Special Report 88-25 MNovember 1988 US Army Corps

of EngineersColo Regions Research &Engineering Laboratory

Ice observations on the Alleghenyand Monongahela Rivers

Michael A. Bilello, Lawrence W. Gatto, Steven F. Daly and John J. Gagnon

00IN

Cf)

RIM

LIVE CE MANAGEMENT

Prepared forOFFICE OF THE CHIEF OF ENGINEERS

Approved for public release; distribution is unlimited. 89 9 2 8 089

C UASS CA T ON OF TiHS PAGE

FomApprovedREPORT DOCUMENTATION PAGE 0M NO, 070-1 0 )88

F Date: Juln 30 1 e

C~SECU',Y CLASSIFICATION 1.B RESTRICTIVE MAR,(:NGS

L nclassified~'Y CLSSIRICATiON AUTHORITY 3. DiSTRBUTIONAVAILABILiTY OF REPORT

*_X's& iCATON. DOWNGRADING SCHEDULE Approved for public release; distribution is unl'Ii lted

2 .~iAG ORGAN;Z.A"ON REPORT NUMBER(S) 5 MONITORING OrGAIAINRPR UBRS

Sjwcial Report 88-25,\.,I., OP PERFORMING ORGANiZATION 6ib OFFICE SYMBOL 7a NAME OF MONITORING ORGANIZATION

U.S. Army Cold Regions Research (if applicobie_)

andT~ F'ngineering Laboratory CECRL Office of the Chief of Engineers-X --,S (C~y Store ond ZIP Code) 7b ADDRESS (City, Stayte. and ZIP Code)

2 1'vrne RoadHaniover, N.H. 03755-1290 Washington, D.C. 20314

AS, - O ,,NDING!SPONSORiNG 8ib OFFICE SYMBOL 9. PROCUREMENT INSTRUMENT IDENTIFICATION NUMBERC):XNZAON I (if appi~cable)

___________________________ ICWIS 32227; CWIS 322283c A D_ SS (CfWv State and ZIP Code) 10. SOURCE OF FUNDING NUMBERS

PROGRAM PROJECT TASK WORK UN 1ELEMENT NO. NO. NO. ACCESSIONNO

-- 'Include Security Classification)

Ice Obs-ations on the Allegheny and Monongahela Rivers

PE7SONAL AUTHOR(S)

Bilello, Michael A. Gatto, Lawrence W.; Daly, Steven F. and Gagnon, John J.3a. TYPE OF REPORT 13b. TIME COVERED 1T.DT FRPR(er ot.Dy 5. PAG_ COUNT

RO ____ To ___ November 1988 47

6 SUPPLEMENTARY NOTATION

17. COSATI CODES _18. SUBJECT TERMS (Continue on reverse if necessary and Identify by block number)

FID GROUP SUB-GROUP Allegheny River River iceIce Winter navigationMonongahela River

19 ABSTACT (Continue on reverse ifrnecessary and Identify by bkbck number)

)I, Corps of Engineers and National Weather Service records of ice conditions on the Allegheny and MonongahelaRivers in Pennsylvania and West Virginia were analyzed for seven recent winters. The on-ground observationsrecorded daily at a number of lock and dam locations were issued in the form of alphanumeric ice codes thatincluded the coverage, type, thickness, structure and extent of river ice. These codes were used to graph iceconditions throughout the rivers to allow easier analysis of historical ice conditions. In addition, comparisonswere made between these observations and aerial videotapes and satellite images of the ice. Results of thesecomparisons illustrate that ice data from these three sources are complementary and should be used togetherwhenever possible.

20 DISTRIBUTION/AVAJILABILITY OF ABSTRACT 21. ABSTRACT SECURITY CLASSIFICATIONIN INCLASSIFIED/UNLIMITED [3 SAME AS RPT. 0 DTIC USERS Unclassified

22a NAME OF RESPONSIBLE INDIVIDUAL 22b. TELEPHONE (Include Area Code) I22c. OFFICE SYMBOL

TLawrence W. Gatto .603-6464100 1 CECRL-RE

DD FORM 1473, 84 MAR 83 APR edition may be Led until exehausted. SECURITY CLAS9FICATION OF THIS PAGEAli other editions are obsolete.

UNCLASSIFIED

PREFACE

This report was prepared by Michael A. Bilello, Meteorologist, Science and TechnologyCo iporation, I lampton, Virginia; Lawrence W. Gatto, Geologist, Geologipal Scipnces Branch,Steven F. Daly, Research Hydraulic Engineer, Ice Engineering Research Branch, and JohnJ. Gagnon, Civil Engineeringechnician, Ice Engineering Research Branch, U.S. Army ColdReimons Research and Engineering Laboratory. The work was funded by the Offlce of theChief'of En-ineers, Directorate ofCivil Works, under the River Ice Management (RIM) Pro-gran, C\VIS 32228, Remote Ice Monitoring System, and CWIS 32227, Forecasting IceConditions on Inland Rivers.

The excellent cooperation received from U.S. Army Corps of Engineers District, Pitts-burgh, and U.S. National Weather Service personnel in Pittsburgh, Pennsylvania, whoprovided the river-ice data, is greatly appreciated. The authors thank the following CRRELpersonnel: Kevin Carey and Dr. George t>hton for their technical reviews of the paper;Mark Hardenberg forhis editorial work; Edward Foltyn for assisting in preparing the tablesin Appendix A; William Bates, Edward Perkins and Eleanor Iluke for drawing the figures;Jacqueline Castor and Donna Harp for typingthe manuscript; and Guenther Frankenstein,Chief, Ice Engineering Research Branch, for providing the opportunity to conduct theresearch.

The contents of this report are not to be used for advertising or promotional purposes. Ci-tation of brand names does not constitute an official endorscment or approval of the use ofsuch commercial products.

11

CONVERSION FACTORS: U.S. CUSTOMARY TO METRIC (SI) UNITS OFMEASUREMENT

These conversion factors include all the significant digits viven in the conversion tablesin the ASTM Metric Practice Guide (E 380), which has been approved for use by the De-partment of Defense. Converted values should be rounded to have the same precisionas the original (see E 380).

