42
PRE LIMINARY ENGINEERING REPORT GRAND FORKS NEW RIVERSIDE PARK DAM s.w.c. PRoJEcr No.520 GRAND FORKS COUNTY I t v a. il '*',. li i ,41 fr ll @ o 34s ¿j (t U ,'-:l l'.-][=lË .;,.. Project Location ide e .'a PARK DA KOTA coMMtsstoN t . ltt.' ¿ AN ORKS Tra iler Par k I I'N !oF *l : ll¡ ..il. I t{ ll P¿rk I TJ t- il f NORTH STATE WATER 0 Radio K ox N or\ f _rl Ef \. \\' l.:'t - - 8t T t + ( ¿ Sher Pä¡k L tt t IE Ì t- ø w ¡S¡ .r8 n9 ¡ ts IA : ¿ .T I ¿ ß AI'E 'H VE APRIL 1984

PRE LIMINARY GRAND FORKS NEW RIVERSIDE PARK · GRAND FORKS NEW RIVERSIDE PARK DAM s.w.c. PRoJEcr No.520 GRAND FORKS COUNTY I t v a. il '*',. li i,41 fr ll @ o 34s ¿j (t U ... {o{@@

  • Upload
    lamkhue

  • View
    217

  • Download
    0

Embed Size (px)

Citation preview

PRE LIMINARY ENGINEERING REPORT

GRAND FORKS NEW RIVERSIDE PARK DAMs.w.c. PRoJEcr No.520GRAND FORKS COUNTY

It

v a.il

'*',. li

i,41

frll

@o

34s ¿j (t

U

,'-:l l'.-][=lË

.;,.. Project Location

ide

e.'a PARK

DA KOTAcoMMtsstoN

t

. ltt.'

¿

ANORKSTra iler

Par k

I

I'N!oF

*l: ll¡..il.

I t{ llP¿rk

ITJ

t-

il f

NORTHSTATE WATER

0

RadioK oxN

or\

f _rl

Ef

\.\\'l.:'t - -

8t

Tt

r¡ + ( ¿ SherPä¡k

LtttIE

Ìt-øw ¡S¡

.r8n9 ¡ tsIA

:¿

.T

I ¿ß

AI'E

'H VE

APRIL 1984

PRELIMTNARY ENGINEERTNG REPORÍ

GRANÐ FORIGS NETü RI\rER,SIDE PARK DAMSWC PRO.]ECT #520

APRTL, L984

NorÈh Dakota State Ítater ComrnissionStaÈe Office suilding

900 E. BoulevardBismarck, ND 58505

PREPARED BY:

ROI{ALD A. S9ÍANSONDesiqn Enginee¡

BY:

r

A.Ðinector of Engineering

APFROVED BY:

VERNON FAIIY, .E.SLate

Prepared for theCity of Grand, ForkS

GRAND FORICS RIVERSIDE PARK DAI\4svüc PRo,fEcT #520

I. INTRODUCTION

In September of L977, the North Dakota State Water Commission

entered into an agreement with the City of Grand Forks to investigate

the feasibility of constructing a new water supply dam on the Red River

(See Appendix A). The purpose of the structure would be to replace the

existing dam which was initially constructed in 1925 and is in an ad-

vanced state of disrepair. It has been repaired several tirnes and is in

such a cond,ition that it could fail at any time.

The purpose of this study is to determine the feasibility of con-

structing a new \^rater supply dam just d.ownstream of the existing dam or

at some other possible location. Surveys and soundings have been made

of the imrnediate downstream area and two test holes have been drilled.

The proposed dam and its features and effects will be evaluated. Con-

clusions and reconunendations will be made regard.ing the possible future

studies and investigations required to make the final decision as to the

details of the proposed dam.

-l-

II. DESCRTPTION OF STTE

Grand Forks Riverside Park dam is an old. rock fitled timber crib

dam located on the Red River of the North in Section 34, Township 152

NorÈh, Range 50 V'Iest, Grand Forks, North Dakota. It was constructed inL925 aE an approxj¡nate cost of $75r700, to provide partial storage forwater supply for the City of Grand Forks. The rest of the Cityrs needs

are meÈ by water from Èhe Red Lake River in Minnesota.

The site presently under consideration is about 600 feet downstream

of the existing d.am and is located in Section 33, Township 152 NorÈh,

Range 50 Vüest. Figure 1 is a map showing the general location of the

d.am and the surficial geology of the area.

The drainage area at the dam is about 301100 square miLes. The

average slope along the main channel is 0.4 feet per mile. Within the

basin the land is used prirnarily for agriculture and the averagie annual

precipitation ís 20 inches.

Near surface sedj:nents along the Red River in Èhe Grand Forks area

are comprised. primarily of clay and. silt deposited by the Red Ríver or

by glacial Lake Agassiz. Along the Red River, up to 20 f.eeL of alluvium

directly underlies Iand surface. The alluvium is prinarily clay and.

silt with some fine sand, reflecting the sediments in the Red. River

drainage basin. The recent alluvium is underlain by approximately 80

feet of lake sediments. The lake sedimenÈs, Iike the alluvium, are

primarily silty clay with occasional sand lenses. In places, reworked

glacial till is present within the l-ake sediments. Approximately 100

feet of glacial tiII underlies the sediments deposited, in glacial Lake

Agassiz. The glacial Èi11, in turn, is underlain by up to 20 feet ofsand and. gravel. The sand and gravel is either glacial outwash or

-2-

preglacial atluvium overtying bedrock. Ordovícian !{innipeg group bed-

rock underlies the area at approxJmately 200 Èo 22O fee| bel-ow land

surface.

-3-

SASXATCHEWAN MANIl ooa---- I fU LE

!

IIv'(tt ì *-la

I

\I \

aqI I L I tlot.l af

a

I1I

R tnrrlla

l!

