Solutions Manual For Surface Water - Quality Modeling

Solutions Manual for
SURFACE
WATER-QUALITY
MODELING
Stephen C. Chapra
University of Colorado
The McGraw -Hill Companies, Inc.

New York St. Louis San Francisco Auckland Bogota
Caracas Lisbon London Madrid Mexico City Milan Montreal
New Delhi San Juan Singapore Sydney Tokyo Toronto
urnriu.
rr:>
3~5
. C. lii (
li<t1
Suppl.
bKS
McGraw-Hill 'E2
A Dil'ision o/ J1ie McGraw-Hill Comp.it1u·s
Solutions Manual for

SURFACE WATER-QUALITY MODELING
Copyright© 1997 by The McGraw-Hill Companies, Inc. Al rights reserved.

Printed in the United States of America. The contents or parts
thereof may be reproduced for use with
SURFACE WATER-QUALITY MODELING
by Stephen C. Chapra
provided such reproductions bear copyright notice, but may not
be reproduced in any form for any other purpose without
permission of the publisher.
ISBN 0-07-011365-3
1234567890BKMBKM909876
http://www.mhcollege.com
1.Jt/lrr---J/ b ~s
40111-\VI
e \ t-(
{ !;t{O(
c..od
NOTE
TO THE INSTRUCTOR AND/OR GRADER
fl 11IJ111 I I I I I I I I I 11 f I I I l I I
This manual has been developed to help you use Surface Water-Quality Modeling more
effectively in the classroom I have solved all the problems in the text and have
reproduced my solutions on the following pages.
In addition, a number of errors have come to my attention during classroom testing. I

have listed these in the back ofthis manual Some ofthe errata relate to homework
problems, and I have kept these errors separate from other types oferrors. McGraw-Hill
should correct all these mistakes when the text goes into a second printing. In the
meantime. you should consuh the errata in the back ofthis manual to mioimiu any
confusion the errors may cause to your students. For example, I have reproduced the list
and distributed it to my students so they can :incorporate the corrections into their own
books.
Fmally. ifyou find additional errors. or just have suggestions or comments you would
like to send me, I've included both my current address and ~mail location below.
Steve Chapra November 3, 1996
Civil, Environmental and Architectural Engineering

Campus Box 428
The University of Colorado
Boulder, CO 80309-0428
Phone: (303) 492-7573

Fax: (303) 492-1347
E-mail: chapra@neptune.colondo.edu
LECTURE 1:
1.1
g I m3
m= cV =8500--2.5 L ·- - = 2125 g
m3 103 L
1.2
3
1 3
(a) Q = 100000capita(650L/ ·1atd) m d 0.7523m
• cap 103 L 86400 s s
. . 365 d kg l mta
W = 100,000 capit(l35 gn,;tm/capta/d)•--·--·--=4927.5 mta
yr 10"1 g 10"1 kg
3
(b) c = W = 135 g/capita/d _10 L = 207_7 mg
Q 650L/capita/d I m3 L
1.3
U = 3 km _1000 m . 1 hr = 0_4167 m
2 hr km 3600 sec s
A=Q= 3 ems = 72 ml
U 0.4167mps
A 12 ml
H=-=--=0206m
B 35m
1.4
(a) a = W = 6,950 mta . 1«>9 mg . kml = 868.751cml

c 8 mg/ ml tonne 109 ml yr
mta
(b) W = ac= 861.1 5 - -•5 µg/L = 4343.75 mta
µg/L
(c) % reduction= 6,950-4,343.75 . l()(J9/4 = 37_5%

6,950
l.S
(a) The total flow is equal to the summation oftbe river flow, Qr and~ waste flow, Qw The river flow
can be determined as
ft 2 1 ml ml
Q, = UAc = 1-·lOOft · - - - - = 2.83-
s (3.281 ft)l s
The waste flow can be converted to the proper units,
l .l
3/8 3
Q = 10 MGD· lm -0.438~
w 22.824S MGD s
and the total flow calculated as
Q=Q, +Qw = 2.831+0.438= 3.269 ems
(b) c= Q,c, +Qwcw = 2.831 cm,(0.2 mg/L)+0.438 CDll(2 mg/L) = _ mg

0 441
Q, + Qw 3.269 L
1.6
IL
(a) Vi= 1 gal· 0.264172 gal 3.7854 L
c= V1c1 +V2 c2 = 3.78S4 L(0.2S0 mg/L)+2 L(2 mg/L) =O.SSSmg

V1 +V2 3. 78S4 L+2 L L
m= cV = 0.85S mg(S.78S4 L) 8 = 4.9464x 10-3 g

L 103 mg
1.7 For the present case. the following balance must hold
Q,.(O.S g/L)+Q8 (0.0S g/L)=4 CIDl(O. l g/L)
Because this equation has two unknowns. another equation is required. This additional formulation
arises from the fact that
Q,. +Q8 = 4 ems
This equation can be solved for Q8 and the result substituted into the first equation to give
Q,.(O.S g/ L)+ (4-Q,. )CO.OS g/ L) = 4 cms(O.l g/L)
which can be solved for ·
QA = 0.2 =0.444 ems

0.4S
and
Q8 = 4 - 0. 444 = 3. SS6 ems
1.2
1.8
(a) c= 10 ml.(100 mg/L) =3.33 mg

300 mL L
(b) m= 3.33 mg·300 mL• lL · g = Ix 10- 3 g

L 103 mL 103 mg
1.9
3 6 3 3 6 3
c= 8.3g/m (12x10 m )+1.0g/m (9xl0 m ) _ _ __!_
5 17 3
12xlo'm3 +9x106 m3 m
1.10 The following balance should hold
S.S = l(IOO)+Q,(0)
Q,.+l
which can be solwd for
Q,. = 1(100)-5.S -17.182 L . ml . min - 2.864 x 10◄ ml

5.S min 1000 L 60 s s
1.11
6 6 2
(a) J = S x 10 tonnes/ yr . I 0 g. km = 500_g_
10,000 km2 tonne 106 m2 m2 yr
(b) U = J = 500 g/ m2yr 200 m

c 2.S g/m3 yr
(c) The rate at which the lake was being filled in would have to balance the flux; that is,
where vb = the rate at which the lediment is accumulating. and c, = the concentration of solid matter
in the lediments. This latter quantity can be calculated as
where ; =porollity and p =density. Therefore, the rate at which the lake fills in can be calculated
as
2 l
J -
vb = - - 500 g/ m / yr . m =0 002 m
(1-;)p (1-0.9)2.5 g/cm3 106 cm3 · yr
or
1.3
vb = 0. 002 m. 103 mm = 2 mm
yr m yr
5 m{lOO- 20) mg3

1 = Ml = VAc = Hile = m 1 mo g = 1115 mg
AAI AA.I At l mo 30.42d 103 mg m1 d
1.13 The flow in the creek and the treatment plant should be equal to the flow below the mixing point,
Q,4 + Q8 = 0.494
The following mass balance should bold
= Q,. (170) + Q 8 (820)

639
0.494
These two equations can be combined to yield,
639 = _Q_,4_(1_10_~_+_(0_
.4_94_-_Q_,._)<_8_20_~
0.494
which can be solved for
Q = 0.494(639-820) = 0.1 3756 ems

A 170-820
Q8 = 0.494- 0.13756 = 0.356 ems
1.14
(a) J= 20g = 2-g-

l m1 (10d) m1 d
(b) v=J =2g/ml/d=2m

c l g/m3 d
3042
(c) M = 2-g-(l mo)l0' m1 · d kg = 6,084 kg
m2 d mo 103g
1.4
~ . . . .. .
LECTURE 2:
l ll&~U IC!tt Ill U I I U UL Mtlid . Rb L lb
2.1 Applying equal-area differentiation
t C iic de
(d) (mg/L)
~
(mgL•l ctl) "'
0 10.S 4
2.7
2 S.1 1.7
I
4 3.1 O.S
0.15
6 2.8 0.27
0.3
8 2.2 0.12
O:lS
10 1.9 0.8
-dc/dt 10
0.1
•
C
0.01 ------+--t--+-+-++++----+------~
1 10 100
Linear regression yields a least-squares fit of
1o{- ~;) = - l6213+2.333logc
Therefore according to the differential approach. the parameters are estimated as n = 2.333 and k =
0.0239.
On the basis of this result. M can try the integral method with n = 2. A fit of the plot m lie versus t
yields a best-fit line ex
-1 = 0.1135+0.042361
C
'
Therefore, this fit yields k = 0.04236.
Finally. a nonlinear regression can be used to directly fit the model (see Lee. 8 problems for
additional details on how this is done). The result is
de = --0. 0326c2.23
dt
The results can be summarized as
integral integral/least-squares differential

2 2.23 2.333
0.04236 0.0326 0.0239
Thus. the integral/least-squares method falls between the integral and differential results.
2.2 de =-kc3
di
c-2
- = - kt+C
-2
The constant can be evaluated by applying the initial condition: c = c0 at t = 0,
Therefore. the solution is
l 1
-=2kt+ -
c2 cl
0
Consequently, if the reaction is third order, a plot cx t/c2 vesus t should yield a straight line with a
slope of2k and an intercept of 1/c,,2.
2.3 Plots of c versus t and 1/c versus t both yield nonlinear patterns. In contrast. a plot ofln c versus
t yielded a straight line (see plot below). A linear regression was employed to fit the following line to
this data:
Inc = 1. 609 - 0. 0351 ( r2 = 0.99993)
Therefore, k = 0.035 min•1.
2.2
1.5
■
1
• ■
.s" 0.5
■
0
20 C) ■ ea
-0.5 ■
t
l.4 Taking the logarithm ofEq. 2.44 yields
log(k(t)) =log(k(20))+ (T - 20) logO
Thus, a plot of log[k(t)] versus (T - 20) should yield a straight line with a slope of logO and an
intercept of log[k(20)]. Linear regression can be used to do this yielding
log(k(t)) = -0.60474 + Q020722(T- 20)
0 = 10°·020m = L0489
klO =10--0.604"74 =0.248

l.5
O=Q~ = 1.9o.1 =1.066
3 20
k30 = 1.6(1.066) 0- =3.04 d"1
l.6 (a) The glucose is decomposing and using oxygen as the oxidizing agent. Equation 2.3 defines the
stoichiometry of the reaction. Therefore, the glUCOle can be expressed in oxygen equivalents as
c = 10 mL(lOO mgC/L) 6(32)mg0 _ 8_ mg0

889
' 300 mL 6(12)mgC L
This is precisely the amount of oxygen depletion that occurs in the bottle. Therefore, if the
decomposition is first-order, the following model could be hypothesized
where Co = the initial oxygen concentration in the bottle. This result can be manipulated to give
(b) Consequently, a plot of the right-hand-side versus t should yield a straight line with a slope of k
and a zero intercept. When this was clone for the first 8 points (the last two seem to have al.ready
reached completion of the reaction) . the result was k = 0.09485 cf1.
2.3
l.7 Because the model follows a first-order process. the natural log of concentration versus time should
yield a str.ugbt line with a slope equal to the removal rate. When this is done. the least•squares fit is
Inc= 7.94175-0.01361/
Therefore. the removal rate is 0.01361 c:f1. This value can be used in conjunction with the depth to
compute a settling velocity
v = kH = 0.01361(55) = 0.07487 dm
l.8 A plot of the natural logarithm of population versus time can be fit with a straight line. The resulting
equation can be employed to compute.
lnp = -290.86+ 0.149985(1995) = 8.35525
p(l 995) =e 1·rn25 =4252
l.9 The rate of growth can be computed as
k = ln(4x 109)- ln(O..S x 109) 0.00693 yr·•

8
300yrs
Thus. in another hundred years (I= 400). the iq,dati,m would be
p = 0.5 x 109 e 0·00693(400> = 8 x Irl
l.10 After overturn. the lake will reaerate and approach saturation. Therefore, the dissolved oxygen can
be modeled by the following mass balance
de
V-=k V(c -c)
dt a "
where k0 = the reaeratioo rate [ct1] and c, = the saturation concentration [ct1]. If c = c0 at t = 0, this
equation can be integrated to yield.
which can be manipulated algebraically to give.
Therefore, a plot of the natural log of (c, • c)/(c8 • c.) versus t should yield a straight line with a slope
of -k0 and an intercept of zero. The values are shown in the table below. The line of best fit is
developed (with a linear regression alpitbm with a zero intercept} to yield an estimate mk 0 = 0.149
ct1. This number can be converted into a transfer coefficient by multiplying it by the mean depth of
the lake,
m
V0 = 0.14924(12) = I. 79)-
d
2.4
~~)
t C
c.-c.
0 4.6 0.000
3 6.3 -0.441
6 7.3 -0.895
9 8.0 -1.343
12 8.4 -1.791
15 8.7 -2.239
18 8.9 -2.686
21 9.0 -3. 134
2.11 Aa:ordi.ng to Eq. 2.46.
which can be used to compute
ku = 0.853 x to8l1'-2 ' = o.311 w1c:·1
2.12 In this problem. we are treating the person as ifhe or she were a batch reactor with first-order decay.
Thus. their level of anaesthetic will be governed by the model
where v is the rrw rpecific concentration of the anaesthetic [mt/k&]. m is the amount of anaesthetic
administered [mg]. Mis the mass of the individual [kg], and k is the first-order decay rate. Thus. the
dole can be determined as
m = Mw"' = 50(10)e0·00'l(ISO) = 675 mg
2.13 The first-order decay model can be expressed in terms of half-life as
1 = t 112 1 j c0 ) = 5730 1J 1 ) - 30.SOO yrs

0.693 "\ C 0.693 \ 0.025
2.14 Both the differential and integraJ mecbod& suggest tbat the rate is second order. If this is correct. the
integral method yields a rate estimate of 0.06 L/(mole min). The plot of 1/c versus t is depicted
below:
2.S
12
10
8
..
-5! 6
4
2
0
0 50 100 150
t
2.15 de = -kc 3
dt
c-l
- = -kJ+ C
-2
The constant can be evaluated by applying the initial condition: c = c0 at t = o.
Therefore, solution is
1 1
- = 2kt+ -
cl cl
0
Consequently. if the reaction is third order. a plot of t/c2 versus t should yield a straight line with a
slope of2k and an intercept of l/c 0 l . For the present problem this yields a slope of0.002. Therefore.
the reaction rate would be 0.001.
2.16 Note that the value of O is independent of the base temperature. Therefore. the value at 20"C can
be computed as
2
k(20) = O.l(l.06)20- ' = 0.074726
and the new formula would be
k(T) = 0.074726(1.06)1 - lo
2.17 Simple divided differences can be used to develop slope estimates at t = 0,
t C
0 1.00 2.00 5.00 10.00
1 0.95 1.87 4.48 8.59
ll.c/ll.t -0.0S -0. 13 -0.S2 -1.41
2.6
The log of the negative of the rate estimate can be plotted versus the log of Co. The result bas an
intercept of -1.30988 and a slope of l.459. Thus. the order is n = 1.4S9 and the rate is
k = 10-u0911 =0.049
2.18 A plot of the loprithm of half-life versus the loprithm ~the initial concentration yields a straight
line with a slope of-0.S0353 and an intercept of 1.199. These values can be employed to compute
n = 1+ slope = 1- 0.503S3 = 0.49647
211-I - } lOA9647-l - l
k= . =------
99
0.031
(n-1)10.-0.,C (0.49647 - l)l0Ll
2.19 The integral least-squares approach (using Heun's method as the numerical integrator) can be
employed to estimate values of k = 0.0S3 and n = 1.46. The fit is shown below.
10
0-+-----l--+---+--+----+--+----~-----+-~
0 2 8 8 10
2.20
r, ::: 1 X 31 mgP = 0.326 mgP04

PIP04 l x 31 mgP+4xl6mgO mgP
0.326 mgP 10 mgP04 3.26 mgP

mgPO.. m m3
Therefore, if the concentration is actually phosphate and you erroneously interpret it as phosphorus.
you would be off by a factor of about 3.
107 32
2.21 (a) r0 = x = 2.69g0
~ 106x 12 gC,
(b) 10ft; x 2.69g0 =26.9~

m gC, m
(c) lOgC x 16x 14 gN 103 mgN = 1761 mgN

m3 106x 12 gC gN m3
2.7
LECTURE 3:
~t.eUU t ltC.U t£ t H t;tlLO l mmm t tit HUU L t l U
3.1 (a) (Note that the nomenclature L is conventionally used to designate BOD)
gal l l lb 8-
P =150 m =0568~ P800 =025-- g =113.4-
q cap · d 264.172 gal cap· d cap •d 0.002205 lb cap· d
L = l l l4 = 199.7__£_
"' 0568 ml
(b) V = 3(2 x 10
5
) = 6 x 105 m1 Q = 6 x IO'= 42657 ml
' 2(14) d ·
ml ml
kV= 0.1(6 x IO')= 60000- vA1 = 0.1(2 x 105 ) =20000-
d d
ml
a= 42857 +60000+20000= 122,857-
d
As reflected by the individual terms of this equation, the most effective purging mechanism is the
decay reaction (60000). followed by flushing (42857) and settling (20000).
(c) The total flow through the lake will be enhanced by the subdivision flow,
3
Q.., = 05678(1000) = 568~
d
ml
Q = 42857 + 568 = 43425T
p= 43425 = 0.352
43425+60000+20000
(d) W,=42857(4)=171429 8 W..,=568(199.7)=113400g W= 284828g

d d d
with: L = 284828 =23 mg without: L = 171429 =14 mg
43425+60000+20000 L 43425+60000+20000 L
3.l
de
(a) - = W(t) + Qc. - Qc- kVc
dt m
(b) = AH = ll x Hf(5) = 55x 106 m3
Q = ~ = 55 x 106 ml = 11. 96 x I <>6 ml

1"111 4.6 yrs yr
6
c = W +Qcin = 2000 x 10 g/yr+ll.96 x 106 m 3 / yr(l5 g/ m3 ) = mg
6 3 6
124 8
Q + kV ll96 x I0 m /yr+0. l/yr(55 x l0 ) . L
3. 1
(c) W =Qc+kVc-QciD
= ll.96x 106 m3 /yr(30 g/m3 )+0.l/yr(55x 106 )30 g/m3 - ll.96x 106 m3 /yr(l5 g/m3 )
= 344.4x 106 g/yr
6
% reduction= 2000xl0 -344.4xt06 ·I00%= 82_78%
2000x 106
(d)
6 6 3 3
(,) c = 0.S·2000x 10 g/yr+ll.96x 10 m /yr(lS g/m ) _ 67.24 mg
l l.96x 106 m3 /yr +0.1/yr(SS x 106 ) L
6 3 3
(;i) c= 2000x 10 g/yr+ll96x 106 m /yr(l5 g/m ) = 94_92 mg
ll.96x 106 m3 /yr+0.l/yr(2-55x 106 ) L
6 3 3
(iii) c = 2000x 106 g/yr+ll96x 10 m /yr(IS g/m ) = _ mg
6 3 6
74 07
2·ll.96xl0 m /yr+0.l/yr(S5x10 ) L
Therefore. judged solely on how much the concentration is reduced. option (i) is the most effective.
(e) There are a variety of other factors that should be considered whell making a decision in the •rea1
world.• These include:
• economics (for example. how much does each option cost)