Multiply By To nbtain

inch 25.4 millimeterfoot 0.3048 meterfoot 3/second 0.02831685 meter 3/secondmile 1609.347 meterdegrees Fahrenheit T°C = (T°F-32)/1.8 degrees Celsius

;I .yk

rD..... ~ '2 t- odes

h7 __

Ice Observations on the Allegheny and Monongahela Rivers

MICHAEL A. BILELLO, LAWRENCE W. GATTO, STEVEN F. DALY AND JOHN J. GAGNON

INTRODUCTION River. These Corps and NWS sites cover therivers from Pittsburgh to West Hickory, Pennsyl-

Detailed information on daily ice conditions vania, about 158 miles upstream on the Al-along entire lengths of navigable rivers is often legheny River, and from Pittsburgh to Opekiska,nonexistent or difficult to recover from data ar- West Virginia, about 115 miles upstream on thechives. In this report ground observations of ice Monongahela River (Fig. 1).conditions recorded at a series of U.S. Army The Corps ground observers use a five-ele-Corps of Engineers Lock and Dam sites along the ment alphanumeric code (Table 1) to describe iceAllegheny Riverin Pennsylvania andtheMonon- conditions each day and send the codes to Corpsgahela River in Pennsylvania and West Virginia and NWS central offices located around Pitts-were compiled from archives, graphed,analyzed and compared to ice data ob- 81 Sao 79'tained from aerial videotapes and Land-sat images.

The objectives of this study were 1) todetermine the annual variability in riverice conditions for selected winters as ob- 4o1 _o

,

served from the ground, 2) to compareice data acquired from the ground, vid- O 6 07

eotapes and Landsat images, and 3) to %e. Ce 3

develop a computer program to graphi- N 2

cally portray the ground data so that 3 2,,

these data, when collected in the future, ,J I'l 4

could be quickly displayed and dissemi- 40 o 1 .. .

nated as an aid for navigation during -oa P) E NN

the winter. This study was a part of the H

CRREL River Ice Management (RIM)program, a program that examined .. N A

several riversin the United States where 0fice causes winter navigation problems.

390

40 0 4

0mi

DATA SOURCES, COMPILATIONAND ANALYSIS AlleghenyR. MonongaheloR.

( Rimerton Braddock Q

Ground observations @ Mosgrove Elizabeth (I(V Kittanning Monessen S

Ground observations of river ice con- @ Clinn Greensboro aditions were routinely obtained from T Freeport Pt Marion Ieight U.S. Army Corps of Engineers Lock @ Natrona

and Dam (L&D) sites on the Allegheny i Acmetonia

River and nine L&D sites on the Monon- 0 Sharpsburg

gahela River, and occasionally from threeNational Weather Service (NWS) sites Figure 1. Location map (circled numbers are L&Dlocated above L&D 9 on the Allegheny numbers).

Table 1. Corps of Engineers alphanumeric ice code.

Amount(coverage) Type Thickness Structure Extcn(

0-None R-Running (floating) In inches B-Breaking In miles1- Scattered A-Stationary H-Honeycombed upstream2-2 tenths P-Stopped T-Rotten3-3 tenths J-Jammed L-Layered4-4 tenths F-Formed locally C-Clear5-5 tenths S-Shore6-6 tenths7-7 tenths Examples:8-8 tenths9-9 tenths 1 S 1/2 T X means scattered shore ice, 1/2 in. thick, rotten and extending an un-10-10 tenths, known distance upstream; unknown data in any category are shown as 'X"; 3 R 2 H 4

full means 3 tenths of the river is covered by running ice, 2 in. thick, honeycombed, andextending 4 miles upstream.

Table 2. Partial record of ice conditions on the Monongahela River, January 1985.

Date Braddoc,. Elizabeth Meonessen Maxwell Greensboro Pt. Marion Morgantown Hildebrand Opekisa

19 7F 1/2 CX20 1F 1/8 CX 1F 14 CX 9A1 CX21 9A1/2CX 2F1/2CX 10A1 CX 10A2CX 1OFI CX 1OF1 CX 1OF2CK 1OF2CX 10A2CX22 10A1CX 6R1CX 10A2CX 10A3CX 1OF1CX 1OF4CX IOF4CX 1OF4CX 1OA3CX23 IOA1CX 5R2CX 10A21/2CX 10A31/2 1OR BX 1OF5CX 1oF5CX 1OF4CX 10A3CX24 10A1 CX 5R2C10 10A3C18 10A31/2CX 1R1 C2 1OF5C11 10F4C6 10F3C7 10A4C1425 9A2C5 6R3C10 10R3L18 10A3C22 1R1B5 1OF5C1O 10F4C6 1OF3C7 1OA3CX26 9A2C5 6R2CI0 10R3L18 10A1C22 10A1 CI 1OF5B1O 10F5B6 10F3C7 10A4C1427 9A2C4 2R2C10 10A3L18 10A3L22 5A1B2 1OF5C10 10F5C6 10F4C8 10A5C1428 8A2B2 2R2C10 10P4L18 10A3L22 8R2B3 1OF5C10 10F41/2C6 10F4C8 10A7C1429 no ice 5R 2 B10 10P 4 L18 10A 3 L22 10A 2 L5 1OF 4 C10 10F 4 C6 10F 4 C8 10A 6 C14

burgh. The data are then issued to users by are shown here. Other methods have been usedcomputer modem and are archived at Corps and in the past to graph river-ice conditions (Bates etNWS offices as chronological listings of the ice al. 1968, Michel 1971, Starosolszky 1985, Cana-observations at each of the sites (e.g., Table 2; dian Coast Guard 1986).Appendix A). The data, however, have two major Our review of the Corps' ice code (Table 1)omissions. The ice observers at some sites often indicated that most of the information given candid not collect data on weekends, and they fre- be displayed graphically, although in preparingquently could not determine how far upstream a the hand-drawn graphs (Fig. 2a and b), it wasparticular ice type existed. We hope that these necessary to drop the ice structure element of thedata gaps can be reduced in the future. Although code, and to reduce the number of amount andthese ground observations are available begin- type categories for the sake of readability. Amountning with the 1961-62 winter, the records for the was reduced from eleven categories to four: 0seven consecutive winters from 1979-80 to (area clear of ice), 1 through 5 tenths (10-50%),1985-86 are most complete and are used in this 6 through 9 tenths (60-90%), and 10 tenthsstudy. (100%). Type was reduced from six to three:

Since it is difficult for a user to visualize and running or floating ice; stationary, stopped,understand the distribution of ice conditions jammed or formed locally (any one of the four);from tables, we developed a way to graph the and shore ice. We also included discharge and airdata. Graphs ofice observations for theAllegheny temperature data to show the relationships be-(Fig. 2a) and Monongahela (Fig. 2b) Rivers dur- tween temperature, discharge and ice conditions.ing the 1985-86 winter that employ our method

