US Army Corps of Engineers Hydrologic Engineering Center

Determination of Land Use from Satellite Imagery for Input for Hydrologic Models April 1980 Approved for Public Release. Distribution Unlimited. TP-71

Standard Form 298 (Rev. 8/98) Prescribed by ANSI Std. Z39-18

REPORT DOCUMENTATION PAGE Form Approved OMB No. 0704-0188

The 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 and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to the Department of Defense, Executive Services and Communications Directorate (0704-0188). Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ORGANIZATION. 1. REPORT DATE (DD-MM-YYYY) April 1980

2. REPORT TYPE Technical Paper

3. DATES COVERED (From - To)

5a. CONTRACT NUMBER

5b. GRANT NUMBER

4. TITLE AND SUBTITLE Determination of Land Use from Satellite Imagery for Input to Hydrologic Models

5c. PROGRAM ELEMENT NUMBER

5d. PROJECT NUMBER 5e. TASK NUMBER

6. AUTHOR(S) R. Pat Webb, Robert Cermak, Arlen D. Feldman

5F. WORK UNIT NUMBER

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) US Army Corps of Engineers Institute for Water Resources Hydrologic Engineering Center (HEC) 609 Second Street Davis, CA 95616-4687

8. PERFORMING ORGANIZATION REPORT NUMBER TP-71

10. SPONSOR/ MONITOR'S ACRONYM(S) 9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 11. SPONSOR/ MONITOR'S REPORT NUMBER(S)

12. DISTRIBUTION / AVAILABILITY STATEMENT Approved for public release; distribution is unlimited. 13. SUPPLEMENTARY NOTES Presented at the Environmental Research Institute of Michigan's Fourteenth International Symposium on Remote Sensing of the Environment, San Jose, Costa Rica, 23-30 April 1980 14. ABSTRACT A land use/Land over identification methodology using LANDSAT imagery has been applied to six watersheds across the U.S. The land use information is stored in a grid cell data bank and is the basis for calibration of hydrologic parameters for watershed models. Flood frequency studies have been completed on four of the watersheds with land use derived from both satellite data and conventional low altitude aerial photography. This paper discusses our experience using the LANDSAT land use classification procedure and compares hydrologic results obtained from the alternative determinations of land use. 15. SUBJECT TERMS remote sensing, LANDSAT, land use, hydrologic models, grid cell database 16. SECURITY CLASSIFICATION OF: 19a. NAME OF RESPONSIBLE PERSON a. REPORT U

b. ABSTRACT U

c. THIS PAGE U

17. LIMITATION OF ABSTRACT UU

18. NUMBER OF PAGES 22 19b. TELEPHONE NUMBER

Determination of Land Use from Satellite Imagery for Input to Hydrologic Models

April 1980 US Army Corps of Engineers Institute for Water Resources Hydrologic Engineering Center 609 Second Street Davis, CA 95616 (530) 756-1104 (530) 756-8250 FAX www.hec.usace.army.mil TP-71

Papers in this series have resulted from technical activities of the Hydrologic Engineering Center. Versions of some of these have been published in technical journals or in conference proceedings. The purpose of this series is to make the information available for use in the Center's training program and for distribution with the Corps of Engineers. The findings in this report are not to be construed as an official Department of the Army position unless so designated by other authorized documents. The contents of this report are not to be used for advertising, publication, or promotional purposes. Citation of trade names does not constitute an official endorsement or approval of the use of such commercial products.

DETERMINATION OF LAND USE FROM SATELLITE IMAGERY

FOR INPUT TO HYDROLOGIC MODELS

R. Pat Webb, Robert Cermak and Arlen Feldman

The Hydrologic Engineering Center U.S. Army Corps of Engineers

609 Second S t r e e t Davis, Ca l i fo rn i a 95616

ABSTRACT

A land usel land cover i d e n t i f i c a t i o n methodology using LANDSAT imagery has been appl ied t o s i x watersheds across the U.S. The land use information is s tored i n a g r id c e l l d a t a bank and is t he ba s i s f o r c a l i b r a t i o n of hydrologic parameters f o r watershed models. Flood frequency s tud i e s have been completed on four of t he watersheds with land use derived from both s a t e l l i t e da ta and convent ional low a l t i t u d e a e r i a l photography. This paper d i scusses our experience using the LANDSAT land use c l a s s i f i c a t i o n procedure and compares hydrologic r e s u l t s obtained from t h e a l t e r n a t i v e determinat ions of land use.

Land usel land cover has been found t o have a s i g n i f i c a n t e f f e c t on t he quan t i t y , q u a l i t y , and timing of storm runoff from urban (and urbanizing) drainage basins . I n an at tempt t o quant i fy t h i s important r e l a t i onsh ip , the Hydrologic Engineering Center (HEX) has i n t e r f aced s tate-of- the-ar t s p a t i a l da t a management techniques with hydrologic planning models, such a s HEC-1 and STORM, t o s imulate storm runoff as a func t ion of land use. The g r id c e l l d a t a banks a l s o conta in informat ion on t he watershed 's environmental, economic, and soc i a l c h a r a c t e r i s t i c s , thus permi t t ing a comprehensive understanding of t he i n t e r a c t i o n between t he water resource system and pos s ib l e fu tu r e urban developments.

HEC has been involved i n a NASA ASVT p ro j ec t which t e s t s and eva lua tes a procedure developed a t t he Universi ty of Ca l i fo rn i a , Davis (UCD) f o r determining land use l land cover from LANDSAT imagery. The UCD procedure was designed with t he ob j ec t i ve of providing Corps of Engineers D i s t r i c t o f f i c e s with a n opera t iona l cos t -e f fec t ive a l t e r n a t i v e t o convent ional methods of ob ta in ing land use da ta . A cons t r a in t on t he procedure was t h a t it not r equ i r e t he use of s p e c i a l image processing o r computing equ ipmnt beyond t h a t which would normally be ava i l ab l e t o the f i e l d o f f i c e ; i .e . , l i n e p r i n t e r , card reader , remote terminal , and access t o a genera l purpose computer. The UCD procedure c o n s i s t s of an i n t eg ra t ed s e t of computer programs centered around an unsupervised c l a s s i f i c a t i o n rou t ine . Data q u a l i t y check, geometric r e g i s t r a t i o n and co r r ec t i on , data c l a s s i f i c a t i o n , symbol map genera t ion , resampling and masking a r e a l l accomplished without t he use of an i n t e r a c t i v e co lo r image d isp lay .

HEC has appl ied t he UCD c l a s s i f i c a t i o n method t o Crow Creek near Davemport, Iowa, and Walnut Creek near Aust in, Texas. Avai lable ground t r u t h d a t a permit ted t he i d e n t i f i c a t i o n of seven land cover ca tegor ies from the LANDSAT imagery: a g r i c u l t u r a l , res ident ia l /h ighways , i n d u s t r i a l / c o m e r c i a l , g rass land , f o r e s t , undeveloped open space, and water. Hydrologic s imulat ions of four a d d i t i o n a l watersheds, prev iously c l a s s i f i e d during t he development of t he procedure, were made, using both conventional and LANDSAT land use da ta . Resul t ing d i scharge frequency curves were compared t o determine t he e f f ec t i venes s of LANDSAT land use i n est imating "true" land use f o r hydrologic modeling purposes.

