflow conditions are met. The main criteria forthese flow conditions is that the bed slope, thewater-surface slope, and the total energy gradeline are parallel. The total energy of the stream isa function of the position of the streambed abovesome arbitrary datum (potential energy), the depth

of the water column (pressure energy), and thevelocity of the water column (kinetic energy). Thetotal energy grade line represents the rate at whichenergy is dissipated through turbulence and bound-ary friction. When the slope of the energy gradeline is known, the various resistance formulasallow computation of mean cross-sectional veloc-ity. When the water-surface slope and the energy

grade line parallel the streambed, the energy gradeline is estimated with the water surface slope.

Under conditions of constant width, depth,area, and velocity, the water surface slope andenergy grade line approach the slope of thestreambed, producing a condition known as "uni-form flow." One feature of uniform flow is that thestreamlines (the traces of the path that a particle ofwater would follow in the flow) are parallel andstraight (Roberson and Crowe 1985). Perfectlyuniform flow is rarely realized in natural channels,but the condition is approached in some reacheswhere the geometry of the channel cross section isrelatively constant throughout the reach.

Conditions that tend to disrupt uniform flow in-clude bends in the stream course; changes in cross-section geometry; obstructions to flow caused bylarge roughness elements, such as channel bars,large boulders, • and woody debris; or otherfeatures that cause convergence, divergence, ac-celeration, or deceleration of flow (Figure 2).Resistance equations also may be used to evaluatethese nonuniform flow conditions (graduallyvaried flow); however, energy-transition consid-erations (backwater calculations) must then befactored into the analysis. This requires the use ofmultiple transect models (e.g., HEC-2).

Figure 2. Some typical channel configurations that disrupt uniform flow.

RESISTANCE EQUATIONS

XSPRO supports three sets of resistanceequations for estimating mean velocity at a crosssection. Each equation or set of equations wasdeveloped from specific sets of data; therefore,use of a particular resistance formula to estimatevelocity is subject to the limitations of the dataused to develop that formula, as well as the as-sumptions of the formula itself. Also, becauseeach resistance equation estimates channel resis-tance or roughness in a slightly different way, thedifferent formulas may require different inputsfrom the user and likely will produce somewhatdifferent results. Selection of the appropriateresistance equation requires understanding theassumptions and limitations in each approach.

Manning's Equation

XSPRO supports the use of Manning's equa-tion for estimating mean cross-section velocity.Manning's equation was developed for conditionsof uniform flow in which the water-surface profileand energy gradient are parallel to the streambed,and the area, hydraulic radius, and depth remainconstant throughout the reach. Lacking a bettersolution, it is assumed that the equation is alsovalid for nonuniform reaches that are invariablyencountered in natural channels, if the energygradient is modified to reflect only the losses dueto boundary friction (Dalrymple and Benson 1967).The Manning equation for mean velocity is givenas:

V = kn R 2/3 S '/2 (1)

where: k = 1 for metric units and 1.486 forEnglish units,

n = Manning's roughness coefficient,R = hydraulic radius, andS = energy slope (water-surface slope).

In Manning's equation, resistance to flow dueto friction at the channel boundary is addressedthrough the use of a roughness coefficient sup-plied by the user of the equation. The roughness

coefficient may be thought of as an index of thefeatures of channel roughness that contribute tothe dissipation of stream energy.

There are three methods for estimatingManning's roughness coefficient for natural chan-nels: direct solution of Manning's equation for n,comparison with computed n values for otherchannels, and formulas relating n to other hydrau-lic parameters. Each method has its own limita-tions and advantages.

The method of direct solution entails measur-ing stream discharge and dividing by the cross-sectional area of flow to obtain a mean velocity.The mean velocity, hydraulic radius (roughly equalto mean depth for wide channels), and water-surface slope may be entered into Manning'sequation, and the equation solved directly for theroughness coefficient, n. This approach gives anestimate of n that is as accurate as the uncertaintyassociated with the measurement of discharge,cross-sectional area, and water-surface slope, butthe n value obtained is only applicable to theparticular stage at which the flow was measured.

Since the features of channel roughness thatcontribute to energy dissipation will vary withwater level, n also will vary with water level;therefore, it is desirable to directly estimate n atmore than one level of streamflow. Most authorscited have found that n values decrease with in-creasing stage, at least up to bank-full flow. Ifstreamflow can be measured at several differentstages, n may be calculated for a range of flows,and the relationship between n and stage deter-mined.

The second method for estimating n values ata cross section involves comparing the reach to asimilar, measured reach for which Manning's nhas already been computed. This is probably thequickest and most commonly used procedure forestimating Manning's n and is usually done fromeither a table of values or by comparison withphotographs of natural channels. Tables ofManning's n values for a variety of natural andartificial channels are common in the literature onhydrology (e.g., Chow 1959; Van Haveren 1986),

while photographs of stream reaches with com-puted n values have been compiled by Chow(1959) and Barnes (1967).

When the roughness coefficient is estimatedfrom table values or by comparison with photo-graphs of natural channels with known n, thechosen n value (nb) is considered a base value thatmay need to be adjusted for local channel condi-tions. Several publications provide procedures foradjusting base values of n to account for channelirregularities, vegetation, obstructions, and sinu-osity (Chow 1959; Benson and Dalrymple 1967;Arcement and Schneider 1984; Parsons and Hudson1985). The most common procedure uses theformula proposed by Cowan (1956) to estimatethe value of n:

n = (nb + n, + n2 + n, + n4)m (2)

where: nb = base value of n for a straight,uniform, smooth channel innatural materials,

n = correction for the effect of surfaceirregularities,

n2 = correction for variations incross-section size and shape,

n3 = correction for obstructions,n4 = correction for vegetation and flow

conditions, andm = correction for degree of channel

meandering.

Table 1 is taken from Aldridge and Garrett(1973) and may be used to estimate each of theabove correction factors to produce a finalestimated n.

While estimating Manning's roughness coef-ficient from a table of values or by comparisonwith photographs of channels with known n is thequickest and most commonly used method, mostexperienced hydrologists and river engineers sim-ply estimate n from experience — often the tablesand photographs are not even consulted. For thatreason, this method is subject to the most variabil-ity between individuals, and is probably the leastaccurate method for arriving at Manning' s n. Also,the method ordinarily is used to produce a single

value for roughness, which is then applied through-out the entire range of flow, often introducinglarge errors in the estimate of high or low flows.

The third method of determining Manning'sroughness coefficient for a cross section usesempirical formulas relating n to other hydraulicparameters. Many of these formulas assume thatthe larger particles in the channel boundary domi-nate the hydraulic roughness; hence, the empiricalrelationships usually correlate n with some statis-tical index of bed-material size distribution. Assuch, these formulas are insensitive to changes indepth of flow at the cross section. To compensatefor changes in roughness with changes in depth offlow, several formulas have been developed thatuse a relative roughness term relating some repre-sentative particle size (e.g., the 84th-percentileparticle size) to the hydraulic radius or meandepth. Changes in depth of flow therefore changeroughness and n values. Also, some empiricalformulas, such as the Jarrett (1984) formula de-scribed below, do not use particle size at all, butrelate the roughness coefficient n to other hydrau-lic parameters, such as slope and hydraulic radius.

Just as Manning's n may vary significantlywith changes in stage (water level), channel ir-regularities, obstructions, vegetation, sinuosity,and bed-material size distribution, n may also varywith bed forms in the channel. The hydraulics ofsand and mobile-bed channels produce changes inbed forms as the velocity, stream power, andFroude number increase with discharge. As ve-locity and stream power increase, bed forms evolvefrom ripples to dunes, to washed-out dunes, toplane bed, to antidunes, to chutes and pools. Ripplesand dunes occur when the Froude number is lessthan 1 (subcritical flow); washed out dunes occurat a Froude number equal to 1 (critical flow); andplane bed, antidunes, and chutes and pools occurat a Froude number greater than 1 (supercriticalflow). Manning's n attains maximum values whendune bed forms are present, and minimum valueswhen ripples and plane bed forms are present(Parsons and Hudson 1985).

Because Manning's roughness coefficientvaries with different flows and cross-sectioncharacteristics, it is important to define the vari-

ability of n over the entire range of flows whendoing cross-section analysis. If possible, this isbest accomplished by measuring discharge at sev-eral different water levels (stages), solvingManning's equation for the true value of n at eachstage, and developing a relationship between stageand Manning's n. If Manning's n is estimatedfrom a table of values or by comparison with

photographs, estimates should be made for severalstages, and the relationship between n and stagedefined for the range of flows of interest. Ifempirical formulas are used to estimate n, it is bestto select a formula that is sensitive to mean depthor hydraulic radius, such as the formulas that usea relative roughness term.

Table 1. Factors that affect roughness of the channel (modified from Aldridge and Garrett,1973, table 2).

Channelconditions

n valueadjustment "

Example

Degree ofirregularity (ni)

Smooth 0.000 Compares to the smoothest channel attainablein a given bed material.

Minor 0.001-0.005 Compares to carefully dredged channels ingood condition but having slightly eroded orscoured side slopes.

Moderate 0.006-0.010 Compares to dredged channels having moder-ate to considerable bed roughness and moder-ately sloughed or eroded side slopes.

Severe 0.011-0.020 Badly sloughed or scalloped banks of naturalstreams; badly eroded or sloughed sides ofcanals or drainage channels; unshaped, jagged,and irregular surfaces of channels in rock.

Variation inchannel crosssection (n2)

Gradual 0.000 Size and shape of channel cross sectionschange gradually.

Alternatingoccasionally

0.001-0.005 Large and small cross sections alternateoccasionally, or the main flow occasionallyshifts from side to side owing to changes incross-sectional shape.

Alternatingfrequently

0.010-0.015 Large and small cross sections alternatefrequently, or the main flow frequently shiftsfrom side to side owing to changes in cross-sectional shape.

" Adjustments for degree of irregularity, variations in cross section, effect of obstructions, and vegetation are added tothe base n value before multiplying by the adjustment for meander.

9

Table 1. Factors that affect roughness of the channel — continued.

Channelconditions

n valueadjustment "

Example.

Effect ofobstruction (n3)

Negligible 0.000-0.004 A few scattered obstructions, which includedebris deposits, stumps, exposed roots, logs,piers, or isolated boulders, that occupy lessthan 5 percent of the cross-sectional area.

Minor 0.005-0.015 Obstructions occupy less than 15 percent ofthe cross-sectional area and the spacingbetween obstructions is such that the sphere ofinfluence around one obstruction does notextend to the sphere of influence aroundanother obstruction. Smaller adjustments areused for curved smooth-surfaced objects thanare used for sharp-edged angular objects.

Appreciable 0.020-0.030 Obstructions occupy from 15 to 20 percent ofthe cross-sectional area or the space betweenobstructions is small enough to cause theeffects of several obstructions to be additive,thereby blocking an equivalent part of a crosssection.

Severe 0.040-0.050 Obstructions occupy more than 50 percent ofthe cross-sectional area or the space betweenobstructions is small enough to cause turbu-lence across most of the cross section.