2

Aerial videotapes Landsat imagesVideotapes (1/2-in. VHS) of the rivers were Five Landsat satellites have provided images

taken vertically with a Panasonic 777 video of the rivers since 1972. Each Landsat has twocamera fitted with a 12:1 zoom lens. A Cessna imaging sensors: either a Multispectral Scanner172 fixed-wing aircraft, flying at an altitude (MSS)withanInstantaneousFieldofView(IFOV)between 2000 and 3500 ft above the river, de- of approximately 260 by 260 ft and a Returnpending on cloud conditions and the width of the Beam Vidicon (RBV) with an IFOV of either 262river, carried the camera. An experienced ice by 262 ft or 131 by 131 ft, or a MSS (same IFOV)interpreter viewed the tapes on a TV monitor and and a Thematic Mapper (TM), with an IFOV of 98visually classified ice conditions into six units by 98 ft. Gray tones and patterns in river ice are(Table 3) that were readily identifiable, that most visible to the eye on images from the 0.6- tosatisfactorily described the range of ice that 0.7-jtrm MSS, 0.580- to 0.680-ltm RBV (Landsat 1usually occurs on these rivers, and that did not and 2), 0.505- to 0.750-prm RBV (Landsat 3), andrequire ground truth data for verification. The 0.63- to 0.69-jtm TM (Landsat 4 and 5).interpreter did not *ttempt to infer characteris- Images of the same location were taken everytics from the tapes that could only be measured 18 days by Landsat 1, 2 and 3. When more thanon the ground (e.g., porosity, strength or ice one was operating simultaneously, images of thethickness), same location were taken about every 9 days.

Bound'aries between the units were mapped During simultaneous Landsat 4 and 5 opera-and the area of each unit was measured. For tions, images of the same location were takenunits comprising both ice and open water-solid every 8 days; images were taken every 16 daysice cover with open-water areas, fragmented ice when one satellite was operating.with open-water aas and ice floes orfrazil slush We anqlyzed black and white Landsat filmand pans-the surface concentration of ice was positives (9 by 9 in.) using traditional photo-also visually estimated, graphic interpretation techniques. No special

computer enhancements or analytical techniqueswere used (Gatto 1985). Reaches of the riversappeared as black, gray or white with textures

Table 3. Ice conditions as observed on vide- and patterns within these tones sometimes ap-otapes (from Gatto et al. 1986). parent, but the subtleties that differentiate the

six ice conditions that are visible on videotapesMap unit Description were not apparent on Landsat images. To deter-

mine which types of river ice usually producedRiver is ice-free, no ice apparent. these tones, textures and patterns, we compared

ice conditions shown on aerial photographs (GattoRiver is completely covered (100%) and Daly 1986) and videotapes taken on dates as

S with ice; no individual ice pans, close as possible to thosefor which Landsatimagesblocks or chunks are visible; ice were available.may be snow-covered. These comparisons show that when the river

Ri ver is partially covered with appeared black on an image and had no discern-solid ice (as described above) but ible textures and patterns, the river was open (icehas open (ice-free) areas, free). It is possible, however, that thin, transpar-

ent ice, which appears black from above andRiver is completely covered (100%) cannot be distinguished from open water in

with ice that has distinct, variably Landst imagesicovered pror of parisized, individual ice pans, blocks Landsat images, covered part or all of particularor chunks. river reaches in some instances. Ice conditions

that appear gray on Landsat images can varyRiver is partially covered with from fragmented ice (usually thin) with large,fragmented ice (as described interspersed open areas to ice floes, pans or slushabove) but has open (ice-free)areas) mixed with open areas. The gray tone usually had

a patchy or mottled appearance, or showed tex-River is primarily open (ice-free) tures or patterns.with floating ice floes, slush or When the river appeared white (or nearlypans. white), ice conditions could vary from solid to

3

06

. -A,

i

0 00 -0

oail si4(/4 ) abvqs'

6 o

%/1pD49 o a.oa a/AO

t: a .

I-N 0

Nd rW/EN

O' N

0006"r~ A.A.

qd, ac ,o 6n0 is .w ?,aOdy aja S- anIOj ON

- C,

.. p p 41

- ~ iii

- ~:2

0! 0 0--D a,

amso sa-v 06,4 ~6

0-0 0

0.I N

5-2

0~ K.0c5 -II -

fragnented ice (usually thicker than gray ice). A or partial ice cover with occasional occurrences ofwhite tui,e cuuld include battered open water upen water throughoutthe winter. Acomparisonareas that are smaller than the Landsat sensor between complete and partial ice covers indi-IFOVs, or fewer open water areas than occur cates that, on the Allegheny River above Rimer-where a gray tone is observed. A white tone could ton, a complete ice cover occurs approximatelyalso mean that the ice was snow-covered. For during 75% ofthe total days when ice is reported.example, thin ice in a Landsat scene may be In contrast, below Acmetonia, a complete icetransparent, appear black and be classified as cover is observed during only about 27% of theopen water. This same ice cover viewed after a total days. On the Monongahela River near Ope-light snowfall would appear white. kiska, a complete ice cover occurs during about

70% of the days when ice is reported, and nearElizabeth and Braddock, about 21%.

RESULTSComparisons of

Ice conditions from river ice observationsground observations It is clear that information on ice type (includ-

The Corps and NNVS ice observations for the ing movement), thickness and structure (Tablewinters from 1979-80 through 1985-86 (Appen- 1) can only be obtained by ground observations,dix A) were examined to determine the dates of although inferences regarding some of theseinitial ice formation and final clearance of ice on characteristics could be made from aerial vide-the Allegheny and Monongahela Rivers. First ice otapes by an experienced interpreter. Because ofoccurred as early as 19 December and as late as the dynamic nature of river ice and the limited20 January on the Allegheny, and as early as 21 view upstream of a ground observer, the groundDecember and as late as 3 February on the observations apply only for the location near theMonongahela. Final ice was observed as early as observation site and only as far upstream as is8 February and as late as 20 March on the visible, although the ice conditions as seen nearAllegheny, and as early as 20January and aslate the dams were usually assumed to persist up-as 4 March on the Monongahela. stream. Sometimes cther upstream observers

Although ice formed on the rivers during all reported ice conditions beyond the view of theseven winters, the severity of the ice conditi-rns observer at the locks and dams.varied each season. Both rivers had the least ice The aerial videotapes give more accurate in-cover in the 1qR2-83 winter, and the most in formation on the area] coverage and extent of1983-84. During four of the winters, ice formed different ice types than do the ground observa-on the Allegheny River earlier than on the tions. Landsat images also show the areal distri-Monongahela, and during all seven winters, ice butions of ice as do the videotapes, but with muchremained on the Allegheny from 1 to 20 days less detail and frequency. We have comparedlonger than on the Monongahela. An inspection data from these three data sets collected duringof Uhe 'otal ni,mber of days that ice was observed 1984-85 and 1985-86 to illustrate their advan-at each of the L&D sites revealed that approxi- tages and disadvantages.mately the lower 20 miles of the Monongahelaand the lower 10 miles of the Allegheny River Winter of 1984-85have the fewest number of days with ice Ground observers reported ice on theAllegheny