• impacts on the biota (for example. dredging might impact bottom organisms which sometimes
serve as important food sources for fish).
• impacts on existing uses (for example. dredging could have an immediate impact by decreasing
water clarity during the dredging period. In addition. increasing the depth could have a longer
tenn impact. For instance, a deeper system would undoubtedly affect the lake's temperature and
possibly interfere with uses such as swimming and sport fishing.
• secondary benefits (for example. in addition to lowering the concentration. doubling the inflow
could affect other quality parameters such as water clarity).
(J)
6
t =3 V =J 55x10 9 45
9
(i) ' Q+kV ll.96x106 +0.1(55x106) · yrs
t =3 2(55 X 106) = 14.37

(ii)
9
' ll.96x106 +0.1(2x55x106) yrs
"":\ t
(,,,, 55-
= 3- - - - X 106
---- 5 608 yrs
9 6 6
' 2xll.96x10 +0.1(55xl0 ) ·
3.2
3 6
= W = 105()() X 10 kg/ yr 10 mg=
3
.J II C
3
495 /JS
~) • Q 212xlo' m /yr kg L
(b) The steady-state mass balance can be written as
0=W-Qc-vA.c
9
v = W-Qc = Q(c., -c) = 212 x 10 (495-21) = 27_4 m =0.0751 m
A,c A.,c l0SOOx 106 (21) yr d
V 1000000 m3
3.4 (a) Q= - = - - - = 1,333,333-
i-,. 0.75 yr
C 0.8
(b) P=-=-=0.10667
c. 15
c J = W-Qc = lx 107 -1333x 106 (0.8) = 893~ mg

() ~ ,.{• 1 X 10' m2 yr
(ti) v~ = J~ = 8933 = 111.667.!!!.=0.305m

c 0.15 yr d
3.5 The values can be used to determine the consumption rate. and ln(r) versus ln(c) plotted and fit.
Q(Uhr) Q(m3/d) c(g/m3) ,(cf')

0. 1 0.0024 23 184.8
0.2 0.0048 31 331.2
0.4 0.0096 41 566.4
0.8
1.6
.
0.0192
0.0384
52
64
921.6
1382.4
10000
1000
10 100
3.3
The resulting slope is 1.968 and the intercept is -0.9553. Therefore, the order is 1.968 (= 2) and the rate
is e49553 =0.385 0 1.
3.6 The first-order model is
C -b
- =e
c,
which can be manipulated (take natural log and algebraically rearrange) to yield
t = _In_(c~,_Ic_) (i)
A,
The parameter ; can be expressed in terms of concentrations as
C
I00-;= - 100
c,
f._ = 100-, 100

c, 100
which can be substituted into Eq. i to give the final result (Eq. 3.40),
3.7 The eigenvalue for this problem can be computed as
=3 m(2 x 105 m2 ) =6x 105 m3

Q= ..!:._ = 6xlo5 m3 _l week =42 857m3
'I",.. 2 weeks 7 days ' d
3
J = Q + k + ~= 42,857 m /d +0.l/d+ 0. 1 mid =0.l04?6 /d
V H 6 x 105 m3 3m
The response times can be computed as
9
t75 = 1.3 =6. 79 d
0.20476
2.3 d
190 = - - - = 11.23
0.20476
3.4
3
t9s = - - - = 14.65 d
0.20476
3.8 Recall from Lee. 2 that the solution for a &e00Dd-order reactioo in a batch reactor is
1
C=C - - -
o}+kci
If the half-life is defined as the time where clco = 0.5,
1
150= -
kco
This is a useful result, if you are strictly interested in determining a 50-/o reduction. HO\W\'er, it
shows that a half-life is not a good general way to summari7.e the kinetics of a second-order reaction
in the same neat fashion as for the first-order case. For the first-order reaction. this number tells you
how long it takes for the quantity to be halved at the set time interval of the half-life. For the second-
order case. since the l,o is coocentratioo dependent. it would take a different time to go from clco = 1
to 0.5, than from 0.5 to 0 .25. Consequently, it's only good for the first halving.
3 .9 The first-order reaction rates can be computed as:
(a) k = 0.693 = 0.0231

30yrs yr
(b) k = 0.693 = 0.086625. 365 d = 3 L62
8d d yr yr
(c) k = 0.693 = 0.05653
12.26yrs yr
3.10 The eigenvalue for this system can be written in terms of the residence time and half-life as
which can be substituted into the response time formula (Eq. 3.42) to give
This formula can be used to develop the following plot,
3 .5
waek month year
Response 1000
time year
(d) 100
.,, -· ..... ~
.
..... ... -............. -.. -- .
- - -- - ------ -- ----- - ------------ month
10
waek
1 +--+-+--Hl-+++l~-+--1t--++-++t+t--+-+-++-t-+++t-+-+-+-H-+++t
1 10 100 1000 10000
Half-life (d)
3.6
LECTURE 4:
4.1 For simplicity, we will formally aolw for the cue where the forcing function is a decaying
exponential. The g,owing exponential woul!d employ a similar solution with appropriate sign
chanps. The differential equation to solve is
subject to the initial condition that c = cO at t = 0. Applying Laplace transforms yields,
Simple algebraic manipulations can be employed to solve for
C=~+ w.1v
s+,t. (s+.P.)(s+...t.)
As usual, the last term requires some algebraic manipulation to express it in a format that is amenable
to an inverse transformation. One way to do this is with a partial fraction expansion,
w.1v A B
----"'----- - - - + - -
(s+.P.)(s+,t.) s+.P. s+,t.
The right-band side can be consolidated to give
w.1v = As+A...t.+Bs+BJ1.
(s+,11,)(s+,l) (s+.P.><s+...t.)
For the equality to hold, the numerators must be equal. Therefore. equating like terms yields the
following pair of equations
w
A A.+ BP,=__!_
V
(A+B)s=O
B = -W,_l_
V A.-,11.
Therefore, the total equation to be inverse transformed is
C=~+s+...t.
W.
V(l-.P.) s+.P.
[-1___1_] s+,t.
Using the information from Table 5.1, this equation can be converted back to the time domain to give
4.1
Thus. the first term on the right mthe equal sign is the general solution and the second is the
particular solution.
4.2 The equation to be integrated is
where Co = 0. Taking the Laplace transform gives
m
sC+J.C=-
V
m 1
C=---
Vs+l
The inverse transform yields
m -Al
c=-e
V
4.3 The equation to be integrated is
de 1.
-+"""=-smmt
w. .
dt V
where c0 = 0. Taking the 1.aplace transform gives
(i)
The rigbt side can be evaluated by a partial-fraction expansion
__ (W__......I_V_)a,
__ = _As_+_B_ +-D- (ii)
(s2 +ai)(s+l) s2 +cv2 s+l
The terms on the rigbt side can be combined and then the similar terms in the numerators equated
4.2
A+D=O A.il.+B=O
w
-!!.m = Bl+ Dm2
V
These three equations can be solved for A, Band D and the results substituted back into Eqs. ii and i
togjve,
The last term within the brackets can be dropped because it represents a transient part of the solution.
The inverse transform yields,
(iii)
Now invoking the trigonometric identity,
sin(~-;)= cos;sin~-sin;cos~
algebraic manipulation can be employed to show that Eq. iii reduces to
where
4.4 (NOTE TO INSTRUCTOR: Tbe followiag pamneten should be used: spill mass = 5 metric
=
tonnes, v. 0.1 mid and outflow• bto' m3/d.J
(a) After computing the lake's volume [V= 5(lxlo') = 500000 m3], the following equation can be
developed to solve for the concentration as a function of time following the spill,
6 j lxlo' 0.1)1
c = 5 x IO e 7. ,;;ioJ+, =1Oe--0.12,
S x IO'
which can be used to compute values. The first few points are tabulated below, and the complete time
series is displayed in the figure.
time (cl) C (mg/L)

0 10.00
I 8.03
2 6.44
3 5.17
4 4.15
5 3.33
4.3
10
•
•
'
2
0 4-+-+--+-+--+-+-+-+~:;:;::;:=i~~...............-4
0 I 10 11 20 21
3
(b) 195 = - = 13.636 d
0.22
(c) t(c = 0.1) = ln(lO/O.l) - 20.933 d

0.22
4.S The lake's eigenvalue can be computed as
The solution for the step loading is
C =~(•- e-AI) = 5 X 10• (1-e-10125(1-1994)) = 12.J.ui1- e-10125(1-1994))

A.V l.O12S(4 x lO7) . ~
This equation can be used to compute the values for the lake over time. These values, along with the
base concentration of S )If/I.. are shown in the following plot
C 20
(Jac/L)
15
10
&
0
1990 199& 2000 200& 2010
4.6 The lake's eigenvalue can be computed as A. = 1.O125/yr. The solution for the exponential loading is
- - -2<MX0.5
c- -7- - -X lo')
- 2.062(e 0.2(1-1997) -e-1.0125(1-1997))
- ' - - (e 0.2(1-1997) -e-1.0125(1-1997))-
4 x 10 (lO125+ 0.2)
This equation can be used to compute the values for the lake over time. These values. along with the
base coocentration of 5 )If/I.. are shown in the following plot.
4.4
eo
«>
c bagll)
20
0 + - - - - - + - -- - - + - -- - - + - ------1
1990 :DX) 2D5 2010
yearw
4.7 The lake's eigenvalue can be computed as ..t = 1.0125/yr. The solution for background concentration,
along with the step and exponential loactings are
C= 5+12.34<{1- e - LOl2'(t-1'94) )+ 2.06~.O.l(t-1997) - e - L012'(t-1997))
This equation can be used to compute the values in the following plot
«>
C (iagll)
20
~-~----·-·························
o--------------
1985
1990 ZD5
:DX> 2010
yearw
4.8 (a) The eigenvalue can be computed as
Q 20 x 106
l = - + k=---+105= 125/yr
V l0x 107
Eq. 4.32 can be employed to determine,
,r
o = a_ -- = 6.283(0.75) - 1.571 = 114159
2
These values, along with a>= 2,r / 1 yr= 6.28/yr and ; = tan-1(6.28/ 125) = 1.374 can be used to
compute the cooc:entration in the system as
6
c= 30 x l0 (t-e-1.l")
125(20 X 107 )
4.5
+ lSxlO' sin[6.283t-3.142-l374]- lSxlO' sin[-3.142-l374Je-u"
2 +6.2832 2 2
20x 101 .Jus 20x 107 .Jus
+6.283
The results are displa)'ed below:
0.3
0.2
c(mg/L)
0.1
0 2 4 8 8 10
dme(yn)
,r + 3.142 + l374
365
(b) t = 2 . d=354d(:Dec. 20)
- 6.283 / yr yr
4.9 The aolutioo must be modified slightly to reflect that the lambda's on either side of the equation
difrer. The mass balance is
de 10,i1 +5A.1Stn(2nt)---
-+A.2C=-- . 20,i1 SI·{4nt+-
")
dt ,r 3,r 2
where l1 = QIV and hi = Q/V + k. The response can be determined as
and the f1s are based on A,z. The result is shown below:
viahoutl"Ndlon
O+----+-----+---------
0 4 2
The major impact of including the reaction is on the mean level (drops from 3.2 to 2.9), wberem the
peak and the amplitude swing are affected negligibly.
4.6
4.10 Taking the derivative ofEq. 4.30 yields
The cosine is 7.efO at ff/2. Therefore,
,r
-+;
t=-2-
(J)
4.11 This problem can be solved in a number of ways. First. it can be reoog,,;:u,d that the function is
linear with respect to the parameters since it can be reformulated using trigonometric identities as
J(t) = J +J_,sinmtcos0-J_Pcosmtsin0
Therefore, general linear least-squares can be used to estimate the parameters (Cbapra and Canale
1988). The approach is particularly efficient because the data is equispaced. In addition, nonlinear
regTeSSion can also be employed. This can be done very conveniently with a spreadsheet solver. The
result in either case is
] =296.5
J...,=251.9
(}= 78.7
The data along with the fit is shown below:
0 11 122 113 243 304 3115
4.ll The lake's eigenvalue is ,t = 0. 75/yr. The step solution can be employed to calculate concentration
from t =0 to 0.5 yrs,
. 6
c= lO xlO (l -e--0·"')
0.75(1 X 105 )
At t == 0.5, this equation can be employed to determine that c(0.5) = 41.695 µe/1.,. This value can
then be used as the initial condition for the general solution over the next interval (t = 0.5 to 1 yr),
4.7
The application of the step and general solutions can be repealed out to t = S yrs. The results are
displayed in the following plot:
200
150
c {µf/L) 100
50
0 1 2 3 4 5
time(yn)
4.13 Tbe lake's eigenvalue isl= 0. 75/yr. The loading can be approximated by tbe Fourier series,
40
W(t)= 10+ [su(2nt)+.!.sin(6nt)+.!.m(l0nt)+ · ..]
tr 3 S
. For three terms, the result 100ks like,
30
20
W(t) 10
0
1 3 5
-10
2
"
time(1n)
Each term in the series can be employed as a forcing function. The resulting solution is
6
c(t)= 10x 10 (t-e--0.1~)
0.7S(lx 10')
+ (400x lO')/,r sin(2nt-14S)- (400 x IO')/,r sin(-145)e--o.n,
I x 1o'Jo.1s +628
2 2
lx 1o'Jo.1s 2
+6.28 2
6 6
+ (400 x 10 )/(3.n") sin(6 nt-l.5J)- (400x 10 )/(3,r) sin(-lSJ)e--0.7,,
Ix 10'.Jo.1s +(3-6.28)
2 2
lx 10'.Jo.75 +(3·628)22
+ (400 x lO')l(S,r) sin(l0nt - 155)- (400x I0')l(S,r) sin(- lSS)e--0·7 "

1x 10'.Jo.1s +(S-6.28)
2 2
1 x 10' .Jo.1s 2
+(S-6.28) 2
4.8
which can be plotted versus time as
200
150
C (µg/L) 100
50
0-- - ---------t------t-----1
0 1 2 3 4 5
time(yn)
4.14 A semi-log fit of lnc1 versus t can be used to evaluate the rate ofthe first reaction
In C
t (d)
The best fit line has a slope of -0.023 which is an estimate of the first substances decay rate.
Therefore, the half-life can be computed as ·
0.693
t 50 l = - - = 30.1 yrs
· 0.023
The decay rate along with the t, = 35 can be substituted into Eq. 4.21 to give
= h(0.023/ k2 )
35
0.023 - k2
which can be solved numerically for "2 ; 0 .035, which can be used to determine
0.693
1,0,2 =-0.035
- =19.8 yrs
4.9
LECTURE&:
5.1 (a) The solutions for the tv.o lakes are
6 jMCJ' 3.6S}
c = 5xl0 e\~+---, =l0e-:z.73t
l S{OJ X 106 )
C:z = L667(10) (e-:z.u11 -e-:z.731)

2.73-2.833
which can be used to compute the trends in the following plot:
10
9
8
7
8
C (JlcfL) 5
4
3
2
6~-----=::::;:::::~--....
0 0.5 1 1.5 2
time(yn)
(b) Equation 4.21 can be used to compute the peak time as
I = lt(2.73/2.833) _
0 359
~ 2.73-2.833 yr
which agrees with the plot.
S.l (a) Steady-state mass balances for phosphorus can be written for each lake and solved for
concentration
4x 101:z 4x 101:z µg
C =--- 9
- - - - - =
6
- - - - =122 9 -
I 67 x 10 + 16(82100x 10 ) L3806x 10 · L
1 1
C = 6.95xl0 :Z =6.95xl0 :Z =
724
µg
9 6 1
:z 36x10 +16(57750xJ0 ) 0,96xl0 :z · L
c = 458x 101:z +67 x 109(2.9)+36x 109(724) = µg

3 45
161x109 +16(59750xl06 ) L
9
C = 18.2 X 101:l + 161 X 10 (45) = µg
4 9 6
3224
182 X 10 + 16(25212 X 10 ) L
5.1
12 9
c = 6.65x10 +182x10 (3224) = µg
5 9 6
2429
212 x 10 + 16(18960 x 10 ) L
(b) Individual solution can be used to aaess bow much« each lake's concentration is attributable to
its own loading as well as the loadings to upstream lakes. The result is
loadlnp
♦ ♦ ♦ ♦ ♦
I m la e
•
I .m [10000%
rnj
10000%
I la
•
0
186%
0.15%
0.07%
5.18% 90.96%
0.20%
0.09%
149% 96.16o/e
L64% 45.08%
Consequently, for total phosphorus. only Lake Ontario is affected stroogJyby an upstream lake (45%
of its concentration is attributable to Erie).
5.3 Using Eqs. 5.18 through 5.20, with lake 1 being Huron, the following time series can be compated for
the three lakes,
0 10 20 30 50
5.4 The following plot is for the 64-reactor case.
The slope of this line can be evaluated as -0.01969. Interestingly. this is approximately equal to the
ratio k/U = 0.02. In fact. it will be shown in Lee. 9 (Eq. 9.9) that the solution as n ➔ oo of the
cascade model is
5.2
5.5 (a) First. the flow can be converted into proper units,
Q =20 ftJ . 86,400 s . 1 mJ =48 924 mJ

s d (3.281)3 ' d
The mass balance for a single reactor can be written as
C Q
P=-=--
cm Q+vA
For our case, a removal rx 60% cot aesponds to p = O. 4 and the mass balance can be solved for
A= Q(l-/J) = 48924(1-0.4) = 3.67 x 105 ml

/J., 0.4(0.2)
(b) The mass balance for two reactors in series can be written as
C
P=-=
c..
(
-Q-
Q+vA
)l
A= Q(l-..fp) = 48924(1-.Jo.4) 1.42 x 105 ml

~ .Jo.4(0.2)
Thus. the total area for the two tanks is 2.84xl0S m2.
(c) The two tanks seem to represent the best option because they require less area to effect the same
removal.
5.6 Pseudocode to implement Di Toro'• algorithm in the format of a fimc:tion is
FUNCTION en ( t, n, lambda() , Cl O)
prodl • 1
DO FOR j • 1 TO n - 1
prodl = prodl • lambda(j + 1, j}
END DO
sum= 0
DO FOR i = 1 TO n
prod2 - 1
DO FOR j • 1 TO n
IF j ~ i THEN
prod2 = prod2 * (lambda(j, j) - lambda(i, ill
END IF
END DO
sum - sum+ cl(t, lambda(i, i), clO) / prod2
END DO
en - prodl • sum
END FUNCTION
S.3
FUNCTION cl (t, lambda, cl0)
Cl = cl0 * EXP(-lambda * t)
BND FUNCTION
5.7 (a) The solutions for the three lakes in series for an impdse flux of Strontium-90 to Lake Huron are
c,. = l.186e-0.07011
c = 24.s(ll86) (e-0.01011 - e-n"')

• 25.89-0.0701
c _ _0._3_5S(_24_.5)_11_8_6[e -o.o,oi, - e-OA04t e -:zs.n, _ e -OA04t ]