\

Tor¡¡r

I

oe4t JI

-

loôr.t .å f o ltlr----1 7

2CENTR LO

l@d Tc Pr c! ta¡¡æ ---lLaoñ

--r¡

/I

2

fn

(D

o

-a

Sh!r'd o ' I ".,,,I c^l"1"'o

¡

li

Slaata

.L O¡vcr

,

II

LI¡

II

itIIIII1t-- t I tddc? lr,,rr-o.&----t-

I

i

Xor Ioñ

I II

It-t

-- s'gf" _ IHalI l

a

,I

, -_-_JG.cã)

IAdoo¡ I nrchlonó---__30uSrour Xc O'c

1H oAl(ofA

RAIRIE

ctú

!fz

9'll- - -1¡

Ë0iñat

iII'1

I

iI

A

{

ELA¡lo^toñ

ve rs iderk

)onê

ATE

¡crLl (taoórfr.d lTor'l Clotton - 1962,o----L¡g----g-.-!c r I i

FIGUR E 1 LOCATION OF RIVERSIDE PARK OAM ANO PHYSIOGRAPHIC PROVI¡¡CES OF THE AREA

IfI. HYDROLOGIC ANALYSIS

A hydrologic sÈudy has been sponsored by the U.S. Department of

Housilg and Urban Development to invesÈigate the existence and severity

of ftood hazards in the City of Grand Forks, North Dakota and to aid in

the administration of Èhe Flood Insurance Act of 1968 and the Flood

Disaster Protection AcÈ of 1973. The results of this investigation were

published in the Flood Insurance Study, <lated September 1977 -

The Water Resources Division of the U.S. Geological Survey fur-nished valuable assistance by providing basic hydraulic data pertaining

to the study area. Survey data, historical flood profiles, and delinea-

tion of areas flooded for past events were obtained by contacting the

U.S. Army Corps of Engineers, St. Paul District. The North Dakota State

Water Conunission and the City of Grand Forks, Engineering Department,

furnished data and cooperated in the stud.y. An interagency group meet-

ing in Fargo, North Dakota, held on December 12, L973, agreed to accept

the 89,000 cubic feeÈ per second as the 100-year flood discharge at

Grand Forks.

The flow-frequency d.ata used for Èhe 100-year flood on the Red

River of Èhe North was coordinated by the states of Minnesota and North

Dakota, U.S. Army Corps of Engineers, U.S. .Geological Survey, and the

Soil Conservation Service, under an interagency agreement.

The Red River flood profiles at Grand Forks are shown in Figure 2.

Ho\^¡ever, as a result of t.he 1979 flood on the Red River, the Corps of

Engineers has now revised, the 100-year flood flow at Grand Forks to be

about l-06,000 cubic feet per second. This revision is defined in the

Grand Forks - East Grand Forks Urban lrlater Resources Study - Flood

Control Append.ix - Corps of Engineers, St. PauI District, JuJ-y 1981.

-5-

9

et -t

¡rEPA

ßil¡E

?rr 0

t ilo

usll¡

G A

ri0 u

n$AN

0Ev

Ér-o

tMttJ

I

ctTY

0t û

tlA}l0

t0Rt

{s, l

l0(G

RAND

FO

RKS

COUN

TY)

RIO

RIY

IR O

I TllE

IIO

RTll

tt00D

Pn0

iltEs

lt

E tï II

jlflf

rril ffi

58 ä

3;??

iä3i

?ii

ojm

-rÍT

=õog

R65

z ó

õgg

er m d) r z I

il1

Li

IJ, rII ïtt ffilIftr

+

ï II H il el r[ï

I1$tï

lt{+

rl

ïil

{{{{

o{@

@oo

oo6@ õo

ó6@

NQÀ

ooo

! o o{ { o

ÈÈËä

@@

ño oo

ELEV

ATIO

N IN

FE

ET (M

.S.L

.)

N @ N @

¡ - m v =

@ @

{ T m = o + 2 o m 2p <3 m @ o m s o c I -

ñ @

-rl r o o IT.

I' vv oo C- T

Irn

=qi

n-l

92 =2c) É ¡rl

-'l I

@ o

l 6) c ¡ m N

IV. HYDRAULIC ANALYSIS

The new dam will be designed to pass the expected flows without

causing any increase in any flood flows. The dam will have a 300 foot

Iong weir set at an elevation of 798.0 msl. The side sloPes will con-

form to the natural ctrannel banks. In order to preselî\¡e laminar flow,

the dam will operate under weir flow control until the weir submerges

and the natural ct¡annel controls the flow. This occurs at an elevation

of 803.6 msl and a discharge of 13,250 cfs as is shown in Figure 3, 4,

and 5.

The changeover from weir control to channel control- shoul-d occur

within an elevation range of l-foot with only a minor amount of tur-

bulence. Thereafter, the flow should continue in a normal channel flow

state with no turbulence or ripples indicating r^¡here the completely

submerged structure is located.

Erosion and damage to the channel and river banks will be kept to a

minimum with reasonabl-e amounts of rock riprapping. Extraneous effects'

caused. by floating tree trunks trapped. at the dam structure, may create

certain turbulent and erosive currents. Tree trunks should be removed

as soon as possible.

Another factor to be consid.ered is ice jarns. The size and effect

of these buildups can vary considerably and are difficul-t to predict and

evaluate.

. Laboratory model studies should be made for the proposed. dam before

final plans are prepared. Results of the hydraulic tests serve as part

of the design criteria.

-7-

CHANNEL CAPACITYEL. 803.6 = SUBMERGENCE

GRAND FORKS NEW RIVE RSIDE PARK DA'Mswc # 520

WEIR SUBMERGENCE STAGESScolc: ltt=5tFreune g

14,700 cFs252 CFS

8.OOO CF.S5,196 CF s2,828 CFS

sEL.798.O

EL.786.O

I@

I

XY TC-7

=t- cll

ÃFGEg___.iNNEI.

*ù 52

--+:-- -NE)R KÍIDF_l-

---/ L--!

-+r,+ !-

'{,É..+tl,-

--T--- 'll

- #

-------!_- -T-r-

=---..1_]-

+-l=lr ]- ?r€f L€'VATIOÀ

-...-..--

--++-¡i-- -/+

=Æ-€r

-].. l- +_

----tir;!:t ã--ø¿

' r-'-__---7- i-l] i_-E

-i-.---_r--_i- -

L-ffi'-T- r+__r+-

f--_!--']__

,tÌ-..:l..r Í---rj-Í--+Ì-.-

1)

e06

804

802

800

798

794

79?