The type and structure of ice given in the ice Riverfor49dayzi ium10Januaiy L25Februarycode (Table 1) made it possible to note the times (Fig. A6) and on the Monongahela River for 37and locations of ice jams and the frequency of days from 14 January to 20 February (Fig. A13).running or stationary ice throughout the winter. Ice was observed on videotapes taken of the lowerAlso, we could statistically summarize the per- 7 rilesoftheAllegheny River on 11 days from 23cent of ice coverage on the rivers. January to 24 February. A 28 February tape

Ice jams were recorded on the Allegheny at the showed no ice. Ice was apparent on videotapes offollowing locations (Fig. 1): above Rimerton in the lower 66 miles of the Monongahela RiverJanuary 1981, above Mosgrove in March 1982, at taken on five days from 28 January to 24 Febru-Parker in January 1985, and near Natrona in ary. A 16 January Landsat image was the onlyFebruary 1985. An ice jam was observed on the one taken this entire winter when ice was pres-Monongahela in January 1984 at Maxwell. ent. There were no days this winter when ground

Ice on both rivers is generally in motion; there observations, videotapes and Landsat imagesare frequently changing intervals of either solid were acquired on the same day.

6

Ground Observations

LLAD?: IORI(1L&D 3: 9RICXLAD 4: 1oRICXL&D 5: 2RICXLAD 6: IOP4LX

L&D 7: IOP3LXLAD 8: No Observation - tL&O 9: 8RICX

I «"

5 0 5

<23 D 5

k 3r

Figure 3. Ice conditions on the Allegheny River on 16 January 1985as observed by ground observers and on a Landsat image (dashedline is gray ice, solid line is white ice).

The 16 January Landsat image showed that The gray ice apparent on the Landsat image70% of the Allegheny River below L&D 6 was was composed of this thin, clear, moving ice,covered with gray ice and 30% was open (Fig. 3). while the white ice consisted of the thicker,White ice and gray ice covered 88% of the river layered ice that was stopped. When used to-upstream of L&D 6 to river mile 72, while 12% of gether, Landsat and ground observations pro-this section was open. Ground observations made vide details of the ice and its extent upstream noton 16 January at the four L&D sites below L&D available from either source alone.6 showed 1-in.-thick, clear, running ice covering The 16 January Landsat image showed only 620-100% (average coverage 80%) of the river miles ofgray ice on the Monongahela River abovesome unknown distance upstream from each Opekiska L&D. The ground observation at Ope-site. Between L&D 6 and L&D 8 was 3- to 4-in.- kiska L&D showed shore ice, 1/2 in. thick andthick, laycred, stopped ice covering all of the clear, covering 70% of the river some unknownriver and extending upstream an unknown dis- distance upstream. Ground observers also re-tance. Above L&D 9 (some unknown distance) ported 1/8- to 1/4-in.-thick, clear, locally formedwas 1 -in.-thick, clear, running ice covering 80% ice and shore ice covering 10% of the river for un-of the river. known distances upstream of L&D 7, L&D 8 and

7

.. .river covered with stationary ice,/ 3 in. thick and layered, and ex-

Ground Observation tending 22 miles upstream (Fig.,1A3L?2 4a). The 4 February tape shows

27% of this reach covered withsolid ice, 62% covered with frag-

N 2 mented ice with interspersed

open areas and 11% being openwater (Fig. 4b). A Maxwell groundobserver reported on 4 Februarythat 100% of the river was cov-ered with stopped ice that was 4in. thick and layered, extending

--- 22 miles upstream.For the first 5 miles upstream

of Maxwell L&D, the tapes and

a. 28 January 1985. ground observations showednearly complete ice cover on both

dates, with the ground observerreporting stationary ice on 28

_ _ _ -January and stopped ice on 4, . February. This suggests that the

/ ice wasmovingbetween 28Janu-Ground Observation ary and 4 February, which would

explain why the 4 February tape(Fig. 4b) showed more frag-mented ice than the 28 Januarytape (Fig. 4a). As with Landsat

S-- and ground observations, the

videotapes and ground observa-tions are also complementary andprovide a more detailed view ofice conditions than either onealone.

Winter of 1985-86Ground observers reported ice

b. 4 February 1985. on the Allegheny River for 63

Figure4. Ice conditions on the lower 5miles ofMaxwell Dam po o l, days from 19 December to 19

Monongahela River, as observed on videotapes and by ground February (Fig. 2a, A7) andonthe

observers (see Table 3 for definitions of ice symbols). from 26 December to 1 February

(Fig. 2b, A14). Videotapes wereMorgantown L&D. No other ground observa- taken ofthe lower 17 miles ofthe Allegheny Rivertions were made. It is not surprising that this and of the lower 13 miles of the Monongahelathin, clear ice below Opekiska L&D was not River on 9 days when ice was apparent from 28apparent on the Landsat image. December to 28 January. Landsat images taken

Ice conditions 5 miles upstream of Maxwell on 3 and 19 January and 4 and 20 February wereL&Don the Monongahela Riveras observed from not useful because the ground was cloud-coy-videotape and the ground were compared for 28 ered. The only Landsat image that showed iceJanuary and 4 February. The videotape from 28 wcks taken on 8 March 1986, after the last vide-January shows 69% of this reach covered with otape was taken and the lastground observationsolid ice, 28% with fra'mented ice with inter- was made.spersed open areas and 3% open water. The The 8 March Landsat image showed gray iceground observerat Max'vell reported 100%ofthe on 92% of the Allegheny River above L&D 8, on

8

Ground Observat ion

9PII-

, a. 28 December 1985.

!~~~- 0 ." ¢ 3 P. .1o - A 1I n R V ¢ W . I /I

Ground Observation

9AICX/

n a a-in-a ~ .0:eco r C.C

b. 8 January 1986.

1_1%- o 03 Pvool AI.gh- River / Nevesrqc.8 *

.9

Ground Observation

C;

c. 22 January 1986.

Figure 5. Ice conditions on the lower 2.5 miles f L&D 3pool, Allegheny River,

as observed on videotapes and by ground observers (see Table 3 for defini-tions of ice symbols).