In addi t ion , two comerc i a l con t r ac to r s provided a land use c l a s s i f i c a t i o n of t he Walnut Creek watershed based on t he same LANDSAT scene t o he lp v e r i f y the accuracy of t he UCD procedure with cu r r en t s ta te-of- the-ar t c l a s s i f i c a t i o n methodologies. Because t he LANDSAT derived land use was placed i n da t a banks which contained convent ional ly c l a s s i f i e d land use , d e t a i l e d c e l l by c e l l comparisons were made between t h e convent ional and a l l of t he LANDSAT land use c l a s s i f i c a t i o n s t o get an i nd i ca t i on of s p a t i a l accurac ies assoc ia ted with LANDSAT da ta and t h e c l a s s i f i c a t i o n procedures.

Based on our experience i n using t h e UCD procedure and commercially derived LANDSAT da t a , recommendations a r e presented regarding the r o l e of remote sensing information i n the Corps of Engineers hydrologic i nves t i ga t i ons program.

LANDSAT LAND USE FOR HYDROLOGIC MODELING

The hydrologic modeling of a watershed, p a r t i c u l a r l y urban o r urbanizing bas ins , r equ i r e s t h a t the d i s t r i b u t i o n of land use be determined. The amount and timing of runoff i s d i r e c t l y r e l a t e d t o t h e i n f i l t r a t i o n capac i ty of a land a r ea with t h e most important d i s t i n c t i o n being between pervious and impervious land su r f aces. Water q u a l i t y parameters have a s imi l a r dependence on land use da t a ; r a t e of accumulation of a p a r t i c u l a r po l l u t an t per un i t a r ea is normally expressed as a func t ion of land use. Water resource planning s tud i e s a r e i n t e r e s t e d i n no t only a n assessment of t he present s t a t e of t he water and r e l a t e d resource system, but a l s o i t s poss ib le f u t u r e conf igura t ion . By expressing hydrologic parameters a s a func t ion of c u r r e n t land use it becomes poss ib le t o r a t i o n a l l y p r ed i c t t h e impact f u tu r e land use changes w i l l have on the quan t i t y and q u a l i t y of f u t u r e runoff .

Manual methods f o r land use i d e n t i f i c a t i o n (e.g., i n t e r p r e t a t i o n of low a l t i t u d e a e r i a l photography and f i e l d surveys) a r e f requent ly used i n watershed s tud ies . With t h i s approach, t he resource requirements , both money and labor , f o r manual c l a s s i f i c a t i o n can be ex tens ive . An a t t r a c t i v e a l t e r n a t i v e i s t he u t i l i z a t i o n of ava i l ab l e remote sensing systems and compute r-as s i s t e d c l a s s i f i c a t i o n techniques.

The LANDSAT s a t e l l i t e s have been shown t o have t h e c a p a b i l i t y of providing land use da t a a t acceptable l e v e l s of accuracy f o r hydrologic modeling purposes, ( ~ a c k s o n , 1977; Ragan, 1975). LANDSAT da t a i s qu icker and l e s s c o s t l y t o ob t a in and i n t e r p r e t than low a l t i t u d e a e r i a l photography, provides r e p e t i t i v e coverage of t h e same a r ea a t l e a s t every 18 days, and is a v a i l a b l e f o r a l l United S t a t e s and many worldwide loca t ions . Addi t iona l ly , LANDSAT'S d i g i t a l format can be d i r e c t l y analyzed by s eve ra l ava i l ab l e c l a s s i f i c a t i o n computer programs, and can be resampled f o r automatic i nc lu s ion i n a geographic da ta bank.

UCD PROCEDURES

An ope ra t i ona l procedure f o r land use c l a s s i f i c a t i o n from LANDSAT da t a has been developed a t the Universi ty of Ca l i fo rn i a , Davis (UCD) f o r use by the Corps of Engineers. Referred t o as t he UCD Procedure, it was designed t o func t ion under t he following c o n s t r a i n t s :

(1) Only output equipment normally ava i l ab l e i n Corps' f i e l d o f f i c e s (e.g., l i n e p r i n t e r ) and batch-mde access t o a genera l purpose computer could be expected. This would e l imina te the need f o r h igh ly expensive, dedicated, i n t e r a c t i v e image processing f a c i l i t i e s .

(2) No a d d i t i o n a l sof tware beyond t h a t provided a s p a r t of t he procedural package would be required.

( 3 ) No spec i a l i z ed t echn i ca l expe r t i s e i n da ta ana ly s i s , computer programming, o r remote sensing wculd be required.

( 4 ) The f i n a l c l a s s i f i c a t i o n would be a usab le product ; i .e . , one t h a t can conveniently be en te red i n t o a g r i d c e l l d a t a bank and t h a t w i l l adequately, from a hydrologic viewpoint , r ep re sen t cu r r en t land use condit ions.

The UCD procedure c o n s i s t s of an organized s e t of computer programs and manual opera t ions f o r the i d e n t i f i c a t i o n of land use from raw LANDSAT da ta . A d e t a i l e d de sc r ip t i on of the procedure is given by Algazi (1979) and Meyer (1978). What fol lows is a b r i e f o u t l i n e of t he primary tasks :

(1) Obtain LANDSAT Computer-Compatible Tapes (cCT), NASA high a l t i t u d e a e r i a l photography, and USGS topographic maps f o r t he l oca t i on and d a t e of i n t e r e s t . Ext rac t a r ec t angu la r a r ea of da ta containing t h e watershed from t h e CCT. Check f o r radiometr ic e r r o r s i n the LANDSAT d i g i t a l d a t a and, i f necessary, co r r ec t .

(2) Determine t h e geometric r e g i s t r a t i o n of t he LANDSAT image with t he coordinate system of the topographic maps. LANDSAT cont ro l po in ts a r e i d e n t i f i e d from the output of a UCD computer program which enhances roads and water bodies found in t h e LANDSAT image. A regress ion equat ion, estimated from the two s e t s of con t ro l po in ts , provides a t ransformation mechanism f o r going between the image coordinate system and the map coordinate system, Universal Transverse Mercator (uTM).

(3) Use an unsupervised c lu s t e r ing algori thm t o p a r t i t ion t h e LANDSAT four-dimens iona l d a t a space. Groups or "c lus te rs" a r e i d e n t i f i e d t h a t contain poin ts with s p e c t r a l r e f l ec t ance values t h a t a r e s i m i l a r t o members of t he same c l u s t e r , and d i s s imi l a r t o the poin ts of other. c l u s t e r s . The c l u s t e r i n g program is allowed t o generate a maximum of 30 c l u s t e r s . Each p ixe l i n t h e watershed da ta is assigned t o a c l u s t e r .

(4) Select f r m a l i n e p r i n t e r map of t he c l u s t e r assignments s i x s e t s of ad jacent p ixe l s ( s p a t i a l groups), a l l belonging t o the same c l u s t e r . Their corresponding loca t ion on the topographic maps i s determined using t h e t ransformation equation of s t e p (2). Visual t r ans l a t i on , from the map t o the a e r i a l photographs, of the s p a t i a l group's loca t ion permits a land use t o be assigned t o each s p a t i a l group. For c l u s t e r s having a cons i s t en t land use assigned t o a l l s i x s p a t i a l groups, a f i n a l land use has been determined. But f o r those c l u s t e r s where c o n f l i c t s e x i s t between the land use i d e n t i f i e d with each of the s i x s p a t i a l groups, f u r t h e r p a r t i t i o n i n g of the da t a space i s required.

( 5 ) Clus ters with con f l i c t i ng land use assignments and c l u s t e r s whose assoc ia ted land use could not be determined from the ava i l ab l e maps and photos a r e rec lus te red by repea t ing s t e p (31, and given f i n a l land use assignments by repea t ing s t e p (4) .