790

788

(o

796

:

7,OOO I,OOOFIGURE 4

786o I,OOO 2,OOO 3,OOO 4,OOO 5,OOO 6,000 9,OOO |O,OOO il,OOO |2,OOO |3,OOO |4,OOO |5,OOO

TURAI-- CH,+-

,lD- E _NE DAM -..E

--t.._

--l

+ t-'eir.'{

'Iitr-.i-+-+-i-

lrt-+-tr+

- - -+i-T--+

-l

I -J-l_Ta+-

rll--l-ï- |-r'- T-:+- ---al

:ir+_l:--!-/

=-<4 -l-

-7

-i--lT--

}_]-!-'ì-t--r+ vE-

ir:

?tc

830

a25

820

8t5

8ro

800

795

79

785

7805,O

I

õI805

OO IO,OOO |5,OOO 2O,OOO 25rooo 3O,oOO 35,OOO 4O,OOO 45,OoO 5O,OOO 55roOO 60,000 651000 7O,OOO 75,OOO 8O,oOOFIGURE 5

V. DAIVT FEATURES

The dam will be a reinforced concrete drop structure shaped to

conform to the configuration of the natural channel. There will be a

flat concreÈe slab with a vertical concrete wall at the upstrea:n face as

shown in Figure 6, 7 and, 8. A ro\Á/ of sheet piling will be driven at the

upstream face of the structure to cut off seepage under the dam. Another

shorter row will be insÈa1led along the downstream face to prevent

undercuÈting by erosive and turbulent currents. The structure will be

supported by structural piling as discussed, in the soils section of this

report.The slab will be placed upon a base of select granular material.

To help relieve uplifÈ pressures that may develop, 2-inch diameter

tubular seeps will be installed between the granular material and the

downstream face of the dam. The upstream and d.ownstream channel and

banks will be riprapped to red.uce erosion. The rock riprap will be

installed upon a base of filter maÈerial and geotectrrical fabric to

reduce undercutting. One feature that should be considered. in additíon

is the possibility of further channel stabilization with a fly-ashportland cement admixture.

Fina1 determination of the dam confignrration will be made after an

evaluation of the information developed in Èhe proposed investigations'

and model studies. These studies and investigations may establish some

criteria that could significantly affect the final design of the project.

Therefore, all detail-s used in this preliminary report are subject to

change and are intended. to represent a starting point from whích the

future studies are to be expand.ed.

-ll-

: El. 798.O

Et. 7 7t.o

Selecl Fill

a3

ao

oõooB

Select Gronulor F¡ll

C,oo ôos

El.78t.O

{ (' {I{ Ii {{ t¡t I

¡r¡{{{ .t 1: ( i{{ {{

t.lc {.1

{t

Ia

{1'

{

Rock Riprop

Selecl F¡ll

Et. 786.O

Sheet Piling

GRAND FORKS NEW RIVERSIDE PARK DAMswc # 52O

TYPICAL SECTIONScoll: l

t'= 5'FIGURE 6

Et.786.O Rock Riprop

EIEr.78l.O

Selecl F¡11

El. 777.O

-

IñI

Et. 762.O

55'

400'

Weir E1.798.O

GRAND FORKS NEW RIVERSIDE PARK DAMswc *+ 520

PLAN A ELEVATION VIEWS - NEW WEIR STRUCTUREScc le : ltt = 5o'

FIGURE 7

35'

@N

PLAN

-I

(rlI Et. 8t5.O !

E1.762.O

EL EV.

-

50' 3oo' 50'

\ ¿Sheel Piling

!8-s

tì':þ¡,' t

tt.9

7é.9

rå.9

77'9

a9.9

-6¡.?\

s3l

t98.oeoao¿gPe4

o5rt

17.?

q6. t

o€7

g',ôr7

89.7

?,'B

1ô.È

t6. ¡

ETÞS AERÊ

¡1.5

o{.?

10,o

t 7.9

l?.4

ßf14r8

4'ál¡.t

06.4

I ?.q

râ.3

ta.9

l ¿.o

12.Z

03.g.¡.

-.,,

IO.?

oa.1

o!. ô FEIEE

o'lr- ; ot'5---- \.-tt*..-

co'ä

gs.7

#I

û0.1 ""{o¿. ô

os 2- Ósl't

".L --?6.O

ol:5

:81'l

89.e

17.9

?l,e

-ì-rsr zì*

oô.1

o6.!

74,t

a6-q

ô4-ã>Ða-nã

cönoðE€ 0r'?'\_..-.-95:4-

6!,0

?5.1

tsI

I

I

l..!Ilãlo

l3plotã¡-

?Þ6

7a.7

69.1

6t.3

7A.O

oJ.l

04 '8FIGURE 8

65'O

6{.?

69.9

6,O

64.?

I

I

I'l-I

I

IJ

I

I

t

I

?5. r

a5- tî

7a,t617'¿ *! "+

797094

tlt6

qq.t

6?'à

72.2

79.7

ao'?

8r.9 ð1,4

-a5.7+-.- u.r/

1ìO'5

oâ.o'9ó'å

o¡'q

03,?

?8.{ ?O'3Bl 7-':\9+*¿l Lt EIEtr\Ë-r *t ãF-\ I "t!"9?.6

' _---

1t.2

7s'4 ?

c¡.s 1

¡jvTuÈJ t.¿' XGâô¡ êuLvERtt,

I ;| -/'

- 99;O

ø4.¿

€oRçaEfÉ

oa-aso208 0!.4

ozt f

GRAND FORKS NEW R]VERSIDE PARK DAMswc ++ ô20

O4.8SITE LAYOUT A TEST BORING LOCATIONS

o8'1

ot'o

o4.4

03,¡

ðit.ltl'l

ôt.i

O9.9

ró{

p.6.

ã9r 6 99.O6¡.t

07.¿

lo.l

¡2"A

VI. CONSTRUCTION REOUTREMENTS

The reinforced concrete weir structure will be constructed. in two

phases. A cofferdam should be constructed out from one bank and. pro-

tected with temporary riprap. The area within the cofferd.am would Èhen

be dewatered. Upon completion of Èhe dewatering, the bottom material

within Èhe coffered area would be excavated. to a depth of approximately

6 feet below the existing elevation. This area would then be backfilledwith a select material. The upstream and d.ownstream rows of sheet

piting and structural piling would then be driven.A select granular material base for the slab would be installed and

the reinforced concrete slab placed. thereon.