9

.... m at i l m n mnlii D ~I

32% of the Allegheny at L&D 4 pool, and on 11% tance upstream. On 22 January (Fig. 5c), videoof the Monongahela River at L&D 2 pool. Since showed 39% covered with ice floes, slush andno ground observations or videotapes were taken pans, and 61% open water. The ground observeron this day, we cannot compare them to the reported30% coverage with runningicethat wasLandsat-derived data. However, we can compare 6 in. thick and breaking, and that extended 9data from videotapes and ground observations miles upstream.from other days.

Ground observers at the Allegheny River L&D Computer-generated graphs3 would have a visual range of at least 2.5 miles It became obvious during preparation of Fig-upstream of the dam, which is the extent o' the ure 2thatbecause of the extensive hand-draftingvideotape coverage for this pool. On 28 December required, use of the future ground observations1985, the videotape showed 82% of this reach would be limited. To expedite preparation ofcovered by solid ice with interspersed open areas, graphs of future data, a computer graphics pro-4% covered by ice floes, slush and pans, and 14% gram was developed to use the same ice codes asopen water (Fig. 5a). The ground observer re- were used to prepare the hand-drawn graphs. Inported 90% of the river covered with 1-in.-thick, the computer-generated graphs (Fig. 6; Appen-clear ice that was stopped, and that extended dixA),theorderoftheL&Dlocationsisreversedupstream 9 miles. On 8 January, the videotape (see Fig. 1), the ice code symbols are slightlyshowed 4% solid ice, 33% fragmented ice, 37% different (see Fig. 2), and ice thicknesses werefragmented ice with interspersed open areas, not included because of space limitations. Theand 26% open water (Fig. 5b). The ground ob- use of a multi-colored diagram will allow thick-server reported a 90% cover of stationary, 1-in.- ness to be added (Bilello et al. 1988).thick, clear ice that extended an unknown dis-

ILJ L&D

HILDEBRLNDEN1>L& n

MORGRNTOWNN

1'-"-- L&D ,LEGEND

PT MRIONC Openwade

l&P IrIO . Dsunce upsream unknrown

GREENSBGRO 15 tenths. Stabory IM

] " 6-9 tenths, Stationry Ice

1 to ,rd . SWWWoWY m

MPXWEW I -S tenths, Running i~

so L&CL 6-9 tenths. Runng ice

10t tenths, Aunnint; iCMONESSEN 1 elm unn

E L IZPBETH ILI"'

' --- BRRODIOCK '

L&0#2

Figure 6. Part of the computer-generated diagram of daily iceconditions, Monongahela River, January 1985.

10

SUMMARY AND CONCLUSIONS combined with ground observations, the two

methods provide an excellent means of recording

The river ice conditions on the Allegheny and and analyzing river ice.Monongahela Rivers were highly variable, as Landsat imagery has the advantage ofprovid-shown by the graphs of the ground observations. ing images of large reaches of a river that can beThe observed ice was largely in motion, although easily interpreted. There is a good data base ofthere was much stationary ice and major periods usable images starting in 1972. Disadvantagesof open water. The graphs provide a convenient are the infrequent coverage, the obscuration byway of showing these wide variations, in space clouds and poor resolution of the images, whichand time. limit the level ofdetailed information. Thin, clear

Each method of observation--ground, aerial ice, for example, is often undetected. Ice condi-video and Landsat-has certain advantages and tions determined from Landsat are recorded asdisadvantages (Table 4). Ground observations either white or gray in tone so that ice detailshave the advantage that data on thickness, that are obtained by either ground observers ormovement and structure can be frequently ob- aerial videos are not apparent from Landsattained, and, generally, ground observations are images.not affected by the weather. The major limitation Despite differences in the detail obtainableof ground observation is the line-of-sight of the from the three methods, they generally agree onobserver, which is often no more than several the overall extent of ice coverage. For example,miles. Given the wide variability of ice condi- the total percentage of selected pools covered bytions, this limitation can be critical. ice as determined on selected dates is shown in

Aerial video observation has the advantages Figure 7. It can be seen that, except for 16ofproviding detailed views oflarge river reaches, January 1985 on the Monongahela River, theatfrequentintervals, andatreasonablecost.The methods are within 15% of each other. Thevideo image is relatively easy to interpret, but Landsat observation on 16 January 1985 (Fig.training or experience is essential. The disad- 7b) indicates much more ice than the groundvantages are the lack of ice thickness and the observation.adverse effect of bad weather, especially low This study has illustrated the importance ofcloud ceilings. Given these restrictions, aerial three observation techniques for monitoring river-video is perhaps the best means of closely observ- ice conditions. Each method provides useful dataing ice conditions on large rivers and, when and, when analyzed together, they give a more

Table 4. Advantages and disadvantages of the three datasets.

Advantages Disadvantages

Landsat - Synoptic view of large - Poor IFOV gives limited,reaches of the river not detailed information

- Good data base of images - Infrequent acquisitionsince 1972 - Cloud cover can obscure

river- Easy to interpret images - Snow cover obscures ice

Video - View of large reaches of - Cannot provide icethe river thickness

- Good IFOV gives as much - Cannot acquire tapesdetail as is required if cloud ceiling is too low

- Easy to interpret, but - Snow cover obscures iceexperienced interpreteris required

- Frequent acquisition

Ground - Detailed ice data - Limited horizontal view- Frequent observations - Data quality depends on

observer- Not weather-dependent - Data must be graphed to

be useful

11

00 L&D 3 Pool L&D 3 Pool

90 Below L&D6 VI3EO

80

70I-.Z 60wC) 50X L&D 3 Poolw. 40

30

20

10

0116JAN85 28DEC85 8JAN86 22JAN86

a. Allegheny River.

Above Opekiska L&D Maxwell L&D Pool Maxwell L&D Pool100-

90 • GFUUND

8• VDEO80 LANDSAT

70

Z 60w(L) 50-

CL 40

30,

20

10

0L16JAN85 28JAN85 4FEB85

b. Monongahela River.

Figure 7. Percent of river ice cover as observed on the ground, from videotapesand from Landsat images.

complete understanding of a dynamic river-ice LITERATURE CITEDregime than would be possible with one methodalone. Bates,R.E.,D. Saboe and M. Bilello (1968) Ice

With the computer-graphics capability devel- conditions and prediction of freeze-over onopedforthis study, there may be increased use of streams in the vicinity of Ft. Greely, Alaska.the ground observations if they are quickly USA Cold Regions Research and Engineeringgraphed and available for rapid dissemination Laboratory, Special Report 121.where navigation on ice-prone rivers throughout Bilello, M.A., J.J. Gagnon and S.F. Daly (1988)the winter is required. This potential for ex- Computer-generated graphics of river ice condi-panded use of these data may resultin the receipt tions. Hampton, Virginia: Science and Technol-of better and more complete information from ogy Corporation, STC Technical Report 2280.the ice observers.