( 6 ) A t t h i s point the watershed da t a f i l e contains a land use c l a s s i f i c a t i o n ( t y p i c a l l y 5 t o 7 ca tegor ies ) fo r a l l i t s p ixe ls . The watershed f i l e i s then resampled a t the g r i d c e l l c en t ro ids using a nearest-neighbor algorithm. The s i z e of the g r id c e l l s i s usua l ly l i n e p r i n t e r compatible with t h e s c a l e of USGS 7-1/2-min. topographic maps.

(7) The resampled f i l e is entered d i r e c t l y i n t o a g r id c e l l da t a bank. A l t e rna t ive ly , a f i l e containing t h e d i g i t i z e d ( i n UTM coordinates) watershed boundary can be used t o mask the resampled f i l e , leaving only the g r id c e l l s wi th in the boundary. Total acreage of each land use c l a s s f o r t h e e n t i r e watershed i s then computed.

HYDROLOGIC LAND USE COMPARISON

The primary reason f o r examining t h e land use c l a s s i f i c a t i o n a b i l i t y of LANDSAT was f o r i t s po t en t i a l app l i ca t i on t o hydrologic mode l ing. The computer program HEC-1 (U.S. Army Corps of Engineers, 1973) has t h e c a p a b i l i t y of e x p l i c i t l y r e l a t i n g land use t o runoff using two procedures : Snyder's un i t hydrograph with percent imperviousness, and the SCS curve number and un i t hydrograph (u.S. S o i l Conservation Service, 1972). The HYDPAR program (u.S. Army Corps of Engineers, 197813) obta ins the necessary information from a g r id c e l l da t a bank and computes t he spec i f i ed hydrologic parameters, which a r e i n t u r n input i n t o an HEC-1 model of the basin. HYDPAR conta ins a regress ion equat ion formulation of Snyder's l ag a s a funct ion of stream length , length t o cent ro id of subbasin, stream s lope , and percent impenriousness. A t ab l e a s soc i a t i ng a percent imperviousness with each land use category i n the da t a bank enables HYDPAR t o compute subbas i n percent imperviousness from subbas i n land use d i s t r i b u t i o n .

I n a s i m i l a r manner HYDPAR can determine the SCS un i t hydrograph parameter from stream length , bas in average land s lope , and subbasin average curve number. Curve numbers represent an empir ical r e l a t i o n s h i p between hydrologic s o i l type, land use, and t h e i r r e s u l t a n t runoff po t en t i a l . From a t a b l e ident i fy ing a curve number with each combination of land use and hydrologic s o i l type, HYDPAR computes subbasin average curve number.

Both hydrologic modeling techniques, Snyder's un i t hydrograph with percent imperviousness and the SCS curve number approach, were applied t o the land use c l a s s i f i c a t i o n s of two w ~ t e r s h e d s : Rowlet t Creek near Dallas , Texas, and Pennypack Creek in Philadelphia, Pennsylvania. For each watershed land use was determined from LANDSAT imagery using the UCD Procedure, and conventional i n t e r p r e t a t i o n of low-al t i tude a e r i a l photography. The following conta ins summary r e s u l t s of the hydrologic land use comparison. Complete d e t a i l s of t he study a r e reported i n (U.S. Army Corps of Engineers: 1979).

RDWLETT CREEK

A ca l i b r a t ed HEC-1 m d e l of a 24.6 square mi le (63.7 h 2 ) por t ion of t he Rowlett Creek bas in , r e f e r r ed t o as Upper Spring Creek, was used t o s imulate runoff from se l ec t ed recur rence i n t e r v a l r a i n f a l l . I n i t i a l assignments of percent i m p e r ~ i ~ u s n e s s t o Rowlett Creek's "LANDSAT and conventional land use ca t ego r i e s permitted HYDPAR t o compute Snyder's l ag f o r t he twenty-three subbasins i n Upper Spring Creek. Nearly i d e n t i c a l va lues of l ag were ca l cu l a t ed from the LANDSAT and conventional land use da ta .

The ca l i b r a t ed HEC-1 m d e l (using t he two land use es t imates of Snyder's l ag ) and syn the t i c r a i n f a l l produced the discharge values p l o t t e d i n Figure 1 f o r se lec ted s t a t i o n s i n t h e Upper Spring Creek drainage. Differences between such discharge frequency curves can be i n t e rp re t ed as a measure of the hydrologic s i gn i f i c ance of LANDSAT's rn i s c l a s s i f i c a t i on of land use. Cons ider ing t h e unce r t a in ty involved i n es t imat ing a frequency curve (even from observed d a t a ) , the d i f f e r ence between LANDSAT and conventional curves i s i n s i g n i f i c a n t .

PENNYPACK CREEK

The SCS curve number method was used t o model t h e Pennypack Creek bas in (55.8 . :2, 144.5 k 2 ) . Curve numbers were assigned t o t he LANDSAT and convent ional land use ca t ego r i e s . For each of Pennypack Creek's s i x ty - f i ve subbasins HYDPAR ca l cu l a t ed subbasin average curve number and subbasin lag. Once again, near ly i d e n t i c a l values of subbasin lag were computed from t h e LANDSAT and conventional land use da ta .

A s an add i t i ona l comparison, subbasin average curve number and lag were ca l cu l a t ed f o r (1 ) land use ca t ego r i e s assigned the i n d u s t r i a l category curve numbers, and (2) land use

ca tegor ies as signed t he n a t u r a l vege ta t ion curve numbers. Parameters est imated i n these two cases , and the discharge frequency curves derived from them, demonstrate the poss ib le extremes ( i n terms of runof f ) t h a t could have been generated from t h e model.

The ca l i b r a t ed HEC-1 model of Pennypack Creek simulated the b a s i n ' s discharge frequency behavior f o r convent ional , LANDSAT, a l l i n d u s t r i a l , and a l l n a t u r a l vege ta t ion condit ions. The r e s u l t i n g discharge frequency curves f o r the e n t i r e drainage a r ea a r e shown i n Figure 2. It i s c l e a r from t h i s f i g u r e , e s p e c i a l l y with re fe rence t o what could have been ( i . e . , a l l i n d u s t r i a l and a l l n a t u r a l vege ta t ion condi t ions 1, t h a t t he d i f f e r ence between LANDSAT had convent ional ly derived frequency curves i s not s i g n i f i c a n t .

ACCURACY VS. SPATIAL INTEGRITY

As previously mentioned, in order t o be a b l e t o conduct t he hydrologic modeling assessment, the LANDSAT land use was processed i n t o e x i s t i n g geographic information system da t a banks. These da t a banks already contained an exhaust ive, s p a t i a l l y accura te r ep re sen t a t i on of the land use which was derived by convent ional means. The g r id c e l l s i z e of these e x i s t i n g da t a banks va r i ed in s i z e from 0.74 acres (pennypack creek) t o 1.148 ac r e s ( ~ o w l e t t Creek and Walnut Creek) t o 1.53 ac r e s ( T r a i l Creek and Castro va l l ey ) . The desc r ip t i on of t he ways i n which t he convent ional land use c l a s s i f i c a t i o n and t h e LANDSAT c l a s s i f i c a t i o n were derived i s presented i n d e t a i l i n HEC's Research Note No. 7 (u.s. Army Corps of Engineers, 1979). Table 1, UCD C l a s s i f i c a t i o n vs. Conventional Land Use, shows a s u m a r y of the cel l -by-cel l comparison f o r the Walnut Creek watershed.