The structure would be constructed in sections of about 20 feet inlength. Each segment will be joined to the adjacent member wiÈh steeldowel bars. The joints will be sealed with a rubber expansion jointmaterial. Construction joints bet\¡¡een the slab and vertical walls would

have a steel p1aÈe installed through the joint to provide a cutoff.After completion of the structure, the ad.jacent area would. be

shaped. to grade, filter material and, fabric installed and rock riprapplaced.

If concrete is placed during the winter, the concrete would have tobe heated. and protected, from the elements during the curing period..

The cofferd.am would require upkeep during this entire project. Itis expected that each project phase wiII require about three months ofcontinuous construction. lfith the demolition and replacernent of cofferdams between phases, the project may require a total of seven months.

Assuming that the work could sÈart in JuIy, after the spring runoffhas decreased, the project could run through January or February.

-15-

surrner storms could cause this sched,ule to be extended by a month or

more, resulting in most of the work being accomplished during the winter

months. This may be feasible but the con5Èruction costs will be pro-

portionatelY higher-

-16-

VTI. VÍATER CONTROL DURING CONSTRUCTTON

One very serious consideration in planning the dam construction

program is whether the site can be protected from high flows during

construction. Previous construction work has been accomplished by

cofferdaming one-half of the river at a time. This has apparently

proved to be adequaÈe for a shorter duration effort and may or may not

be sufficienÈ for a project which will require considerable more time.

The layout for the proposed first phase cofferdam is shown in Figure 9.

The assumption is that the horseshoe shaped cofferdam will be

constructed by hauling in a fat, cohesive clay fill and dozing it intothe stream in the d,esired. configuration. The temporary earthen dike,

will be protected with a riprap facing. The cofferdam will require

continual upkeep and repair until the first half of the dam is com-

pleted. At that time, the first phase cofferdam will be demolished. and

the second phase cofferdam will be constructed. It will come out from

the opposite bank in a manner similar to the first cofferdam. Since

this phase will involve flow over the completed weir, instead. of through

a constricted open channel, the elevation of the cofferdam will have to

be proportionally higher.

A design flow of 51000 cfs was used as a trial f1ow rate. It \nras

routed through the natural channel and the channel as constricted by the

first phase cofferdam. The details are shown in Figure 10.

The figures show that in the natural channel situation, the fl-ow of51000 cfs would result in a water surface elevation of 79L.ll- msl and a

velocity of I.24 fl/sec. I/üith the cofferdam in place, the \^rater surface

would be 790.91 msl and the velociÈy would be 4.62 ft/sec. This indicates

that the cofferdam would be a feasible proposal, but it will require

-L7 -

rock riprap facinq and considerabl-e maintenanee to survive the erosion

caused. by high ïrater vel-ocities-

Ttre cofferdan will have to be constructed with Èwo or three feeÈ of

freeboard to provide proteetion against sudden rises in river elevation

due to heavy rainfalls. Assr:rning a three month firsÈ Phase consÈruction

schedule, at least one or two of these short intense rainfalls should be

anticipated and planned for in advance. This situation will pose a

probtem for potential contractors-

-L8-

ra.a

tf.t

o5.a

oo. a

FEÙè€

IOaiz-

t6.!

E-!

ETDS]NÊRÈ

---\

l?9

09 I

77.9

t5.9

7?.9

77.9

E,S

\g¡-l- .

a€.3

80804f{

o5.7

06

06.?

so.?

8Þ'?

?r.9

t5.z

7 t.6

75.7

6r.a

'-9!.r--.,

sq.6

Oã'O-

03.7

!7'1

t5'3

7:7.2

l?'3

ll'6

o,4'2

ol.5

90.6

t{-8l?rst4 'l

t¡.? t€.9

ó4.7

09.6oQ,8

9l,O co'7

a3,7 -*

-

6¡.1 =\t|.f --\

t0! o

ol.a

à0'3

8 a.$

oo:3

9¡.o

GRAND

ol'â

oÊ. I

0€.3.-....-..:sr.t

tg-9I l'tÕ

II

-s.7*3Þ .- sr'z.-.ã\q6S:3

dc.?r oo,t

64"2

o8i" ¡¡rxore s'¡'-aBovE äRoorD

---=,-al'1cwcñETf

ÀPÊ0ll

o?'996-6

o4.5

z

I

t,si¿e¡l.l

--dåm7It7

Ò.4 -4+.O--

7'l-à

7t2 I

IIl-ItI

?èi1

0.9'l

67.?

I

I

I1f

tb

leI

9à.3I

¡9l.tðtri

e¡,?I

I

III

I

l

7€'O

o4,2

DAMê '.1

73.8

't.5, t?

v4.o

tg' 4:l

!1.6 ?86.O

95,ô

?t.¡

?l'9 t5. I7+.9

I

õI

to.o

72,2

ao'2

Ê4.7 .a-.Éq€r€

6?'9

Ê9.9

7t.965. O

7¡, ?

70.3s-?--".-45'?\ll9 Þe:ß àsàlfi "æ\@'? a2.4

7970987t.è

a!.9,

9t.5

06.l

o€. ã

lz't

67.¡

ro.0

73.48l'7 .-3\9t\oule?'6

-8t',¿"'

ôl;0

G.5

st.4-/

ò4.+

I ?.O

90. I

ao208o!. t of- t

o2'7FORKS NEW RIVERSIDE FARK

sWC r+ 520PHASE I COFFER DAM LAYOUT

(PHASE tr TO BE REVERSED)ôú. o

õ7. O

o. t.9

04.6to.l

t?6

ot.6 æ.ool. I

FIGIJRE 9o?,9 oa.9

o

Ê5

{lrl-.8 f0 X l0 ìO fllE l¡lctl!, \\ l{) l'¡, rlr'lñ\"é xr.t lr I r:t- & F'i1I R ri() tr1,rr r{ ,r 1

^

4Ét 0780t

5.O

4.O

3.O

2.O

t.o

(I/

I

I

)I,

IIIIiI

I

jI

t\

/. \It

IIt\

/ a//

l

i

I

I

III

I

IIl

Illil

r

l

I

1

aI

I

I

)

-tI

rl(I

II(

\\

I

1

l

\IN

I

0Y

III

I

iI

D

il

Lf tA )vIti

^I

I

I

(E'l

l/

I

ì

iIIIq

l

I

I

I

I

I

I

792

790

788

786

784

782

780

I

tuoI

SrOOO4|OOO3,OOOFIGURE IO2,OOOI,OOO77

VIII. SOILS INVESTIGATION

A prelirninary subsurface investigation was performed by a soils

engíneering consultant. Íræ d,rill hoJ-es were mad,e¡ onê on either síd.e

of the river just downstrean of the existing dam as sho!ûn in Plgure

8- The soils investigation report is lncluded in Appendix B.