Amount ofvegetation (n4)

Small 0.002-0.010 Dense growths of flexible turf grass, such asBermuda, or weeds growing where the aver-age depth of flow is at least two times theheight of the vegetation; supple tree seedlingssuch as willow, cottonwood, arrowweed, orsaltcedar growing where the average depth offlow is at least three times the height of thevegetation.

"Adjustments for degree of irregularity, variations in cross section, effect of obstructions, and vegetation are added tothe base n value before multiplying by the adjustment for meander.

10

Table 1. Factors that affect roughness of the channel — continued.

Channelconditions

n valueadjustment I/

Example

Amount of .vegetation (n4)—(continued)

,

Medium 0.010-0.025 Turf grass growing where the average depth offlow is from one to two times the height of thevegetation; moderately dense stemmy grass,weeds, or tree seedlings growing where theaverage depth of flow is from two to threetimes the height of the vegetation; brushy,moderately dense vegetation, similar to 1- to2-year-old willow trees in the dormant season,growing along the banks and no significantvegetation along the channel bottoms wherethe hydraulic radius exceeds 2 feet.

Large 0.025-0.050 Turf grass growing where the average depth offlow is about equal to the height of vegetation;8- to 10-year-old willow or cottonwood treesintergrown with some weeds and brush (noneof the vegetation in foliage) where the hydrau-lic radius exceeds 2 feet; bushy willows about1 year old intergrown with some weeds alongside slopes (all vegetation in full foliage) andno significant vegetation along channel bottoms where the hydraulic radius is greater than2 feet.

Very Large 0.050-0.100 Turf grass growing where the average depth offlow is less than half the height of the vegeta-tion; bushy willow trees about 1 year oldintergrown with weeds along side slopes (allvegetation in full foliage) or dense cattailsgrowing along channel bottom; treesintergrown with weeds and brush (all vegeta-tion in full foliage).

" Adjustments for degree of irregularity, variations in cross section, effect of obstructions, and vegetation are added tothe base n value before multiplying by the adjustment for meander.

Table 1. Factors that affect roughness of the channel — continued.

Channelconditions

n valueadjustment if

Example

Degree of mean-dering v (Adjust-ment valuesapply to flowconfined in thechannel and donot apply wheredownvalley flowcrosses mean-ders.) (m)

Minor 1.00 Ratio of the channel length to valley length is1.0 to 1.2.

Appreciable 1.15 Ratio of the channel length to valley length is1.2 to 1.5.

Severe 1.30 Ratio of the channel length to valley length isgreater than 1.5.

If Adjustments for degree of irregularity, variations in cross section, effect of obstructions, and vegetation are added tothe base n value before multiplying by the adjustment for meander.

Resistance Equations Suggested byThorne and Zevenbergen

Resistance equations that include a term forrelative roughness have an inherent sensitivity tochanges in depth because the ratio of mean depth(or hydraulic radius RD to large-element particlesize is incorporated into the relative-roughnessterm. Thorne and Zevenbergen (1985), in a reviewof resistance equations developed for mountainstreams, tested several formulas using relative-roughness terms for estimating mean velocity in

steep, cobble/boulder-bed channels. For smallrelative-roughness values (i.e., R/d m greater than1), Thome and Zevenbergen recommended anequation developed by Hey (1979) for estimatingmean cross-section velocity:

= 5.62 log ( 49113(gRS)Y2 35 d84)

R r314

S = energy slope (water-surface slopein uniform flow),

dm = intermediate diameter for the 84th-percentile particle size, and

D = maximum depth at section.

Similarly, for large relative-roughness values(i.e., R/dM less than or equal to 1), Thorne andZevenbergen recommended Bathurst' s (1978)equation for estimating mean cross-sectionvelocity:

(gRS)/2 = 0.365 dm D

R )2.34 ( vvy ( AE -0.08)

AE = 0.039 - 0.139 log (ci-4)

where: 15 = mean depth,W = water surface width, and

all other variables are aspreviously defined.

XSPRO supports these formulas as an optionfor calculating mean cross-section velocity. How-ever, in applying these formulas to a cross-sectionanalysis, the assumptions of the equations must beconsidered, i.e., that channel gradients generally

ai = 11.1max

where: U = mean cross-section velocity,g = acceleration due to gravity,R = hydraulic radius,

12

exceed 1 percent, channel beds are predominatelycobble and boulder substrate, and relative rough-ness is large. Thorne and Zevenbergen (1985)reported average errors of only 6 percent whenusing the Hey equation for small values of relativeroughness (i.e., R/d84>>1), but even the best equa-tions overpredicted mean velocity by as much as30 percent for the highest values of relative rough-ness (i.e., R/d84<1 ), an error they attributed todifficulties in measuring bed-material sizes.

Jarrett's Equation for Manning'sRoughness Coefficient

The previous discussion of Manning's equa-tion alluded to the existence of empirical formulasfor n that do not make use of particle-size data asan index of relative roughness. These formulastend to relate the roughness coefficient to otherhydraulic parameters. Jarrett (1984) developedthe following equation for n, relating the rough-ness coefficient to water-surface slope and hy-draulic radius at the section:

n = 0.39 Sa" R-"6 (7)

Jarrett's equation for n has no explicit term forrelative roughness; however, he reported a posi-tive correlation between water-surface slope andcoarse bed-material particle size. Thus, althoughparticle size is not an explicit part of the equation,it is still implicit in the slope term. Jarrett alsoreported a slightly stronger correlation betweenManning's n and slope than the correlation be-tween n and dm particle size.

Jarrett (1984) also compared n values calcu-lated with the above equation to actual n valuesobtained from cross sections with measured hy-draulic-geometry and flow data. The averagestandard error of the estimated n values was 28percent, and ranged from -24 percent to +32percent. Jarrett found the equation to slightly

overestimate n, with the greatest errors typicallyassociated with low-flow measurements when theratio of R/d50 is less than 7.

XSPRO supports the use of Jarrett's equationfor estimating Manning's roughness coefficientand mean cross-section velocity. Again, the limi-tations of the data from which the equation wasdeveloped should be considered when doing across-section analysis. Specifically, the equationis limited to the following conditions:

Natural channels having stable bed andbank materials (gravels, cobbles, andboulders),Water-surface slopes between 0.2 and4.0 percent,Hydraulic radii from 0.5 to 7.0 feet,Cross sections unaffected by down-stream obstructions (i.e., no backwater),and

5. Streams having relatively small amountsof suspended sediment.

Because Jarrett's equation includes hydraulicradius as a parameter for estimating Manning's n,it is inherently sensitive to changes in depth. Thenegative coefficient associated with the hydrau-lic-radius term indicates diminishing resistancewith increasing depth. However, the relativelylow value of this coefficient (n is only sensitive tothe 1/6th power of hydraulic radius) means that nwill change only slightly through the normal rangeof stage at a section. An independent evaluation ofthe equation on Idaho mountain streams con-firmed this (Potyondy 1990). Jarrett's n appearedto fit the measured data best at flows at or abovebank-full stage; the poorest fits occurred at lowflow. Potyondy concluded that the equation wasbest applied to bank-full flow estimates, with low-water n values supplied from field measurementsof hydraulic geometry and discharge.

13

SUBDIVIDING CROSS SECTIONSNatural channel cross sections are rarely per-

fectly uniform, and it may be necessary to analyzehydraulics in a very irregular cross section. Fre-quently, high-gradient streams have overflow chan-nels on one or both sides that carry water onlyduring unusual high-flow events. Even in chan-nels of fairly regular cross section, overbank areastransport water at discharges above bank-full.These areas usually have hydraulic properties sig-nificantly different from those of the main chan-nel. Generally overflow channels and overbankareas are treated as separate subchannels, and thedischarge computed for each of these subsectionsis added to the main channel to compute totaldischarge.

When subdividing a channel cross section intomain channel, side channels, and overbank areas,XSPRO assumes frictionless vertical divisions("smooth glass walls") between individual sub-sections. The assumption of negligible shearbetween subsections avoids the formidable task ofestimating small energy losses due to frictionbetween adjacent, moving bodies of water. XSPROalso assumes that flow can access each subsectionas the stage reaches the lowest elevation of thatsubsection, that is, the overflow channel oroverbank area is not blocked off from the flow atsome upstream location.

14

FIELD PROCEDURES AND TECHNIQUES

A good cross-section analysis depends on goodfield data, which requires careful reach selectionand proper field techniques. Whether a critical orrepresentative reach is to be analyzed must bedetermined, and the uniform flow assumptions of

Manning's equation must be considered in reachselection. Proper field techniques must be fol-lowed in survey procedures, particle-size determi-nations, and streamflow measurements.

REACH SELECTION

The intended use of the cross-section analysisplays a large role in locating the reach and the crosssection. The user must decide whether the sectionis to be located in a critical reach or in a reach thatis considered representative of some larger area.The reach most sensitive to change or most likelyto meet (or fail to meet) some important conditionmay be considered a critical reach. A representa-tive reach will typify a definable portion of thechannel system and will be used to describe thatportion of the system (Parsons and Hudson 1985).

Once a reach has been selected, the channelcross section is sited in the location consideredmost suitable for meeting the uniform flow re-quirements of Manning's equation. The uniformflow requirement is approached where width,depth, and cross-sectional area of flow remainrelatively constant, and the water-surface slopeand energy grade line approach the slope of thestreambed. For this reason, marked changes inchannel geometry and discontinuities in the flow(steps, falls, and hydraulic jumps) should beavoided. Generally, the section should be locatedwhere it appears the streamlines are parallel to thebank and each other.

Straight channel reaches with perfectly uni-form flow are rare in nature, and in most cases,uniform flow is only approached to varying de-grees. If a reach with constant cross-sectional areaand shape is not available, a slightly contractingreach is acceptable, provided that there is nosignificant backwater effect from the constriction.Backwater occurs where the stage-discharge re-lationship is controlled by the geometry of a singlecross section or a break in bed slope a shortdistance downstream of the area of interest (sec-tion control). Manning's equation assumes thestage-discharge relationship is controlled by thegeometry and roughness of a long reach of channeldownstream of the section (channel control); thus,Manning's equation will not produce an accuratestage-discharge relationship in pools or otherbackwater areas. In addition, expanding reachesalso should be avoided, as there are additionalenergy losses associated with channel expansions.When no channel reaches are available that meetor approach the condition of uniform flow, it maybe necessary to use multitransect models (e.g.,HEC-2) to analyze cross-section hydraulics.

FIELD PROCEDURES

The basic information to be collected in thereach selected for analysis is a survey of thechannel cross section and water-surface slope, a

measurement of bed-material particle-size distri-bution, and a discharge measurement.

5

Intermediateflow

Figure 3. Diagram of longitudinal profile and plan view of a pool-riffle sequence. Water surface profilesin upper figure represent high, intermediate, and low flow conditions.

Survey of Cross Section and Water-Surface Slope

The basic data required for a channel cross-section analysis are a surveyed channel cross sec-tion and water-surface slope. The cross section isestablished perpendicular to the channel, and thepoints across the section are surveyed relative to aknown or arbitrarily established benchmark el-evation. The distance-elevation paired data asso-ciated with each point on the section may beobtained either by sag-tape or rod-and-level sur-vey. The intricacies of correct survey proceduresare beyond the scope of this document. For detailsof the sag-tape procedure, the reader is referred toRay and Megahan (1979). Benson and Dalrymple(1967) present an excellent overview of rod-and-level surveying procedures, including guidanceon equipment, field notes, and vertical and hori-zontal control.