For a cel l -by-cel l comparison of t he Walnut Creek watershed it was necessary t o e s t a b l i s h an e x p l i c i t aggregat ion of the conventional land use ca t ego r i e s i n t o the fewer LANDSAT land use ca t ego r i e s , Table 2. The R I A computer program (u.S. Army Corps of Engineers, 1978a) generated t he coincident mat r ix , Table 1. The s t r u c t u r e of t h i s c ro s s t abu l a t i on t a b l e is s i m i l a r t o o t h e r s t h a t w i l l be presented l a t e r i n t h i s paper. Each element of t he t a b l e (row and c o l u m canbinat ion) r e f e r s t o g r id c e l l s wi th in the watershed da t a bank t h a t have the concurrent &?SAT conventional land use spec i f i ed by t h e row and column headings of t h a t p a r t i c u l a r element. For example, the 2nd row, 1 s t column of Table 1 r e f e r s t o a l l g r i d c e l l s i n t he Walnut Creek da t a bank t h a t a r e c l a s s i f i e d both commercial / industr ia l by LANDSAT r e s i d e n t i a l by the conventional c l a s s i f i c a t i o n . For each element of t he t a b l e , four numbers

EXCEEDANCE FREQUENCY, in percent EXCEEDANCE FREQUENCY, in percent

CONVENTIONAL LAND USE ---- LAMDSAT LAND USE

Figure 1

RBWLETT CREEK Figure 2

ANNUAL PEAK DISCHARGE FREQUENCY PENNYPACK CREEK BASIN (SELECTED STATIONS )

SUBBASINS 1 -65 (55.74 sq. rn i )

TABLE 1

W a l n u t C r e e k L a n d Use C o m p a r i s o n UCD LANDSAT vs. CONVENTIONAL

*To o b t a i n h e c t a r e s , m u l t i p l y a c r e s by 2.471. d

ROW TOTAL

9136 100.0

25 .O 25 .O

1861 100.0

5.1 5.1

- 223 100.0

0.6 0.6

15229 100.0

41.8 41.8

9989 100.0

27.3 27.3

70 100.0

0 .2 0.2

36578 100.0 100.0 100.0

- A c r e s * % Row % c01 % T o t a l

I

4

t

UCD Z E D SAT

RES

COM/ IND

QUARRY

CRO P / PASTURE

FORE ST/ RANGE

WATER

COLUMN TOTAL

WATER

6 0.1 9.5 0 .O

0 0.0 0 .O 0.0

0 0 .0 0.0 0.0

28 0.2

44.4 0 .1

13 0.1

20.6 0 .O

16 22.9 25.4

0.0

63 0.2

100.0 0.2

RES

3801 41.6 59.8 10.4

471 25.3

7.4 1.3

13 5.8 0.2 0 .O

1581 10.3 24.9

4.3

49 3 4.9 7.8 1.3

0 0.0 0 .O 0.0

6359 17.4

100.0 17.4

C ~ P / P A S T U R E /RANGE

3147 34.4 19.1

8.6

366 19.7

2.2 1 .O

17 7.6 0.1 0 .O

9324 60.9 56.4 25.5

3655 36.6 22.1 10.0

9 12.9

0.1 0.0

16518 45.2

100.0 45.2

FOREST

859 9.4 9.4 2.3

78 4.2 0.9 0.2

2 1 9.4 0.2 0.1

2947 19.3 32.2

8.1

5233 52.4 57.2 14.3

6 8.6 0.1 0.0

9144 25.4

100.0 25 .O

COM/IND

1098.8 12 .O 31.6

3 .O

8 16 43.8 23.5

2.2

61 27.4

1.8 0.2

1106 7.2

31.8 3.0

361 3.6

10.4 1 .O

31 44.3

0.9 0.1

3473 9.5

100.0 9.5

CONVENTIONAL

QUARRY

225 2.5

22.0 0.6

130 7.0

12.7 0.4

111 49.8 10.9

0.3

313 2.0

30.7 0.9

234 2.3

22.9 0.6

8 11.4

0.8 0.8

1021 2.8

100.0 2.8

TABLE 2

WfiwJT C U E K LAND USE CATEGORY MAPPING

GE LANDSAT

r e s i d e n t i a l 1 r e s i d e n t i a l 2

urban

h ighly r e f l e c t i v e

vege ta t ion / crops dark f i e l d s open a rea

- woodland

B a t t e l l e LANDSAT

r e s i d e n t i a l

i n d u s t r i a l / commerc i a 1

t r anspo r t a t ion

bar ren land

crop land /

pas ture range land

f o r e s t r i p a r i a n

water

Conventional Land Use

Low dens i t y s i n g l e family r e s i d e n t i a l Medium dens i t y s i n g l e family r e s i d e n t i a l High dens i t y s i n g l e family r e s i d e n t i a l Mult i family r e s i d e n t i a l Mobile home parks

S t r i p commercial Shopping c e n t e r s I n s t i t u t i o n a l I n d u s t r i a l I n d u s t r i a l and commerc i a 1 complexes Public Use: cemetar ies , publ ic assembly

a r ea s , waste disposa 1 a r ea s Transpor ta t ion , comnunication, u t i l i t i e s

Barren land/quarry

Crop land

Pas ture /range 1 and Deve loped open space Undeveloped urban land

Forest

Water

UCD LAND SAT

r e s i d e n t i a l

commercial/ i n d u s t r i a l

bar ren land/

crop land /

pas ture

f o r e s t / range land

water

a r e given: (1) t o t a l acreage of a l l g r i d c e l l s represented by the appropriate j o i n t land use c l a s s i f i c a t i o n ; (2) row percent , o r the precent of g r id c e l l s with the given LANDSAT c l a s s i f i c a t i o n t h a t have also the given conventional c l a s s i f i c a t i o n ; (3) column percent , o r the percent of gr id c e l l s with the given conventional land use t h a t have &o the given LANDSAT land use; and (4) t o t a l percent , o r t h e percent of t he e n t i r e watershed t h a t has t he given jo in t land use c l a s s i f i c a t i o n . Continuing our example above, 471 ac re s (1186 Hectares) were found t o be c l a s s i f i e d both commercial / industr ial by LANDSAT and r e s i d e n t i a l by conventional means. This acreage represents 25.3% of a l l the a r ea (1861 ac re s ) t h a t was c l a s s i f i e d by LANDSAT as commercial / industr ial , 7.4% of a l l t he a rea (6359 ac re s ) t h a t belonged t o the conventional r e s i d e n t i a l category, and 1.3% of the t o t a l watershed a r ea (36,578 acres) .

The f a r r i g h t column (row t o t a l ) and the bottom row (column t o t a l ) of Table 1 give t h e marginal d i s t r i b u t i o n s of LANDSAT and conventional land use, respec t ive ly . These represent t h e acres and percent i n t he d i f f e r e n t land use ca tegor ies without the condi t iona l requirement described above f o r the body of the tab le .

Figures 3 and 4 show the s p a t i a l l oca t ion of both t he conventional and t h e UCD c l a s s i f i e d land use which a r e pa r t of the Walnut Creek g r id c e l l da t a bank.