• (25.89-0.0701) (0.404- 0.0701) (0.404- 25.89)
These equations can be used to compute the following time series,
1.2 Huron
St Clair
r o.a
3
~
o.e
:;.
~ 0.4
0.2
0
0 10
...,,..,
2D
(b) For our problem. the mass balances are
-de,. =- .l,.,_c,.
dt
The following dimensionless variables can be defined
t• = i,.,.t
• -c,.
c,.= • c.
c.:a:- • c.
c,,. c,,. C
• = -c,,.
1,.. A.. = -l•• 1· = i .. • i ••
A.~. = - .. A, -l •• = -
A.,.,_ " A. A.""
"" ""
These can be substituted into the differential equations which can be 10lved for
.
c,. =e
_,•
S.4
,. 1• ( _,. -J.".,t• -A.".,t• -A.".,t• ]
• "'k"'• e -e e -e
c=----,----------
• ,t• -1
ff
,t• -1 ,t• - ,t•
ff ff ff
Now if "M ....,,-ngniui that ..t~ and ,t~ are approximately equal and much larger than unity, the
solutions for Lake St. Clair and Lake Brie reduce to
Consequently, the solution shows that the three lakes in series can be represented by two. That is,
Lake St. Clair can be omitted with negligible impact.
S..S
LECTURE 6:
n u a 1 i . U"!ft01rJ!m ll4!u n .n~~a t~~~o ma;; i t 1:11 t H. Ii t m t =st u uurn. 1 t t u urn
18 24]
6.1 [X][Y] = 28 40
40
[X][Z] = 66 83
sol [Y][Z] = [ 26 33
12 6] 13
[Z][YJ =[ 44 32
4]
[ [
46 16 38 39
6.l (a) The steady-state mass balance for Great Lakes phosphorus can be written as
l.381 X 1012 0 0 0 0 Pi 4x 1012

0 0.96 x 1012 0 0 0 Pl 6.9Sx 1<>12
-0.067 X 1012 -0.036x 10 12
LI 17 X 10 12
0 0 P3 = 4.58 X 10
12
0 0 -0.16l x l0 12
o.sss4 x 1012
0 18.2 X 1012
0 0 0 -0.182 X 1012
0.S 154 X 1012 ,,,
p4
6.6Sx 1012
The matrix inverse is
7.243 X 10- IJ 0 0 0 0
12
0 l.042 X 10- 0 0 0
1
[Ar = 434S x 10-14 3.3S7 x 10- 14
8.9S3 x 10-13 0 0
l.19Sx 10-14 9.233x 10-1' 2.462x 10-13 1708x 10-12 0
4.22x 10-1' J.261 X 10-I' 8.69Sx 10- 14 6.033 X 10-ll 194x 10-12
The inverse can be multiplied by the forcing functions to compute the phosphorus concentrations.
2.9
7.24
[p] =(Ar {W} =
1
4.S
32.24
24.29
(b) The effect of Huron on Ontario is quantified by a;J= 8.695 x 10- 14 (mg / m3 ) / (mg/ yr). Thus,
the coocentration effect is
The impact is negligible.
(c)
C0 = 6.033 x 10- 13(0.75)18.2 x 1012 +8.69S x 10-14 (05)6.65 x 1012
= 8.235 + 0.2 = 8.43 ~

L
6.1
Thus, the effect of Erie is much more significant.
6.J The Laplace transforms ofEqs. 6 .3S and 6.36 results in
sC1 -C10 = -a 11C1 +a 1:z½

sC:z -C:zo = <Z:21C1 - <Zn½
or collecting terms
(s+a 11 )C1 -a12C2 =c10

-a 21C1 +(s+a:z:z)C:z =c20
½ = SC:zo +a11C:zo +a:z1C10

(s + .lI Xs + .l1 )
where the lambdas are defined as in Eq. 6.43. Next._.a partial fraction expansion can be applied to
each equation. For example, for C1,
SC10 +a:z:zc,. +a1:2C:zo - A +__!!_ (i)

(s+.l1 Xs+.l1 ) s+.l1 s+.l1
sc10 +a:z:zC10 +a1:2C:zo = A(s+.l.)+B(s+,i.1)
(s + .lI Xs + .l1 ) (s + A.1 Xs + .l1 )
Equating like terms on each side of this equation results in the following simultaneous equations
sc10 =(A+ B)s

a:z:zC10 +a1:zC:zo = Al. +B.lf
These equaticm can be solved for
a :z:zCto + a i:zC:z0 - A. f C10

A= - - - - - - - - " - -
.l. -A.,
B = .l.c1o - a:z:zc10 - a1:zc1.
A.• -A.I
When these terms are substituted back into Eq. i and the result inverse transformed. Eq. 6.41 results.
Similar manipulations on the equation for C2 will result ia Eq. 6.-12.
6.2
6.4 (a) Steady-state mass balao~ can be written for the lake (1) and the aedimenta (2) as
Substituting the parameter values. these equations can be expreued in matrix format.
45000000 -25007Jc• }-{2,000,000,000}

[ -2500000 7500 Jlc:z - 0
The matrix can be inverted
2.73 X 10-1 9.09 X 10-']

[ 9.09 X 10_, 1.64 X 10◄
which can be multiplied by the right-hand-side vector to give c 1 = 54.545 and c:z = 181818 .,ag L"1•
(b) The dependency of the sediment concentration on the inflow concentration can be formulated
based on the matrix invene as
which, using a target c2 = 100,000, can be solved for
c. = 100,000 =55µg
• 9.09 x l0-' •20x106 L
(c) Eqs. 6.41 and 6.42 can be applied and aolved. The~ is shown below:
C1 0.2
(µg/L)
0.1
0
0 10 20 30
C2 16
(µg/L)
10
6
0
0 10 20
6.3
6~ Mass balanceis for the pesticide can be written for the lake (I) and the sediments (2) as
or collecting terms
where
If c 1 = m/V1 and c2 = 0then the general solution (Eq. 7.41 and 7.42) is greatly simplified
where the b are eigenvalues which are defined as
;.., = (J..11 +J..n)±J(J..11 +J..22)2 -4(J..11J..22-J..,2J..2,) = 0.24795

;.,. 2 0.006634
The respome can be computed and is displayed below:
6.4
31 / .water(ng/L)
c:.1> /
10
,sediments {µg/L)
//
/
o~!!i::==+===~=--..---
0 100 150
t (years)
The time for the sediment to reach a peak can be determined by differentiating the sediment solution.
setting it equal to uro and solving for,
1
= lt( AI I A1 ) = Ir(0.24795 / 0.006634) =IS yrs
P A, - Al 0.24795-0.006634
6.6 The mass balances for the three forms can be written as
':" = -k,.c,, +k6,,c6 -k.c,. +k"'cc

de,.
-=-k,.,.c• +k,.c,. -k6 cc• +k_.cc
dt
dee = -k_.cc +k11cc, -k"'cc +k.c,,
dt
In addition. at all times, the total mass must be conserved since there are no system losses.
(a) At steady state, these equations can be solved for c., = 48.9, c., = 27. 7, and c., = 23.4 m,/L.
( b) The equations can also be integrated over the initial IO days to give the following trajectories. As
can be 1een. the solutions are moving towards the Sleacly state levels.
100
A
50
..................................... ll .... .
C
0 +-~~-+---+-----f---+--+--+----+--+-~
0- 2 4 10
• •
6.5
6.7 The mass balances for the three forms can be written as
(a) At steady state. these equations can be solved for c., = 29.5, c1,o = 21.1, and Coo= 42.8 mg/L.
(b) The equations can also be integrated over the initial 10 days to give the following trajectories. As
can be seen, the solutions are moving towards the steady state levels.
100
so C
A
B
0 ...-------t-----t---+------~----t------1
0 2 4 6 8 10
6.8 (a) Steady-state mass balances can be written for each reserwir. For example. for Lake Powell, the
balance would be
where Q = outflow, c = chloride concentrati~ and the subscripts are Taylor Park ( 1). Blue Mesa (2),
Morrow Point (3), Fontenelle (4), Flaming Gorge (5), Navajo (6), Powell (7), Mead (8), Mohave (9)
9
and Havasu ( 10). The entire system in matrix form is (note that all equations are divided by l x 10 .
0.226 0
-0.226 0.853
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0.9
4
c,
Ct
C3
0 -0.853 0.914 0 0 0 0 0 0 0 0.4
0 0 0 l 43 0 0 0 0 0 0 C4 5.1
0 0 0 - l43 l52 0 0 0 0 0 C) 20
0 0 0 0 0 125 0 0 0 0 c, = s
0 0 -0.914 0 -152 - l25 13.4 0 0 0 C7
7145
0 0 0 0 0 0 -13.4 12.3 0 0 300
0 0 0 0 0 0 0 - 12.3 12.4 0 c, 102
0 0 0 0 0 0 0 0 - 12.4 11.8 C9 30
C10
The matrix inverse can be determined as
6.6
4.425 0 0 0 0 0 0 0 0 0
1172 1172 0 0 0 0 0 0 0 0
l.094 l.094 1.094 0 0 0 0 0 0 0
0 0 0 0.7 0 0 0 0 0 0
0 0 0 0.659 0.659 0 0 0 0 0
0 0 0 0 0 0.8 0 0 0 0
0.075 0.075 0.075 0.075 0.075 0.075 0.075 0 0 0
0.082 0.082 0.082 0.082 0.082 0.082 0.082 0.082 0 0
0.081 0.081 0.081 0.081 0.081 0.081 0.081 0.081 0.081 0
0.085 0.085 0.085 0.085 0.085 0.085 0.085 0.085 0.085 0.085
(b) The loact;ngs can be multiplied by this matrix to give the concentrtations in the individual
reservoirs. The results are
Concentntion Concentration
dment
4.000
Blue Mesa Powell 55.916
Morrow Point Mead 85.741
3.989 Mohave 93. 116
16.930 Havasu 100.237
(c) The effect of Mead on Havasu is reflected by the matrix inverse element. a,,10 = 0.085
(mg/L)/(H>91/Yr) which can be used to compute the effect of halving the load as
.6c10 = 0.085 ~/L •0.5-714.5(109 g/yr)=26.617 mg

IO g/yr L
(d) If all outflows are halved, the matrix must be recomputed. The resulting concentrations would be
Conceatntioa Concentration
dment dment m
6.25
Blue Mesa 87.368
Morrow Point 133.971
Fontenelle 6.233 Mohave 145.4942
F 26.453 Havasu 156.621
(e) The Flaming Gorge-Lake Powell spill case must be modeled dynamically. The mass balances for
the general case of a first-order reacting substance (with a half life of to.,) would be
Thus, the solution for two reactors in series can be used to look at the spills. For the case of a
conservative substance, the result in units of pglL are
6.7
2
0+---+----+------,t--- + - --+----+--t------t
0 10 20 30 40
(j) For the case of a substance with a half-life of 2 yrs, the result is
0.5
0.4
0.3
0.2
0.1
0
0 10 20 40
6.8
LECTURE 7:
7.1 Substituting the parameters yields,
dp = 0.015p
dt
where p = I 0,000 at t = 0. Euler'& method (with a step size of 0.5) can be employed to compute
p(0.5) = 10000+ (0.075x 10000)0.5 = 10375
p(l.O) = 10375+ (0.075 x 10375)0.5 = 10764
The process can be continued with the result
0 2 4 6 8 10 12 14 16 18 20
10000 11587 13425 15555 18022 26882 24194 28033 32480 37633 43604
These results are displayed below. In addition. a plot of lnp versus t is shown. This data can be fit by
linear regression to give
lnp = 9.21+0.0736/
Thus, the slope is close to the rate of growth (0.075). Note that the discrepancy is due to the error in
Euler's method. If a more accurate numerical solution were obtained (by using a smaller step size or
employing a more accurate method such as 4th-order RK}, the slope would approach 0.075.
100000 r,,,::-,--,.....,..=::--:,--...,...,,:--,,
20000
10000 ~ --+~ -+----41---'I

0 10 20 0 10 20
7.2 Substituting the parameters yields,
where p = I 0,000 at t = 0. Euler's method (with a step size of 0.5) can be employed to compute
5
p(0.5) = 10000+ [ 10- (20000- 10000) 10000]0.5 = 10500
p ( I.0) = 10500+[10-5 (20000-10500)10500]0.5 = 10999
The process can be continued with the result
7.1
0 2 4 6 8 IO 12 14 16 18 20
10000 11983 13832 15431 16723 17709 18429 18937 19288 19527 19687
These results are displayed below along with the. exponential growth model from Prob. 7. I . Observe
how the logistic model asymptotically approaches Gmax as t approaches infinity.
0 10 20
7.3 The Heun method with a step size of a half month can be employed to compute the following
concentration time series for the lake. The peak occurs early in January.
o.a
0.2
0.1
0 2 4
• 8 10
7.4 The Heun method with a step size of an hour can be employed to compute the following concentration
time series for the lake. Note bow the midnight discharge takes longer to deplete because the lake is
colder at that time. Thus, the reaction is a less effective purging mechanism for a night spill.
midnight discharge
or:=~~-=~::::::::::.:::=:i
0 0.2 0.4 o.a o.a 1
7 .5 The Heun method can be applied with one-yr time step. The results for a few steps are shown below:
time slooel predict slonel correct

1930 -0.0342 17.3658 -0.0070 17.4000
1931 -0.0178 , 17.3616 -0.0037 17.3794
1932 -0.0093 17.3594 -0.0019 17.3687
1.2
These, along with the remainder of the values can be plotted along with the inflow concentration
below. Notice bow the lake response is smoother and lap the forcing function.
200
150
p (pg/\.) 100
11120 11MO 1111JO 1111JO 2000

tlmt(y. .,)
It is also interesting to contrast the Heun solution with other solutions. In the next plot we contrast it
with the Euler and the 4th-order RK methods. Notice how the Euler solution is much more jagged than
the higher order methods.
100
75
25
0 + - - - --+--- -+-- - - t - - - - - - 1
1~ 19«) 19(1) 19(1) am
tlme(years)
Although it looks like the Heun and 4th-order RK methods yield very similar results, a blow-up of the
concentrations from 1970 to 1980 indicates some discrepancy.
4J
10
0 +--- - - + - --+--- + - - - - - - t -- - 1
1910 19n 1974 1978 1978 1eeo
time (years)
7.3
7.6 The 4th-order RK method should be applied to the second lake with care. In particular, it is
important to use the proper intermediate values ftom the first lake when forming the k's. The correct
values for a series of steps are tabulated below. The entire calculation is shown in the figure below.
As can be seen, the peak for the second lake occurs at about 0.38 yr.
I c, l11 lt, is, A.1 Ii C2 lu l22 lu A.2 ;,.

0 10.000 •27.300 -26..S.S.S -26..57.S -2.S.849 -26..568 0.000 16.667 1.5.731 I.S.771 14.872 l.S.7.57
0.02 9.469 -2.S.849 -2.S.144 -2.S.163 -24.47.S -2.S. 1.56 0.31.S 14.873 14.013 14.0.SO 13.224 14.037
0.04 8.966 -24.476 -23.808 -23.826 -23.17.S -23.820 0..596 13.22.S 12.436 12.470 11.712 12.4.58
0.06 8.489 -23.17.S -22..543 -22..560 -21.944 -22..S.54 0.84.S 11.713 10.989 11.021 10.326 11.010
0.08 8.038 -21.944 -21.34.S -21.361 -20.778 -21.3.56 1.06.S 10.327 9.663 9.692 9.0.56 9.682
0.1 7.611 -20.778 -20.2ll -20.226 -19.673 -20.221 l.2.59 9.0.57 8.449 8.476 7.894 8.467
10
C
I
(µg/L)
'
6
4
2
0
0 0.2 0.4 0.6 0.1 1
t(yr)
7.7 The calculation can be implemented with the results summarized in the attached table:
,, /{,11, c1) C;+tf2 ./(.f;+112, C;+112) c,
0 44.15 37.38 36.92 15.00

1 31.83 67.83 26.26 51.92
2 22.64 89.50 18.68 71.18
J 16.10 104.90 13.28 96.85
4 11.45 115.86 9.45 110.14
5 8.15 123.66 6.72 119.58
6 5.19 129.20 4.78 126.30
7 4. 12 133. 14 3.40 131.08
8 2.93 135.95 2.42 134.48
9 2.08 137.94 1.72 136.90
10 1.48 139.36 1.22 138.62
00 0 142.86 0 142.86
As can be seen, the results are identical to the Heun method for the amount of precision displayed. It
should be noted that the two results do typically yield different results. However, for the present case,
the divergence would only be manifested for greater than 4 or 5 significant digits. The student can
consult Chapra and Canale ( 1988) for examples that illustrate such differences.about
7.4
LECTURE 8:
:x...·.w,;;,_,.~~i~u~~iN~o,,;,~~1,aa~,a,~.1,:-."'-....'''!-->-"~~;i,~>,·~:i:·~u;_o i;:.~1~~-;i;i,.,.:~.:m~,.~.i.":Q~il!•-•".i.;Quo~ .:.c_:~~m.,~.;QQ~; · ~:·~ ~ ~ _ ; u ~ 1a1:1::i.c~;~·~~~)cs -~'~::-: H»oQ~Q».»~~aw,.i:.~i•~r-.-s~
8.1 (a) A mass balance for the dye can be written as
deb E'(C -c. )

V-= O
dt "
where the subscripts o and b designate the ocean and the bay, respectively. Because the ocean
concentration can be assumed to be zero, the equation becomes,
Therefore, a plot of ln(c6) versus t should be linear (if our model holds) with an intercept of ln(c6cJ
and a slope of -(EIV). As shown by the following plot, the model seems valid
3.5
Inc (ppb)
3
slope= -0.00744
2.5 - - - - - - - i - -- - - - t - - -- - ,
0 100 150
t (d)
Therefore,
3 2
E = E' l = 18599 m d(lOO m) =
/ m
92 955
A.. 20 m2 ' d
(h) A steady-state mass balance for the bay can be rewritten to include loading and solved for
3
W=(E'+kV)c6 =[11S99m +0.01/d(2.Sxl06 m3 )]1-\=0.436xl05 !
d m d
8.l The mass balance for the bay can be written as
8.1
or
where the eigenvalue is
E V
l= - + -
V H
The bay's wlwne is

3
V = ,r(S00)1 ·S= 3.93 x 106 m
The bulk dispersion coefficient can be computed as
1
E'= EA~= 100000 m /d(20 ml)= 40000 m3
t SO m1 • d
Therefore, the eigenvalue can be computed as
40000
l = + 0.1 = 0.0302 d"1
3.93x 106 S
and the 95% response time estimated as
19 , =-3 =-3-· = 99.4 d (0.272 yr)

l 0.0302
8.3 Steady-state mass balances can be written for the system as
where the subscripts 1, 2, 3 and 4 refer to lower Green Bay, upper Green Bay, Lake Michigan and
Lake Huron, respectively. Substituting parameters yields,
[
36.836
0
-20
- 25.4 99.92 o
-30
-40.8 778.444 c3
]c·
Cl
l = {1200}
200
5808
Nore that we have used the following units: length [km]. nws [tonne]. time [yr]. The matrix inverse
can be computed as
0.031575 0.006421 0.000247]

0.008155 O.oI 1826 0.000456
[ 0.000427 0.00062 0.001309
8.2
and the concentrations as
40.6l}µg
{c}= 14.80 -
{ 8.24 L
8.4 (a) The bulk diffusion coefficient can be computed as
5 3
E'= l x 10 (1500) = l.S x 106 m
100 d
During the runoff' event, the following mass balance holds
or substituting parameter values
de
- + 0.58333c = 20
dt
Therefore, the solution would be
which can be used to compute the peak concentration.
c = 34.28~l - e--0.51lll{l)] = 23.603 ~
(b) After the storm. the following balance holds,
Substituting parameter values and solving yields
C = 21609e--0.41333(1-2)
The time to reduce to 5% of the peak value can be computed as
3
195 = 2 + - - - = 8.207 d
0.48333
8.5 (a) The bulk diffusion coefficient can be computed as
E'= Qsn -s = S x 106 70-30 =8.7 x 106 m3

s-so 30-7 yr
8.3
and the diffiaaion coefficient calculated as
E' l 8.7 x 106 (100) 104 an2 yr an