Information on water-surface slope also isrequired input for a cross-section analysis. Thesurvey of water-surface slope is somewhat more

complicated than the cross-section survey in thatslope of the individual channel unit at the locationof the section (e.g., pool, run, or riffle) must bedistinguished from the more constant slope of theentire reach. (See Grant et al. 1990 for a detaileddiscussion on recognition and characteristics ofchannel units.) Water-surface slope in individualchannel units may change significantly withchanges in stage and discharge (Figure 3), whilethe slope of the entire reach will remain essentiallyunchanged. Thus, at low flow, the slope of theindividual channel unit will have a strong influ-ence on the stage-discharge relationship, while athigh water, the average slope of the reach willcontrol the stage-discharge rating. This is animportant distinction for the XSPRO software,which allows the user to specify different slopesfor high- and low-water stages. For this reason,when water-surface slopes are surveyed in thefield, low-water slope may be approximated bythe change in elevation over the individual chan-nel unit where the cross section is located (ap-proximately 1 to 5 channel widths in length), while

as described above. Additional guidance for bed-material sampling in coarse-bed streams is pro-vided in Yuzyk (1986) and Church et al. (1987).

high-water slope is obtained by measuring thechange in elevation over a much longer reach ofchannel (usually at least 15 to 20 channel widths inlength).

Bed-Material Particle-SizeDistribution

Computing mean velocity with resistanceequations based on relative roughness, such as theones suggested by Thorne and Zevenbergen (1985),requires an evaluation of the particle-size distribu-tion of the bed material of the stream. For streamswith no significant channel armor and bed mate-rial finer than medium gravel, bed-material sam-plers developed by the Federal Inter-agencySedimentation Project (FISP 1986) may be used toobtain a representative sample of the streambed,which is then passed through a set of standardsieves to determine percent-by-weight of particlesof various sizes. The cumulative percent of mate-rial finer than a given size may then be determined.Particle-size data are usually reported in terms ofdi, where i represents some nominal percentile ofthe distribution and d i represents the particle size,usually expressed in millimeters, at which i per-cent of the total sample is finer. For example, 84percent of the total sample would be finer than thedM particle size. For additional guidance on bed-material sampling in sand-bed streams, the readeris referred to Ashmore et al. (1988).

XSPRO supports resistance equations forestimating velocity in steep mountain rivers withsubstrate much coarser than the medium-gravellimitation of FISP samplers. For these streams, themethod used to measure substrate particle size is apebble count (Wolman 1954), in which at least100 bed-material particles are manually collectedfrom the streambed and measured. A grid patternof sampling points is paced or staked along thestream, and at each sample point, a particle isretrieved from the bed and the intermediate axis(not the longest or shortest axis) is measured. Themeasurements are tabulated as to number of par-ticles occurring within predetermined size inter-vals, and the percentage of the total in each intervalis then determined. Again, the percentage in eachinterval is accumulated to give a particle-sizedistribution, and the particle -size data are reported

Discharge Measurement

When analyzing channel cross-section data, itis desirable to have at least one good measurementof discharge at the section. Although a dischargeestimate is not required to run XSPRO, the avail-ability of streamflow.data will greatly improve thequality of the input and provide a good check onthe accuracy of the output (i.e., data on stage,discharge, and other hydraulic parameters). Ifonly one discharge measurement is obtained, itlikely will occur during low water and will beuseful for defining the lower end of the ratingtable. If two measurements can be made, it isdesirable to have a low-water measurement and ahigh-water measurement to define both ends of therating table and to establish the relationship be-tween Manning's n and stage. If high water cannotbe measured directly, it may be necessary to esti-mate the high-water n using the Jarrett formula(Jarrett 1984) or the resistance equations recom-mended by Thorne and Zevenbergen (1985). Ifseveral discharge measurements can be made overa wide range of flows, relations between stage,discharge, and other hydraulic parameters may bedeveloped directly without the use of XSPRO.

It is beyond the scope of this document todiscuss all the intricacies of correct streamflow-measurement techniques. The reader is referred toBuchanan and Somers (1969), and Rantz andothers (1982) for an in-depth treatment of thissubject. Also, Smoot and Novak (1968) presentprocedures for calibration and maintenance ofcurrent meters to ensure accurate measurement ofvelocity and discharge. When equipment is func-tioning properly and standard procedures are fol-lowed correctly, it is possible to measurestreamflow to within 5 percent of the true value.The data gathered from a standard discharge mea-surement also include information on top widthand cross-sectional area, from which mean veloc-ity and mean depth may be computed. Thisinformation is extremely useful for improvingquality of output from any channel cross-sectionanalysis program.

17

PROGRAMOPERATIONS

INTRODUCTION TO XSPRO

GETTING STARTED

XSPRO is a user friendly program for analyz-ing cross-sections of small mountain streams.

Requirements

XSPRO can be run on any 100 percent IBM"PC/XT/AT compatible computer with the follow-ing requirements:

DOS 2.1 or higher256K of RAMEGA, VGA, Hercules,or AT&T 400 graphicsOne floppy drive

XSPRO can be run directly off of your distri-bution diskettes or you can copy the files to adirectory on your hard disk. You will notice asignificant increase in performance if the programis run from a hard disk.

Program Installation

You should first make a copy of the XSPROdistribution disk. The procedure will be differentfor one and two floppy systems.

One Floppy Systems

If you have only one floppy drive, place thedistribution disk in drive A and type:

DISKCOPY A: B: <ENTER>

When the computer prompts you for thediskette for drive B, remove your distribution diskand place your backup disk in drive A. Yourcomputer is now using your floppy drive as drive

" Any use of trade, product, or firm names in this publica-tion is for descriptive purposes only and does not implyendorsement by the U.S. Government.

B. When the copying is done, type:

A: <ENTER>

The computer will prompt you for the diskettefor drive A, press ENTER and your system will beback to normal with your floppy acting as drive A.

Two Floppy and Hard Disk Systems

To install XSPRO on either a two floppy orhard drive system, you need to copy all of the filesto either a second floppy or your hard drive. Abatch file (INSTALL.BAT) has been included tomake this easier. To use the batch file to install theprogram, insert the disk in the A drive and type:

INSTALL <D1:> <D2:\DIRECTORY><ENTER>

replacing <D1:> with the source drive and<D2:\DIRECTORY> with the drive name anddirectory of where you want to install the files.The install program will create the directory if itdoes not already exist and then copy all of thefiles to it. For example, to install XSPRO fromfloppy drive A to hard drive C in the directoryXSPRO type:

INSTALL A: C:\XSPRO <ENTER>

More information can be found by typing:

TYPE README.1ST <ENTER>

21

ross-Section Protestienal AnalyzerRun

4.,S_OurCeFilanarne:Data dollactiOData rdi►at:Units

01.1tPt#'..,File outputUnits:: liters

qe

RUNNING XSPRO: AN OVERVIEW

This section is meant to be an overview of theXSPRO program. All of the software elementsdescribed here are explained more completely inlater chapters. This section is not meant to be asample run or example. For more complete ex-amples see the section on "Putting XSPRO toWork."

To run XSPRO, be sure you have a DOScommand prompt and that the XSPRO files are inthe current drive and directory. To start the pro-gram type:

XSPRO <ENTER>

Press ENTER after viewing the title screen,and you should see the XSPRO opening screen(Figure 4). The opening screen displays a menuwindow across the top supplemented by a secondwindow displaying all of the current controllingparameters. The menu is active upon startup and

Figure 4. Main screen.

is controlled in the usual manner by using theARROW keys to highlight a selection and press-ing the ENTER key to activate that selection.Furthermore, you may activate a menu selectionby pressing the key corresponding to the first letterof that selection (i.e., press R to activate the Runselection). The controls are the same for thesubmenus.

Context-sensitive help is available at any pointin the program by pressing F 1 .

The first time XSPRO is run, it will bring upthe default values for all of the controlling param-eters. After the first time, the most recently se-lected controlling parameters will be rememberedand used on the next run of the program. This willmake it easy to compare formulas, data sets, or anyparameter from one run to the next. If these dataget corrupted for any reason, the program will startup with the hard coded defaults again and willwarn you that it is doing so.

Input

The first step is to match the controlling pa-rameters with the data being used and the outputdesired. Keep in mind that the program will startup with the parameters selected on the last run ofthe program or the hard coded defaults. An inter-active input process is provided to edit all control-ling parameters and input data. Selecting

Modify from the main menu brings upa submenu containing the Parametersand Data options.

Modify-Parameters

Selecting the Parameters submenuallows editing of the fields in the con-trolling parameters window.

The parameters are edited in twoways. The first method is to simplytype the appropriate entry into the field.The other method is a menu selectedinput and is used when there are a finiteset of responses. Highlight the fieldyou wish to change and pressSPACEBAR to see the selection menu.These menus work like all other menusin XSPRO.

Modify-Data

Selecting the Data submenu brings up anentry window for entering or editing cross-section data. You will specify the number ofpoints to use and then enter the distance-eleva-tion pairs. The manually entered data can be

controlling paramehKeyboard Data Enby/r7dif

Enter 1henomber of so rvitt!tf points< <

F2 to Delete last pOint:.,r3 to Add a pointPGDN for next 10 00166poup . fOr prey 10 points

Data units: feetANAL" Stable 5Pt: 410 (Not De

Survey date Nyinirnidd): $1/10124

INPUSour

. DataData:-.Unit

iotitPrilenUnit

POiddistance elev

4- 6.37.6

"5.4

' 4 - 4.28 ' 2.1

6 10 4.8

8 14 7.3

Figure 5. Keyboard entry sreen.

saved to a file in position-elevationformat. Figure 5 shows the keyboarddata entry screen with a sample set ofdata.

Running the Analysis

The default values that will appearon the first run of XSPRO are enoughto run an analysis of a sample data filethat is included with your package.Selecting Run from the main menustarts the analysis process. This is theheart of the XSPRO program. Here theprogram brings together all of the in-formation in the parameters windowand analyzes the cross section you havedefined. The first screen you see will be a graph ofthe cross section. This portion of the program isalso interactive. You will be asked to input thelow- and high-flow stages and correspondingslopes, and the increment to use to analyze thecross section. You will also be asked to selectsubsections along the cross section that should beanalyzed separately. Furthermore, if you chooseto supply your own Manning's n, you will enterthat data here. The low- and high-flow stages andthe section boundaries you select will be displayedon the screen.

(For a quick run of the sample data, enter 0.0for the low stage, 2.3 for the low-stage slope, 7.0for the high stage, 4.0 for the high-stage slope, and0.1 for the increment. Leave the section boundaryblank and the analysis will start. See the section on"Putting XSPRO to Work" for more completeexamples.)