COMMERCIAL C APAB ILITY

The UCD procedure provided complete l y acceptable land use percentages f o r t h e subbasins and t o t a l watershed a r eas ( t h e marginal va lues i n Table 1) which were input i n t o t he hydrologic models. Spurred on by t h i s i n i t i a l success, the HEC fu r the r pa r t i c ipa t ed i n t h e Water Management and Control ASVT by comparing commercially prepared LANDSAT land use c l a s s i f i c a t i o n s . The objec t ives of HEC's p a r t i c i p a t i o n were: (1) t o compare t h e r e s u l t s of the UCD procedure t o the r e s u l t s from commercial vendors, (2) t o provide cos t da t a t o a cos t -e f fec t iveness study of LANDSAT determined land use which was p a r t of t h e ASVT and, ( 3 ) t o evaluate the a b i l i t y of commercial vendors t o c r e a t e a computer f i l e which could be d i r e c t l y i n se r t ed as a da ta va r i ab l e i n t o an e x i s t i n g g r i d c e l l data bank. I n order t o execute t h i s po r t i on of the ASVT, the Walnut Creek da t a bank was se lec ted as the t e s t a r ea because (1) the HEC was confident i n t h e s p a t i a l i n t e g r i t y and accuracy of t he conventional land use i n t h e d a t a bank, (2) the g r id c e l l s i z e was 1.148 ac re s which was comparable t o a p i x e l s i z e of 1.1 acres and, ( 3 ) the g r i d p a t t e r n f o r t h e study a r ea was o r i en t a t ed t o a l i g n with t h e UTM coordinate system. The commercial cont rac tors which were se lec ted t o p a r t i c i p a t e i n t h i s po r t i on of t he study were General E l e c t r i c and B a t t e l l e ' s P a c i f i c Northwest Labra tor ies i n Richland, Washington. Both commercial firms crea ted resampled tapes which contained the land use value which corresponded t o t he cent ro id loca t ion of each of t h e Corps rec tangular g r id c e l l s (200 f e e t x 250 f e e t ) . Table 2 contains the commercial LANDSAT land use c l a s s i f i c a t i o n s which correspond t o t h e conventional c l a s s i f i c a t i o n . Tables 3 and 4 show t h e r e s u l t s of t h e cel l -by-cel l comparisons f o r the B a t t e l l e and General E l e c t r i c c l a s s i f i c a t i o n s , respec t ive ly .

Using t h e diagonal percentages i n Tables 1, 3 and 4, the o v e r a l l c e l l by c e l l accuracies of the UCD procedure, B a t t e l l e 'and General E l e c t r i c were 53, 44 and 44 percent , respec t ive ly . In evaluat ing t h e s p a t i a l i n t e g r i t y of t he LANDSAT derived land use from the var ious c l a s s i f i c a t i o n procedures, there a r e t h r ee p o t e n t i a l sources of e r ro r : (1) d i f f e r ences r e s u l t i n g f r o m t h e land use, land cover c o n f l i c t , (2) geometric co r r ec t ion and resampling e r r o r and (3) mi sc l a s s i f i ca t i on of the p i x e l s p e c t r a l s igna tures .

When comparing LANDSAT and conventional land use c l a s s i f i c a t i o n s , it i s important t o recognize t h a t the same land cover can be i n t e rp re t ed d i f f e r e n t l y ; i .e . , conventional land use ca tegor ies a r e no t always compatible with LANDSAT land cover ca tegor ies . For example, the conventional category "transportation/comnunication/utilities includes major highways, r i g h t -of-way fo r r a i l roads and power t ransmission l i n e s , communication towers, a i r p o r t f a c i l i t i e s ( including bui ldings, runways, and vacant land wi th in the a i r p o r t l i m i t s 1, and sewage treatment p lan ts . In con t r a s t , LANDSAT w i l l recognize t h e t reatment p lan t s e t t l i n g tanks as "water bodies", the open f i e l d s surrounding a runway as one of the vegeta t ion ca tegor ies , and right-of-ways as whatever land cover c l a s s is nearby. Even though t h i s is a problem with any LANDSAT land use c l a s s i f i c a t i o n , i n the Walnut Creek watershed only a very small por t ion of t he watershed had t h e p o t e n t i a l f o r t h i s kind of e r r o r .

TABLE 3

W a l n u t C r e e k L a n d U s e C o m p a r i s o n B a t te l l e LANDSAT vs. C o n v e n t i o n a l

-0 o b t a i n hec tares , m u l t i p l y acres b y 2.471.

L

A c r e s * % Row % C O ~

% T o t a l ROW TOTAL

6513 100.0

18.8 18.8

2286 100.0

6.6 6.6

1416 100.0

4.1 4.1

-

181.36 100.0 52.3 52.3

6295 100.0

18.1 18.1

54 100.0

0.2 0.2

-'

34700 100.0 100 .o 100.0

BATTELLE LAND SAT

WATER

12 0.2

19.0 0.0

1 0.0 1.6 0.0

0 0.0 0.0 0.0

3 7 0.2

58.7 0.1

12 0.2

19 .O 0.0

1 1.9 1.6 0 .o

63 0.2

RE S

COM/IND

QUARRY

Clop/ PASTURE RANGE

FORE S T

WATER

100.0 0.2

FORE ST

763 11.7 8.7 2.2

319 14.0

3.6 0.9

319 22.5 3.6 0.9

4922 27.1 56.3 14.2

2410 38.3 27.6

6.9

9 16.7 0.1 0.0

8742 25.2

RES

2723 41.8 45.6

7.8

661 28.9 11.1 1.9

212 15 3.5 0.6

1720 9.5

28.8 5 .O

6 36 10.1 10.6

1.8

2 3 42.6

0.4 0.1

100.0 25.2

100 .o 8.9

COM/ IND

976 15.0 31.6

2.8

336 14.7 10.9

1 .O

160 11.3 5.2 0.5

1137 6.3

36.8 3 .3

464 7.4

15 .O 1.3

13 24.1 0.4 0 .o

3086 8.9

100 .o 2.8

*

100 .o 45.7

CONVENTIONAL

QUARRY

70 1.1 7.2 0.2

7 9 3.5 8.1 0.2

11 1 7.8

11.3 0.3

51 7 2.9

52.9 1.5

20 1 3.2

20.6 0.6

0 0 .0 0.0 0 .o

978 2.8

CROP/PASTURE /RANGE

1969 30.2 12.4

5.7

890 38.9

5.6 2.6

6 14 4.3.4

3.9 1.8

9803 54.1 61.8 28.3

2572 40.9 16.2

7.4

8 14.8 0.1 0 .o

15856 45.7

TABLE 4

W a l n u t C r e e k L a n d U s e C o m p a r i s o n G. E. LANDSAT vs. C o n v e n t i o n a l