The limited scope of this initial investigation serves to indicatethat a much more detailed investigation will be required bef,ore the

final design of this projecÈ can be completed. The tlpe and, J-ength ofstructural piling required was d.iscussed, buÈ not in detail in the soí1s

report. A second area that will require study is the tlpe and lenEth ofsteel sheet piling that will be needed for upsÈrean and downstream

cutoff purposes. The Èhird concern Ls the slope stability of the riverbanks. This is certainly an important factor that must be fully eval-uated. A subsidiary consideraÈion could be possible of stabilizaÈionwith fly-ash and portland cement.

-2t-

IX. PROJECT COSTS

Since the project is still in the preliminary investigation stage,

any cost estimaÈe for the project must be of a preliminary nature. f?le

basic concept for the design is reasonably close to what the finat

design will be, thereforer the costs should be close enough for general

project financial- planning. As is shordn in the detailed cost esÈimate

below, all anticipated costs have been identified. The most reasonable

quantit,ies and þrices (1984 construction season) have been used. If the

project is d.elayed several yearsr. inflation may escaJ-ate these prices.

La,nd acquisition costs are not inch.rded in this estimate anp must

be added. It is anticipated that the city will be abJ.e to acguire the

].and,.

Tt may be desired to look at the feasibility of constructing the

dam at a different site than the one considered. This may also change

the costs.

Hov¡ever, it is still considered that Èhe estimate should be cl-ose

enough to cover some reasonable changes or relocatíons.

GRA}T.D FORKS NEV{ RTVEESIDE PARK DAI4

,Preljminarl¡ Cost Estimate

No.

1.

2.

3.

4.

5.

6.

7-

8.Ò

10.

1r.

L2.

Item

SiÈe Pfeparation & Restoration

Mobil-ization

Cofferda¡ns and !Íater Control

concrete (Reinforced)

Sheet Piling

Structural PilingRock RiPraP

RRR. Filter Material

Excavation

Fi11

Drainage Fi1lMisc. Material-s

guantitlr

rsLS

Unit Price

LS

L,ooo cY 30o.0o

L6,50o LE 20.00

24, 000 r,r 20. 00

1,600 cY 25.00

500 gs 10.00

1on0o0 cY 2,OO

101000 cr 1.00

L,ooo cY 12.CI0

LS

Subtstal

Contingencies

Engrinee::ing

Contract Ad¡ninlstration

Total

$ 10,000

10, o0o

10o, 000

300, ooo

33O, 000

480, 000

40, 000

5, 0oo

20, 000

20,0o0

12, o0o

5,000

$1, 332, 0oo

200,00o

134, ooo

134, O00

$1,80O,O00

-23-

X. RECOMMENDATIONS AND CONCLUSIONS

It appears that it is possible to build the proposed new Grand

Forks Riversid.e Park Dam. There are basically three questions to resolve

in regard to the construction of the dam. These are: should a new d'am

be built, how will it be financed and who will share in the cost? If

the project is to be pursued, the technical details, such as the location'

soils invesÈigation, piling requirements and hydrologic and hydraulic

characterisÈics' must be fully evaluated.

This program should begin with consideration of the expected costs

and the decision of the sponsoring agencies whether they wish to proceed

with a more thorough investigation of the project which would enable a

final design to be PrePared.

Because of the apparent condition of the existing dam, a fulI

investigation should be initiated as soon as possible. Because of the

complexity of the studies and the financing d,ifficulties, it is expected

thaÈ the studies shoul{ be started this year. This will allow the

project to be scheduled for construction in 1986-

-24-

. APPENDIX A

Project Investigation Agreement

StrC Project #1536

Prel imi nary I nvesti gat ionby the

North Dakota State Water Cornmission

I . PARTI ES

THIS AGREEMENT is between the North Dakota State llater Commission,

after referred to as the Commission, acting through the State Engineer,

Fahy, and the City of Grand Forks, hereinafter referred to as the City,through its Hayor, C. P. 0rNei I l.

ô!_tE_E r_!.I-l

here i n-

Vern

act i ng

il. PRoJECT, LoCATIoN AND PURPoSE

The City has requested an lnvestigation to determine the feasíbility and

design for a new water supply dam on the Red River of the North. The proposed

location would be in the vicínity of the existing Riversidè Park Dam.

lhe City currently receives its municipal water suppìy from the Red River.

The water is stored within the channel of the river hy a ìov¡head char¡nel dam.

The existing dam is in a state of temporary repeir and Ís structuraìly unsound.

The repairs made are not considered adequate to meet future water supply demands

I I I. PRELIMINARY INVESTIGATION

The parties agree that further information is necessary concerning the

proposed project. Therefore, the Commission shal I conduct prel ìminary investi-getíons consisting of the fol lowing: hydrologic and field surveys, topographic

mapping of the dam site, office studies, design and specifications and cost

est ¡ ma tes .

Soil foundation investigations, laboratory test¡ng and stability analysis

shall be a responsibility of the City.

IV. DEPOSIT - REFUND

The City shalì deposit S1500.00 with the Commíssion. Upon completion of the

preliminary investigation, upon receipt of a request from the City to terminateproceeding further with-the preliminary investigation, or upon a breach of th¡sagreement by the city, the commlssion shall provide the cíty with a statement ofall expenses incurred in the preliminary investîgation and shall retun any un-

expended deposlt funds.