2
E = - = ------------------------ = 551.5-
~ 500 m2 86400 X 365 S S
(b) The mass balance can be written and solved for
0=W-Qc-kVc-Ec
W=Qc-kVc-E'c
W =5x 106 (10) + 1(981748)10+8695652(10) =
50x 107 87 X }06 9.8 X 106 = 147 X 106 ~ =146,774 kg

10 g yr
(c)
A.= Q + E +k = 14 95 yr"1
V V .
3
t95 = .8 =02 yr(= 2.4 mos)
14 5
8.4
LECTURE 9:
U l t ~ Q l Q t
9.1 A thin element can be drawn around the reactor inflow,
~ I~~.
1-~~
advection -i--+- j advecdoa/dlapenion
. If this element is infinitesimally thin and at steady state, the mass in must equal the mass out,
Clearly, because of dispersion within the tank,. the pipe 000centratioo cannot be equal to the
concentration immediately within the tank. Thus, the only way that co = ci would be if E = 0. A
physical argument relates to the fact that dispersive mixing in the tank wou1d dilute the concentration
just inside the inlet.
9.2 The Peclet numbers (and the Dahmkohler number) for the four cases are
k, hr"1 2 2 2 2
U, m/hr 100 100 100 100
l,m 100 100 100 100
E, m2/hr 0 2000 10000 00
P.=LU/E 00 s 1 0
D. = kl!U 2 2 2 2
9.3 Using the same approach as in Example 9.2, but with E = 1000 and 100,000, the results are
0.5
.. ..... .........
P =10
•
0-------------------
0 20 40 80 10 100
9.1
9.4 The velocities should first be computed in mid.
- t1 Se2ment2
~
t3
distances. m 0 to 8046 8046 to 16093 > 16093
flow. m'ld 97.849 97.849 195697
area. m2 18.58 18.58 55.74
velocitv. mid 5267 5267 3511
Employing distances in meters, the following equations define the concentration profiles in each
reach:
c ( x) = 5e --0.•-5267
I
" 1o[ 1-e--0.1-,,
+-
0.1
5267
l
The results are displayed below using distance in. kilometers for the abscissa:
25
20
15
C (nvl)
10
0 10 20 30 40 50 eo 70
The concentration falls below 5 mg/I.. at about 46.7km.
9.5 First solve for the velocity and convert the dispersion coefficient into units that are commensurate
with the rest of the problem
4
U= !l._ = 5 x l0 = 250 m E = 10' cm2 m2 86,400 s = 864,000 m2
A~ 200 d s 1<>4 cm2 d d
Eq. 9.57 can be solved for
W=S x l0'4 ml (10mg/m3) l+ 4(0.11/2)864000 . kg = lkg

d 2502 106 mg d
9.2
(b) Solve for the velocity
U=_Q_=5xla4 =125m
A.. 400 d
Eq. 9.57 can be solved for
3
W =5 x l04 m (20mg/m3 ) l+ 4(0.29S)l,OOO,OOO · leg =8.79kg
d 1252 106 mg d
9.1 c,. = lOe- 3':!,ooooo) + 25 [1-e- 3':!,<•0000>] =5.738+53.281 =59.018 mg

U L
To find the level at B, we must determine how much the concentration bas depleted by this point
- E._ lS000-10000
c(25000) = 59.0lSe 3600< > = 25.649 mg
L
Then we have to perform a mass balance to determine the concentration,
cs = 25.649(43200)+ lO x 103 = _94 mg

12
86400 L
Finally, the 000centration at the water intake can be calculated as
- E..(3,000-2,000)
Cc = 12.94e 4'00 = 8.5) mg
L
The entire solution is shown below:
0 +--- - ---+----+-- - - + - - - - - - !
0 10 :i!J 40
X (km)
9.3
.
9.8 We will assume that all the additional flow derived from the source is reflected in an increase velocity
downstream. Although this would not be strictly true (some would be reflected by an increase in
depth), it is a valid first approximation. ~ in the development of Eqs. 9.58 and 9.59, the solution
will remain
for XS 0
for x > 0
However, the eigenvalues are modified to reflect that the velocities and flows are dift'erent above and
below the discharge ·
1
J 1 =u-
2£
Plk.E)
( l+ 1+--
Ull
The mass balance at the outfall now becomes
9.9 IfEq. 9.61 is used to determine the eigenvalues, the solution for the downstream segment becomes
x:2::0
Thus, it can be seen, that as E approaches zero, this solution becomes
W -!z
c=- e u x:2::0
Q
which is the plug-flow result. Note that this approach does not help for x < 0 because a singularity occurs
in that direction for the plug-flow case.
9.4
LECTURE 10:
10.1 {a) After converting the english units into commensurate metric units (B = 1000 ft= 304. 79 m~ H =
10 ft= 3.0479 m. Q = 500 cfs = 1,223,109 m3/d), the velocity and the loss rate can be computed as
3
V= 1223109m /d =l3l 6_7 m k(total) = k +~ =0.05+ 0.3/ 3.0479 =0.148 d. 1
304.79 m(J.0479 m) d H
which can then be used to calculate the estuary number
6
= kE = 0.148(1 x 10 ) = 0.0856
11
u2 1316.12
Therefore, the system can be classified as being mildly advective.
(b) Eq. 10.24 can then be employed to compute the profile at the downstream station (x = 8 mi =
12874 m),
2
6 _ (12174-(1316.7)1) _ O 1411
c(xt)=I0xl0 /928.9e 4 c1x1o'), ·
• 2✓,r(lxl06 )t
The results are shown below.
C 0.3 12174 m
(J&O/L) ..,_mfrom
I
0.2
0.1 t(d)
0 6 10 16 20
10.2 The data can first be plotted
120 T
i
100 t
I
80
c (mg/L) m
«>
20
0 200 400 CD) 1000
t (min)
10. l
Next. the following summations can be developed:
Station I Station 2
(x= 1 km) (x= 5 km)
11- l 15S20 15600
~)c, +c,.,x,••• -t,)
l•O
11-1 2,317,600 l l,32S,OOO
~)c1t1+ c1• 111• 1 Xl1+1 -11 )
1-0
11-I 3S9,316,000 8,284,2S0,OOO
I<c1tl +C1+1tl•• x,••,-11)
1-0
These values can be used to compute:
c= 15520 = 38_8 µg 15600

1
c2 = = 19.5µg
2(260- 60) L 2(9S0- 550) L
86 400 5
M 1 = 3.7 x 10' ml (38.8 mg\.260-60)min • kg - L994 kg
s ml f d l06 mg
M:2 = 3.7 x 105 ml (19.5 ~ ll'9so-550)min ~

86 400 5
• = 2.004 kg
s m f d 10 mg
Thus. the mass of the conservative tracer is conserved.
f 1 = 2317600 = 149.3 min l 2 - 11325000 726.0 min

lSS20 - 15600
~,., -- 359376000
15520 149.i = 8S6.3 min2 ~ - 8284250000 726.0:2 = 4021.5 min:2
t.l - 15520
U = (S - 1) km 103 m 1440 min=

9989
m
(726.0-149.~ min d d d
5 3
A= 3.7xl0 m /d= 37 m2 H= 37 m:2 =0.6l73m
9989 m/d 60m
2 2 2 2
E=(9989mld)2(4021.S-856.3) min d d 10000cm =llOllcm
2(726.0-149.3) min 1440 min 86400sec m2 s
103 2000
10.J c(O) = lO g mg = 4 µg ilpoac1 =--=0.8/d
2500ml g L 2500
1000
c,, = 4e-O.&I f" L--• = - - = o.s d
· - 2000
10.2
A mass balance at the end of the channel yields the following concentration at C.
Cc= 2000 4e--e.l(,- o.,) = 0_4e--e.l(H,.,)

· 20000
The results are shown below:
c 4 ·.._A
(f.11/L)
o.l--...C:::;:=:=;::==1=1...,.~~~~
0 2 4
l0ppb
(b) m=l0gx--=~=2S0g
0.4ppb
10.4 c(0)= lOg lOl mg=4µg 2000

l.,_. =--=0.8/d
2SOOm3 g L 2500
1000
f"---, =--=0.Sd
2000
C
... o1._, -- 4e-O.l(r-o.,)
Mass balance for lake (assuming t • 0 at O.S d):
de + 2c = O.Se-0.11 c=- ( - lt -e -ea,)

0.8- - e ·
dt 2(0.8-2)
c (NIL) 2
o ~k;::::::=:i~~;;;;;=:!!1:111-.
0 2 3 4 5
t (. .ys)
10.3
• ... - • - • - - • · • ..... A -~·., .·,.c; ,., , ._. . . . ..... -... •••
The time of the maximum coocentratioo in the lake can be calculated as:
I =0.S+h(O.l/ 2) 1.26days
C 0.8-2
10.S (a) Recall from Pig. 10.S that the distance encompassing 99-/o of the area under the bell-shaped
curve due to pure dift'usion can be defined as a = s2J2Dt . Therefore, the time required for one side
of the distribution. defined by this bound. to move a distance x is
t = <2x I 52)2 = [2(10) /

5211
= 7,396,450 s (= 85.6 d)
2D 2(10°"')
The distribution in the sediments at this time would look like (for the plot we assume an initial
condition of mp= 1 g/cm2) .
0 0.05 0.1
0 ....,...-+-t-+-1-t--1-t--1-+-IH
10
20
30
(b) The flux can be calculated by substitutingEq. 10.19 into Pick's law,
d m - ;- ) z m - r2
-
J=-D-
dz ( 2.J
P
,r/)t
, 4Dt =--'-,
4t .J,rDt
4Dt
which can be evaluated at the sediment-water interface as ( expressing all times and rates on a day
basis)
(10-~)2 (10-0)2
J =m 10- z , - ~ =m 10- 0 ,- 4(0.0864)8'.6 = 0_000206m
'4t.J,rDt '4(85.6)J,r(0.0864)85.6 '
If the initial m, is assumed to be unity (i.e .. 1 g/m2), then the flux would be 0.000206 g/cm2/d.
10.4
LECTURE 11:
b Ul UUt:tQ L t
11.1 Inlet condition:
Outlet condition:
0 = W11 + Q11-1.II c,._1 + c,. n E' ( ) k VC
~,11+lC11 + 1t-1.II c,._1 -C11 - 11 11 11
2
W,. = -(Q,._1.,. / 2+ E~_1.,.)c11_ 1 +(Q,.,,.+1-Q,._1.,. /2+ E~_1,,. + k11V,.)c,.
11.2 The matrix inverse for the case where n = 10 is
0.00068 0.000392 0.000226 0.000)3 7 ..53E-O.S 4.36E-O.S 2 ..54E-O.S I.SE-0.S 9E-06 .S.62E-06
0.000.588 0.000627 0.00036) 0.000208 0.00012 6.986-0.S 4.~ . s 2.39E-O.S l.44E-O.S 9E-06
0.000.508 0.000.542 0.00060J 0.000347 0 .0002 0.000116 6.76E-O.S 3.986-0.S 2.39E-O.S l..5E-O.S
0.00044 0.000469 0.000.52 0.000.588 0.0003-4 0.000197 0.0001).S 6.76E-O.S 4.~ . S 2..54E-O.S
0.000381 0.000406 0.0004.S 0.000.Sl 0.000.583 0.000338 0.000197 0.000116 6.98E-O.S 4.36E-O.S
0.000331 0.0003.53 0.000391 0.000443 0.000.507 0.000.583 0.00034 0.0002 0.00012 1..S3E-O.S
0.000289 0.000309 0.000342 0.000387 0.000443 0.000.S) 0.000.588 0.000347 0.000208 0.00013
0.0002.56 0 .000273 0.000302 0.000342 0.000391 0.0004.S 0.000.52 0.000601 0.000361 0.000226
0.000231 0.000246 0.000273 0.000309 0.0003.53 0.000406 0.000469 0.000.542 0.000627 0.000392
0.000216 0.000231 0.0002.56 0.000289 0.000331 0.000381 0.00044 0.000.508 0.000.588 0.00068
The results of the calculation are shown in the :figure and table below. Halving the spatial interval
results in approximately a halving of the percent relative error of the result.
1
C
(mg/L)
0.5
n•5
0
0 20 40 80 80 100
n=5 n= 10
X analytical numerical error numerical error
0 0.766
5 0.709 0.680 4.16%
10 0.651 0.609 7.28%
15 0.609 0.588 3.46%
20 0.564
25 0.522 0.508 2.74%
30 0.484 0.462 4.54%
35 0.449 0.440 1.98%
40 0.416
45 0.385 0.381 1.16%
50 0.358 0.353 1.38%
11.1
55 0.332 0.331 0.24%
60 0.308
65 0.287 0.289 0.83%
70 0.268 0.275 2.68%
15 0.250 0.256 2.09%
80 0.235
85 0.223 0.231 3.46%
90 0.213 0.229 7.40%
95 0.207 0.216 4.57%
100 0.204
11.3 The matrix inverse for the problem is included in the solution for Prob. 11.2. Note that the key
element needed to answer pat (a) is a 10~1 : 0.000216. This resuh can be used to calculate.
C
0
-
-
c, : 0.000216(1000)
"11.1(Q)
OJ ::; 0.463
Thus. the result is higher than was originally computed in Example 11.2a. The reason for the
discrepancy between the two results is illustrated by a graphical comparison among the two numerical
and the analytical solutions below. Note that at the end of the tank, the numerical solutions
overestimate the analytical result. and the coarser scheme is higher than the finer one. Further, also
notice that the last value for the coarse case corresponds to a broad region from x = 80 to l 00
centered on x = 90. In contrast. the fine case corresponds to a region from 90 to 100 centered on 95.
It is both these biases that lead to a higher reduction for the finer case.
1
C
(mg/L)
0.5
n•5
0
0 20 40 IO 10 100
11.4 Following the same general approach outlined in Box 11.2, the first derivative can be approximated
with a centered difference,
de:; c1+1 -c1_ 1

dx 2Ax
This approximation. along with the centered difference second derivative approximation, can be
substituted into the steady-state advection-dispersion-reaction equation to give
0 : E c1+1 - 2c, + c,- 1 U c,+1 - c,-1 kc

Ax2 2/u I
Collecting tenns and multiplying by V; = A Ax gives
11.2
11.5 Backward differences for each of the regiom can be written as
km Oto 8: W, + Q,_uc,_1-QIJ+IcI - kV,c1+ S,,V, = 0

km 8 to 16: w, + Q,_uC1-1 - Q,J+ICI - kV,c, = 0
16km: W, +G-uc,_I -QIJ+ic, -kV,cI = 0
>16 km: Q,_uc,_I -GJ+IcI - kV,cI = 0
(a) and (c) The numerical and analytical solutions are compared below.
0 20 40 80
(b) The numerical dispersion for the calculation can be estirnated as
Ax SOOrn rn m2
km Oto 16: E =-U=--4320 -=1.080000-
,. 2 2 d • d
Ax SOOm m m2
> 16km: E = -U=--3240 -=810000-
,. 2 2 d • d
These results can be put into perspective by using then to compute an estuary number for the
segments
km Oto 16: 77 = ~ = O.l(l.0 ~000) 0.00579

3
U 432
3
> 16 km: = kE = O.l( lO.OOO) = 0.00772
'1 U 2 32402
These results are quite small and indicate that the model is still advection dominated.
11.6 The following general centered difference equation can be written for the interior segments. as well
as the first segment.
0 -_ W.I + Q,HJ c1_1 +c1

2
Por the last segments. the uro derivative condition can be included by approximating the derivative
between segments 18 and 19 as
11.3
dc11,19 = C19 - C11
dx ~11,19
or, because the derivative is zero, c19 = c1,. Therefore, the balance for the last segment is
The solution to the resulting set of equations along with the analytical solution is shown below,
0.02
-20 -10 0 10 20 30
11.7 [NOTE TO INSTRUCTOR: The point source bu UDits of c/d.)
(a) The concentration at the mixing point can be computed as
7
W ---;:::U=x:!::10= === =19_93 mg
c= Q ~ = 2.Sx lo' l+ 4(0.075)1x 10' L
fTvz 25<>2
(b) The estuary number is
6
1/ = kE = 0.075(1 x 10 ) = 1.2
u2 is<f
(c) The segment size required to attain a desired degree of accuracy can be determined as
l
1+(1/12) l I
(1- 0.05)2 - -Jii
- - - - - - - - - - - ~ ~ = 750.05 m
0.075
11.4
(d) A backwardl. difference scheme can be employed to compute the profile using a Ax = 750 m. The
concentration at the injection point is 19.12. which corresponds to a percent error of 4.1 %.
20
-20000 0 20000 40000
11.8 (a) The system can be developed in matrix form as in (note that we have employed backward
differences for the advective terms),
[
-L77xl011
0
1
L82 X 1<>1 -6.S X 10
L99xl011
-6.Sxl09
9
0
L65xl011
2.2lxl011
ic, l
c1 = 2xl011
c3 3xl011
11
{2.2 X 10 l
l
The matrix can be inverted to give
13
5.68 x 10-ll l9 x 10- 1.42 x 10-13
5.17 X 10-ll 5.34 X 10-ll 3.99 X 10-ll
[ l52x10-12 l57x10-11 4.65x10-11
and multiplied by the right-hand-side forcing function to yield concentrations of c 1 = 12.91, c2 =

23.24 and C3 = 20.43.
(b) The element of the inverse needed to answer this part is ajJ = 1.57 x 10- 12 , which can be used to
determine
11.9 Using the same approach as for Example 11. 1, a backwards difference can be used to characterize
the tank with n = 5. The following equations result,
11,5
with the resulting solution: 60.92, 46.21, 35.26, 27.48, 22.90. Altematiwly,, ·• in Example 11.2, a
centered difference approach can be used to develop
3.8 -1 -1
-3 4.8 0 00 00 C2
Cl {2()()}
0
0 -3 4.8 -1 0 C3 = 0
[0 0 -3 4.8 -1 c 0
O O O -3 3.8 / 0
5
with the resulting solution: 65.34, 48.28, 35.75, 26.74, 21.11. Both results are shown in the plot
below. Notice bow the backwards difference exhibits numerical dispersion. MOl'e precise results can
be obtained by using a greater number of segments.
100
75 Backwards
25
0 -t--t---t--t----+--+--t---t--t----+----1
0 5 10
U.6
LECTURE 12:
252
12.1 (a) At < - - - = 0.002083 hr(= 0.125 min)
2(150.000)
252
(h) M < - - - = 0.001042 hr(= 0.0625 min)
4(150.000)
252
(c) M < - - - = 0.000694 hr(= 0.041667 min)
6(150.000)
3
(d) n = 000:0.6) = 103 680 calculations
25(0.000694) '
The result in (d) is over SO times as much as in Example 12.1. Thus. for quartering the spatial step
and minimizing the truncation error. the calculation will take over SO times longer to run. To put
that in perspective, suppose that the calculation originally takes 10 sec to run. The new run would
take over 500 sec. Waiting 10 minutes versus 10 seconds is a big difference!
12.l First. we have to compute the velocity and convert the rate to consistent units
7
U= L = 0.47222 m k=l d U6x10- 5 s·1
12(0.3) s d 86400s
(a) n = kE = U6x 10-5(45) 0.000234
U2 (0.47222)2
Therefore. this is a very advec:tive system.
(h) t:.x = lE = 4
2( 5) 19.06 m
~ U 0.47222
(c) M = (t:.x):z = (l 9.06}2 =40.34 s