* T o o b t a i n hectares, m u l t i p l y acres by 2.471. 12

A c r e s * % Row % C O ~

% T o t a l

G.E. - LAND SAT

ROW TOTAL

7833.9 100.0

22.3 22.3

3174.2 100.0

9 .O 9 .O

1894.2 100.0

5.4 5.4

16013.6 100.0

45.6 45.6

6203.8 100.0

17.7 17.7

0 100.0

0 0

35119.6 100.0 100.0 100 .O

RES

COM/ IND

QUARRY

AGRICUL- TURE /

OPEN SPACE

WOODLAND

WATER

R E S

3573.7 45.6 57.4 10.2

464.90 14.7

7.5 1.3

358.2 18.9

5.8 1 .O

1587.6 9.9

25.5 4.5

242.2 3.9 3.9 0.1

0 0 0 0

6226.8 17.4

100.0 17.7

COLUMN TOTAL

CIIOP/PASTURE /RANGE

2308.6 29.5 14.8

6.6

1678.3 52.8 10.8

4.8

433.9 22.9

2.8 1.2

8264.6 51.6 53.1 23.5

2871.1 46.3 18.5 8 .2

0 0 0 0

15556.6 44.3

100.0 44.3

COM/ I N D

1060.8 13.5 31.6

3 .O

693.4 21.8 20.6

2 .O

690.1 36.4 20.5

2 .O

677 -4 4.2

20.2 1.9

238.9 3.9 7 .1 0.7

0 0 0 0

3360.5 9.6

100.0 9.6

CONVENTIONAL

QUARRY

205.5 2.6

20.8 0.6

102.2 3.2

10.4 0.3

289.3 15.3 29.3

0.8

261.8 1.6

26.6 0.8

127.4 2.1

12.9 0 - 4

0 0 0 0

986.2 2.8

100.0 2.8

FOREST

679.6 8.7 7.6 1 .9

231.9 7.3 2.6 0.7

106.8 5.6 1.2 0.3

5208.5 32.5 58.3 14.8

2701.2 43.5 30.3

7.7

0 0 0 0

8928 .O 25.4

100.0 25.4

WATER

5.7 0.1 9.2 0 .O

3.4 0.1 5.5 0.0

16.1 0.9

26 .O 0.1

13.7 0.1

22.1 0.0

23.0 0 .O

37.1 0.1

0 0 0 0

62.0 0.2

100.0 0.2

A

A comparison a t t h e g r i d c e l l l e v e l can be thought of a s a comparison a t t he p i x e l l e v e l ; both u n i t s of a r ea a r e near ly the same s i z e . A t t h i s s ca l e e r r o r s introduced during t he g e m e t r i c co r r ec t i on and resampling s t e p s w i 11 be erroneously i n t e rp re t ed a s LANDSAT m i s c l a s s i f i c a t i o n e r r o r s . This w i l l be p a r t i c u l a r l y t r u e of land use ca t ego r i e s defined by a smal l number of ad jacent p ixe l s . The border a reas of t he l a r g e r land use ca t ego r i e s a r e a l s o su scep t ib l e t o geometric problems. I n order t o a s se s s t he magnitude of the reg is t ra t ion / resampl ing e r r o r , t he RIA program was used t o c a l c u l a t e t he l i n e a r d i s t ance each g r id c e l l was from the neares t r e s i d e n t i a l category. These d i s t ances were then over la id on t he a c t u a l r e s i d e n t i a l l oca t i ons from the t h r e e LANDSAT c l a s s i f i c a t i o n s . The reason f o r t h i s overlay was t h a t i f t he major problem i n the ce l l -by-ce l l comparison was i n the geometric correct ion/resampling procedures, a l l of t he LANDSAT r e s i d e n t i a l a r ea s should be w i th in one o r two gr id c e l l s of an a c t u a l r e s i d e n t i a l a rea . The LANDSAT r e s i d e n t i a l a r e a s which loca ted beyond t h i s d i s t ance could be a s sumd t o be e r r o r s i n mi sc l a s s i f i c a t i on .

Figure 5, LANDSAT RESIDENTIAL VS DISTANCE TO ACTUAL RESIDENTIAL, is a histogram of t he s p a t i a l i n t e g r i t y of t he t h r e e c l a s s i f i c a t i o n s f o r r e s i d e n t i a l land use and i s shown a s a percentage of the r e s i d e n t i a l a r ea f o r each c l a s s i f i c a t i o n . The f i gu re shows t h a t a t l e a s t 60 percent of a l l these c l a s s i f i c a t i o n s f e l l wi th in 2 gr id c e l l s of t h e ground t r u t h r e s i d e n t i a l a r ea s . The B a t t e l l e c l a s s i f i c a t i o n was more accura te f o r the r e s i d e n t i a l a r ea s , but when compared f o r a l l land use ca tegor ies it was l e s s accura te than t he UCD procedure. There were a t l e a s t two reasons for B a t t e l l e ' s h igher r e s i d e n t i a l accuracy when compared t o UCD and G.E.; f i r s t , they had personnel from the Corps who were very f ami l i a r with t h e study a r ea t o he lp with the c l a s s i f i c a t i o n , which was no t the case i n the UCD and G.E. c l a s s i f i c a t i o n s and second, they had t h e c a p a b i l i t y t o predefine urban a r ea s and r e c l a s s i f y c l u s t e r s which were g iven a r e s i d e n t i a l value but were loca ted ou ts ide "known" r e s i d e n t i a l a reas .

The t h i r d source of e r r o r , m i s c l a s s i f i c a t i o n , was looked a t f o r only t he UCD c l a s s i f i c a t i o n . A R I A coincidence t a b l e was generated f o r the conventional land use and t he c lus te r . values t o s ee i f any of t he c l u s t e r s were mi sc l a s s i f i ed . The HEC could no t i d e n t i f y any obvious mi sc l a s s i f i c a t i ons , therefore , it was assumed t h a t the remaining e r r o r (40-60 pe rcen t ) in t he cel l -by-cel l comparison was due t o t he l i m i t a t i o n s of e i t h e r t he c l u s t e r i n g techniques o r the s a t e l l i t e sensors themselves.

F'lgure 5

LANDSAT' RESIDENTIAL VS DISTANCE TO ACTUAL RESIDENTIAL

DISTANCE

1 3

RE SAMPLED TAPE S

The resampled tapes supplied by UCD, B a t t e l l e , G.E. and Bendix ( f o r t he Castro Valley Watershed) were required t o conform t o the following spec i f i ca t i ons :

(1) The f i r s t record on t h e tape would be a header record which would give t he number of rows and the number of columns i n the da t a matrix.

(2) The data would be supplied a row ( l i n e ) a t a time, fo r however many number of rows there were i n the study area.

(3) The f i r s t land use value would correspond t o t h e g r id loca t ion 1, 1 on the Walnut Creek or Castro Valley Basemap.

(4 ) I f t h e con t r ac to r windowed out t h e watershed, c e l l s ou ts ide the study area wculd be given a land use value of -1.

In add i t i on , a l l con t r ac to r s were warned t h a t i f they windowed the study a r ea , t o include a t l e a s t some buffer a rea on the resampled tape t o make su re t h a t a l l g r id c e l l s i n the ex i s t i ng da ta base would be given a value.

The HEC was ab l e t o process t he UCD and B a t t e l l e tapes d i r e c t l y i n t o t h e e x i s t i n g da ta base. The B a t t e l l e tape, however, had severa l problems. After the tape had been provided t o HEC, it was discovered t h a t t he wrong corner g r id c e l l was used i n t h e resampling. A l l land use comparisons of the B a t t e l l e c l a s s i f i c a t i o n i n t h i s paper w i l l be a f fec ted by t h i s known resampling e r r o r . In add i t i on , when the B a t t e l l e resampled f i l e was entered in to t he Walnut Creek g r id c e l l da t a bank only 34,697 acres (54.21 sq. mi.) were located within the geographical ly co r r ec t watershed boundary, the l a t t e r being defined in t he data bank a s containing 36,574 acl;es (57.15 sq. m i . 1. The t o t a l c l a s s i f i e d land area l i s t e d on B a t t e l l e ' s color-coded map was 35,869 acres (56.05 sq. m i . ).

The G.E. tape was not produced t o the o r i g i n a l spec i f i ca t i ons , so t h a t a new resampled tape had t o be generated, and it a l s o had resampling problems so t h a t it only contained 35,120 ac re s (54.9 sq. mi.).