12

Canadian Coast Guard "19S6) Ice maps for lakes. USA Cold Regions Research and Engi-Trois Rivieres Sorel et Grondines. Quebec, Can- neering Laboratory, Monograph III-Bla.ada: Garde Cotiere, Bureau des Glaces. Starosolszky, 0. (1985) Ice and river engineer-Gatto, L.W. (19S5) Ice conditions on the Ohio ing. In Developments in Hydraulic Engineering-and lllinois Rivers, 1972-85. IEEE International 3 (P. Novak, Ed.). New York: Elsevier AppliedGeoscience and Remote Sensing Synmposium Science Puhlishers, pp. 175-219.Digest, 2:856-961. U.S. Department of Commerce (1985 andGatto. L.W., and S.F. Daly (1986) Ice condi- 1986) Climatological Data, Pennsylvania-tions along the Allegheny, Monongahela and Monthly and Annual Summaries. Asheville, N.C.:Ohio Rivers, 1983-84. USA Cold Regions Re- National Oceanic and Atmospheric Adminstra-search and Engineering Laboratory, Internal tion Environmental Data Service, National Cli-Report 930 (unpublished). matic Center.Gatto, L.W., S.F. Daly and K.L. Carey (1986) U.S. Department of Interior (1985 and 1986)Ice atlas, 198.1-85, Ohio, Allegheny and Monon- U.S. Geological Survey-Water Resources Datagahela Rivers USA Cold Regions Research and for Pennsylvania. Water year 19oo5 ai d 1986,Engineering Laboratory, Special Report 86-23. Allegheny River at Natrona, Pennsylvania, andMichel, B. (1971) Winter regime of rivers and Monongahela River at Braddock, Pennsylvania.

13

APPENDIX A. ICE CODE RECORDS ANDCOMPUTER-GENERATED GRAPHS OF DATA

LEGEND

C Open waler

Distance upstream unknown

] 1-5 tenths, Stationaiy ice

6-9 tenths, Staltionary Ice

1o tenths, Statonaiy Io

[] 1-5 tenths, Running ice

] 6-9 tenths, Running ice

10 tenths, Running ice

15

Allegheny River

19 79-80

t; z

HICKORY Ft

1401F

FRANK1LIN I l

120j

PARKER . . . . . . . . . .

RIMERTON62 L&DJ49

MOSGROVE

20

FigureTAl

&r)16

-P 2452 4rw7 j"I 49(2 "'No"02 (2:I0 97'92 95550'j P 1769'04 999 FR992114 U. HC

991(3

'91 111672

62E 41D

-S RID9 222p 61'4IC 521( i D(3

'25X ER 8'X 91 4

b 81 2( 921(9X 1299(9 lGJ6L3

IO 192( 12D3IOS9 1Df

21 1002(3 1092(9 191(3 '095(2 1tL63

-13 '42(3 292(3 103911

2.2, 644( IP!X lo~x 9 1022(30261

2,13 5(4(9 1091(9 1092(9 1096(9) 12151

?1 69499 594(3 100( X,6( 7913

42929, 294(3 10A92

21'3 1791(9 1094(6 1093(3 1096(9 1029L9

220 1(9 094(6 1093(9 1095(9 113L

OREN96( 9426 1092(3 1095(9 02j9692.2 9426 09LV4(9 1 Il 99

P.33219 103912:24 10219N/2 134( pi 913

2'61094(7 102u9L X

129 3p1(3 l094(2 1013(3212q 991[9 3p(3 [. 991(9lpie 124( 109133/ 1091(9 392(9

192(f9 ID226 109532 I037I0m33:5 9(19 9926 109438 10113B

3"s 7P2'6 109498 10110193/7 10F432 1032013/9

3.10 '9419 39223 89419 12512

!:12 59419 SRO(8.' 3 5119 3221913,4 25419 11(59

116

3,1!8 2S419 24

8/22 '5419

17

1980-81

ww, P7 vl

T'2

FFigure A

22 ?? ?71?

'q' ER i C-*Oh)1 McIRCSN FR1EEPORT0.0 CLINTON OPIIN!T(A 3'i 81 8CNIJ PNONE -RaN1,14w'N

':OF

'05.25.

05FHX

6p~cx OF3LX101 6110

9 i~plx !FIL6!Oflclo !J30:0 5DI20L

IOPIC !Y416 T~fCIO I. ZG( I!V4. 00 135.20

!OP9201'05.2i 057800

/I 09505X 105.L6 l05.3(1

'0971 05.76 rareco207TX~10 101f 11 11

1/3! 5991 0.71 10.1 £91031

361 '091 9.0610.01

Z09*2 MIX 70 105.8(

213 09701X 05.716 05.80

2192 1097L2 1066 O305.9(10T

11:25?2179 7.1 159

217 117119 75.616 10.913 SF1

:/29 65.16 05.(

130 I13 03. 109180 13 946 1m40 70.7.1 1021010SXXlaa

21 1081(MI212283 183.

23185111( 8(

2/N 1805. 1085.1 80 19

1981-82

I T I II

?I,

4,;4

L&D*34

5rHRRPS BURGL&D42

FVigure A3.

20

p, o

2 a 3 a UB

-

m~~ ~ ~ ~ szm z g 9a

21

1982-83

i Fi

PRhRKER

RI MERTON UL

MOSGROVE

KIT TANNING

L&D#6 T

FREEPORTL&0#*5

NRT RONAL&0#4

20?ACM1ETOM IA

L&D*3

SHRRPSOUR GL&IJ*2

PITT. PA.

Figure A4.

22

pl R, 4S7 X '1xxX

'S -4'>F 2SI X irS!A

K35! x

*p 239'

:p1l>23'4

Y Ex'2V V

'3

23

1983-S-1

18 HICKORY . 'ir

FFRRNWLIN i :

00

PRRKER

802;? ?? 97

RIIERTON60 L&D*9

h1OSGROVE-A 1

Figue A5

KITTRNN24

7 V'f k 40Y9 2,~'4! '429 go 699 I 2.0IEP 410 ON '49 m "6 -S6909 VR5 O R 92F90 I9 Q441. 4 WICK

~~'111 24?L Is,2 "2> 2244'77124 29> 42' :~2< 0A204 09405 '29AF

7: ' IS!9t'>il c

-9 &L - 240 '09005 10j ; F1

31' 'P. 7PX0 ,Pr '2A 10 t !08 iM<

* P 'P ( IPC .. ~c !2 '90;.4 )09004 !S4 C9I l

'2930!MS io4x '244: 'OPIoS 12249' 20204OPEL" 17 '041 340 2s2

- 24706 IOP716 ''4131 I'f5L 1093 1029 7 ! '0KL!'5G1 l 230 6917C 2X'2400 '095'084'25 791041L7 OLX21X FXP"[ 1040 30.<

592 'CX0 '0945 209310 '0402. 94103 1019L? ''O5 024

'5 MEN0 IQC'5 6 13( OM0 1995[5 99200S 10190 '723 444101X3" < < MEN6 MENo 094(4 0083 SOL M '1144L 25910 09114

212034 '961 '9

* 1910 69 16 9(4 1050 014593 IOU87 1041104 25140 09144x

10411:

2.45

1984-85

W Li 8Jt7,-iT

1401

j NIRTORL&5*9

200#

L&D*2

NFigure A6.