Ut:.x(a- /I)+ 2E + k(t:.x}2 0+ 2(4.5)+ Ll6x 10-5(19.06)2
(d) E,, = CJ.AJ(a-05)- UM]= 0.4722(19.J =-4.S ml

~L 2t:.x vv1..0- 0.4722(40.34)]
2c19.06) s
Thus, the numerical dispersion negates the physical dispersion.
( e) Simulation results are shown below. for the calculation. At = 10 s and t:.x = 20 m were used.
12.1
C 1
(mg/L)
0.5
0 +-+--+--~...-..--+.......--+-......---+"'-+-~-+-+-+--~i--4--+--4
0 10000 1&000 20000
x(m)
12.3 Both the analytical and numerical results are shown below. For the numerical calculation, M = 0.01
min and tu = 400 m were used.
C 0.018
(mg/L)
0.008
0.000 ~-=::+-.....;..i---+-~~::::.......,;::::::...,.:::-.,_;;:~
0 20000
x(m)
12.4 Simulation results are shown below. For the calculation, At = 0.002 min and Ax = 0.25 m were
used.
C 100
(mg/L)
80
80
40
20
0
0 2 4
x(m)
• 8 10
12.2
-
12.5 The movement of the spill through the system takes about 20 d as shown in the simulation results
below. Note that a time step of O.OS d was employed for the calculation.
C
(µg/L)
0 5 10 15 20
t (d)
12.3
LECTURE 13:
lt4U L alt bl
13.l The analytical solution to the problem is displayed below. The results for all three methods should
converge on this solution.
C 0.018
(mg/L)
0.008
0.000 ...j,,11111~+--- + - -- l--~~-.....::::::::=--.;=a:a....~::...i
0 20000 40000
x(m)
13.2 Simulation results are shown below. The results for all three methods should converge on this
solution.
C 100
(mg/L)
80
60
40
20
o L---+-----+--_:::::::::::i~::::::==s:::;::=;~
0 2 4 6 8 10
x(m)
13.J First. we have to compute the velocity and convert the rate to consistent units
7
U= l. = 0.47222 m k = _!. d = 1.16 x 10-5 ,·
1
12(0.3) s d 86400 s ·
Simulation results are shown below. For the calculation. tu = 20 m was used. The results for all
three methods should converge on this solutioo.
13.1
C 1
(mg/L)
0.5
0 +-t--t-+-t-1+-+-"".......~--+'--+-~i--+--+--P-+-t--t-4
0 &000 10000 15000 20000
x(m)
13.2
LECTURE 14:
- - - · - ·........ bl!.
6 b...________ ,_m_,.,.-------u-u,.u,--t
. l-ut-•--u:□-l:~ll0~~',),,~'«Y,,
14.1 (a) The croa-aectional area can be estimated by evaluating the following integral
where B is the width. Because our data is equi-spaced. Mi can use the multi-segment trapezoidal rule
to evaluate the integral as in
A = (S4 -0) O+ 2(13+ l2S+·· ·+0.8)+0 = 76.S ft:z

' 2(18)
(b) The mean depth is obtained by merely dividing the cross-sectional area by the width to give
16.S/S4 = 1.41667 ft.
(c) The ftow can 1>f! esrirnated by evaluating the following integral
Q =fo'n(x)U(x)dx
Again. the multi-segnient trapezoidal rule can be employed to evaluate the integral as in
Q =(S4 - 0) 0+ 2(0.39+ 0.4375+•·· + 0.064)+0 = 33.28S cfs

2(18)
(d) The mean velocity is obtained by merely dividing the ftow by the cross-sectional area to give
33.285/76.S = 0.435 fps.
(e) The longitudinal dispersion coefficient can be estimated employing formulas such as Eq. 14. IS.
A.-= 7.1 m2 H=0.432m Q=0.942cms U=0.133mps B=l6.4S8m
Eqs. 14.16 and 14.1.S can be employed to compute
32 16 462 2
u• = J9.8(0.432)0.004 = 0.13 mps E = 0.011 0.13 ·
0.432(0.13)
= 0.933..!!!._
s
The McQuivey and Keefer (1974) formula can also be employed to provide an alternative estimate
2
E = 0.05937 °·
943
0.004(16.46)
= 0.85 m
s
(j) Using Eq. 14. 18,
ml
E1,,, = 0.6(0.432)0.13 = 0.034-
s
14.l
14.l Using Eqs. 14.27 through 14.30, relationships can be developed between depth. velocity and flow.
Log-log plots of depth versus flow and velocity versus flow can be developed and fit with power
functions as shown below.
y
I
\
\
\
log(>,, U)
0 .
~
r/
..... ..... - • ~· \
,_
I
\
. \,
lot (Q) u
2
1.5
y,U 1
0.5
u=0.365Q 0.353
0
0 5 10 15 20
Q
14.3 The flows can be sorted ancJ tabulated as shown below:
Rank Flow Probability Recurrence Rank Plow Probability Recurrence

Interval Interval
1 0.010 0.050 20.00 11 0. 127 0.550 1.82
2 0.017 0.100 10.00 12 0.142 0.600 1.67
3 0.023 O.ISO 6.61 13 0. 187 0.6S0 l.S4
4 0.024 0.200 S.00 14 0.311 0.700 1.43
5 O.OSl 0.250 4.00 IS 0.340 0.750 1.33
6 0:062 0.300 3.33 16 0.453 0.800 1.25
7 0.065 0.350 2.86 17 0.651 0.850 1.18
8 0.089 0.400 2.50 18 0.849 0.900 1.11
9 0.091 0.450 2.22 19 1.331 0.950 1.05
IO 0.105 O.SOO 2.00
When plotted on pl'obability paper, a 7Ql0 of about 0.017 ems can be estimated.
14.2
14.4. (a) u• = J9.8(_0.9)0.C)002 =0.042 mps
ml
E1• = 0.6(0.9)0.042 = 0.02268-
s
301
L,,, = 0.4(0.9) 0. = 14.286 m
02268
A 21
(b) R=-=-=0.849
p 3l8
1
n = _!_R213S 112 = - -0.8492130.0002112 = 0.014
U 0.9
(c) Using the formula from Fischer et al. (1979)
0 92301
E = 0.01 I · = 212 ml
0.9(0.042) s
or the McQuivey and Keefer (1974) formula
2
E = 0.0S931 0.9(27) = 240 m
0.0002(30) s
14.S This problem is complicated because an initial state of the stream is not specified. However, because
dispersion is so weak, the signal at KPS.S essentially refects the history at KPO.S prior to 5:00 AM.
Therefore. the initial state can be reconstructed by extending the boundary conditions at KPO.S back
in time based on the information at KPS.S. As shown below, this means that the first S hours of the
simulation are used to get the system so that it roughly conforms to the actual initial conditions at
5:00AM.
The flow simulation fairly nicely matches the wave at KPS.S. In contrast. although the shape of the
concentration profile is captured well (suggesting that the diapersion is correct). the model response
occurs quicker than the data Thia may in part be due to a number « factors including the
inadequacy of the kinematic wave model or the fact that the system may not be truly one-dimensional
(due primarily to backwaters and dead zones) for this stream.
Q 2
(ems)
1.1
0.1
x(km)
0 -----------◄---+--------.---+--l
0 I 10 11
14.3
C
700
IOO
x(km)
300 +-+--+--1-----ii---+--+--+-+-+--+--t---+--T--t--t--t--t
0 5 10 15
14.6 Por a rectangular channel
P= B+2y
Therefore, Manning's equation can be written as
2/3
Q=_!. By Sl/2
n ( B+2y )
The values can be substituted and the equation r&-eXpressed as a roots problem
2/3
1 3
/(y)=0=-- 0y (0.0005)112 -20
0.025 30+2y )
(
This equation can be solved for y = 8.996 m. which can be used to compute
Ac = By= 30(8.996) = 270 m 2
Q 20 m
V=-=-=0.014-
Ac 270 s
14.7 For a trapeu>idal channel, Manning's equation can be Mitten as in Eq. 14.31
13
/(y)= 1 [(30+15y)y]'5 -20
0.025(30+2~
This equation can be solved for y = 7.848 m. which can be used to compute
Ac = By= 30(7.848) = 328 m 2
Q 20 m
V=-=-=0.061-
Ac 328 s
14.4
LECTURE 15:
15.l After converting all the parameters to metric and determining that U = 14.1S6/(30Sx3.0S) =
0.015239 mis. ln(sls0 ) can be plotted versus distance up the estuary. The regression fit of the data
yields a slope m0.000155 which can be employed to compute,
1 l
E = 0.015239. HY' cm 9.8xlO'~
0.OOOlSS m1 s
:c(m)
-28000 -18000 -IOOO
15.l (NOTE TO INSTRUCTOR: The x-values in tbe table are in kilometers, not miles.)
(a) After determining that U = 0.3/100 = 0.003 mis, ln(s/s0 ) can be plotted versus distance up the
estuary. The regression fit of the dala yields a slope m0.000108 which can be employed to compute,
E = 0.003 . 10'' cml =2.77 x 10' cml

0.000108 m1 S
x(m)
-30000 -20000 -10000 0
-2
-4 loa(sls.)
J
(b) First, we can express the parameters in commensurate units,
E =276714 cml · 86400 5 ml =2390812 ml

s d 10" cm1 d
3
u =0.003.!!!. 86400 s =2S9.2~ Q = 0.3 m • 86400 s = 2s92.'!
s d d s d d
The estwuy number can then be computed as
11
=(v
1 / H)E = (02/5)2390812 =l4l
U1 2S921
15.l
(c) The loading to attain a coocentration ex 1 v/m 3
is
W = QcJt + 47] = 25920(,l)Jl + 4(L42) = 67061 g

d
3
15.J Q = 2 2.5(13)-2.1(11) =0.249~
" ,r 13+ 11 s
15.4 The ln(s/s0 ) can be plotted versus distance up the estuary. The regression fit cxtbe data yields a slope
of 0.000266/m which can be employed to compme,
E= 3 cm/s . 1<>2 cm= ll 26 x 106 cml

0.000266 / m m s
-10000 0
l-----+----ll-----+----l----------1------ 1
x(m) 0
-1
-2
• -3
15.5 (a) Steady-state mass balances can be written as
0 = G19o - Q13s1 + Ql1-'l + E~(s3 -s.) + E;. (-'l - .s;)

0= G1so -Qus1 +Ql1-'l + E~(s3 - s.)+ E;1("1 - S1)
Collecting terms yields,
which can be solved for E'n =E' 13 = 2.54x106 andE'12 = 4.20xl0'.
(b) Steady-state mass balance for segments 1 and 2 can be written and expressed in matrix fonn ~
9400000
[ -400000 -2900000J{c•}
S400000 =
cl
{2000000001
0 J
which can be dved for c1 = 2.18 and c2 = 0.161 mg/L. Thus. the highest concentrations occur in the
surface layer.
15.2
l{
15.6 Equation lS.24 can be developed where
8.73 1186
-8.73 0 00 -2.99
0 {E'} _ 14007500
S23SOOO]
0 -1186 7.67 0 - 14647500
[ 0 0 . 0 2.99 4487500
{E'}={l5)1500000
15.7 Mass balances can be written for the three segments as
where/= fluoride concentration (mg L·11. Q = outflow [m3 cf1J. V= volume [m3] . and E' = a bulk
diffusion coefficient (m3 cf1]. These equations can be integrated either analytically or numerically to
develop time series for the three concentrations. These results can be used along with the
measurements to form the sum of the squares of the residuals between the predictions and
measurements. Trial-and-error or an oprimi:zer can be then used to mioirni:,e the sum of the squares
by adjusting the bulk diffusion ooefficieni.. The plot below shows the result (for this case we used the
Excel solver) with values of 4xla4 and 5xta4 m3 cf1 for the epi-hypo and the epi-bay coefficients,
respectively.
0 40 IO 120 118
15.8 NOTJ: TO INSTRUCTOR: TIie lcale on 1111, PIS.I is in error. Absciaa sllcNdd be fro• 0 to
4750.
(a) Steady-state mass balances can be written (using centered dift'erences for the adwction terms) as
lS.3
1 1 +1 2 '
O=Qo1so-Q12-2-+E12(s2-s1)
O=Qi, '1 +s2 -Q23 "2 +s3 +E~2(s1 -s2)+E;3(S3-12)

2 2
•
•
•
Which can be solved in sequence for
•
•
•
Using the values from the problem statement. the results are
Segment E E
(ml /yr) (cm2/s)
0
1
S.OOxlo' 3171
2
6.67do' 4228
3
6.63 xlo' 4203
4
14.4 xto' 3814
s
17.8 xto' 4707
6
18.0 xlo' 4754
7
ISO xlo' 4754
8
ISO xl06 4754
9
Note that we cannot calculate the dispersion across interfaces 7-8 and 8-9 because of the lack of a
gradient Consequently, 'M have assumed that the diffusion coefficient for these interfaces is equal to
that for 6-7.
lS.4
(b) The results for phosphorus are based on similar mass balances as for chloride, with the exception
that the settling loss is included. The resulting concentration profiles are as shown below,
X 250 150 1250 1750 2250 2750 3250 3750 4250

c(mg'm3) 25 20.53 18.35 25.25 40. 18 33.32 22.28 20.43 19.47
TP 60.00
(mg/ml) 40.00 ■
■
30.00
■ ■
■
20.00 ■
■
■ ■
10.00
0.00
0 1000 2000 3000 4000
15.5
I ....,..._:-,..-
LECTURE 18:
·.~ UI. l l NQlQ,~lU Q,~um.tm~~QIQ
16.1 (a) The total volume can be obtained by integration
Because the data is discrete. a numerical approach such as the trapewidal rule can be applied to
evaluate the integral
V =(6 -0) 2830000+2405500 + ···+(31 _ 28) 424500+0 =4_2 x 101 ml

2 2
(b) The epilimnion and hypolimnion can be evaluated in a similar fashion by using linear
2
interpolation to estimate an area at the tbermocline of Af..10) = (2405500+1188600)/2 = 1797050 m .
V =(6 _ 0/830000+2405500 +(l0-6)2405500+1797050 = 2 _4 llxl0' m3

• 2 2
Vr, = (l 4 - l0) 1797050+ 1188600 +(28 _ 14) 1188600+424500 +(JI _ 28>424500+0 = }.79 X I07 m3
2 2 2
16.l (a) The total volume can be obtained by integration
1473
V=
!14'3
0.1675333(z-1453)4- 7.446(z-1453)3 + 115.5717(z - 1453)2 - l0l0S(z-1453) dz
~
=[0.0335(z-1453)' - l862(z-1453)4 + 38.52(z-1453)3 -50.S25(z-1453)2 ]14,l =97363 x 104 m1
(b) The easiest way to do this is to first detennine the hypolimnion volume and then get the
epilimnion by difference
1463
v,. =[0.033S(z-1453)' -1862(z-1453)4 + 38.52(z -1453)3-50.52S(z-1453)2 1
14,1
=18207 x 104 m1
v. = (97363-18207) x 104 m3 = 79156x 10'4 mJ
(c) V(z) = 0.0335(z - 1453)' - l.862(z -1453)4 + 38.52(z- 1453)3-50.525(z- 1453)2
16.3 The solution to this problem involves differentiating the volume-depth function
A(z) = dV(z)
dz
There are a number of ways to approach this problem. The one adopted here involves using finite-
divided differences to evaluate the derivatives. Using Eqs. E.23 (for the interior estimates) and E.24
(for the estimates at the ends) from the numerical methods outlined in App. E, the results are as
shown below. The volumes are also included for completeness. ·
16.1
. -
depth (m) volume(m3) area (m2)
0 53000000 9142500
2 36040000 7817500
4 21730000 6426250
6 10335000 4842875
8 2358500 2451250
10 530000 589625
12 0 0
Note that the last area (at depth= 12) actually evaluated as -59625. This is indicative m the inherent
errors involved in numerical differentiation.
A(t:) 1.00E+o7
(m2)
5.00E-+08
O.OOE+G0+----1-------+--...,;;;::-.,.
0 4 I 12
t:(m)
16.4 For this system, the steady-ltate water balance would be
Q.._ + Q,.... + Qp,,aad =Q.,.. + Q.,..
The known tenns are
Q., = 1000 Qi- =0.001~(1

d
xlO' m2 ) =100
.
Q_, =0.0025 m(l x to' m 2 )O.7 =175
\d
which can be employed to estimate the groundwater inflow as
ml
Q.,_. = 3000+ 175- 1000-100= 2O75-
d
16.5 Using the formulas from Lee. 16 (16.12 through 16. 15), the evaporation flow can be estimated as
L. = 597.3-0.57(30) = 580.2
2
f(Uw)= 19.0+0.95(1) = 19.95
17.27(30)
e., = 4.596e237.3 +Jo = 3L927
!?.27(2')
e0 = 4596e 237-3 • 2' = 21836
16.2
3
Q = l 9.9S(31.927 - 23.836) x , = m
1 10 278
• 580.2(1)100 d
16.6 The pit will till until the evaporative loss balances the groundwater inflow,
The surface area of. the pit is related to its depth according to
A• = ,,,. 2 = ,r(20H)2
Combining the two equations gives
Q.,.,.....s = 0.01 f(U.,Xe.-e.) ,r(20H)2

L,p.,
H = _l Q.,.....i L,p.,
20 0.0lf(U.,Xe1 -e.)tr
or substituting values,
H = _l Ix to" (58L34)1 = 52_08 m

20 o.ot(l9.9SX28.44- t9.89)tr
16.7 Using the same approach as in Examples 16.4 and 16.S, the hydrograph and pollutograph can be
generated for the system as shown below.
0 100 150 200
24
C 16
out
8
0 100 150 200
16.3
--- -· . - - ---·· - ·•.. . ·-·-·· -
16.8 The solution to this problem is based on Eq. 16.30,
O.tr
1
l
l
8.64 X 10'
o.os= x,( 0.1(100)