CONCLUSION

The LANDSAT derived land use c l a s s i f i c a t i o n percentages a r e well within an acceptable e r r o r t o be used f o r hydrologic modeling. The LANDSAT derived land use c l a s s i f i c a t i o n may a l s o be successfu l ly placed i n t o e x i s t i n g g r id c e l l da t a banks t o be used i n o ther types of ana lys is , such a s environmental assessments. Caution needs t o be made though on the s p a t i a l i n t e g r i t y of the LANDSAT derived land use fo r g r id c e l l da t a banks which have a g r id c e l l s i z e which i s a t about t h e same r e so lu t ion as a p ixe l . The HEC i s cu r r en t ly inves t iga t ing the accuracies assoc ia ted with a 4.6 and a 10.3 acre g r id c e l l s i z e with the same data.

The UCD procedure was succes s fu l i n e l imina t ing t h e need f o r expensive image processing equipment and i t s accuracy was as good or b e t t e r than the commercial firms which supplied land use data f o r t h e same area . This paper presents problems encountered with t he commercial products during the conduct of HEC's por t ion of the ASVT. The authors f u l l y r e a l i z e t h a t because of continuous technology changes t h a t these problems may have been el iminated.

REFERENCES

Algazi, V.R., G.E. Ford and D. I . Meyer, 1979, "A Non-Interactive Approach t o Land Use Determination," 1979 Machine Processing of Remotely Sensed Data Symposium, Purdue Univers i ty , June 27-29.

Cemak, R . J . , A. Feldman and R.P. Webb, 1979, "Hydrologic Land Use C l a s s i f i c a t i o n Using LANDSAT," Proceedings of the AWRA S a t e l l i t e Hydrology Symposium, Sioux F a l l s , South Dakota, June 11-15 ( i n press ) .

Jackson, T. J. , R.M. Ragan and W.N. F i t ch , 1977, "Test of LANDSAT-Based Urban Hydrologic Mode l ing ," Journal of the Water Resources Planning and Management Divis ion, ASCE, 103 (wRI), pp. 141-158.

Meyer, D. and V.R. Algazi, 1978, "Land Use C l a s s i f i c a t i o n of LANDSAT Data Using a Clus te r ing Algorithm, " Department of E l e c t r i c a l Engineering, Universi ty of Ca l i fo rn i a , Davis, Ca l i fo rn i a ( d r a f t ) .

Ragan, R.M. and T. J . Jackson, 1975, ' 'Use of S a t e l l i t e Data i n Urban Hydrologic Models," Journa l of the Hydraulics Divis ion, ASCE, 101 (HY12), pp. 1469-1475.

U.S. S o i l Conservation Service, 1972, "Hydrology," SCS National Engineering Handbook, Sect ion 4, Washington, D.D.

U. S. Army Corps of Engineers, 1973, "HEC-1 Flood Hydrograph Package, Users Manual," The Hydrologic Engineering Center , Davis, Ca l i fo rn i a

U.S. Army Corps of Engineers, 1978a, "Resource Information and Analysis Using Grid Ce l l Data Banks: Users Manual," The Hydrologic Engineering Center, Davis, Ca l i forn ia .

U.S. Army Corps of Engineers, 1978b, "HYDPAR: Hydrologic Parameters, Users Manual," The Hydrologic Engineering Center, Davis, Ca l i forn ia .

U.S. Army Corps of Engineers, 1979, "Determination of Land Use from LANDSAT Imagery: Applicat ions t o Hydrologic Mode l i ng , "Research Note No. 7 , The Hydrologic Engineering Center, Davis, Ca l i forn ia .

Technical Paper Series TP-1 Use of Interrelated Records to Simulate Streamflow TP-2 Optimization Techniques for Hydrologic

Engineering TP-3 Methods of Determination of Safe Yield and

Compensation Water from Storage Reservoirs TP-4 Functional Evaluation of a Water Resources System TP-5 Streamflow Synthesis for Ungaged Rivers TP-6 Simulation of Daily Streamflow TP-7 Pilot Study for Storage Requirements for Low Flow

Augmentation TP-8 Worth of Streamflow Data for Project Design - A

Pilot Study TP-9 Economic Evaluation of Reservoir System

Accomplishments TP-10 Hydrologic Simulation in Water-Yield Analysis TP-11 Survey of Programs for Water Surface Profiles TP-12 Hypothetical Flood Computation for a Stream

System TP-13 Maximum Utilization of Scarce Data in Hydrologic

Design TP-14 Techniques for Evaluating Long-Tem Reservoir

Yields TP-15 Hydrostatistics - Principles of Application TP-16 A Hydrologic Water Resource System Modeling

Techniques TP-17 Hydrologic Engineering Techniques for Regional

Water Resources Planning TP-18 Estimating Monthly Streamflows Within a Region TP-19 Suspended Sediment Discharge in Streams TP-20 Computer Determination of Flow Through Bridges TP-21 An Approach to Reservoir Temperature Analysis TP-22 A Finite Difference Methods of Analyzing Liquid

Flow in Variably Saturated Porous Media TP-23 Uses of Simulation in River Basin Planning TP-24 Hydroelectric Power Analysis in Reservoir Systems TP-25 Status of Water Resource System Analysis TP-26 System Relationships for Panama Canal Water

Supply TP-27 System Analysis of the Panama Canal Water

Supply TP-28 Digital Simulation of an Existing Water Resources

System TP-29 Computer Application in Continuing Education TP-30 Drought Severity and Water Supply Dependability TP-31 Development of System Operation Rules for an

Existing System by Simulation TP-32 Alternative Approaches to Water Resources System

Simulation TP-33 System Simulation of Integrated Use of

Hydroelectric and Thermal Power Generation TP-34 Optimizing flood Control Allocation for a

Multipurpose Reservoir TP-35 Computer Models for Rainfall-Runoff and River

Hydraulic Analysis TP-36 Evaluation of Drought Effects at Lake Atitlan TP-37 Downstream Effects of the Levee Overtopping at

Wilkes-Barre, PA, During Tropical Storm Agnes TP-38 Water Quality Evaluation of Aquatic Systems

TP-39 A Method for Analyzing Effects of Dam Failures in Design Studies

TP-40 Storm Drainage and Urban Region Flood Control Planning

TP-41 HEC-5C, A Simulation Model for System Formulation and Evaluation

TP-42 Optimal Sizing of Urban Flood Control Systems TP-43 Hydrologic and Economic Simulation of Flood

Control Aspects of Water Resources Systems TP-44 Sizing Flood Control Reservoir Systems by System

Analysis TP-45 Techniques for Real-Time Operation of Flood

Control Reservoirs in the Merrimack River Basin TP-46 Spatial Data Analysis of Nonstructural Measures TP-47 Comprehensive Flood Plain Studies Using Spatial

Data Management Techniques TP-48 Direct Runoff Hydrograph Parameters Versus

Urbanization TP-49 Experience of HEC in Disseminating Information

on Hydrological Models TP-50 Effects of Dam Removal: An Approach to

Sedimentation TP-51 Design of Flood Control Improvements by Systems

Analysis: A Case Study TP-52 Potential Use of Digital Computer Ground Water

Models TP-53 Development of Generalized Free Surface Flow

Models Using Finite Element Techniques TP-54 Adjustment of Peak Discharge Rates for

Urbanization TP-55 The Development and Servicing of Spatial Data

Management Techniques in the Corps of Engineers TP-56 Experiences of the Hydrologic Engineering Center

in Maintaining Widely Used Hydrologic and Water Resource Computer Models

TP-57 Flood Damage Assessments Using Spatial Data Management Techniques

TP-58 A Model for Evaluating Runoff-Quality in Metropolitan Master Planning

TP-59 Testing of Several Runoff Models on an Urban Watershed

TP-60 Operational Simulation of a Reservoir System with Pumped Storage

TP-61 Technical Factors in Small Hydropower Planning TP-62 Flood Hydrograph and Peak Flow Frequency