226

"P9r[ AMPS32 J0'3T0802 883208 58(09389 N77 W05 705F20(5 ?1-089 00~~ U2817'W - 11,1

0 32303 '0230

'i 3203x 3310.8 221x 01 1292iC 8203. 737 310 5'!

3*I1Y R 027 3810x 9080X ;80 1.11 13 3( 22V829 _N g~-1

'2E3103 53173X 02'08X 7818 9iLX '1.3 E8'

*i 7' 0933X 39703 'P308'230':PS9 ('OPEC?7

774[5 .9( '780'700'7( ,0C 'PH9'0

*73 [ 79F(0 IR900 , OF,4t'3 I73( '0M!y r 74 79P7.331900 '236 0~~ 'PS,'

*E :2 [ '7900'90 '_724(6 '78405 090 '3V 713 33 o7

70 *.PIES '104C3 '29405 '7933

''9'(5:Cq[3 '79000

73 7415 29109 '79300 '28406 '792(2S '095(72~x

73 ']94[6 '29403 29005 [ 1P%( ' 9 '09(0'210 21

"I ^9416 'Z9IC9 29000 i29305 1.7900 305403 TV13

77 ic 03403 '03 9 '79005

74 '0405 39203 09005 '08505 0S165 '082

73 09305 99009 29605i 08300S ME3 '2820? '712 33!0 '09138PJ

70 39503L 99009 99HS 02306E 09300 28207

7 39304 29009 99005 '0836 093(0i 820?

7.2 I 7537 0909 99000 108505 093(0 '3204l '3133 09122

7. 1300[ '2350[9 9900070 391^6 '29009 997[6

IN' 0910 859 29105 108305 595(5 73204

I80 HE B 432IR56 EMS 85203 !09122 73182

2.14 38l729 29305 '0805 084(5 3103C

2,5 870 29'03 1083(5 309115 3023( 331XM

7 S 2105 29892 P 3907/37 283056 22709 3970051 i

713 3A860 3870 294T3

217.2 2 23

27

1985-86

85/8' . .- .-

w8 ES7HICKORY II

1I 4

FRRNKLT ?I

122

I ERTON 22L&0#

MOSGROVE

K! T TRNNT NG

CLINTON

F RE EP ORT

NRTRONR

20 ;

ACMTO'1IAL&D#3 7

SHRRPS~uRG

PITT. PR.

Figure A7.

28

- N l,.w' !991162 932' '9<3L WI 'l91x 127 111

-. '3'1Y 12<6, )'[2O03 9?7 IA ~ SM

''s 9C[ pcx 9C'>RC FE2 9i, 9P0 qf~

*p" 2 pN ~ ' 26 512b 9p733 '2387 "5> 9$X o

*, 1 3

1 222 S2:33 Ru R230-9398 SX sfa S1

3/ 3 I3923 '423 '29326 WES2 29238 2907 VIM( 3222 22D

'4l" 93323 IR329 2336 I323:633228 3L7 1136OX '03223

!"s 2 10393 1316 356 POS 227 342 P3332

3 3 ' 9 22 92223 9PI 32I6 923(6 99409 923(27 993(3X 39553 )?

136 9323e 992M23 31 922( 936 9R2( 9 21&(7 914[9 9P2t3

123 42226 MIN( 806( 3538? 394(9MIOS 99653 3 15333

'/8

2.19

21 32632 9263 923 206 929 92C68 92362 233323

3I8 323 32639 323M 222(6 3588 228? 568 25359

261 12316 IH R SL

327 35792 5~293 568 393

Monongahela River

1979 -80

20

BRA DEBJOC

FigreA8

MORGRNT30

5E1 MUMCC EL25E! H 4OWES5[M "4D5L; 35!ISBOD PV. 4kR 5P7TN NLJBRRHO YUEXSIR

23

7

31

':3S 1FIIIRXX IRX

',,S~~~~053 10M53510i~lol 192 BM IR~

DuDX 1053330 3*3 51I l~lX 05235 10064

,/9 WO105735 0 33 13011 30533MAR 10011

1/0OW G2S RTU 99 I53 P2351 325317/21 55K 3~s91 SI3 lRX 93 4

Z/17qlx IAX il 10 3 1f2X 1OAZ/13 lpx DII GI1 OM lRX o3

2/1 o,2laI GI11 7245A~x 1R 1

DuDS IHX 0lxl SI2 19 0 1

"i7 OMu25a~X 0 1

21/I

z/21

3/31

1 9S0-81I

--- - - - ------------------ - - - - -

-------------------------------

- - ----

Figure A9.?

32 ???

R~' Rr42121' 1.12 3 94221 4t Pi 3S322O ' 9t"A'J9 4t2204 )2K2684

: 6 Y! '2

'''6 '2228I

2p,'16 'FP2

P.* ?,,6 1092V

P2, P 2381's 21'''' 2''6 X12

'4x '222 :F2'2 C1901 3 '224' '924X6 122>'

*ix inpl 0. l94 9"Kx 22..26 '33455 '92262314

'S 332 94 :93223 '423 6 19446 :224 6 1003012R!X3 0263X '923 23 39385 '9226 36 10O441'9462 l02632 '123 3 92222X '0444X6 10146 1024 12

* 3 363 0622 '00323 1021AS '92436 i026 6 1044 10ORS '3 0203 106 1093 23 100! 4 '02636 '026 X6 10242'

* . 3632 :932 100323 %!AS3 '0429% 3 10 24212Dixl39503 '2222 '943223 39126 10N43 '02526 102421

22P8x 932 2 ~ l '223 194 2 2126 :042 1236'4123 292 2 '0283 A0M331 :924 6i 04 6 02f4 2''2 2240 x '983 19236 9125 102416 1024 6 '02448