-t-

V. RIGIITS OF E}'ITRY

The Oity agrees to obtain wrltten permission from any aff,ected landov'lner

for surveys or subsurface lnvestl.gations by the Conrnission (or any contt'actor)

which are requlred for the preliminary lnvegtlgation .

vl. tNDEt4NlFtcATlol'¡

The City her.eby âccepts r:eponsiblllty for, and holds the Comnission free

from, all clalms and damages to public or private propert¡esr riEhts, or persons

åris¡ng out of this Investigâtlon. ln the event a sult ls l'nltlated or judgment

entered against the Commisslon, the Board shall indefünîfy it for" any judgmçnt

arrived at or judgment sat¡6fled.

CITY OF ORAND F.ORKS NORTH DAKOTA STATE T.IATER CO}IMISSION

Septe$ber 2A, L9'17Dâte

tl--,:J"t --Vern Fãliy -State Englneer

September 8, 1977Date

0

APPENDIX B

T\lvin City TestingSubsurface Exploration Report

REPORT OF SUBSURFACE EXPLORATION PROGRAM

PROPOSED LOI{ HEAD DAM

RIVERSIDE PARK

GRAND FORKS, NORTH DAKOTA

#120-11173 and #54-1065

I NTRODUCTI ON

l,le understand that a nerú reinforced concrete low head dam will be constructeda short distance downstream of the existing dam in the Riverside Park area ofGrand Forks, North Dakota. l,'le further understand that present plans call forthe concrete dam to be supported on a driven pile foundation.

l'le have performed a subsurface exp'loration program and engìneering review forthe proposed cosntruction. The scope of our exp'loration program and engineer-

ing revie¡{ on this project is as follows:

1 To put down two standard penetration test borìngs in thelocation of the proposed dam.

To perform laboratory tests on representative so'i1 sampìesin order to estimate pertinent soi'l parameters.

To recommend driven piìe types and estimate requìred pilelengths and recommended harnmer energies for the propirsedpile foundation.

2

To perform a preììminary review of the potential embankmentstabìlity at the site.

This report presents the results of the field and laboratory testing. Alsopresented are our recommendatjons based on a review of the field and ìaboratorydata.

3

4

Èumcltl, È¡l¡Èrnc¡

Page 2 - #54-1065

EXPLORATIOII PROGRAI.I RESULTS

Site Cond'itions

The site explored is approximately 150'-200'downstream of the existing weir jnRiverside Park jn Grand Forks, North Dakota. One boring was done on each side

of the Red River.

Based on a topographic map prov'ided to us, the river is about 200'wide at theproposed dam site. The river is approximateìy 20'deep. The south bank of theriver is approximately40' to 50' above the river bottom. The north bank risesto about 50' or more above the bottom of the river. The side s'lopes-on the

south bank appear to be as steep as 1 on 1.5 (vertical to horizontal) and about

1 on 2 on the north bank.

Subsurface Conditions

The subsurface cond'it'ions at the boring location on the north bank jndicate

soft alluvial and Lake Agassiz clay and s'i'lty c'lay deposits down to a depth of35'. From 35'to 40'is found a layer of loose si'lty sand. underlying thesilty sand are more fat clay Lake Agass'iz deposits down to a depth of 49*'. Atthis depth a lean clay washed tilì deposit js encountered and extends from 49*'down to 89'. From 89'to 103' are stiff to medium stiff fat clay deposits.Below a depth of 103' are stiff to very stiff c'layey sand and sandy c'lay tÍì1depos i ts .

The so'il boring performed on the south bank of the river indicates a surficia]layer of fill from the surface down to a depth of 15'. Th'is fjll is a mjxture

Èu.rl Ctr, ÈEJtrtìcl

Page 3 - #54-1065

of silty sand and clayey sand with some gravel and other debris. Under'lying

the fill are deposits similar to those encountered in boring 1. The depth tothe stiff till deposits is 108'.

The attached boring ìogs indicate the layering of the soil deposits encoun-

tered at the two borings.

Ground water measurements were made at the times and levels as ind'icated on the

attached boring ìogs. Due to the cohesive nature of the soils, the measured

water levels may not be reliable ind'ications of the true steady state water

table. Normally, extended periods of time are needed to estabìish steady statewater tables in very cohesive soìls such as found on this site.

The river level undoubtedly governs the location of the water table on both

sides of the river.

ENGINEERING REVIEt.l AND RECOMMENDATIONS

Pro.'iect I nf ormati on

Our understand'ing of the proposed construction is rather limjted at this t'ime.

l,le understand that the structure will be a low head dam mainìy constructed ofreinforced concrete. We understand that it is p'lanned to support the structureon a driven pile foundation. l.le have no information regarding any earthworkor other loadings to be placed on the river banks.

The scope of this report is just to address the loads, recornmended depths, and

har¡mer types for the p'lanned pi'le foundation. tle also w'ill address jn a

Èu.rìctv Ec!¡trrt¡rÉG¡ImErIEr¡E

Page 4 - #54-1065

preliminary ¡{ay any potentiaì embankment stabiìity problems. Additionalreview and possjble testing may be required to better determine the stabìljtyprob'lems after we have a chance to review the construction plans.

Foundati on Reconunendati ons

It is our opinion that a driven pile foundation for the proposed concrete dam

is the optimum foundation support system. l.le have no information re-garding the required loads for this project; however, we feel that any piìingdriven would have to penetrate at least a few feet into the hard clay ti'l'ldepos'its found at a depth of 103' in boring 1 and 108'in boring 2.

In our opìnion, the most feasible pìle types would be a heavy walìed 7-518" 0r

9-518" 0.0. steel pipe pììe. The 7-518" p'ile can accornmodate loadings of up

to about 60 to 75 tons per pile. This pile would probably penetrate about 3'

to 5' into the hard clay till to develop these loads. Piling of this type

should be driven with a har¡mer havinE an energy rating on the order of 18,000

to 25,000 ft-lbs.

If h'igher loads are required, then 9-518" 0.D. steel pipe pìles with a minjmum

wall thickness of 0.4" is reconmended. These pi'lings can readily develop

loads of up to 100 tons and would probabìy penetrate about 5'to 6' 'into the

hard glacial till. P'i'ling of this capacity should be driven with a harnmer

having an energy rating on the order of 30,000 to 45,000 ft-lbs.