When you enter the last section boundary orManning's n value, XSPRO will begin the analy-sis of the cross section. After the analysis isfinished, the graph will be cleared and you will beshown the results of the analysis in the form of atable. This table can be scrolled through, back andforth, and you are provided with a highlight bar toease viewing. An output file (named in parameterswindow) is filled with the same stage-to-dischargedata seen on the screen. The format of this outputfile differs slightly from what was on the screen.The character identifiers for the different sections

are replaced with negative numbers (A becomes-1, B becomes -2, etc.), so that this output file canbe used with your favorite spreadsheet program.Press SPACEBAR to continue.

Finally, the program will graph a log-log re-gression of discharge vs. radius. You are shownthe regression formula used. If the regressionanalysis was not selected in the parameters win-dow, XSPRO will move directly to the mainscreen from the table. Pressing SPACEBAR willreturn you to the main screen from the regressiongraph.

Printing Results

The output file generated is more suited to-ward inclusion in a spreadsheet and is not as easilyviewed on a piece of paper, so a separate printingfacility has been included with XSPRO. Theoutput file you select to print will be printed inapproximately the same format as the table yousaw after the analysis was performed.

To print the data, select Print from the mainmenu. A new window will appear on your screen.Here you can select what, where, and how to print.There are defaults that will come up each time.The default file to print is the current analysisoutput file, the destination is LPT1, and a formfeed will be inserted between each set of data. Ifany of these defaults vary from what you want todo, you may change them.

You may choose to send the output to a printeror to a file. If you print to a file, you must specifya file name. This file will contain exactly whatwould be sent to the printer.

Be sure your printer is connected and turnedon. To start printing after you have made theappropriate changes, press F10. To abort printingat any time during the printing process, press ESC.

24

XSPRO BASIC PROCEDURES

XSPRO has many modern software conve-niences, including consistent operation of functionkeys and use of windows. In addition, XSPRO

provides context-sensitive help, menu control, andsimple error correction.

SPECIAL KEYS

The function keys and the ESC key havespecific and more-or-less consistent operationthroughout the program. The F1 key is alwayshelp, and the ESC key is always an abort key. F10,on the other hand, has two functions depending onwhere you are in the program. In an entry window,F10 will act as the completion key. When you aredone entering, you press F10 to save the data andcontinue the program. The other function of theF10 key is as an undo key in the analysis part of theprogram. When you are entering the stage, sectionboundary, or Manning' s n data, F10 will back youup a step or undo the last thing you did.

The SPACEBAR has a dual role as well. In thecontrolling parameters window, pressing theSPACEBAR will bring up a selection menu for thehighlighted parameter. In the analysis phase, theSPACEBAR is used to continue to the next phaseof the analysis. You press the SPACEBAR toclear the output table and move to the regressionanalysis, and to move from the regression analysisback to the main menu.

From the main menu, the F2 key will allow youto save data entered from the keyboard to a file.

HELP

Throughout the program, pressing the Fl keywill bring up a context-sensitive help window.The help window will give you information aboutwhat is highlighted by the cursor and provideinformation for making decisions. The only way

to access help for a particular topic is to be at thatspecific point in the program and press Fl. Thereis no help index facility in this version. To clearany help screen, press Fl again.

MENUS

XSPRO is a menu-driven program. All menusin XSPRO work identically and are consistentwith the menus you are familiar with from otherprograms. For horizontal menus you use theLEFT/RIGHT ARROW keys, and for verticalmenus you use the UP/DOWN ARROW keys tomove between selections. Once the selection ishighlighted, you may activate that selection by

pressing ENTER. To directly activate a menuselection, press the key corresponding to the firstletter in the selection; you will not have to pressENTER. For example, to select Run from the mainmenu you just need to press R from anywhere inthe main menu and the program will activate thatselection. To abort any selection menu, pressESC.

25

ERROR CORRECTION

If you select the wrong option from a menu,pressing ESC will bring you back a level. Thisaction is an abort; any changes made will be lostand the affected data will return to the values thatwere present before that selection was made. If themistake was made in an input menu (from thecontrolling parameters window) you may, ofcourse, go back and change your selection.

If you make a mistake in typing, the BACK-SPACE, INSERT, and DELETE keys will work asexpected.

Error in Stage/Slope/Interval Data

If you make a mistake during input of the stagedata, you can press F10 to restart the stage speci-fication procedure. All stage values will be clearedand you will be back to entering the low-flow stagevalue. After you enter the stage interval you muststart over to make a change in the stage/slope/interval data. If you want to return to the mainmenu, press ESC.

Error in Section Boundary Data

You may clear any previous section boundaryentered by pressing F10. If you keep pressing F10,you will step back through the sections that you'veentered, clearing the previous one. After you entera blank section to continue the program, you muststart the analysis over to change any section bound-ary data. If you want to return to the main menu,press ESC.

Error in Manning's n Data

You may back up through the sections to clearthe Manning's n values you entered by pressingF10. If you back up all the way to the first section,you may reenter the high-flow stage for theManning's n values. After you enter the high-stage Manning' s n for the last section, the programwill proceed with the analysis. You must thenrestart the analysis to make any changes. If youwant to return to the main menu, press ESC.

26

XSPRO INPUT

Input comes to XSPRO in three forms:cross-section data from a file or the keyboard;parameters to control the program, such as

filenames, data format, resistance equations, etc.;and 3) specifics about how to analyze the cross

section, such as low- and high-flow stages, slope,boundaries for defining separate channels in thecross section, etc. These are explained in detailbelow.

DATA ENTRY

XSPRO can read in cross-section data from afile or directly from the keyboard.

Input From the Keyboard

Data can be entered from the keyboard byselecting Modify-Parameters from the main menuand selecting Keyboard as the source of data,pressing F10, and then selecting Modify-Datafrom the main menu. You are asked for thenumber of position-elevation points that occur onthe cross section. This can be increased ordecreasedlater; it is just used as a starting point. Afterentering this, you will see entry fields for the firstten points. Use the PGDN key to see the next tenpoints. To go back a full screen, use the PGUPkey. You can move about through the entry fieldswith the ARROW keys, TAB key, or ENTER key.

Up to six digits will be retained for the positionand elevation data.

F2 will delete the very last point. F3 will adda point to the end of the list. You may add pointsup to 200. You may delete all but one point.

When all input data are correct, press F10. Ifyou want to save these data now (they can be savedlater), answer Y to the question and give XSPROa file name to save to. The data will be saved inposition-elevation format. To save the data later,use the F2 key from the main menu.

Editing previously entered data is done throughthe Modify-Data menu selection. The only differ-ence is that you won't be prompted for the numberof points to be entered.

Input From a File

If File is selected for the source of data, youwill need to specify a file name to use. When youselect Run from the main menu, data will be readfrom the file and put through the analysis. The filemay contain up to 200 data points. Be sure that thedata file you are using is formatted correctly.Refer to Appendix A for information on input fileformats. If you are using a file that was saved byXSPRO as data entered from the keyboard, the filewill be in position-elevation format.

CONTROLLING PARAMETERS

XSPRO allows you to modify all of the impor-tant parameters that control the input, analysis,and output. All of the current values are displayedat once on the screen. You can adjust all of theseparameters quickly and easily for comparisons ofdifferent formulas or different values. The firsttime the program is run, hard coded default valuesfor the controlling parameters will be used. Fromthen on, the most recent set of parameters will bethe defaults when you start the program. XSPRO

saves these parameters in a file named"DEFAULTS.XSP." If this file is ever corruptedor missing, XSPRO will use the hard coded startupdefaults again for the controlling parameters.

To modify any of the controlling parameters,select Modify from the main menu. Then selectParameters from the modify menu. The mainmenu will disappear and your cursor will be posi-tioned on the input source field in the controllingparameters window. You can move from field to

INPUT:Source of data:Filename:Data collectidi meData Forthat:::

OUTPUT:

FOS dutptit mode:(Mitt

ANALYSIS -AnalystS procedureResistance Equationd84 Units: Feetd84 partible diameter- 0.003/337Survey date (Yy/mtiVdd): SO/01/01Cross-Section number:

Figure 6. Menu selection fields.

other, as in the first example, triggers a differentchain of events. A survey done with a sag tape will

field using the UP/DOWN ARROW keys. All ofthe fields and how to edit them are explainedbelow.

Parameter Editing

There are two ways that parameterediting is done, depending on the fieldyou are editing. The most basic way isto simply highlight the field you wishto edit and type in your response. Thefields that allow this are: filename, dMdiameter, date of cross section, cross-section number, and all numerical en-try fields. These fields do not have afinite number of responses, and there-fore, anything within reason can beentered here.

The second method of editing in-volves fields that have a finite set ofpossible values; you will get a selection menu tochoose from for these. See Figure 6 for an exampleof a field that uses menu selection. The DataCollection Method field has three possible re-sponses: Sag Tape, Rod and Level, or other. Toedit that field, just arrow down to highlight thefield and press SPACEBAR. A menu box willappear with the possible selections for that field.This menu box is controlled exactly like the othermenus. To select other, press 0. The selection forthe Data Collection Method will change to readother and the cursor will be on the next field. Youcould have just as easily highlighted other in theselection menu and pressed ENTER to select it.Most of the fields in this window work this way forease of typing and for control of the input. Thefields that use the selection menu method aremarked in this documentation with an asterisk.

Another way that XSPRO controls what needsto be entered is by hiding particular fields whenthey won't affect a particular analysis. An ex-ample of this is the Resistance Equation field. IfThorne and Zevenbergen is selected as the resis-tance equation, the dM fields will appear on thescreen. For other resistance equations, these fieldsare not needed and you will not even see them.

There are some parameters that require addi-tional information. Choosing Sag Tape instead of

require other information about the conditions ofthe survey, and you will be prompted for thesedata. XSPRO first brings up an input window withfields for data that are specific to sag tape surveys.This is one way that additional information isentered. Here you must complete the inputs andpress F10, or ESC to abort. You will return to thecontrolling parameters window and the cursor willbe on the next field.

Input Parameters

Source of Data*

Data for a cross section can come from apreviously created file, like a text dump from aspreadsheet, or you may input the position-eleva-tion pairs from the keyboard. In this field, youselect the source for the cross-section data. Tomodify the selection, highlight the field and pressSPACEBAR. Then, select the data source fromthe menu.

Filename

The file name can be up to 40 characters longand can contain drive names and directory names.Furthermore, the wildcards '*' and '7' will work as

28

Controlling Parameters fF1;1714110;INPUT: Select Data Format

Source of data: File Position-Elevation free formFilename: IN.DAT Elevation-Position free formData collection ser Defined File FormatData Format:Units:

OUTPUT:Filename:File output moUnits:

ANALYSISAnalysis procedResistance Equa

•

Position data field: Elevation data field:

column 12 columnlength 4 length

1.10.n1:700.00.11.

they do in DOS. Simply type in the name of the fileyou wish to use. You may use the wildcards tocreate a directory mask, and XSPRO will displaya list of all the files in the specified directory thatmatch that mask. The current directory is thedefault. For example, to obtain a listing of all ofthe files in the current directory that have a "PRN"extension, enter:

*.PRN <ENTER>("*.PRN" is a directory mask)

for the file name. A display window will appear onthe screen with a listing of all the matching files.From here you can enter a file name or a newdirectory mask. Pressing ESC will return you tothe parameters window without changing the maskyou entered.