1
8.64 x IO'
and after evaluating the Bessel function in the denominator
K(
O
O.lrl
8.64 X 10'
l= 0.0SK (
O
0.1(100)2
8.64 X 10'
l
The right-hand-side can be evaluated as approximately 0.1749. The argument of the left-side is,
therefore, about 1.6564. Therefore, the distance where the concentration falls below 5% is
1
1.6564 ·8.64 X 10' = m
r = 4869
0.1
16.9 I developed this solution on Excel because it includes the modified Bessel functions as intrinsic
functions. The following spreadsheet printout and contour plot displays the results for (a). As can be
seen, the bacteria die-off rapidly and are below I 0000 number/mL at a distance of about 5000 m from
the outfall.
-1<XXX> .8XX) .8XX) -<OJ .axx:, 0 2lXX) 4XX) EKXX) EKXX) 1<XXX>
4750 112 425 1500 ,&;J7 1CE96 141651 1CE96 ,&;J7 1500 425 112
4230 131 512 1880 6186 15291 21983 15291 6186 1880 512 131
3750 150 ooe 2342 8154 21883 33188 21883 8154 2342 ooe 150
3250 170 706 2847 1CE96 31417 50497 31417 1CE96 2847 706 170
2750 189 ooe 3388 13473 44467 77f«J 44467 13473 3388 Eal 189
2250 20B Ea> 538 18861 61981 121045 61981 18861 538 Ea> 20B
1750 2Z3 893 4454 1QB97 84174 1927t8 84174 1QB97 4454 893 2Z3
1250 236 1066 4888 23)75 100700 317321 100700 23)75 4888 1066 236
750 2415 1118 5225 2>483 134293 557713 134293 2)483 5225 1118 2415
250 250 11415 S3B6 28615 1SCXfi8 1191184 15CXfi8 28615 S3B6 11415 250
..
1200000 ~aw.~ ·: :-'

1vv,,,vv,..,, !%:::
.. '/ ".
111000000-1200000
800000 /ff#'
■ BOOCXX}..1000000
600000
8 600000-800000
400000 250 D 400000-600000
200000 1750 ■ 200000-400000
0 ■ 0-200000
16.4
(b) Including a cross-flow has a marked effect on the extent of the bacterial contamination. Although the
concentration immediately by the outfall is significantly reduced. values of over 50000 number/mL
persist at distances over 10000 m along the shore. In fact. it takes a distance of over 35000 m until
the bacterial levels drop below 10000 number/mL.
-1CXXX> -aooo ..8CXX) -«XX> -2CXX> 0 2(XX) «XX> 8CXX) 8CXX) 1CXXX>
4750 0 0 0 0 0 0 0 2 23 1~ 200
4250 0 0 0 0 0 0 0 11 1Cl5 376 826
3750 0 0 0 0 0 0 2 89 432 1178 2126
3250 0 0 0 0 0 0 17 376 1569 3278 4834
2750 0 0 0 0 0 0 147 1700 4EJ32 8033 10200
2250 0 0 0 0 0 3 1118 6689 1~ 17:.!5 10035
1750 0 0 0 0 0 38 7062 21~ 29700 31964 31378
1250 0 0 0 0 0 588 33800 S3004 55n) 51179 45-nE
750 0 0 0 0 0 0019 1CB448 10C643 84285 70256 5ro16
250 0 0 0 0 0 186:DJ 206854 138885 104284 82396 67027
~,.r·-,..-,._. . ... Y\'7
::::
150000
~""""'--~ ~ . ■ 200000-250000
~ 150000-200000
100000
250 D 100000-150000
50000 1750 ■ 50000-100000
■ 0-50000
..-I §
..,.
..-
(c) Including the constraining effect of the opposite bank has a minor effect on the extent of the bacterial
contamination. Th.is is due to the rapid die-off and strong flow keeping the impact of the discharge
localized. About the only noticable impact is a slight heightening of concentrations downstream at
the far shore. For example, the concentration at x = 10000 andy = 4750 increases from to 290 to 383
number/mL. The contour plot looks quite similar to the one for part (b).
-1CXXX) -0000 ..8CXX) -«XX> -2CXX> 0 2(XX) 4CXX) 8CXX) 8CXX) 1CXXX)
4750 0 0 0 0 0 0 0 2 27 136 383
4250 0 0 0 0 0 0 0 11 1Cl5 383 853
3750 0 0 0 0 0 0 2 89 432 1180 2134
3250 0 0 0 0 0 0 17 376 1569 3278 4935
2750 0 0 0 0 ro 0 147 1700 4EJ32 8033 10200
2250 0 0 0 0 0 3 1118 6689 13n 17:.!5 10035
1750 0 0 0 0 0 38 7062 21~ 29700 31964 31378
1250 0 0 0 0 0 566 33898 53934 553X) 51179 45796
750 0 0 0 0 0 0019 1CB448 10C643 84285 70256 5ro16
250 0 0 0 0 0 186:D3 206854 138885 104284 82396 67027
16.5
-----------------------------------
LECTURE 17:
17.1 Stokes law can be employed to calculate settling velocities and the following plot developed. Note
bow the stationary and growth estimates for the plankton are connected with lines to highlight the
variation due to growth condition.
measured
plankton
(growth) 10
~
• • 0.01
■■t----1C
0.001
Several observations can be drawn from this plot:
• Stokes law does a respectable job for the inorganic particles.

• In contrast, there is much more variability for the plankton. In particular the cell status interjects
almost an order-of-magnitude difference in the settling velocity.
• Plankton settle faster when they are not actively growing.
• The major systematic difference for the ·plankton seems to be that Stokes law overpredicts settling
for the smaller plankton.
17.l Using the smaller deposition zone area.
v. = - - - - · - - - - - =
212x 109 21-0.5
lOOOO x 106 (l-0.9)2.4 x l06
0.001814 -
m
yr
mm
= 0.929-
yr
This result can be substituted into Eq. 17.21 to determine
v = 912.S O.S 0.001814 = 0.0000869 m = 0.0869 mm

r (1- Q9)2.4 X 106 yr yr
17.3 (a) Steady-state mass balances for solids can be developed for the water and bottom sediments as
0= W1 +W2 -Qm-v.A;n+vrA.,(l-;)p
0= v_.A_.m-vrA.,(l-;)p-v6 A.,(1-j)p (i)
17.1
These equations can be added to develop a balance around the entire system.
v = Wi +K-;-Qm = 0.lx to' +0.2x lo' -0.0S x lo' =0.OOl m

• A.,(l-;)p 106 (1-0.9)2.S x 106 yr
(b) A model of the sediment traps can be expressed as
which can be employed to detennine
(c) Eq. ; can be rearranged to detennine
V =V m V = 160 S 0.001 = 0.0022_!!!,

r .r (1- ;>; • (1- 0.9)2.S X 106 yr
17.4 A steady-state mass balance for a diffusing,'reacting substance in a sediment with negligible burial
can be written as
which can be integrated with the boundary conditions
c(O) = c0
c(<X>) = 0
to yield
Therefore, if the model bolds, a semi-log plot of c/c0 versus depth should yield a straight line
with a slope of -Jk ID. Such a plot, which is shown below, yields a slope estimate of-0.00979.
The reaction rate is
0 693
k= · = 0.02476 yr·•
28
and the diffusion coefficient can then be estimated as
D = 0.02476 cm2 . yr =8.19xl0-6cm2

0.009792 yr 86400 s-365 d s
17.2
0 50 100 150 200
1
0.1
17.5 According to Eq. 17. 31. a semi-log plot cl concentration versua depth ahould yield a straight line
with a slope of -k/vb, Such a plot. which is shown below. yields a slope estimate of -0.5555. The
reaction rate is
0 693
k = · = 0.0357 yr·•
19.4
and the burial velocity can then be estimated as
vb = 0.0351 cm. m = 0.000643..!!!,

0.5555 yr 100 cm yr
0 2 6
0.1
0.01
· The layer at 3 cm then corresponds to
0.03 m
A
u.l = - -- -- = 466. yrs
0.000643 m I yr
It, therefore, corresponds to 1975 - 46.6 =1928.
17.3
17.6 A centered-difference finite-difference model of the diffusion reaction equation was coupled with a
model of the lake. Mass balances for the lake (segment 0) and the sediments (segments 1 through n)
are
•
•
•
These equations can b; integrated with the initial condition c0 = m/V0 = 5 Ci/10xl0 m = SOO nCi/L.
6 3
The results for the sediments are displayed below.
c (nCi L·1)
0 250 500
0
10----►• \
m~
1
z(m)
1.S
As can be seen. the signal penetrates to about 1 m into the eediments. After about 10 yrs, the
concentration drops to about 125 nCi/L near the surface.
17.7 Using the same approach as in Example 17.5, a range of wind velocities can be employed to
generate resuspension velocities as in the plot below. Two fealures of the graph bear mention. First,
nothing happens until a rather high wind speed (15 mps) imparts enough energy to the water so that
the critical shear stress is exceeded. Second, once the critical level is exceeded. the resuspension
velocity grows rapidly with increasing wind. This is the result of the strong dependence on wind
velocity (sixth pmwr) of the underlying equations. The general conclusion is that resuspension is a
sporadic process that will be dictated by a few high-wind events.
17.4
Vr 0.04
0.02
0 5 15 20
LECTURE 18:
18.1 Condition numbers can be developed for the flow rate and the inflow concentration as
Q c-.
0.4415 1
-0.1245 1
0.1098 1
Not surprisingly the values for c;. are equal to unity since the model is directly dependent on the
inflow concentration. The results for Q are a little more illuminating Notice how increasing the
flow increases c1 because it is contained in the inflow. Thus the pin from the inflow is greater than
losses due to increased outflow. In contrast. because "2 is not in the inflow, the increased flushing
leads to a diminution of~-
18.2 (a) From Okubo's diffusion diagram, the mean and bounds for the dift'usion coefficient are
Emm. = 500
E_ =12000
cml
E = .}500(.12000) =2500-
s
Thus, the range is considerable.
(b) The mass balances for the main lake and the bay can be written as
The following parameters apply (these reflect the additional loading due to the new source)
V. = 2xl01 V2 =5xl06 <h1 = 3xlo' Q1 = l.lxl01

6
Qi.-= 8xl0 Qz.;. = lxlo' C1;. = 10 C2;. = 100
The bulk diffusion coefficient and reaction rates are
86400 365 250 0 693

E'12 = 2500 x = 9.855 x 106 m3 / k = · =0.693 yr·•
lx 10'4 200 yr I
Substituting the parameters and solving the simultaneous equations yields c 1 = 5.89 and c2 = 9.68 µg
L•l.
(c) Before performing the first-order error analysis, it is useful to determine the range of
concentrations that would be obtained using the range of diffusion coefficients from Okubo's plot.
These result in c1 = 5.41 to 6.22 and c2 = 13.11 to 7.33 µg L. 1 for the lower and upper bounds,
respectively. Thus, the percent relative variation is
&1 = (6.22-5.41)/2 ·100% = 6.76%

5.89
18.1
r-
& = (7.33-13.11)/2 _100%=-29.5%
l 9.68
The first-order error analysis yields
Ac1 = a,1 AE = 0.0001178 12000-SOO 0.676

IE 2
12 5
Ac2 = a,2 AE= - 0.000845 000- 00 = -4.85
IE 2
which correspond to percent relative errors ~ U .48o/e and -50.1%. respectively. Thus, the actual
ranp is less than indicated by the first-order enor analysis.
(d) A triangular distribution to represent the distributions of diffusion can be written as
(ii )
From the peak to 7.ero a similar approach yields
/2 (x) = - - - - - - -2- ( x l l l l & J l -x) (iii)

(x_ -xmiaxx_ -.r.,_)
(c) The formulas for F(_x) can be developed by integrating Eqs. ii and iii (a la Eq. 18.10).
(d) These functions can be used to generate random values of the diffusion coefficient. For the
present case this is easy, because the two equations for F(_x) are quadratics. Thus, their roots can be
obtained by the quadratic formula We therefore employed the following algorithm,
1. Generate a random number between Oand I.

2. Depending on whether the value is below or above F<..r,..), use either F,(.r) or Fix).
3. Substitute the random number for the left hand side of the equation.
4. Solve for the root af this equation with the quadratic formula. This value will be the di1fusion
coefficient
5. Use this diffusion coefficient along with the other model parameters to solve for c1 and c2.
6. Repeat the process.
The results for the diffusion coefficients for 1000 reali7.ations is displayed in the following plot.
As can be seen, the results are skewed to the left. but they seem to be shifted a bit too much to the
right. This might be an indication of the inadequacy of the random-number generator we employed
to develop the simulations.
18.2
100 1. . .
IO ~
IO ~
i .,.
f 40
20 20%
0 ft
0 2000 4000 IOOO IOOO 10000 12000
The resulting values for c 1 and c2 are displayed in the following histograms. In addition. the
statistics for the simulations are summari7.ed in a table. The results exhibit the same range as the
first-order estimate~ that is. they indicate that the diffusion variability bas a much g,eater influence 011
the bay concentration than for the lake. However, they provide additiooal information in the form of
the shape of the distribution.
180 1~
120
1ft
I' IO'I
I'
I eo
.,. i
l 40
2ft
0
IM UI .,. .... ,.1
K
FM IM IM 11M 11.11 1L1t
C1
C1 C2
Minimum 5.528 7.330
25th percentile 5.940 7.962
median 6.044 8.578
75th percentile 6.130 9.326
maximum 6.218 12.281
18.3 (a) The total phosphorus concentration in Lake Ontario can be estimated as
I 0000 X Io' 22.34 µg

212 x l09 +12.4(19000 x l06 ) L
(b) The derivative of concentration with respect to settling velocity can be evaluated as
18.3
Ac= a, Ava = -0.948(40-4)/2100%=-76%
C a,1 C 22.34
Thus. the concentration error would be on the order of ±76%. Because this model is so simple, we
can check this result versus the actual range of c = 34.7 to 10.3 that is computed with the range v, = 4
to 40 m yr-•. This range corresponds to an error of ±4941/4. The discrepancy is attributable to the
curvature of the functional relationship of c to Va variations displayed below:
- . - . --- - - . - . - - - .. - . . - . -- ...-....... -~-,· ...... -.. -. --.... -

'--········· . -. . . .
.... ....
....
I 20
'--- - - - - - - _Q ----_-_'-1_1-_-_-_-_+-_-_-_-_..._-_-_-_-_..._-_-_-_+_-_-_-_..._-_-_-_-_..._-_. ._..,....,
..
The condition number for the settling velocity can also be determined for this case. The result is
12 4
CN = · (-0.948) = -0.527
~, 2234
Thus. this result indicates that a unit variation in the settling velocity will result is about half as much
change in concentration.
18.4 The Excel solver was used to determine the following best-fit estimates of the coefficients: k 12 = 3,
~, = 0.49 and v, = 0.238. The resulting fit is displayed below:
20
15
.....
- - - •• ¥ ••••••
10
5
• ■
■
0+----+----+----+-----t--------=.-
0 2 4 6 8 10
I lt .d.
18.5 (a) The general representation ofthe triangular function is depicted below. (a) The value af/(x,,,_)
that gives a unit area under this function can be calculated by
1 = (x - x . ) f (x,,_)
- - 2
2
/(xpai) = - - - (i)
x - -xmiD
-•------2----
~,_) ................. :
(b) The equations for/(x) up to the peak can be expressed as
Substituting Eq. i gives
2
Jj(x ) = - - - - - - - - ( x - x m i D ) (ii)
(x- - x. .xx,,_ -xmiD)
From the peak to zero a similar appioach yields
(iii )
(c) The formulas for F(x) can be developed by integrating Eqs. ii and iii (a la Eq. 18. 10),
Both/(x) and F(x) are displayed below for the values of the diffusion coefficient used in Problem 18.2.
18 ..5
- ... - .-
/{_x) 0.0002 F~) 1
0.0001 0.6
0 0 .,
0 6000 12000 0 6000 12000

X X
LECTURE 19:
nl!(lm~_,;,u:au ti Q ·"l.llllli!!l.:Q~ju~., ~-~ .4 • muut t • H btt . rn~ ti U.h4 w. U1 I .. : ,Q.H H ~ OO~O.)ml Hll (:.m: HIZ)U~lQ~l;iQ_);~nq,4;,~~~.;,~:,,.~,;.»»,I❖.•.~
19.l Separation of variable can be applied,
which can be integ,ated to give
or
~
g(glucose)) 6 x 32 gO 6
19.2 (a) mcBOD = 30000 1 0 0 - - - - - - - - - - - - ' " - - - - - - = 3.2 x 10 g
L (6x 12+ 12 x 1+6x 16) g(glucose)
5 7 10 11
(b) lno =_ .3 ll+ 1.575701 x 10 6.642308 x 10 + U43800 x 10 8.621949 x 10 = 2 _424
139 44
sf 283.15 283.152 283.153 283.Is4
Os/ = el.4l4 = 11288
o, = l l.288(1 - 0.000035 x 11000] =6.94 mg/ L
19.3 (a) The diffusion coefficient can be estimated with a chloride balance
0= Qs., -Q-'l + E; (s 2 1 -s2 )
E,·.. = 4800 50-IO = 38 400 m3 d"1

' 10-5 •
(b) The BOD in the lake and the bay can be computed, without the subdivision. as
Qi.;,,c1,.,[Q2 + E;1 +(k +v, I H1 )V1 ]-Qz,.,c1 •.,(- E;1 - Qz)

Ci = ' ' I '
[Q1 + £ 11 +(k +v, I H1)Vi][Qz + £ 11 +(k +v, / H1 )V1 ]-E11 (-E12 -Q1 )
=:--------------....,:__;_;. ___
o- 48~57X38400 ___;;_____________
+ 4800)
(60477.5+ 38400+ (OJ+ OJ / 8)1.28 X 107 1(4800+ 38400+ (OJ +OJ/ 3)1.2 X 106 ] - 38400(-38400-4800)
= 0.038 mg L"1
_ Qz,.,c1 ,.,[Q1 + E;2 +(k+v, I H1)ViJ2 -Qi..,Ci.;,,E~