Steei H piling could also be considered and loads of 50 to 100 tons can be

accommodated on appropriate s'ize sections. Steel H pìf ing wouìd probably be

substantially more expensive than the heavy walled pipe.

Èurl4't, EE¡ttlt¡

Test Borinqs

Page 5 - #54-1065

The above estimates are intended only for prelimìnary planning. The actual'lengths should be predicated on a test program that would'involve drivingtwo to four test piles throughout the site. Each of the piles should be

mon'itored with the dynamic piìe analyzer. Thìs would provide informatjon re-garding the actual required length, the efficiency of the hammer, and the re-quired final penetration to achieve the design load with an appropriate safetyfactor.

Embankment Stabi I itv

A preliminary review of the slope stability of the two river banks was per-

formed. Based on the laboratory tests and estimated soil parameters in the

existing fil'l on the south slope, it'is our opinion that the stabi'lity of the

banks should definiteìy be reviewed. Any addìtìonal filI p1a.ced, steepening

of the slopes, or seepage developed after construct'ion of the dam, could

render the banks 'instabl e.

l,Je recommend that a more detai'led analysis be performed on both banks. Once

the final design of the dam js comp'leted, we recommend that you prov'ide plans

and other informat'ion to us so that a review of the stability can be performed.

EXPLORATION PROCEDURES

Two standard penetratìon test borings were put down during the period ofDecember 1,1983 through December 7,1983. The borings were located as shown

on the attached sketch. Surface elevations were referenced to the benchmark

also indicated on the attached sketch.

CU.rl CÊ\' EE!¡ClrrCI

Page 6 - #54-1065

Soi I Samol inq

Soil samp'ling was performed in accordance with ASTÞI: D 1586-67. Using thisprocedure, d 2" 0.D. sp'lit barrel sanpler is driven 'into the soil by a 140

lb weight fa'lìing 30". After an initial set of 6", the number of blov{s re-quired to drive the sanpler an add'itional 12" is known as the penetration

resistance or N vaìue. The N value is an index of the relative density of

cohesionless soils and the consistency of cohesive soils. Thin wa'll tube

sampìes were obtained according to ASTM: D 1587-67 where indicated by

appropriate symbol on the boring logs.

Soil Classification

As the samples were obtained in the field, they were visua'lly and manua'lly

classified by the crew chief in accordance with ASTM: D 2488-69. Representa-

tive portions of the sampìes were then.returned to the laboratory for furtherexam'ination and for verification of the fjeld classjfication. In addition,selected sampìes were submitted to a progran of laboratory tests. Logs ofthe borìngs indicating the depth and identificat'ion of the various strata,the N value, the laboratory test data, water level informat'ion and pert'inent

information regarding the method of maintaining and advancing the drill holes

are attached. Charts illustrat'ing the soi'l classification procedure, the

descript'ive terminology and symbols used on the borìng ìogs are also attached.

Èr¡rlcñr,InÉæ EE¡ÈNCI¡EÈIÀr

Ou

ca

PLLL

30i5

o

98

LABORATORY TESTS

w

25

TYPE

SB

SB

SB

SB

SB

SB

SB

SB

3T

SB

FA

3T

SAMPLENO

'ì'l

0

2

5

3

4

6

7

I

9

'l

1

I

2

WL

V

N

6

3

5

5

4

8

3

4

8

GEOLOGICORIGIN

FI LL

LAKEAGASS I ZDEPOSITS

{-sunrace ELEvArtoN ?ö'd:T9' MATERIALOESC

FILL, mixture of SILTY SAND andCLAYEY SAND, a trace of gravel,some concrete, brick, andcinders, black and dark brown,frozen to 1Þz'

SILTY CLAY, 9Fây, soft tomedium, a fevJ lenses of siìt,and wet to waterbearing sand (CL)

l4EDIUll FAT CLAY, 9FdV,medium to rather stiff

Continue on next page

( cH-cL)

pFoJEcr PR0P0SED L0lrl HEAD DAM-RMRSIDE PARK-GRAND FORKS. NORTH DAK0TAVERTICAL SC^LE54-'1065 2BORING NO

LOG OF TEST BORING

JOB NO ltt = 5r

OEPTHIN

FEET

3t

l5-

28.

EUTN Cl¡Tt, EESÈlrllE¡sE-3 (77-8).5 rrÉ --!rr'lÌ¡ aEErlEÉlI r-.

OuLLPL

80E

37T

o

o4Ï

#10

70

1

LAEOFATORY TESTSW

25

51

22

TYPE

3T

SB

SB

SB

SB

SB

SB

SB

SAMPLENO

l3

15

l6

17

18

l9

20

4l

WLN

-12

'10

5

5

7

7

IGEOLOGIC

ORIGIN

LAKEAGASS I ZDEPOS ITS( Cont)

LAKEt.lASHEDTI LL,LAKEAGASS I ZDEPOS ITS

MEDIUII FAT CLAY' (Cont) (CH-CL)

FAT CLAY, 9rây, medium (CH)

LEAN CLAY, a trace of gravel r gFô!,medjum to rather stiff (CL)

Continue on next page

OEPTHIN

FEEf

PRoPoSED L0tl HEAD DAM-RTVFRSTnF pARK-fìÞANn FnÞkç NnDTLt n^rnr,l5r54-'t065 lrt 2 (Cont)

LOG OF TEST BORINGBORING NO.

PROJECT

JOB NO vERlrcAL scALE

47.

58_

70-

J5

Eul¡n cñrv EcsEmclrrE' -rgrrtll-¡ ¡EõlcAlll r-sE.3 (77.8;.t

OuLLPLD

104

LABORATORY TESTSw

22

TYPE

SB

SB

SB

SB

SB

SB

SB

SAMPLENO.

21

23

27

2

2

2

2

WLN

0

I

8

I

7

8

l0

10

I

EOOR

LOG CIGIN

G

LAKEt.lASHEDTI LL,LAKEAGASS I ZDEPOS I TS( Cont)

AGASS I ZDEPOS ITS

OESCRIPTION OF MATEHIAL

LEAN CLAY, (Cont) (cL)

FAT CLAY, 9rô.y, rather stiffto medium to rather stiff (CH)

Continue on next page

DEPTHIN

FEET

pFoJEcr PROPOSED LOlrJ HEAD DAM-RMRSIDE PARK-GRAND FORKS. NORTH DAKOTA

2 (Cont)BORING NO.VERT¡CAL SCALE54-'t065JOB NO.