This file must contain some columnar form ofdata representing a distance measurement and anelevation measurement. Each line in the file willidentify a particular point on the cross section. Thespecific formats available are described in Appen-dix A.

Data Collection Method*

XSPRO supports three selections of data sur-vey methods: Sag Tape, Rod and Level, or other.Other is included for data that needs no correction.For the two specific survey types, some datacorrection must be made for uneven tape endelevations, tape physical characteris-tics, and tape sag. The sag tape forms

tension, and a tape weight need to be entered. Youmay move between the edit fields using the arrowkeys. The other function keys work as before: F1for help, F10 to complete, and ESC to abort.

Data Format*

Each selection corresponds to a different for-mat for the input file. The files will normallycontain two columns of numbers representing aposition-elevation pair on each row that definessome point on a cross section. There are currentlythree formats available; they are outlined in Ap-pendix A. Format selections are made from amenu.

The special case of selecting User Definedfrom the data format field involves specifyingexactly which columns contain the two data valuesin your file and the length of each value. This isprovided for an input file that may contain morethan just the position and elevation data. Forexample, if you have four columns of data in a fileand one of those columns contains a position valueand another contains an elevation value, but theyare not the first two columns or are not adjacent,you should select User Defined from the menu toextract the correct numbers. All you will need totell XSPRO is which column numbers to searchfor the data and how wide those data columns are.

Figure 7 shows the screen after selecting UserDefined as the data format. The values for position

a catenary curve, and if elevation dataare measured from the streambed tothe tape, these data will be incorrect. Ifthe survey method was rod and level,the tape is assumed to be without sag.If the tape ends for either method areuneven, distance measurements usingthe tape will be incorrect.

If you select either of the two spe-cific survey methods, you will beprompted with an input window toenter the data to be used for correction.For rod and level, just an elevationdifference needs to be entered. For sagtape, an elevation difference, a tape Figure 7. User defined data format screen.

811624 -161',811024 „101

,M1024811024 .101

:811024eilet4a:101811024

::1111024,: ter.811024 ":101.811024

:"is lei :-.811024.101111024 7Aoi :r

;,:811024011024 :101,... •811024 ,101

.°:81102401.1811024 101..

292;334

data column and length and for elevation datacolumn and length have been entered. When youselect User Defined, an entry window appears andyou are prompted for the column positions andlengths of the data. Figure 8 shows a portion of asample input file where the data are not in aspecific order. In this particular file the distancevalues start at column 12 and are 4 columns wide.The elevation values start at column 21 and are 3values wide. As you can see, these values havebeen entered in Figure 7.

Figure 8. Sample input file.

Units*

The units of the input data can be either centi-meters, meters, or feet. This is another menuselection field. The units you select will be usedfor all inputs. If centimeters are selected, the dataread in from the input file will be converted tometers, and all inputs will be in meters. All of theinternal computations will be done in Englishunits.

Output Parameters

Filename

The output file will contain the stage-to-discharge relationship data. The format of the data

allows easy importing to your favorite spread-sheet. The file name can be up to 40 characterslong and will accept drive and directory names.Like the input Filename field, you may enter adirectory mask to get a listing of matching files.

File Output Mode*

If the file named in the Filename field exists,you will be prompted as to how you want thewriting to that file to take place. You may selectto Overwrite the file or Append to it. Overwritingwill completely erase the file and then write thenew data to the file. Appending will simply addthe new output to the end of the file, preserving allprevious data in the file. If you print an appendedfile, all of the data in that file will be printed. Thereis no limit on the size of the output file except foravailable disk space.

Units*

The units of the output data can be either inmeters or feet. This is also a menu selection field.All output will use the units selected here. All ofthe internal computations will use English unitsregardless of the input and will be converted tometers for output if meters are selected as outputunits.

Analysis Parameters

Analysis Procedure*

You may choose how the program will per-form the analysis. The selections are:

• Geometry onlyThe only values computed are perimeter,area, width, hydraulic radius, and averagedepth. This option is useful for comparingcross-section changes in geometry throughtime. The output data are reorganized torepresent the depth of the stream and willgo from the specified stable point to thelowest elevation on the cross section. Thisoption is explained in detail under "Spe-cific Uses" in the section on "PuttingXSPRO to Work."

• Hydraulics onlySelect this to skip the regression analysis.All the values computed for the geometricanalysis are included in the output, alongwith slope, Manning's n, average velocity,and discharge. The data are organizedfrom the low-stage value up to the high-stage value.

• Both Hydraulics and RegressionThis selection gives you the hydraulicanalysis and will also perform a regressionon the discharge and hydraulic radiusvalues.

Resistance Equation*

You may choose to use the equations sug-gested by Thorne & Zevenbergen (1985) or theJarrett equation to compute a Manning's n, or youmay choose to supply your own Manning's nvalues to use in the analysis (User suppliedManning's n). The Thorne & Zevenbergen modeluses a user-supplied du diameter. If you chooseThorne & Zevenbergen, the fields for du diameter

and units will appear on the screen; the other twoselections do not use these fields and so, for thoseselections, the fields will be invisible.

dM Units*

The diameter of the du particle can be in feetor millimeters. Only if you select the Thorne &Zevenbergen equations will you see this param-eter displayed on the screen or be able to select avalue for it.

dM Particle Diameter

This is a number representing the diameter ofthe 84th percentile size class in the bed material.Only if you select the Thorne & Zevenbergenequations will you see this parameter displayed onthe screen or be able to enter a value for it.

Survey Date

Enter the date the survey was taken (optional).The date format is in year/month/day. This isprovided to annotate the output.

31

106.00

Fin to restartFl for helpESC to exit

RUNNING AN ANALYSIS

After modifying the controlling parameters,you may now run the analysis. Choose Run fromthe main menu. If you are entering the cross-section data from the keyboard and did not enterthe data through the Modify-Data selection, thedata entry window will pop up. Follow the direc-

tions for entering data. If you chose a file as theinput source, XSPRO will read in the data fromthat file and move directly to graphing the crosssection. Fl continues to provide context-sensitivehelp throughout this portion of the program.

CROSS-SECTION SPECIFICSXSPRO will plot out the cross section on your

screen after analyzing the position-elevation pairsand doing any data corrections. The lowest eleva-tion point will be found and will be assigned as thezero elevation point in the graph. Remember thatthe sag tape and rod and level correction formulas

will have an effect on the appearance of the graph.You will be prompted for more information. Anexample of the graph of a cross section is shown inFigure 9. Your screen should look similar to thisafter you select Run and input any cross-sectiondata that is needed.

7.00 -

6.00 -

5.00 -

5 4.00>

3.00 -

2.00 -

1.00 -

0.00 -•

- 0.00 25.00 50.00 • 75.00horizontal position (ft)

Specify lower and upper bounds of output discharge stages:(low stage values can range between 0.00 and 7.72 feetlow stage (ft)

Figure 9. Sample cross-section graph.

33

high7.00 -

6.00 -

5.00 -

g 4.00as

3.00

2.00

1.00 -

0.00 -- 0.00 25.00 50.00 75.00 100.00

horizontal position (ft)

low

Specify lower and upper bounds of output discharge stages:(high stage values can range between 0.00 and 12.72 feetlow stage (ft) 0 slope (%)high stage (ft) 7.6 slope (%)output table stage increment (ft)

2.34

1_

F10 to restartF1 for helpESC to exit

Figure 10. Stage, slope, increment entry.

Stage Data

In this section, the F10 key provides you ameans to restart the stage specification portion ofthe program. At any point in this section, pressingF10 will back the program up to the point ofredrawing the graph and prompting you for thelow-stage value. There is no mechanism formoving around the various entry fields, so tochange a value after entering it, you must restartthis section and reenter the values. Refer to Figure10.

High and Low Stage

If you selected to analyze just the geometry ofthe cross section, the high- and low-stage valueswill be computed from the highest and lowestelevation in the cross section. If XSPRO is beingrun in normal mode and you have selected toanalyze the hydraulics of the cross section, youmust enter the high- and low-flow stage values youwill use.

As you enter the low- or high-stage values,they are graphically represented on the cross-section plot as dashed lines at the stage entered.The low- and high-flow values can be the same ifyou only want to analyze a single stage.

For the high stage, a number that is 1.5 m, or 5ft, above the highest input elevation may be used.The program will assume vertical banks and ex-trapolate discharge for values above the highestinput elevation. Data that are extrapolated arerepresented with an "" after the discharge valueon the output.

Percentage Slope

A slope value (in percent) is needed for bothstage values. XSPRO will linearly interpolate aslope for each incremental stage. If you are justanalyzing the geometry of the stream, no slopevalues are needed and you will not be prompted forthem. The same low- and high-flow slope can beused if slope does not change.

34

100.00- 0.00 25.00 50.00 75.00horizontal position (ft)

72#1

Enter each section boundary (blank to continue):Entered Actual Entered Actual

72.000 #2

high

low

F 1 {) In 1r !t It s (rItAt'it

t Wry

F1 forEHL, IC) -xiI

7.00

6.00 -

5.00 -

2.00 -

1.00 -

0.00

Stage Increment

This is the increment that XSPRO will use tostep through from the low- to high-flow stages.The analysis will start at the low-stage value andwill increase by the value you input in the incre-ment field, up to, but not past, the high-stage value.In the output you will see an evaluation of the crosssection at every increment. For stages that containmore than one defined channel (see "SectionBoundaries" section), the data for that stage willbe split for each subchannel, and a sum of thevalues across the stream will be computed.

In Figure 10, all of the values for stage, slope,and increment have been entered.

XSPRO does not provide a mechanism to printthis graph or the regression graph. However, youmay use other software. See the "Printing" sectionfor more information.

Section Boundaries

After you have entered the stage increment,XSPRO will clear those values from the screen

and prompt you for the section boundaries (Figure11). The ability to break up a cross section intomultiple channels that can be analyzed separatelyis one of XSPRO' s major features. To define asection boundary, enter the horizontal distance tothat boundary. The boundary will appear as avertical dashed line at the closest surveyed point.(The actual horizontal distance to this point isshown.) Up to five sections (four unique sectionboundaries) may be defined. The order in whichthe section boundaries are entered does not matter.In the output, each section will be analyzed sepa-rately, and a cross-sectionwide summary for thecurrent stage will be the sum of all of the separatesection numbers. Figure 11 is the same graph asFigure 10 after stage, slope, and increment valueshave been entered, and shows an example ofwhere a section boundary should occur. Aboundary at the 72 foot mark was entered.

If you make a mistake, the F10 key will clearthe last entered section boundary. You may clearall of the boundaries that have been entered.

Figure 11. Section boundary selection.

35

To continue after you have entered all of thesection boundaries, press ENTER. If you have nosection boundaries, leave the field blank and pressENTER.

Manning's n Values

Finally, you will be prompted to enter yourown Manning's n values if you selected Usersupplied Manning's n in the parameters window.Otherwise, XSPRO will calculate the resistancebased on the equations specified. BecauseManning's n varies with flow depth, you will beasked to specify both a low- and high-stage nvalue. These low and high stages can be differentfrom the ones used to define the boundaries of theanalysis in the previous screen. For example, nvalues calculated from field discharge measure-ments at low and high stages can be used. Both the

low- and high-stage n can vary from section tosection, thereby accommodating different compo-nents of flow resistance over the cross section.