C2 - I ' ' '
[Q1 +E12 +(k+v. I H 1)V1][Qz +E12 +(k+v. I H 2 )V1 ]-E12 (E11 +Q2 )
= 48~57)(54800+ 38400+(0.I + 0.I/8)1.28 x 10 )-0
7
[60477 .S + 38400+ (0.1 + 0.1 I 8)1.28 x 107 )(4800+ 38400 + (0.1 + 0.1 I 3)12 x 106 ] - 38400(38400+ 4800)
19.1
= l.35mgt·•
The loading and flow from the subdivision can be computed as
Q...., = 0.568(10000) = 5678 m3d"1
Q.,.a, = 113.4(10000) = 1,134,000 gd·'
The BOD in the lake and the bay can be 00111puted, with the subdivision, as
0-(48~57)+ ll34x 106 )(-38400-4800)

c-----------=--...:,..--'-----=------'--~-------
7 6
1 - (60477.S+ 38400+ (OJ+ OJ /8)128x 10 )(4800+ 38400+(0.l + 0.l/ 3)12 x 10 ) - 38400(-38400-4800)
= 0.22 mgL"1
6 7
(48~57)+ 1134 10 1(548()()+ )8400+(0J +QJ/8)128 10 )-0
c---
2 -
- - -~-~~--~~- X
X
- - -~--~--~------
7
{60477.5+ 38400+(0.l +OJ/ 8)128 10 1(4800+ 38400+ (OJ+ 0.l/ 3)12 10 X
X
6
38400(- 38400-4800) )-
= 6.78 mgL"1
19.4 The basic approach \Wllld be to regress the In of the particulate and dissolved 00111ponents versus
distance. This yields
2
In LP= 3.056- 0.lx (adjusted r = 0.997)
lnl, = 2.768-0.0137x (adjusted r 2 = 0.937)
Because the fit of the particulate fraction looks O.K., the slope can be converted to a settling velocity,
v. = 0.1 1cm·1(6600 m cf 1XI km/1000 m)/2 m = 0.33 m cf1. [A caveat to this manipulation is that the
disappearance of particulate BOD may be due to mechanisms other than settling. For example, some
of it can be converted to dissolved form.]
100
10
1
60 80 100
•
0.1
19.2
Although the i's appear good for the diuolved BOD, inspection of the fit indicates that the
dissolved component seems to systematically deviate from the straight line. [Aside: This is a good
example of how a ..good" or "high" ,:Z does not always tell the whole story. It also illustrates why
plots must always be developed and scrutinu.ed]. The fits suggest that part of the dissolved
component seems to decompose readily, whereas some of it aeems to be refractory. This hypothesis
can be expressed as an alternative model.
where F, = the fraction of the dissolved component that is refractory, and "'' and "'1 are rates for the
refractory and labile fractions, respectively. The model parameters can be ~mated in a number of
ways. For example, nonlinear regression can be used to estimate,
where the fit is shown in the following plot.
20
.. . .. . . labile
... . .. ....
. . .. . .
0 +--+--+---+-----.-~·-··-·-t·--·-··--·-~ ............-+--+------4
0 20 40 60 80 100
Thus, from the fit, it appears that the dissolved BOD is about SO% refractory. Using the stream
velocity. the exponents can be expressed as rates.
1
k"' = 0.00784 / kn(6600 m d"1) km = 0.052 d"1
1000m
1
k = 0.0618/ "-"6600 m d"1) km 0.408 d"1
• ....."" 1000m
19.5 (a) As was done for oxygen in Example 19.2,
c= .L= 0.781 =ll418 mole

n. 0.684 m1
c= l.1418 mole 28 g-nitrogen = 3197 mg

m1 mole L
(b) As was done for oxygen in Example 19.2,
19.3
c = _p_ = 0.000314 = 0.0llS mo!e
H• 0.0272 m
c = 0.0115 mole 44 g- nitropn = 0.5079 mg

m3 mole L
19.6 (a) For convenience in later calculations. the flows can be converted to daily units (Q,.,= 172800. Q,
= 432000. Q = 604800 m3 cf1). Then the mass balance at the input along with the conversion factor
of CBOD,ICBOD. = 0.31. can be used to compute
1 = 172800(10)+432~0) 1 = 9.14 L"I

L1) 604800 0.31 mg
(b) The distance downstream to attain 95% reduction can be 00lllpUted as
k,.21 = 0.2(1047)21- 20 = 0.2888 d"1
3 3
x,, = k/U = 0.2888/24192 =2Sl,.300m
19.7 Using Eqs. 19.32 and 19.33. the results are
distance 30 20 10
temperature. ·c 25 22 18
salinity. ppt s 10 20
o. mgL"1 5 6.5 1.5
O,t, mg L"l 8.263 8.744 9.467
o... mgL"1 8.031 8.250 8.398
% saturation 62% 79°/4 89°/4
19.8 Using Eq. 19.32. the saturation concentration at sea level can be computed as 8.263 mg/L. This
value can be corrected for salinity (Eq. 19.33) to give 7.81 m,/L. Finally, Eq. 19.39 can be used to
adjust for the altitude effect to give 6.91 m,/L.
19.4
LECTURE 20:
•a t u u 1•u t a;•1aaat uu t. t ~ .a mm uu HIQUUQ:U a ut u t ~• a:t u nn nm n 1
20.1 (a) The amount of CBOD spilled can be calcu1atecl as
M caoo = 30000 r ( 100 g-glucoee) 6 mole. 32 &02 / mole = 3.2 x l 06 g- CBOD

\. L 1 mole•(6·12+12·1+6•16)g-glua,se/mole
(b) The mass balances for CBOD and oxygen can be written as
dl
V-=-QL -k4 VL
dt
1fL = Lo and c = c, at t = 0, these balances can be solved for
The remaining model parameters can be evaluated as: o. = 6.942, k,, = 0.08925, k,1 = 0.126. These
can be substituted to give the following profiles:
15
10
mg/L satll"ation
,. _ - - - - - - _ -!deficit I__
o t---- L --+--~==::::+==;;;;~;.:.:~-·
0 10 3J
lime (dllya)
( c) The time of minimum 00 can be calculated as
t = In(0.15968 - 0.122583) = ?.1 l days

~ 0.15968 - 0.122583
20.l First the key model parameters can be calculated as: V= 2.5xl06, Q = 3.57x IO', k,, = 0.11, and k,1 =
0.158. Next, the mass balances can be formulated in tenns of population,
20.1
where Pop= population, PL= per capita BOD generation rate(= 12.S B/cap/d). and PQ = per capita
flow generation rate(= 0..57 m3/capd). The first equation can be solved for .
L= PopxPL
(Q+Popx P0 )+k4 V
This result can be substituted into the second equation which can be solved (with c = 6 mg/L) for Pop
= 14,417 people. This can in tum be used to compute
W =~x Pc,= 14417 x 12.S= 18x 106 !

d
Note that the corresponding waste flow is
ml
Q,., = Pop x PL= 14417 x 057 = 8217-
d
Finally, the inflow concentration can be calculated as
6
C;,, = 18x\0 = 4.93 mg
351 x 10' +8217 L
20.3 Using the same approach as in Example 20. l.
2
KL= 2.09 x 10-' cm /s. 1 m . 86,400 s = 0.U06 m
0.1 cm 100cm d d
Thus. the use of a much thinner boundary layer (1 mm as opposed to the original 2.6 cm). raises the
mass-transfer coefficient by a factor of 260. Next. we can calculate the reaeration coefficient
2 6
k = 0.l806m/d(,r(0.030.S)2]m • 10mL =0.l&06d"'
a 300mL m1
Now the exchange is increased further because of the increased surface area for the open beaker. This
all leads to a much shorter response time,
3
t9 , = - = 1.71 d
1.76
which is more like the type of exchange actually observed in nature. Although opening up the
constriction of the bottle neck made some difference, the major factor here was the much smaller
thickness of the boundary layer.
20.2
20,4 The saturation can be calculated with Eqs. 19.32 and 19.39 to be 6.27 JDt/L which corresponds to a
deficit above the dam of D = 6.21 - 6 = 4.21. Using coefficients &om Table 20.2, Eq. 20.44 can be
employed to calculate
r = l + 0.38(0.65)0.7(2Xl - 0.11 x 2Xl + 0.046 x 26) = 159
which can be used to determine the deficit below the dam as
421
n. - = =2.68 mgL"1
~ow 159
which can be used to compute aw-= 6.27 - 2.68 = 3.6 JDt/L.

20.5 Equation 20.56 along with the diffusivity of oxypn (2.09xl0"~ can be substituted into Eq. 20.53 to
yield
5
K = v _ [12.919r·
I &"1,()_z M°·7l
which can be manipulated to give
which can be simplified to the same general form as Eq. 20.58
36.74r_J55
K, = K,,0z [ M
The two formulas are compared below,
2
KIIK1,02 Eq. 20.58
1.5
/ Eqs. 20.53 and 20.56
1
0.5
0
0 100 200 300
20.3
10.6 NOTE TO INSTRUCTOR: The time units for Station 2 should be boun.
(a) The data can be integrated to determine the mau rA ethylene passing w:h point.
M1 =Ql c dt = 222 nun

12
,1
3
~ (100-0 min) 0+ 2( 9 + 69 + 8 l +···+SO)+O m!
2(10) m
= 1,380,840 mg
3 .
M =
l
Q1 t
12
c dt = 13320 m (16-95 hr)O+l(I.4 + 2 -9 +···+0.6)+0 mg= 191800 m
hr 2(13) m3 • g
1
which can be used, along with the trawl time and Eq. 20.60 to calculate,
k = _l_ln 1,380,840 = 1948 d-1

05 d 191,808
The correction for ethylene (Eq. 20.62) can then be used to estimate the reaeration
k0 = 1.15(3.948) = 4.54 d"1
(b) The velocity betMen the two points can be calculated as
U = 13500-6000 m = 15000 m = 625 m

05 d d hr
Inspecting the values at KP 6 indicates that the centroid of the dye patch passed at about SO min.
This can then be employed to determine the distance upstream at which the dye "Was introduced.
x = 625 m 50 min hr - 520.83 m

hr 60min
Therefore, the two measurement points are at x = 520.83 and 8020.83 m. Further, the initial mass of
ethylene discharged can be ~mated as
This value can be nonnali7.ed to the croa.sec:tional area (calculated from the ratio of QIU as 21 .3 m2)
and the result. along with the other parameter values. substituted into Eq. 10.24 to compute the
following.
8
0 10 20
20.4
20.7 The travel time between the two stations can be determined by
- - 4000-500 m Id
t2 - t1 = - - - - = 0.202S d
0.2 m/ s 86400 s
This value. along with the concentrations can be substituted into Eq. 20.61 to yield
400
k= I ,..( ) = 4.842 d. 1
0.2025\150
which can be converted to a reaeration rate by Eq. 20.62
k1 = RJc =1.38(4.842) =6.68 d. 1
20.8 The reaeration rate can be detennined in a number vf ways. Nonlinear regression can be used to
estimate ka = 0.09 0 1. Alternatively. Eq. 20.49 can be manipulated.
Thus. a plot of the right-hand-side versus t should yield a str.ligbt line with an intercept of z.ero and a
slope of-k.,. As shown below. such an analysis yields an estimate of0.088 a 1.
0 10 20 30
-0.5
-1.5
-2.5
•
-3.5
20.9 (a) According to Pig. 20.7, the OM!ns-Oibbs formula should hold.
040.67
k 0 = 5.32 . l.85 1.02423- 20 = 28.68 d· 1
03
(b) For carbon dioxide. Eq. 20.SS provides a first estimate
iJ2)0.2S 1
k a,C02 = 28.6\ = 265 d.
44
20.5
20.10 Manning's equation can be formulated for a rectangular channel as
l/3
1
1.5 = - -20
0.03 1{ 20+2H)
lOH 0.0011/l
which can be solved for H = 0.206 m. The velocity can then be computed as U = I.S/(0.206x20) =
0.363 mps. The Owens-Gibbs equation (which is chosen because the stream is shallow) can then be
employed to give,
7
k = S.32
o
°·0.2061.U
363
°" = SO d-1
The average temperature below the discharge is T = (lxlo+0.Sx25)/I.S = 15°C. The rate can be
corrected for temperature as in
20.11 (a) The Banks-Herrera formula can be used to compute,
K1 = 0.728(2)°-' -0.317(2)+0.0372(2)2 = 0.544 m d-1
which can be converted to l 6°C by
K, =0.544 x l02416- 20 = 0.495 m d - 1
Finally, the result can be expressed as a reaeration raae by
k
0
= K, = 0.495 = 0.707 d-1
H 1
(b) For ammonia, we must also estimate the gas.film transfer rate. This can be done with Eq. 20.55
and20.S9
K 1 ,H1o = l68U,. = 168(2) = 336 m d-1
Then Eq. 20. 14 (along with the value cl H. = l.37xl0..s from Table 19.3) can be used to detenniAe
1 1 8.2lx I0-'(289.IS) = lS4+S.0S = 6_92

-=--+
v,. 0.495 1.37 X 10- '(340.8)
1
v = - - = 0.145 m d-1
• 6.92
20.6
10.ll (a) The velocity can be calculated as U = 800/(2. 7x200) = 1.48 mps. For this velocity and depth,
the Churchill formula is used to compute
148
k
0
=S.2S 2.7L67 =1.48 d.. 1
The Wanni.nkhof formula can be used to compute
K1 = 0.098(i(LS)1.64 = 0.1917 m d.1
which can be combined with the Churchill result to obtain
k0 =1.48+ O.l 9 l 7 =l.SSS d.1

2.7
which can be temperature corrected to give
(b) The reaeration rate can be expressed as a mass transfer coefficient by multiplying it by depth. KL
= 1.75(2.7) = 4.726 mid The gas-film exchange coefficient for water vapor can be estimated as
m
K g .H2O = 168(1.5) =252-d
Both these rates can be scaled to apply them to ammonia,
32)0.2.5
K1 = 4.12 -
17 1
18)0.2.5
= 4.035 m
d
K8 =252( - =255.6m
17 d
Then. using the Henry' s constant from Table 19.3, Eq. 20.15 can be used to determine the ammonia
mass-transfer coefficient,
V = 4.035 1.37 X }0-.5 = 0.138 m

v 1.37 x 10- +8206x 10-5 (298.15)4.035/ 255.6
5
d
Note, that this shows that ammonia is predominantly gas-film controlled.
20.7
LECTURE 21:
l 1.1 The amount of CBOO spilled can be calculated as
Mcso0 =30000, ( lOOg-glucose) 6 mole-32 gOi /mole =32 x 10' g-CBOO

\. L 1 mole•(6·12+12·1+6·16) g-glucose/mole
Mass balances for CBOD and oxygen can be written and solved for
k4L ( -(llr,.+l )t -, -(1/r,.+l,)t)

o=o - - -0- , 0
• lc4 -/co
The remaining model parameters can be evaluated as: o. = 6.942. k = 0.08925. k, = 0.126. These
0
can be substituted to give the following profiles:

15
10
Slllurllllon
20 30
21.2 NOTE TO INSTRUCTOR: Part (c) requires tbat tbe deptla of tbe reactor be specified. The
followiq IOlutioa 1auaes u H • 1 m.
(a and b) The BOD can be computed simply as
where co = the initial orpnic carbon CODtent and k., = 0.2(1.047 1j.~ = 0. 159. The oxygen can be
calculated in an identical fashion to Eq. 19.14
Together with a saturation value (calculated from the temperature and elevation dala) of 7.79, the
following plot of the prosreaion in the bottle can be developed.
12
0 ~1--+-+-+-1--+-+-+-li-t--+-+-t-+-4
0 4 8 12
21.l
Notice that after about 8.2 days. the bottle runs out of oxygen. which terminates the BOD
decomposition. In point of fact. tho systelll migbt switch to anaercbic decomposition at this point.
Assuming that the reactor is 1 m deep. the Banks and Herrera formula (Eq. 20.52) can be used to
compute a temperature-correcte reaeration rate w 0.4 er'. This value can be employed to C101Dpute
the oxygen in the open reactor as
12
8 saturation
deficit
4
0 4 8 12
21.3 The first step is to estimate k, by fitting a straight line to a plot of In L versus t,
lnL 2
0 -f---- - - + -- - - - - - 1 - - -- -4---------<
0 5 10 15 2)
t (d)
The best fit yields a slopeof-0.1281 er1• Assuming a theta of 1.047, this can ·be~rted to a 20°C
rate of
20 10
k,,20 = 0.1281 X L041 - =02028
Next. the value of 0.1281 can be substituted into Eqs. 21.13 and 21.14 (along with D0 = 11.288 - lO =
1.288. Lo"" "SO, I~== 8, and D~ =9.988) formulated as a toots pRlblem.
21.2
r k k }=
JI( •• '
1 1 j~[
k.-0.1281 ..l0.1281
1_ l.288(k. -0.1281)]}- 8 = 0
k.,(S0)
OJlll
r (k k )= k4(S0){~[1- l288(k.-0.1281)]}- A.-0.1211 -9.988=0

n •. ' k. 0.1281 k.,(50)
These equations can be solved simultaneously to yield estimates of k,. = 0.10992 and k. = 0.06118
which can be converted to 20°c rates of k,. = 0.139 and k4 = 0.091. These values can be used to
simulate the sag. The results are compared with the data below.
Note that the data for this problem was generated with values of k, = 0. 126, k4 = 0.063, and k., =
0.118, which correspond to 20°C rates of k, = 0.2, k4 = O. l, and k0 = O. lS. The simulation results
were contaminated with random error generated between • l and I.
10
6
C (mc/L)
4
0
0 5 10 15 20
t (days)
21.4 Mass balances can be used to determine values below the mixing point: Q =9 rn'ls, T0 = 11. 7778 °C,
S0 = 4. 111 g/L, and 0 0 = 10 m&'L- The saturation concentration [with temperature (10.86 m&'L),
salinity (10.59 m&'L) and elevation corrections] can be calculated as o. = 8.767. Therefore, the initial
=
deficit is D0 8. 767 - 10 = -1.236 m&'L-
21.5 Mass balances can be used to determine values below the mixing point: Q =122311 ml/d, T0 = 20
°C, l 0 = 101 m&'L, and c0 = 6 mg/I.. The saturation concentration can be calculated as o. = 7.34.
Therefore, the initial deficit is D0 = 7.34 - 6 = 1.34 mg'L. The reaeration rate can be determined as
1
Both this rate and the deoxygenatioo rate (k4 = l cf ) do not need to be corrected for temperature
because the mixed temperature is exactly 20 °C.
The deficit solution is therefore,
21.3
:ZU + l(lOl) ( e-26333.s
D=l.34e-:Z6333.s :ZU -l34e-26333.s
I )
1-21.6
which can be used to determine the pn,file" shown below:
mg/L 4
0 +--- + - -- - - + -- - - - - + - -- - - - - + - - - - - - - - - 1
0 10000 20000 30000
X (m)
The minimum DO is equal to 3.25 mg/Land occurs at about Xe = 3.5 km.
21.6 The first step is to perform mass balances at the point source. The results are
waste river mixed

Q 1 4 5
T 25 15 17
L 50 5 14
0 2 7 6
Next. the reaction rates and hydraulic characteristics for each reach can be established. Note that Eq.
19.28 was used to set"'
and Eq. 20.41 to set k•.
1 2 3
U, mps 0.5 0.2 0.4
H,m 3 5 7
U, mpd 43200 17280 34560
r. c 0
17 17 17.5
kno, Id 0.3 0.3 0.3
kn.Id 0.261385 0.261385 0.267457
k.ao, Id 0.534805 0. 1572 0.134207
k,.r, Id 0.498076 0. 146404 0. 126481
01, mglL 9.694024 9.694024 9.593855
The BOD and DO deficit equations can now be developed. The junction at KP 30 merely necessitates
a reinitialization of the variables and a change in parameters. A formal mass balance must be
21.4
developed at KP 60, because of the point source that enters there. Some results are listed below along
with a plot.
L D 0 o.
0 14.00 3.67 6.03 9.69
10 13.18 4.04 S.6S 9.69
20 12.40 4.33 S.36 9.69
30 11.68 4.SS S. IS 9.69
40 10.04 S.1S 3.95 9.69
so 8.63 6.63 3.06 9.69
60 7.42 1.2S 2.44 9.69
60 7.85 6.73 2.87 9.S9
70 7.26 7.06 2.54 9.S9
80 6.72 7.33 2.26 9.S9
90 6.22 1.S6 2.03 9.59
16
12 DO saturation
8
4
0 +----+------f---+------+-------j
0 100 200
21.5
LECTURE 22:
•••uuw .HU M. .l t U
22.1 NOTE TO INSTRUCTOR: This problem requires a cl"OIHedioaal area. The followinc
solatioa lllt!I aaAc • 10 m2•
The solution for a distributed source with no flow is
where S,/' = the distributed source expressed as a line load (recall Fig.. 22.2a) with units ofmg m·1 d.
1
. This equation can then be solved for
s; = cAck
-~z
= 2(1 0X0.693 / 5)
0.693/5 10000
=lS 7 mg
. md
I- e u 1- e 0.1(86400)
22.2 The depth of the stretch must first be determined with the Manning equation, which for a
rectangular channel is
2/3
U=.!_ HB sJJ2
n ( B+2H )
Substituting the parameters, the resulting equation can be expressed as a roots problem,
f (H) = _I_( SOH )2130.0002112 - 0.2