LOG OF TEST BORINGlrt = 5t

105-

IU

94

Irac¡o'rÉLr^vr*re SCIILg}(PLC¡FtaðClf| sr P UuMN sstt.

OuLL.PLo

126i¡

LABORATORY IESTS

w

21

TYPE

SB

SB

SB

SB

SAMPLE

NO

2B

29

30

3l

WL

4 :00I 2-6-83STAFIl

@

COMPLEIE

MErHoo 6FA 0-lgtr'4DC O-24r'Dl4 I 9!6 r -l'l 92 |

N

55

3I

00

98

18tr<lNI'IRNIIR

WATENLEVEL

GEOLOGICORIGIN

LAKEAGASS I ZDEPOSITSIc0NT- ITI LL

BAILED OEPTHS

lotolo

CAVE-INoEPTH

I g'None24ulNone

CASINGOEPTH

SAMPLEDOEPTH

21 |

121'121',

ftrrE

ll:004:004:30

FAT CLAY, (Cont)

End of Boring

#Cal cul ated

SILTY SANDY CLAY, a trace ofgravel, very stiff (cL)

I

JoBNo 54-1065 vERrrc^LscALE ltt = 5 |

PR.JE.TPRoPoSEDf .o_MEÃ,D_o-ÃM-RIVERSIDEPARK.GRAffi_SJ,IORTHõiläiÀ.2 (Cont)

RINGLOG OF TEST BO

DEPTHIN

FEEI

105

108.

121 -

CREW CHIEF

TATER LEVÊL IE SUREXEXÎS

DATE

lo

sE.2 (77-8ì .5 Eurn ctFt ÈrsÈrlcl

CoNFTNED - STRESS - STRAIN CURVES - utrcorurtrueo

December 28. 1983Date 120-11173Project No,

Proiect: DAKOTAPARK.GRAND FORKS. NORTHPROPOSED LOt^l HEAD DAM-RIVERSIDE

1.5tlU'F.t)Ør¡JÉ.l-v)fEoF5tUo

1.0

0.5

0 0 5 10 15AXTAL STRATN (o/o)

20 25

sl€tch ol sampl€afler lailure.

Boring No. 1 Sample No. B

Depth (ft) 17-19* ( too Isoitrype SILTY CLAY (CL)

(inches)-L80x 5.85U-U

21.405.5

?Á6LL(yù

Diameter-HeightType ofTest

11.6 osiConfining Pressure1.15 tqfMax. Deviator Stress

Moisture ContentDry Density

PL (V") 12 -7!-l

I

I

,l :lI

-+-J

t-Ft::'-:

=

---.1

-l

¡I

+ïtl

If

II

I

rlrI -.r.rl¡

I

I

I

II

!-t-1

i

I

I

l

t'1 ,j l¡i'''l'-l

Ijlli1l iI r-1 :

L I l.i

l,t,lilll¡rltr:

',ii

r-l iirll;lil!'ìirl.¡

lrØFoØt¡JfEØfroFulo

1.0

0 0 5 10 15AX|AL STBATN (%)

20 25

Sketch ol sampl€ette¡ lallure.

1 sttple No. 1127-29+ (bot. )

Diameter-Height (inches)-fu!f, x -L96

Dry Density (pcfl 97.4 '

LL Øl 40 .7 PL (%)

SILTY CI AY ICI ISoilType

Moisture Content

16 _4

lo/^l 26 -?

lt-ilType ofTest16.6 psiConfining Pressure0-R7 fcfMax. Deviator Stress

Boring No.Depth (ft)

iI

J

I'1

I

I-.t

I

-t.l

II

,J

-1

-1'+

I

J

I

i.l_t

I

_a.-¡

I

,,¡,i.iL l.,l

'l'll itt;, .:.r

1..ll

ir'-¡-iiíl-ir. I

i i -i'"!iii,iiii...r . ! -:

.: -. l-l,'t.i.

lr;lilr Irilirlii'"t ;

'l;'

i :. ì -[-1liriir I .¡-.1il,.ir-lll,li.i l.;'ilrl,

0.5

662 CROMWELL AVENUE sc¡rt exPloRatron ST. PAUL. MINNE9OTJ

lr{ :.

i:t,¡.I

{'.

-,l.r*t

llt;Il,

r't

It.ì

t_

rtl(-lii

Ë.II

l

I

I

l!U)bu)ØulfEFØGo5IUo

0 5 10 15AXIAL STRAIN (o/o)

Sketch ol sampleetler lsilure.

Boring No.-Sample No.-Depth (ft)Soil TypeDiameter-Height (inches)- x

-Type ofTestConfining PressureMa¡<. Deviaùor StressMoisture Content (%)

Dry Density (pcf) - "LL (%) PL(%')

''i-il:i ¡

J

I

f.I

_t

I

l

l¡:-t't_

'i-1

,.i

t

i

{

t

20 25

l!rnbØU)UJccl--U)fEoF[rJo

1.0

0.5

0 0 5 10 15AXIAL STRAIN (o/o)

20 25

Skelch ol sampl€afler lailure.

2 Sample No. 922-24+ (bot. )

(incnes)-Z'87.x 5'55U-U

Confining Pressure 17 ' 6 Ps'iMax. Deviator Stress 0 .78 tsf

LL wl 30 -4 PL wl 16 -2

sIt TY Ct AY (Cr )

?q7Moisture Content (%)

Dry Density

Diameter-HeightType ofTest

Boring No.Depth (ft)SoilType

¡

-tt.!l'+-{.++

-l_1.-

I+++

.J. I

++

-LI

++r

+t.-t --+--f

i

I

II

l

I

I

j-i 'r-i'

I

II

:

III

i--lr1.l l1i"i

lllFI

I

t.

I

II

f

coNFTNED - STRESS - STRAIN CURVES - uxco¡rn¡¡eoPROPOSED LOt^l HEAD DAM.RIVERSIDE PARK-GRAND FORKS, NORTH DAKOTA

12î-1117"Pro¡ect No,Project:oate: December 28. 1983

A662 CROMWELL