The previous stage lines will be cleared andyou will be prompted to enter a low- and high-stage n for each section. A blue or solid (if you donot have a color monitor) vertical line will markthe right edge of the current section. XSPRO willcalculate the low and high point of each sectionand not permit stage values outside this range.XSPRO will permit calculation of stages up to 5feet above the highest surveyed point.

After the last high-flow n is entered, the pro-gram will start the hydraulic analysis. XSPROwill linearly interpolate a Manning's n for eachincremental stage for each section, and will com-pute a flow-weighted average for the output sum-mary. Manning's n will not interpolate to lowerthan 0.01.

TABULAR HYDRAULIC OUTPUT

After you finish entering the parameters forthe analysis, XSPRO will produce a table showingstage-to-discharge data. The data start at the low-flow stage and increase at the increment youentered up to the high-stage value. You can scrollthrough the table using the UP/DOWN ARROWand PAGE keys. A highlight bar, current linenumber, and total lines of data information areprovided to ease viewing. Pressing SPACEBARclears the table and continues to the next screen.

If more than one section was defined, thesections will be assigned letters starting with thesection that has the lowest elevation. So 'A' willbe assigned to the section with the lowest eleva-

tion, then 'IV to the next, etc. In the summary fora specific stage, the section column will show thenumber of sections summed.

The data that were written to the output file arethe same as the data in this table and may be printedlater. However, the section boundaries in theoutput file are designated by negative numbers(-1 = A, -2 = B, etc.) and there is a stage value oneach line. Both changes are to make the output fileeasier to import into a spreadsheet. If an `*' ispresent after the discharge for a given section orstage, that number was extrapolated beyond therange of the input survey data.

REGRESSION ANALYSIS

When the data table is cleared, you will see agraph representing a regression analysis of dis-charge to hydraulic radius for the current crosssection. The graph will show a scatter plot of theactual radius:discharge pairs and then will plot theregression line. At the bottom of the graph, theregression formula used is shown. Figure 12shows a sample regression graph.

There is the potential for failure of the programat this point if out-of-bounds data were used for theanalysis. DATA MAY BE LOST if this occurs.XSPRO will stop analyzing the discharge valuesat the first sign of a value less than zero. This is tohelp keep the program from crashing. Still, someerrors may occur.

XSPRO does not provide a printer utility forprinting this or the cross-section graph. However,there are ways of getting this graphical output toprint. See the "Printing Graphics Screens" sectionfor more information.

Press SPACEBAR to clear this graph andreturn to the main menu. You have just executeda complete run of XSPRO' s main feature.

Note: If the error message "insufficient datafor analysis" appears, decrease the stage incre-ment in the cross-section window.

3000 -

2500 -

To'15 2000 -w

1500 -4:1

1000 -

500 -

0.00 0.50

Q = 101.510 R 2.714r2 = 0.9972, n = 35

1.00 1.50 2.00hydraulic radius (ft)

2.50 3.00

(press SPACEBAR to continue)

Figure 12. Sample regression graph.

PRINTING

Controlling PararnettOM

Print data from file: Fri:NW:RN:Destination: LPT1

Insert form teed between cross sections (Y/N)? No

OFinleittkckl";it mode:"OverVaite' Feet

_ - f t'30 at completiOn'".',." P-1 , ,.E.SC.fd Abort "

ANALYSISprocedure:

reion: :. , Ai,a.hisis

---Resistan— ce Equat ion:, )

C' ''' Wiliiitiidd,

Suveiidatieon number..,

Both Hydraulics and RegressionJanett

90/07/31

Figure 13. Print setup screen.

XSPRO OUTPUT

OUTPUT FILE

A file named in the parameters window isfilled with the analysis data while the analysis istaking plaáe (see Appendix A for a more completedescription of this file). At the beginning of theoutput file there will be some header informationpertaining to the parameters used. This can beignored and is only used by the printing program.

The real data in the output file are in columnform with titles. This file is meant to be importedinto a spreadsheet program. The section bound-aries are renamed to negative numbers so conver-sion to numbers in a spreadsheet will be possible.For a less cryptic output, you can print the datafrom this file using the print facility of XSPRO.

Data From the Output File

The output from an analysis run issaved to the file specified in the con-trolling parameters window. This filebecomes the input file for the printingfacility. To print the output to a printeror file choose Print from the main menu.You will see a new window with theparameters that will be used for print-ing (Figure 13).

These fields work like the fields inthe controlling parameters window.Some are of the menu-selection varietyand are marked with an asterisk afterthe field name.

Source File Destination*

The source file to the printing facility is the filethat contains the output from your run of an analy-sis. If the file contains data from more than one runof an analysis, all of the data in the file will beprinted. The default value for this is the file namedas the output file in the controlling parameterswindow. However, you can use any XSPROoutput file.

The field works just like the Filename fields inthe controlling parameters window. You may usedrive names, directory names, and wildcards for adirectory mask. The field is 40 characters long.

You can direct the output to a printer or to afile; the text will be identical. The DOS devicenames for parallel and serial ports are used todirect output to a printer. Most likely your printerwill be connected to LPT1, the first parallel port,and LPT1 will come up as a default each time youprint. Your printer may be connected to a differentport; if so, you will need to select that port from theselection menu. If you want to print to a file, selectFile from the menu and you will be prompted forthe name of the file.

39

Filename

If you select File in the Destination field, youwill be prompted for a file name. This file will befilled with exactly the same data as would be sentto the printer, including all form feeds. If the fileexists, you will be asked how you want XSPRO towrite to the file (appended or overwritten).

This field works just like the Filename fields inthe controlling parameters window. You may usedrive names, directory names, and wildcards for adirectory mask. The field is 40 characters long.

File Output Mode*

If the file named in the Filename field exists,you will be prompted as to how you want thewriting to that file to take place. Overwriting willcompletely erase the file and then write the newdata to the file. Appending will simply add thenew output to the end of the file, preserving allprevious data in the file.

Form Feeds*

You can also select to have a form feed issuedafter each data set. This will only apply when theprinted data contains data from more than one run.If you choose No, the separate sets of data will allrun together. If you choose Yes, each data set willstart on a new page. A form feed will be issuedafter the last page for all printouts.

Printing Graphics Screens

XSPRO does not provide you with an internalutility for printing graphics screens. However, apublic domain software, "PRTSC.COM," is pro-vided with XSPRO to enable use of the SHIFT-PRINT SCREEN key sequence to send graphicsscreens (i.e., cross-section plot, regression analy-sis) to either IBM, Epson, or Laserjet printers. Theprogram can also detect which type of printer isbeing used.

To print graphics screens, PRTSC.COM mustbe loaded before loading XSPRO. At the DOSprompt, type:

PRTSC [PN] [RNNN]

where [PN] is the printer number (i.e., LPT1) and[RNNN] is the resolution. If you do not includeanything on the command line, LPT1 and theEpson printer are assumed. You are given a choiceof resolutions, but higher resolutions may affectprinting speed. The resolution number must bethree characters: " 75," "150," or "300." There-fore, to configure for LPT2 and the Laserjet at 300dots per inch, type:

PRTSC P2 R300

After this command, type "XSPRO" to beginthe program as usual. Once inside XSPRO, press-ing the SHIFT and PRINT SCREEN keys simul-taneously will send the graph to the named printer.

Additional information about this programcan be obtained by printing the file "PRTSC.DOC"supplied with XSPRO. This software is in thepublic domain; however, if you find yourself us-ing the utility often, the author requests that yousend a donation of $5-$10 to:

Jonathon ZuckUser Friendly, Inc.

P.O. Box 21164Kalorama Station

Washington, DC 20009(202) 387-1949

sn

Exit Cross-Section Professional Analyzer

Run Print

Contrang ParametersINPUT:

Source of data: FifeFilename: SAMPLE..DATData collection method: otherData Format: Posittion-Elevation free formUnits:

OUTPUT:Filename:

"File output mode:Units:

Feet

SAMPLE.OUTFeel

ANALYSISAnalysis procedure: Both Hydraulics and RegressionResistance Equation Jarrett

Figure 14. Main screen.

PUTTING XSPRO TO WORK

EXAMPLES USING SAMPLE DATA

CD XSPRO <ENTER>Quick Run Then, to run XSPRO type:

To let you get the feel of using XSPRO, let usrun through an example using data from a sampledata file. Two sample data sets are provided withyour distribution disk. They contain actual datacollected from mountain streams in Oregon.

In the examples, only the mechanics of howXSPRO works will be explained. If you have anyquestions about the contents or purpose of any ofthe parameters discussed here, please refer to thecomplete explanations in the "Controlling Param-eters" section.

Starting XSPRO

To run XSPRO, be sure you have a DOScommand prompt and that the XSPRO files are inthe current drive and directory.

Floppy Drive

XSPRO <ENTER>

If XSPRO was installed on any other drive ordirectory than shown here, just change the aboveinstructions to match your case.

Analyzing the Sample Data

Following the title screen, your screen shouldlook like Figure 14. If this is the first time XSPROhas been run on your machine, the default valuesfor all of the parameters should be set to run withthe data file "SAMPLE.DAT." You can verifythis by matching what is in Figure 14 and what yousee on your screen for the parameters window. Ifthe parameters do not match up, you may need togo to the next example in the "Changing theControlling Parameters" section where there areexamples of changing the parameters so you can

If you are running XSPRO from afloppy drive, boot your machine as younormally do and then place the XSPROdisk in drive A. From the A drive type:

XSPRO <ENTER>

Hard Drive

If you installed XSPRO on yourhard drive, you will need to know thedrive and directory in which XSPROwas installed. Let us use the exampleof XSPRO being installed on drive Cand in the subdirectory XSPRO(C: \XSPRO). In this case, to change tothe XSPRO drive and directory, type:

C: <ENTER>

- 0.00 25.00 50.00 75.00horizontal position (ft)

Specify lower and upper bounds of output discharge stages:(low stage values can range between 0.00 and 7.72 feetlow stage (ft)

uxi.00

F10 to retortFl forESC to exit

Figure 15. Sample data cross-section graph.

match up your screen and Figure 14.Starting the Analysis

After the parameters are all set, you are readyto run the analysis. To run the analysis, highlightthe Run selection from the main menu and pressENTER, or simply press R on the keyboard.XSPRO will read in the data from the file andgraph it on your screen. Your screen should nowlook like Figure 15.

Stage/Slope Data Entry

In this section, you will be entering data thatwill control how the cross section will be ana-lyzed. This section can be run over and over againuntil you get what you want out of XSPRO. Theoutput file will be filled with the data from thisanalysis, so if you are going to run this over andover, it is suggested that you use the Overwrite

option for the output mode.Your cursor should be in a highlighted box in

the lower left corner of the screen. The promptreads Low stage (ft). Type:

0.0 <ENTER>

to use the lowest point on the graph as the low stageof the analysis. You will notice that the stagesselected are represented as horizontal lines on thegraph on your screen.