Analysis TP-63 HEC Contribution to Reservoir System Operation TP-64 Determining Peak-Discharge Frequencies in an

Urbanizing Watershed: A Case Study TP-65 Feasibility Analysis in Small Hydropower Planning TP-66 Reservoir Storage Determination by Computer

Simulation of Flood Control and Conservation Systems

TP-67 Hydrologic Land Use Classification Using LANDSAT

TP-68 Interactive Nonstructural Flood-Control Planning TP-69 Critical Water Surface by Minimum Specific

Energy Using the Parabolic Method

TP-70 Corps of Engineers Experience with Automatic Calibration of a Precipitation-Runoff Model

TP-71 Determination of Land Use from Satellite Imagery for Input to Hydrologic Models

TP-72 Application of the Finite Element Method to Vertically Stratified Hydrodynamic Flow and Water Quality

TP-73 Flood Mitigation Planning Using HEC-SAM TP-74 Hydrographs by Single Linear Reservoir Model TP-75 HEC Activities in Reservoir Analysis TP-76 Institutional Support of Water Resource Models TP-77 Investigation of Soil Conservation Service Urban

Hydrology Techniques TP-78 Potential for Increasing the Output of Existing

Hydroelectric Plants TP-79 Potential Energy and Capacity Gains from Flood

Control Storage Reallocation at Existing U.S. Hydropower Reservoirs

TP-80 Use of Non-Sequential Techniques in the Analysis of Power Potential at Storage Projects

TP-81 Data Management Systems of Water Resources Planning

TP-82 The New HEC-1 Flood Hydrograph Package TP-83 River and Reservoir Systems Water Quality

Modeling Capability TP-84 Generalized Real-Time Flood Control System

Model TP-85 Operation Policy Analysis: Sam Rayburn

Reservoir TP-86 Training the Practitioner: The Hydrologic

Engineering Center Program TP-87 Documentation Needs for Water Resources Models TP-88 Reservoir System Regulation for Water Quality

Control TP-89 A Software System to Aid in Making Real-Time

Water Control Decisions TP-90 Calibration, Verification and Application of a Two-

Dimensional Flow Model TP-91 HEC Software Development and Support TP-92 Hydrologic Engineering Center Planning Models TP-93 Flood Routing Through a Flat, Complex Flood

Plain Using a One-Dimensional Unsteady Flow Computer Program

TP-94 Dredged-Material Disposal Management Model TP-95 Infiltration and Soil Moisture Redistribution in

HEC-1 TP-96 The Hydrologic Engineering Center Experience in

Nonstructural Planning TP-97 Prediction of the Effects of a Flood Control Project

on a Meandering Stream TP-98 Evolution in Computer Programs Causes Evolution

in Training Needs: The Hydrologic Engineering Center Experience

TP-99 Reservoir System Analysis for Water Quality TP-100 Probable Maximum Flood Estimation - Eastern

United States TP-101 Use of Computer Program HEC-5 for Water Supply

Analysis TP-102 Role of Calibration in the Application of HEC-6 TP-103 Engineering and Economic Considerations in

Formulating TP-104 Modeling Water Resources Systems for Water

Quality

TP-105 Use of a Two-Dimensional Flow Model to Quantify Aquatic Habitat

TP-106 Flood-Runoff Forecasting with HEC-1F TP-107 Dredged-Material Disposal System Capacity

Expansion TP-108 Role of Small Computers in Two-Dimensional

Flow Modeling TP-109 One-Dimensional Model for Mud Flows TP-110 Subdivision Froude Number TP-111 HEC-5Q: System Water Quality Modeling TP-112 New Developments in HEC Programs for Flood

Control TP-113 Modeling and Managing Water Resource Systems

for Water Quality TP-114 Accuracy of Computer Water Surface Profiles -

Executive Summary TP-115 Application of Spatial-Data Management

Techniques in Corps Planning TP-116 The HEC's Activities in Watershed Modeling TP-117 HEC-1 and HEC-2 Applications on the

Microcomputer TP-118 Real-Time Snow Simulation Model for the

Monongahela River Basin TP-119 Multi-Purpose, Multi-Reservoir Simulation on a PC TP-120 Technology Transfer of Corps' Hydrologic Models TP-121 Development, Calibration and Application of

Runoff Forecasting Models for the Allegheny River Basin

TP-122 The Estimation of Rainfall for Flood Forecasting Using Radar and Rain Gage Data

TP-123 Developing and Managing a Comprehensive Reservoir Analysis Model

TP-124 Review of U.S. Army corps of Engineering Involvement With Alluvial Fan Flooding Problems

TP-125 An Integrated Software Package for Flood Damage Analysis

TP-126 The Value and Depreciation of Existing Facilities: The Case of Reservoirs

TP-127 Floodplain-Management Plan Enumeration TP-128 Two-Dimensional Floodplain Modeling TP-129 Status and New Capabilities of Computer Program

HEC-6: "Scour and Deposition in Rivers and Reservoirs"

TP-130 Estimating Sediment Delivery and Yield on Alluvial Fans

TP-131 Hydrologic Aspects of Flood Warning - Preparedness Programs

TP-132 Twenty-five Years of Developing, Distributing, and Supporting Hydrologic Engineering Computer Programs

TP-133 Predicting Deposition Patterns in Small Basins TP-134 Annual Extreme Lake Elevations by Total

Probability Theorem TP-135 A Muskingum-Cunge Channel Flow Routing

Method for Drainage Networks TP-136 Prescriptive Reservoir System Analysis Model -

Missouri River System Application TP-137 A Generalized Simulation Model for Reservoir

System Analysis TP-138 The HEC NexGen Software Development Project TP-139 Issues for Applications Developers TP-140 HEC-2 Water Surface Profiles Program TP-141 HEC Models for Urban Hydrologic Analysis

TP-142 Systems Analysis Applications at the Hydrologic Engineering Center

TP-143 Runoff Prediction Uncertainty for Ungauged Agricultural Watersheds

TP-144 Review of GIS Applications in Hydrologic Modeling

TP-145 Application of Rainfall-Runoff Simulation for Flood Forecasting

TP-146 Application of the HEC Prescriptive Reservoir Model in the Columbia River Systems

TP-147 HEC River Analysis System (HEC-RAS) TP-148 HEC-6: Reservoir Sediment Control Applications TP-149 The Hydrologic Modeling System (HEC-HMS):

Design and Development Issues TP-150 The HEC Hydrologic Modeling System TP-151 Bridge Hydraulic Analysis with HEC-RAS TP-152 Use of Land Surface Erosion Techniques with

Stream Channel Sediment Models

TP-153 Risk-Based Analysis for Corps Flood Project Studies - A Status Report

TP-154 Modeling Water-Resource Systems for Water Quality Management

TP-155 Runoff simulation Using Radar Rainfall Data TP-156 Status of HEC Next Generation Software

Development TP-157 Unsteady Flow Model for Forecasting Missouri and

Mississippi Rivers TP-158 Corps Water Management System (CWMS) TP-159 Some History and Hydrology of the Panama Canal TP-160 Application of Risk-Based Analysis to Planning

Reservoir and Levee Flood Damage Reduction Systems

TP-161 Corps Water Management System - Capabilities and Implementation Status