12IX I 42 102 9 43A 919 104316 102l326 !008I16 '942 104333 '033 f6 321 6 10230 I0236 0265

'9' 912 X 022 '033 10436 636 66111

1 '23 794 23286 561 1

'301923 '912 T0V2 2 6 sx

0'8?X 10112 940 6RI 1021 6 0914

2.4 101 TO2 9P26 1021 3 1014I23 DI221

2/6 104235 104023 99126 034 2 1804 1D36 102831241922' 04203X 99126 1024 3x 10034i 1023 6 1090214

OPP 3' '39232 99356 09633x 124 4 0336 ARM 12 09322X O211 p2 104603K 10424 104336i :109226

'022 R 0222 992n 104222 103 106 1043232.'1 10203 33131 9fi 6 042 10 2323 1028 6i 32313OR

2I'N'N32 99186 103282 100213 19.43 A 9431226 10203!f v13s U321 ?0.336 102201

19211 13186 100111218 S2211

7,'9

2-29

33

1981-82

t2 ~OPEKISKR

HILDEBRRNE -

'MORGRNTOWN I

to L&D

PT. MRRION

GREENSBORO

60 -4

60 RIXWELL

j0 L&DT

MONE SSEN lI i40 L&D#4

ELI ZRBETH 'i

20 LD3V

BRAIJUOCK V

Figure AlO.

34

'W* i~il NH PL 1%~ t '547p -Re ".5

o

.4

1 s y3P14 4

,10 18,21.2 VMS4 'Cqr5 !cs 100:ls IS. '5 '34X 'Op AX RT23 9RiM CR235x 44235c ifq36 100253

'' 234 2'RX C0R40 '01R23 19RO( C234x CR4 I NN4 CR33515 '5f35 35334 '25334 'CSCLC3 71295 CR334X 'CR235 as4CS CR237

'9 44445x CR954 CR423 59346 'CR685 CR44 1CRIRS 'CR49.3 MI3C &334 'CR534 'CR4339 3R396 339R CR435s 00CR46 309434"I 48RC54 COR55 994414 99 CR9409 CR445 CR406 10944522 ~ 45X CR545 91)(0 COR549 138595 09I OR96 1f3RR

234Rxx 'CR':44J 49159 92356i CR549( 109545 COR559 52139.4 496439 49754 29545 794*'

:6 914 CR94OM3 2149X

3929 39'95 3195l 45396 SF339 SFE3 SF334 3V3C4

101'1R24 39345 39155 5513* 9 '9234 3939630

'31

35

T-4

4- b

IH LLj~

P>< ri-- ~ ± 93.2 l I 'i c+

Iiur A 11

3.2#' 7 F 7 t-36

237

2,2

t t

L? ?

I Ij

Fiur A [2

38I 1 I

'F'3 Y02EX

* ~ ~ ~ : Y7 HM'Y~ r ,

2 ~ ~ W3 1.!. YC rr

E X: I' UH-1 L'A iSE Y t m!

:2:' YM or!> f ML:sqi' KS x ' FIC ':3P 0~FrzCC

9PL :93s' x ioraC x C2X I f'CX 'MY arc~

'*PI ' 5 'OPEL - ':r'> !O:r-:ix o3( N ir[

:7 !Pe 1'7P'. losa *2'>e ! 3E CSle 'R'X O39 larrtp

'A .9'7 !L 29329 79CM I 97'! 7[x 22929x WHY2

2 92E9 P4L!Th~ 39'£ J"9 23 039 '3C

1984-85

ILa L;1

_&D _

MORGFRhTON

PT. MRRION 12-AL&D#8? ?? V

EL I A ET

L&IJ*3V

BV VR 0 CCL&D#J X

PITT. PR

Figur A13

40?

--PAC.'c -A :F[ q .' w 't "ji53Qff ~OR "Sol IN~ '26545 ! 1 5

tEN

F -Z E'M: X Z': ''7 Y v 21> '23

-S 'Y:,S V ..:' 2A!" 71 E 275spl 'IF'ES INC,' ,R4['4

'E IEE li JA'9L5 2~'2 9SOW FES 272 7 '25329

12i91 sIE PtS ER: EE5 252 -Hf45 'mac '3R3114siq*8289 310 "P! '274l2 'N41 ''2423 1431

4 -2B £543 'E54'2'NS7 59325 '7947O '3(STE 100SL14s 13S: 47 '''Q E!' 52 ''6' '31'4li 37psl8 '05321

:P9 .4055' 'I'' 31577 '55 ENSES '26412 '2312 575784 725422

ES 4 OSP43 24174 !2Us7 2832 2622.0 '0p2L5 '2921 !NMI75 :;327 5542'? '?M, i '27L 3274S '97273 l232 "fits '0541

ES '620 16'67 '3UM1 229222 36755 7293122 115 '2767 053124I

ED '6115 '5491s l'79 '253122 23385 127485 IONLS 707515 O0A1i3' 325 1646121 '295:9L! 293372 2622 727466 707563 IO31s 72551

'295620al '0P367 16115 36262 87315 207312 0751309582cse '27368 2Q115 27315 27315 0748Xinn 2726 0394 M612 07283 27f315 208305 '07585

2" 51116 3413 2'29sh 6'23a0 19284 67222 SF311 209361 0748"76 'Pig5 '4873 'C0566 0936213 67284 237224 '0751 209352 227483

'2 '181 '0586 1963810 122 IT8129I 2031 70931 066EARso '255 9310 18284 709161 707312 2933 91386

i7492P~3 126383 AR2M 20312 66386

41

1985-86

cm 85/86 ......

OPEKIXSKR

HILDEBRRNDL&ID

M-ORGRNTOWN 06

PT. MRRIONL&D#8

GREENSBORO

L&D*3

MRADDOCK V.L&D

Fiur A14.S

4 L&D42

:~E~8222 7 !39 04(22(4 99122. 12458080 P'. 99 "A "2J 184L0mi Z92h2SX4

TEXx928

W '9 91, .311120 RIES5 ?F'?2 Ell!5 9, 1 BF 42HI

*Rs e4281 8' 3R!8x 79111 Vi1l1 3418144291 31188 1F!1. 8418

9pi9x 911,918 58283 54125 9"D '['25 Vill 9222173'VP !081118 99188 9RIES 821( 9221l grill 9r21

I 21 I1

/72

!14'25

,/30 92181 109111 topiei 51111/31 9M18 1PIL9 RIES £8122 721(1 75121 82,212/1 IPI710 SPItZ 3RIIS 7MC2 F121 ?Si2l 8110212

2/3

U U. S. GOVERNMENT PRINTING OFFICE. 1989--600-0G7--82057

43