The cursor will move right, and now you areprompted for the slope (in percent) of the channelat that low stage. For this example type:

2.3 <ENTER>

The cursor will move down and left and nowyou are prompted for the high stage. Enter the highstage as:

7.0 <ENTER>

42

and the high-stage slope as:

4.0 <ENTER>

The last entry on this screen is the analysisincrement between the selected stages. For thisexample, enter:

0.5 <ENTER>

Section Boundaries

The lower portion of your screen should clearand a new prompt for a section boundary distancevalue should appear. You should notice that atapproximately the 70 foot mark (horizontal posi-tion), a small subchannel starts and continues tothe right end of the cross section. You will isolatethis subchannel from the rest of the cross sectionfor this analysis. To tell XSPRO to create a sectionboundary here type:

70 <ENTER>

and you will see that XSPRO marked the point onthe graph closest to the 70 foot mark. This is notwhere you wanted the section boundary to be, soyou must clear this boundary and enter a new one.To clear the boundary, press F10. The line willdisappear and XSPRO will be ready to accept anew entry. Try a few values if you like; you shouldfind that 72 will work. To continue to the actualanalysis, leave the Section Boundary field blankand press ENTER. A small status window willappear to show you that XSPRO is in fact workingand has not crashed your machine.

Stage-Discharge Data Table

After a bit, the graph will disappear and anoutput table will appear on the screen. The data aresorted by stage and show the changes in hydraulicvariables that occur as the stream "fills" up fromthe low stage to the high stage you specified. The

UP/DOWN ARROW keys and PAGE keys willwork as expected to let you scroll through thisdata. The data shown here are the same data thatget written to the outputfile. Press the SPACEBARto continue to the regression.

Regression

XSPRO will plot a radius to discharge regres-sion. The values used for the regression will bewritten to the output file. Press SPACEBAR toclear the graph and you will be back at the mainscreen.

Changing the ControllingParameters

XSPRO lets you change all of the controllingparameters easily so you can see how they mayaffect your output, compare multiple sets of data,or correct a mistake. For this example, you willchange the value in a menu selection field and usea straight keyboard entry field.

Start XSPRO. Select Modify from the mainmenu. You will get a submenu that allows you tobranch off to Modifying the Parameters or Key-boarded Data. From here select Parameters.Your cursor should move to the Controlling Pa-rameters window and the Data Source field shouldbe highlighted. You may move between the fieldswith the ARROW keys. Details of the editingprocess are explained in the "Parameter Editing"section, so only an example of each will be pro-vided here.

Selection Menu

Most all of the parameters have a finite set ofpossible values. For these parameters, we havemade selecting a value just a matter of making aselection from a menu. Highlight the Data Collec-tion Method field under the input section. PressingSPACEBAR will bring up a selection menu for

43

that field. The menu shows all possibleentries for that field (see Figure 16). Inthis case you can select from Rod andLevel, Sag Tape, or other. Highlight-ing other and pressing ENTER, or justpressing 0 will select other for theData Collection Method field.

Keyboard Entry

The other type of parameter is theone that you are probably most famil-iar with. It simply involves filling inthe blank. All of the numerical fieldsand the Filename fields will work likethis because there are an infinite num-ber of responses. For an example,highlight either of the Filename fieldsand type in the name of a file. XSPROwill allow any character combinationto be entered here. Figure 16. Menu selection fields.

SPECIFIC USESAnalyzing Geometry Changes from

Year to Year

Suppose you have a set of cross sections forwhich you collect data annually and you wish tocompare the annual changes in channel geometry.XSPRO makes it easy to do this kind of analysis.Setting the Analysis Procedure parameter to Ge-ometry Only will reduce the output of XSPRO toonly the geometric data: area, perimeter, topwidth,average depth, and hydraulic radius. The outputwill be ordered from the top of the cross section(default) or a specified stable point down (oppo-site of the ordering of a hydraulics analysis). Thiswas designed specifically for analyzing changesin streambed geometry, since the upper part of thecross section is assumed stable and can serve as areference point for changes in the lower part of thecross section. XSPRO will automatically selectthe high and low stages for the analysis, based onthe highest and lowest points in the cross section,so all you have to do is enter the stage incrementand the section boundaries. However, there are

several options for specifying stable points towhich all elevations will be referenced. In thisexample, you will see how to analyze the geom-etry of a single cross section, and then have XSPROappend the output data of consecutive runs to asingle file. You will then have a data file that spanssome timeframe and can easily be imported into aspreadsheet program or printed out for furtheranalysis.

Step 1: Set the Input and OutputParameters

For this example you will again use the file"SAMPLE.DAT" that came on your distributiondisk. The data in the file "SAMPLE.DAT" usesthe following input parameters (these will be thedefault settings when XSPRO is run for the fasttime):

Data Collection Method: OtherData Format: Position-Elevation

Units: Feet

Our first step is to set the controlling param-eters to match the data. Select Modify and thenParameters from the main menu. The cursor willnow be positioned on the Data Source field. Fol-low the instructions in the previous section tomatch all of the input and output parameters withthose in Figure 14. You should be aware of whatoutput file and mode are selected so you don'toverwrite any important data.

Step 2: Set the Analysis Parameters

Move down the window to theAnalysis Proce-dure field. You want to analyze only the geometryof the stream so you must change the AnalysisProcedure field to Geometry Only. To do this,press the SPACEBAR and select Geometry Onlyfrom the selection menu. Notice that the resis-tance equation information will disappear. Noneof that will matter in this analysis.

Press F10 to complete the parameter setup.

Step 3: Run the Analysis

Select Run from the main menu and XSPROwill graph the cross section. You will then beprompted to supply the upper boundary (stablepoint) for the analysis, along with the stage inter-val. A line indicating the upper boundary will bedrawn on the cross section. The upper boundaryrepresents the stable or reference level and will beindicated on the output as elevation 0. The channelgeometry will be calculated from this point down-ward to the lowest elevation on the cross section atintervals indicated by the specified stage interval(Note: This may result in the first interval beingless than the specified interval, depending onwhether the specified interval divides evenly intothe difference between the stable point and thelowest point on the cross section).

As with the hydraulic procedures, you can alsosubdivide the cross section into subchannels. In-dicate where the section boundaries should beplaced by typing a number corresponding to thehorizontal distance; a vertical line will appear atthe data point closest to this value. When all

section boundaries have been entered, hit ENTERto start the analysis.

The output will be ordered from the specifiedupper boundary down to the lowest elevation onthe cross section, and identified in the left-mostcolumn as distance below datum. PressingSPACEBAR will return you to the main menu,and complete the analysis of this set of data.

Step 4: Run the Analysis Again With NextYear's Data

From the main menu, select Modify and thenParameters and change the input file field so itcontains the name of the file for the next set of data.Also change the Output Mode field to read Ap-pend. This will append the next year's data to thesame output file. If you don't want to do this, youmust choose a new output file name or the previ-ous output data will be lost. Exit back to the mainmenu.

Select Run from the menu and hit ENTER twotimes to accept the same upper boundary and stageincrement. Enter the same section boundaries aspreviously specified and hit ENTER to begin theanalysis. The output table will be appended to thesame file as the previous analysis. When theanalysis finishes, you have a file with data in tableform for 2 years of a cross section. This last stepcould be repeated with many more years of data.

To complete the analysis, follow the directionsin Appendix B for importing files into Lotus 123.Year-to-year changes in channel geometry can beplotted on a single graph as long as the samereference level was used in each cross section.

There are two other ways of specifying stableor reference points for conducting Geometry Onlyanalyses besides indicating them directly on thecross section as described above. One is byidentifying a stable point when entering data fromthe keyboard (see next section). The other is byusing a text editor program to add an uppercaseletter S to the end of the line of data (no spaces)corresponding to the stable point in an ASCII datafile. So, for example, if your cross-section data isin an ASCII file with each line consisting of a

INPstclur

- I enter the riuin

DatatO DewUnit :„ 'I 6.2 to Add a point

au-rF I PGDN for next i0 pointsFilen ' POUF! foi pr1y,10FileUnit

Data unite:ANAL .m I - Stable pt: 10 (Not Defin e

Anal , I 'Resis

Figure 17. Keyboard input.

single distance-elevation pair of numbers (sepa-rated by spaces), appending an S to any line willlabel the elevation of that point as the stable point.You will then see a line corresponding to thatelevation in the graphics screen when the crosssection is plotted. That point will also be identifiedas the stable point in the Modify-Data screen (seebelow). The advantage of either of these two waysof specifying stable points is that they are auto-matically written to the input data file. Identifyingthe stable point using the graphics screen alonedoes not alter the input file.

Entering Data With the Keyboard

Thus far, you have been using data from a filethat already exists. XSPRO also allows you toenter the data for a cross section rightfrom the keyboard, and it will save thatdata to a file that can be used someother time. You can also edit anyentered data, as in the next example.

Start up XSPRO. This exampleuses made-up data; as always you maysubstitute any real data you have.

Step 1: Tell XSPRO You Want toKey In the Data

From the main menu, select Modifyand then Parameters. Your cursor willbe in the Data Source field, which isthe one you want to change. PressSPACEBAR and select Keyboard as thesource. Make sure all other parametersmatch what you want to do and pressF10.

Step 2: Enter the Data

The Modify submenu should be active. SelectData and an entry window should appear on yourscreen. The fast prompt is how many points youwant to enter. This need only be a rough estimateas you can increase or delete points later. Themaximum number of points per cross section is200. For the number of points enter:

15 <ENTER>

XSPRO will now give you a table of distanceand elevation points and your screen should looklike Figure 17. The entry screen can only displayten points at a time. To move between sets ofpoints, you must use the PGUP/PGDN keys. Tomove around within the set of points that are beingdisplayed, you can use the UP/DOWN ARROWkeys, TAB, and ENTER. If you need more pointsthan you have, you may press F2 to add a point tothe end of the list. If you have too many points, youmay press F3 to delete a point from the end of thelist. These functions are provided only to adjustthe number of points; you cannot insert or deletepoints from anywhere but the end of the list.

(Note: The data can be in any order as XSPROwill sort the entries by elevation.)

Type in the following list of distance-eleva-tion pairs:

Distance Elevation0.0 4.61.5 3.23.0 1.84.5 0.66.0 0.17.5 0.09.0 0.3

10.5 1.012.0 1.913.5 1.7

Press PGDN to get the next set of entryfields. Then enter:

15.0 2.116.5 3.318.0 4.019.5 4.321.0 4.5

Press F10 to complete the entries. You willthen be asked to specify which observationnumber corresponds to the stable point; this infor-mation is used in the Geometry Only analysisdescribed above. The default value of 0 indicatesthat no stable point is to be specified. Press F10again. You will be asked if you want to save thedata to a file now. You can save the data to a filefrom the main menu by pressing F2, but to be safe,go ahead and save the data now. Press I' and thentell XSPRO what file name to use. After you typein the file name, press F10 to save the data.

Step 3: Analyzing the Data

The rest of the analysis works just like it doesfor data from a file. SelectRun from the main menuand you will see a graph of the cross section. Youenter all of the pertinent data and the analysis willcontinue.

Step 4: Editing Data From Keyboard Input

You can edit any mistakes made or make anychanges needed to the data. When you get back tothe main menu, select Modify and then Data again.You should get a filled editing screen with all ofthe values you used last time. You can now moveto any cell and edit it. When you complete theediting, XSPRO will again ask if you want to savethe data. If you have made changes, you shouldanswer with a Y and tell XSPRO what file name touse.

Saving Keyed In Data From theMain Menu

When you complete either entering or editingdata, XSPRO will ask if you want to save the datato a file. You can answer Y and save the data here,or you can save the data later from the main menu.From the main menu, press F2. You will beprompted for a file name and when you finishentering the name, press F10 to save the data. If nodata have been entered, you will get an error.

47

LITERATURE CITED

Aldridge, B.N., and J.M. Garrett. 1973. Rough-ness coefficients for stream channels inArizona. U.S. Geological Survey Open-FileReport.

Arcement, G.J., Jr., and V.R. Schneider. 1984.Guide for selecting Manning's roughness co-efficients for natural channels and flood plains.U.S. Dept. Trans., Federal Highway Adminis-tration Tech. Rept. No. FHWA-TS-84-204.

Ashmore, P.E., T.R. Yuzyk, and R. Herrington.1988. Bed-material sampling in sand-bedstreams: Sediment Survey Section, WaterResources Branch, Inland Waters Directorate,Environment Canada, Report IWD-HQ-W'RB -SS-88-4. 90 p.

Barnes, Harry H., Jr. 1967. Roughness character-istics of natural channels. U.S. GeologicalSurvey Water-Supply Paper 1849.

Bathurst, J.C. 1978. Flow resistance of large-scale roughness. Journal of the HydraulicsDivision, Am. Soc. Civil Engr., Vol. 104, No.HY12, pp. 1587-1603.

Benson, M.A., and Tate Dalrymple. 1967. Gen-eral field and office procedures for indirectdischarge measurements. Techniques of Wa-ter-Resources Investigations of the UnitedStates Geological Survey, Book 3, ChapterAl.

Buchanan, Thomas J., and William P. Somers.1969. Discharge measurements at gagingstations. Techniques of Water-Resources In-vestigations of the United States GeologicalSurvey, Book 3, Chapter A8.

Chow, Ven Te. 1959. Open-Channel Hydraulics.McGraw-Hill Book Company, New York680 p.

Church, M.A., D.G. McLean, and J.F. Wolcott.1987. River bed gravels: sampling and analy-sis. Department of Geography, University ofBritish Columbia, Vancouver, British Colum-bia, Canada. pp. 43-88.

Cowan, W.L. 1956. Estimating hydraulic rough-ness coefficients. Agricultural Engineering,Vol. 37, No. 7, pp. 473-475.

Dalrymple, Tate, and M.A. Benson. 1967. Mea-surement of peak discharge by the slope-areamethod. Techniques of Water-Resources In-vestigations of the United States GeologicalSurvey, Book 3, Chapter A2.

Federal Inter-Agency Sedimentation Project. 1986.Catalog of instruments and reports for fluvialsediment investigations: Federal Inter-AgencySedimentation Project, Minneapolis, Minne-sota. 138 p.

Grant, Gordon E., Frederick J. Swanson, and M.Gordon Wolman. 1990. Pattern and origin ofstepped-bed morphology in high-gradientstreams, Western Cascades, Oregon. Geo-logical Survey of America Bulletin, Vol. 102,pp. 340-352.

Hey, R.D. 1979. Flow resistance in gravel-bedrivers. Journal of the Hydraulics Division,Am. Soc. Civil Engr., Vol. 105, No. HY4, pp.365-379.

Jarrett, Robert D. 1984. Hydraulics of high-gradient streams. Journal of Hydraulic Engi-neering, Am. Soc. Civil Engr., Vol. 110, No.11, pp. 1519-1539.

Parsons, Stephen C., and Shirley Hudson. 1985.Stream channel cross section surveys and dataanalysis. U.S. Bureau of Land ManagementTech. Rept. No. TR-4341-1.

d9

Potyondy, John P. 1990. An analysis of the use ofroughness coefficients in the modeling ofchannel hydraulics: a comparison of flowscalculated with Jarrett's equation to flowscalculated using Manning' s n-values. Unpub-lished report to the Boise National Forest.

Rantz, S.E., and others. 1982. Measurement andcomputation of streainflow: Volume 1. Mea-surement of stage and discharge. U.S. Geo-logical Survey Water-Supply Paper 2175.

Ray, Gary A., and Walter F. Megahan. 1979.Measuring cross sections using a sag tape: ageneralized procedure. USDA Forest ServiceGeneral Technical Report INT-47.

Roberson, John A., and Clayton T. Crowe. 1985.Fluid Mechanics. Houghton Mifflin Com-pany, Boston. 712 p.

Smoot, George F., and Charles E. Novak. 1968.Calibration and maintenance of vertical-axistype current meters. Techniques of Water-Resources Investigations of the United StatesGeological Survey, Book 8, Chapter B2.

Thorne, Colin R., and Lyle W. Zevenbergen. 1985.Estimating mean velocity in mountain rivers.Journal of Hydraulic Engineering, Am. Soc.Civil Engr., Vol. 111, No. 4, pp. 612-624.

Van Haveren, Bruce P. 1986. Water ResourceMeasurements. American Water Works As-sociation, Denver. 132 p.

Wolman, M. Gordon. 1954. A method of sam-pling coarse river-bed material. Transactionsof the American Geophysical Union, Vol. 35,No. 6, pp.951-956.

Yuzyk, T.R. 1986. Bed material sampling ingravel-bed streams: Sediment Survey Sec-tion, Water Resources Branch, Inland WatersDirectorate, Environment Canada, ReportIWD-HQ-WRB-SS-86-8. 74 p.

50

APPENDIX A

FILE FORMATS

INPUT FILES

Input files are ASCII text files that, in general,will contain two columns of data. One column willcontain distance, the other elevation. Each line ofthe file will represent one point on the crosssection. However, XSPRO can use several sepa-rate formats for the data files. Following is a listof the different formats and a description of howinput files must be formatted for each.

Position-Elevation Free Form

This format requires a file that has at least twocolumns of data. The first column should have aposition or horizontal distance, and the secondcolumn should have an elevation number. Allother columns in the file will be ignored. The filewill be read from top to bottom. Each line in thefile will be processed, and the two values on thatline will represent one pair of distance-elevationpoints on the cross section.

If you key in the data and save it to a file, thefile will be saved in the position-elevation format.

Elevation-Position Free Form

This format requires a file that has at least twocolumns of data. The first column should have anelevation, and the second column should have aposition or horizontal distance number. All othercolumns in the file will be ignored. The file will beread from top to bottom. Each line in the file willbe processed, and the two values on that line willrepresent one point on the cross section.

User Defined

If none of the above formats match your needs,you can specify the column number and length ofthe position and elevation data in your file. If youhave a file that was created by a spreadsheetprogram and it has the position and elevation dataat columns other than the first two, you can specifythose columns and the length. The program willsearch through the entire file for the data.

OUTPUT FILES

The output file that the program generatescontains all of the hydraulic data as well as theregression numbers. The first five lines of the filewill contain header information that is used by theprinting facility. The next line has the columntitles followed by the analysis data.

The letters you saw representing the sectionboundaries when the data was displayed on thescreen are changed to numbers in the output file soa spreadsheet program won't have difficulty withtext in a column of numbers. The relationshipbetween the letters and numbers is simple; whatused to be A is now -1, what used to be B is now-2, etc.

Another difference from the on-screen data isthat the stage numbers are included on each line tofurther enable you to import the data into a spread-sheet. The asterisk at the end of the line for datathat were extrapolated will usually not interferewith a spreadsheet. The fmal line of the output fileis the numbers used in the regression analysis.

Of course, if you have the program append tothe output file, there will be more than just one setof output data. Each set of data will start with a lineof asterisks. The files can get as big as the diskspace available.

51

APPENDIX B IMPORTING XSPROOUTPUT INTO LOTUS 1-2-3

XSPRO produces an output file that is easilyimported into Lotus 1-2-3 for further analysis.Following are steps to import that data.

Start Lotus 1-2-3 and select File/Import/Text. The file you want to use is the outputfile from XSPRO. Specify the correct filename and Lotus 1-2-3 should load the file.

Remove the header information (every-thing but the column headers) for each dataset. Leave a blank row between each set ofdata. You should now have all the data andthe headers for each column.

Now you have to convert the data to num-bers. Move the cell pointer to the first rowof data, not text (should be the first row

after the top column title row). SelectData/Parse from the menu. SelectFormat-Line. This will be the format usedfor all of the columns of data and willformat the headers as well.

Select Input-Column and mark all of thedata.

Select Output-Range and mark the upper-left cell.

Select Go.

The data and column titles should all be in theirown columns now. Don't forget to save the filebefore you finish.

53

REPORT DOCUMENTATION PAGE IForm ApprovedOMB No. 0704-0188

Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering andmaintaining the data needed, and completing and reviewing the collection of inlormalion. Send comments regerdIng this burden estimate or any other aspect of this collection of information.including suggestions for reducing this burden, to Washington Headquarters Sonless, Directorate for Information Operations and Reports, 1215A/bison Davis Highway, Suite 1204, Arlington,VA 22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project (0704-0188), Washington, DC 20503.

1. AGENCY USE ONLY (Leave blank) I 2. REPORT DATEAugust 1992

3. REPORT TYPE AND DATES COVEREDIFinal

4. TITLE AND SUBTITLE

XSPRO: A Channel Cross-Section AnalyzerS. FUNDING NUMBERS

•

AUTHOR(S)

Gordon E. Grant, Joseph E. Duval, Greg J. Koerper, and James L. Fogg

PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES)

U.S. Department of the InteriorBureau of Land Management—Service Center •P.O. Box 25047Denver, CO 80225-0047

B. PERFORMING ORGANIZATIONREPORT NUMBER

Technical Note 387BLM/SC/PT-02/001+7200

9. SPONSORINGIMONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSORINGIMONITORINGAGENCY REPORT NUMBER

11. SUPPLEMENTARY NOTES

12a. DISTRIBUTIOWAVAILABILITY STATEMENT 12b. DISTRIBUTION CODE

ABSTRACT (Maximum 200 words)

XSPRO is an interactive, menu-driven software package designed to assist watershedspecialists in analyzing stream channel cross-section data. It has been specificallydeveloped to handle channel geometry and hydraulic conditions for single transects insteep (gradient > 0.01) streams. Several resistance equations are supported, includingthose specifically designed for large roughness channels. Analysis options includedeveloping stage-to-discharge relationships and evaluating changes in channel cross-sectional area. Both graphical and tabular output can be generated. XSPRO can assistresource specialists in analyzing instream flow needs, performing hydraulic reconstruc-tions, designing effective channel and riparian structures, and monitoring channelchanges.

SUBJECT TERMSStream channels Watershed Riparian areasChannel cross sections Instream flow

NUMBER OF PAGES68

PRICE CODE

17. SECRURITY CLASSIFICATIONOF REPORT

Unclassified

18. SECRURITY CLASSIFICATIONOF THIS PAGE

Unclassified

19. SECRURITY CLASSIFICATIONOF ABSTRACT

Unclassified

20. LIMITATION OF ABSTRACT

ULStandard Form 299 (Rev. 2-891