0.03 50+2H
which can be solved for H = 0.279 m. Beca111e the stream is so shallow, the Owens-Gibbs formula
can be used to compute k. = 19.14 cr1• This value can then be used in conjunction with Bq. 22.2 and
22.4 to solve for the deficit over the stretch.
s I (
-1c.-x
. )
D= , ; 1--e U
0
The results are displayed in the following plot,
10
aaturatlen
a
8 DO
'
2 deftclt
o~~--+--,~--+--+---+--+-~--+-~
0 I 10
22. l
22.J The oxygen surplus (negative deficit) due to plant growth can be modeled as
P'-R= Ilk"~ = ___0_.5(_2_~___-_2)__ = 2.917-~-

_:ix 10000 m d
1- t U 1_ t 0.2(86400)
22.4 After converting the velocities into mpd. the BOD, deficit and oxygen for the first stretch can be
00lllpUted as
Because DO new flow enters at the second stretch. no mass balance is required and the following can
be computed
- ._
2
<x-20000> k 1, (20000) ( - ~<.r-20000> - k.2 <x-20000>)
L\(x)=(o.r2 -0i(20000))e u2 + 6 e U2 -e Ui
k,2 - ktil
o2 (x) = oa2 - ~ (x)
Similarly for the last stretch.
- .., (.r--40000)
~(x) = £.i{40008)e u,
D:,(x)=(o.i - o.i(40000))e u,
-!ucx--40000> k r
+ 43""'2
(40000)(e -~C.r--40000)
u,
- t., C.r--40000)) R
- e u, + 1
- P, [
_1
-
l - e u1
11
" <.1--40000)]
~-~ ~
The resulting values are displayed in the following table and figure.
22.2
L1 · Li L3 SOD1 Point1 01 Oa Point2 <>2 013 RrP3 Point3 03
0 80.000 0.000 0.000 10.000
10000 63.469 1.714 4.791 3.495
20000 50.353 50.353 2.253 5.306 2.441 6.559 0.000 2.441
30000 46.614 2.599 2.424 3.977
40000 43.153 43. 153 1.029 3.205 4.766 3.234 0.000 0.000 4.766
50000 38.436 0.806 -0.626 2.516 5.303
60000 34.235 0.201 -0.782 2.868 5.712
IO
L
(msfL) IO
40
20
0
0 20000
10
0
(ms/L)
•
•
4
0
0 20000 40000 80000
x(m)
22.5 NOTE TO INSTRUCTOR: This problem requires tbe per-capita geaention rate of BOD from
Table 19.2. That table (p. 3S8) bu incorrect units. The units of CBOD should be efcapita/d,
3
not m /capita/d.
A good way to solve this problem involves the control-volume app1oach implemented on a
spreadsheet. Flow balances for each segment can be written as
where Q,1, = the natural distributed inflow for segment i -(=qAx), Ax = volume length. and Q,, = the
new distributed flow due to the squatters, which can be calculated as
Pop x Pq x .1x
Qpl = ----=---
Xtot
2 J
where Pop= the squatter pnpaJation. P, = the per-capita rate of flow generation for a developing
=
country (recall Table 19.2), andx_ the total distance nfthe stretch (i.e., 20000 m).
The depth can then be calculated baled on Manning's formula Thia can be done on a spreadsheet by
setting up the following circular calculation (i.e.• one-point iteration or successive substitution as
delcn"bed in Cbapra and Canale 1988),
H
= [ Qn (B + 2H)l/3
s Vi s5n
r.s
Once the depth is determined, the croa-sectional area (A,, = BH). velocity ( U = QIA,,). reaeration
(O' Connor-Dobbins) and decay rates (Eq. 19.28) can be determined for each cell.
Next, BOD and oxygen baJaONll ,:an be written for each volume,
which can be solved serially for
Note that the concentration of BOD from the squatters can be computed from Table 19.2 as 60/0.19 =
31S.8 mg/L.
After this calculation is implemented on a spreadsheet, a squatter population can be determined that
leads to a minimum oxygen of 2 mg/L. For the present case. I determined a population rL I. Ix I 06
people. Some rL the resulting parameters and the variable distributions are shown below:
2- >- 20
···,,. k"
k"
k,(/d) 1 _ ' •· .. _ (/d)
k ...
u (mps)
I
_··u·····-·----•-·-
' ••• • • • ••••
·-- -----------
H (m) ::::-- H
0 - - - - + - - - - - +,- - - - - I f - - -- + ,-- ----4- -- - - 4
0 20 40 60
22.4
12
100
a
0 L
(mg/L)
4 ----1 (mg/L)
oxygen
0--------------------
0 20 40 60
21.6 NOTE TO INSTRUCTOR: This problem requires a base flow. In tbe f0Uowin1 solution, I've
1
used a value ef 10 ems. I've also made tbe distributed inflow smaUer, 0.00001 m /s.
(a) Using Manning' s equation and other formulas (e.g., the Churchill formula for reaeration), we
can determine,
U = 0.6158 m3/s H=0.812m k,, = 4.38/d
The uniform load can be determined as
S = qL11 = 0.00001(86400)50000 = 2660.3i

L BH 20(0.812) m3d
Therefore, the profiles of Land<> for x = 0 to SOOO m. can be computed with Eqs. 22.6 and 22.9.
Beyond that point. the standard Streeter-Phelps model bolds with the last values of L and o (at x =
5000) serving as the initial conditions. The resulting profiles are shown below.
(m;i,)12
..· ........ .... .. ........ ........ .... ...... ............... L .
-,----··.:.:··..:.:··:.z- .•.•.•.•.•.•.•.•.•.•.•.•.•.•.•.•.. .• ...•.•...... ...•.... .. .. ... 200
.· 0
:
4 L
(mg/L)
0-+----+-- -+-- - + - - - + - - - - + -- - + - --+----+----+---f
0 10 20
(b) A gpod way to ave this problem involws the control-volume approach implemented on a
spreadsheet. Flow balances for each segment can be written as
where Qd1 = the distributed inflow from the landfill for segment i (=qtu where Ax = volume length),
22.5
The depth can then be calculated based on Manning's formula. This can be done on a spreadsheet by
setting up the followillg circular calculation (i.e., one-point iteration or soocessive substitution as
described in Chapra and canale 1988),
H ={ Qn (B+2H)2/3l3/5
sV2 B513
.J
Once the depth is determined. the croa-sectional area (A .. = BH), velocity ( U = QIA..), reaeration
(Churchill) can be determined for each cell.
Next, BOD and oxygen balances can be written for each volume,
which can be solved serially for
L. = Q;_t.l,-1 +Q.L,1
I Q; +k"'V,
Note that the result is almO&t identical to case (a). This is due to the fact that the load has low flow
but a very high conceutlatiun. Tbembre, the simplifications aft'ered by Eqs. 22.6 and 22.9 are
justified.
22.6
LECTURE 23:
23.1 Mass talances at the injection point can be uaed to generate initial conditions for organic N and
ammonium as (2/7)6 = 1.714 mg'L. Then the mass balance equations (Eq. 23.9 through 23.12) can
be integrated (either numerically or with Bqs. 23.14 through 23 .17) to gjw the profiles in the
following plot.
N 4
Total N
(mg/L)
3
Nitrite Amm~onla .................----~ltrata
........
.......
2
,.,_/,.... .... Organic N
1 --·--.
0 10 20 30 40 50
t (d)
The hydrogeometric parameters can be used in conjunction with Manning's equation to determine
that U = 0.664 mps. This can be used to express the prafiles on a distance rather than time reference.
Thus, the SO day aeries shown above would correspond to a distance of,
m s km
x =Ut =0.664-(50 d)86400 - - = 2870 km
s d 1000m
23.l NOTE TO INSTRUCTOR: The units of U should be mis, not mid, u in the problem
statement. This is not a critical error, as the studellts should recopbe that U is not required to
obtain the solution.
(a) Eqs. 23.9 through 13 can be integrated either analytically or numerically. The latter is easier.
The results are displayed below:
12
q/L
• 0
0 10 15 20
t(d)
(b) The maximum ammonia concentration is 2.49 at a travel time of about 3.7 cl The pK for the
reaction can be computed as
23.1
2729 92
K = 0.090i8+ · -9564
p 288.15
This can be substituted aloog with the pH into Eq. 23.30 to determine
1
F . = - - - - - - 0.0266
11 1+ 10-8 / 10-9.564
which can be used to calculate a maximum unioniz.ed ammonia as 0.0266(2.49)=0.066 mg/I..
23.3 Equations 23. IS and 23.16 can be substituted into Eq. 23.13 to give,
The right-hand-side can be made maoageable by clustering terms to give
Then the solution simply becomes (recall Eq. 4.18),
23.4 Solve Eq. 23.27 for,
Substitute this result into Eq. 23.25
23.2
which can be solved for Eq. 23.29
This result. in turn can be substituted back into Eq. i
13.S After determining the volume (V= (lm)lxlo' m2 = Ix I OS m3] aod,low [Q = lxJOS m3/(2 wu)(wk/7
3
d) = 7142.9 m /d]. mass balances for tile four forms of nitrogen can be written and solved in series (recall
Lee. 5) for
N ,.. QNoja ,., 7142.9(5) 0S3mg

0 2
Q+k00V 7142.9+0.1(100000) . L
N = QN11;m +k00VN0 7142.9(2)+0.1(100000)2.083 = .3 mg

0
0 2 98
Q+ ka1V 7142.9+0.075(100000) L
N -= k,.VN0 _ 0.075(100000)2.398 ,.. mg

1 0560
Q+ k;nV 7142.9+0.25(100000) L
N = k111VN1 = 025(100000)0560 =1.959 mg

n Q 7142.9 L
23.3
LECTURE 24:
1:ti UMl bUlttK .... .•.
24.1 First. the rate can be conwrted to a volumetric basis and oxygen units as in
P. =r -p.:. ~ = 2.61-
JO ff)
= 3.3315-
•·0 " H 8 m3 d
Then the average daily and daylight photosynthesis rases can be determined as.
p = 333112(0.58333)) =12394 ,P
\ n m3d
P0 = 3.337J 3.) = 2.125 md

\.ir
~
24.2 [NOTE TO INSTRUCTOR: TIie ant to last eqaatioa in Eumple 24.2 should be:
Equation, 24.12 and 24.14 can be used to compute
P. = (14-S)mg/L = 14_4 _,!_

,_ (10/24) d m3 d
and
R = (8- 5,S)mg / L = 6 _,!_

- (10/ 24) d m3d
These values can then be substituted into Eq. 24.13 to give
}1 = 14.4+6=20.4+
md
The plant respiration can be determined as
R,, =6-0.05(4)=5.8 +
md
Next, the maximum pbatosynthesis rate can be determined by
Finally. Eq. 24.10 can be used to determine the average daily rate
24.1
p = 26.ni 2(0.58333)) = 9.68_!_
\ ,r m3d
24.3 The extinction coefficient can be calculated with Eq. 24.20
The average daylight solar radiation can be determined with Eqs. 24.10 and 24.11,
11l2 I 500
I 0 =--=-=-=1000ly/d
2/ tr J 05
The light attenuation factor can be computed with Eqs. 24.17-19,
1000 1000
a 0 = - - = 2.851 a 1 = - - e--0-'<7 > = 0.08628
350 350
;, = 2.718(05) (,--0.08628 - , - 2.857) = 0.334
0.5(7)
The photosynthesis rate can then be calculated with Eq. 24.16,
P = 0.15(2)1.06622- 20(0.334)5=0569 g/ m3 / d
and converted to areal terms by
P'= PH= 0569(7) = 3.98 g/ m2 / d
Assuming an average value for the respiration rate, k,. = 0.15 cr1, the respiration can be estimated as
(Eq. 24.23)
R =0.15(0.15)1.0822- 20(5) =0.131 g/ m3 / d

R'=0.131(7)=0.919 g/m 2 /d
24.4 The data can be converted to deficit by using Eqs. 19.32 and 19.39. The resulting time series can be
plotted and used to determine the parameters required for the delta method.
..
D
(mc/L)
,_ •14.S
2A
24.2
/=0.5 A=7mg/L D = 0.1517 mg/L ;= 1430-1300= 1.5 hrs
Using Pig. 24.5, an estimate of k. = 14.5 cf1• T.ben. this value along with Eq. 24.6 can be employed
to determine AIP = 0.19 which can be u,ed to compute P = A/0.19 =- 7/0.19 = 36.8 m,/1.Jd. This
daily value can be fraNlated into a maximum value by
36 8
P. = ,,p = n( · ) =106.7 /L/d
"' 2/ 2(13/24) mg
Finally, Eq. 24.31 can be used to calculale,
R =36.8+ 14.5(0.1517) =39 mg / L / d
24.S Solar noon is 7 + 10/2 = 12. Using Pig. 24.5, the lag can be determined as ; = 3.85 hrs. This
means that the peak occun at about 1550 or 3:50 P.M.
24.6 The awrage rates can be determined by integration. For example for the photosynthesis rate,
J.H P<.z)A(z')dz
P=-o........,_ __
1: A(z)dz
Simpson' s 1/3 rule can be applied to make the evaluation numerically,
a• x 106 (1)+4(0.95x 106 (6)+ 0.83x l<t(S)]+ 2(0.9 x 106 (10))+0.7 x 106 (5)
p= 12 =6766
8 lx 106 +4(0.95x 106 +0.83 x l<t)+2(0.9 x l<t)+0.7 x 106 .
12
The respiration can be evaluated in a similar fashion as 2.484. The profiles of P and area are shown
along with the awra&,e value OD the following plat.
:A
: $
I
4 _ ,
I
P:
24.3
LECTURE 26:
25.1 (a) First. the volume and the flow can be determined.
V = A,H = 1 x 105(2) = 2 x 105 m 3
Q=~= 2xl05 =28571 m3

r,.. 7 d
Next. the decomposition rate can be temperature c:orrec:ted. "'=

0.05xl.04i,_20 = 0.0397 cf1• These
values can then be employed to compute the pond's BOD concentration.
L= 28571(50) = 39.1 2 mg L-1

28571 + 0.0397(2 X 105)
(b) The table from App. B indicates that the saturation concentration is 10.084 mg!L.
(c) Assuming that the Banks-Herma formula holds. the oxygen mass-transfer rate of 0.546 mid can
be calculated. This can then be temperature corrected. 0.S46 x 1.0241,.20 = o.~5 mid, and expessed
as a reaeration rate by dividing by the lake's depth. k. = 0.485/2 = 0.2426/d.
(d) The oxygen concentration of the inflow can be calculated as o-. = o.,. -
I>. = 11.288 - 3 = 8.288
mt/I... The SOD can be .temperature comc:ted. 0.2xl.08 s-» = 0.136 ,Jm2/d. Then an oxygen
1
balance can be written and solved for
= 28571(8288)-0.0397(2 x 105)39.12 +0243(2 x 105)10.084-0.5(1 x 105)0.136 = 5297 mgL·J

0
28571 + 02426(2 X 105)
Note that oxygen transfer formulas other than eau.Herrera could have been used. For the formulas
in the book. the results are.
Banks-Herrera Broecker Wanninkbof

K1:i.o 0.546 1.728 0.307
K1.u 0.485 1.535 0.273
ka.u 0.243 0.767 0.136
0 5.297 8.057 3.478
25.l (a) The BOD balance for the lower layer can be written as
Lh = (v, +v1 )A1 L, = (0.05+0.05)(1000000)2.67(2) = 1.526 /L

v,A, +v1 A1 +kdVh 0.05(1000000)+0.0S(l000000)+0.0S(5000000) mg
The oxygen balance for the lower layer can be written as
25.1
h = v1 A1o, -k4 V1,L1, = 0.0.5(1000000)9-0.0S(SOOOOOO)l.526 = mg/ L

0 1371
v1 A1 0.0.5(1000000)
(b) If all the lett1ing BOD is ewntually oxidized. this is equivalent to setting the SOD equal to the
settling loss, SOD= v,A,l11• Therefore, the DO balance can be solved for
h = v1 A1 o, -k4V11L1,-SOD = 0.05(1000000)9-0.0S(SOOOOOO)l.526-0.05(1000000)1.526 = -0.IS mg

0
v1 A1 0.05(1000000) L
which is clearly impoesible. Thus, we know that o,. = 0. What does this mean regarding the SOD
and the BOD? The BOD should be O.K. However. it means that not all of the settling BOD is
converted to SOD. A way to formulate this is as
= v,A,o, -k4V1,L11 - F x SOD = 0.05(1000000)9-0.05(5000000)1.526- F x 0.05(1000000)1.526

0
v,A, 0.05(1000000)
where F = the fraction of the BOD that will be exerted which is determined as
F = v1A1o, -k4 V11L11 = 0.05(1000000)9-0.0S(SOOOOOO)l.526 = 0.

8989
v1 A1 Lh 0.0S(lOOOOOO)l.526
Thus, the SOD is
SOD= 0.8989(0.3814) = 0.343 gm-I d-1 0.8989(0.0SX

100000 .
25.3 The function to be solved is
f (CSOD) = 2{1- sec{ 0.575 c~ ]]- CSOD
which can be solved iteratively for CSOD = 2.965 g m·2 <f1
25.4 (a) The flux [Jc- = 2.67(0.2SX5) = 3.3375) can be substituted along with the other parameter values
into Eq. 25.56
f (SOD)= J2(0.00139)100(3.3375){ I -sec{ 0.575 s~ ]} + l 714(0.0654)3.337~ •-sec{

0.897 S~D ]} - SOD
This equation can be solved for the root, SOD= 1.083 ,Jm2/d. The thickness of the aerobic 7.0lle can
be calculated on the basis of Eq. 25.S0 as
/_,... = 2.lx 10-5(_ 4 __1_ 10 86400) = 0.67 mm

~ 1.083 100 1 1 ·
25.2
(b) A numerical approach can be developed by writing hahmcw for the aerobic layer as
for the 4 anaerobic layers rL equal me. balances can be written as
•
•
•
where H. =the total thickness of the active anaerobic sediment. Because the aerobic layer is so thin.
the aerobic equations can be solved at steady state for
(ii)
Thus. only the anaerobic equations need to be integrated and Eqs. ; and;; provide the means to
determine m1 and n1. Now the next major step in solving these equations involves determining H1.
This can be done by recognizing that
and that
25.3
Given values of m1 and ni. these equati0111 can be solved simultaneously for H 1and SOD.
A nice way to aolve the l)'ltem is to integrate the anaerobic equations until they reach a steady
state. At each time step, the methane in the anaerobic layers can be QOIDpan,d with the methane
saturation value. If it is exceeded. the methane value in the layer would be set at saturation.
At each time step, Eqs. i through iv are solved for the aerobic layer characteristics. The final
results for the present cootext are an SOD = 1.23 f/m2/d and a thidmeu, H 1 = 0.S89 mm. The other
variable values and tnmsf'ers are shown below.
f1.2111! caoo■ o.ar f•.N 1 !

■-oD ■o.w
1
2
· · · · · · · ·· · · · · ··· ·
-
UN
~~l;:;:;"'! ~ -------------
71.1
O.GNa
-
~1:,....... -- ~ ,
....
··············· ·····+~··························l··~······ ···········
UN UII O.GNa
a -
l t.1•
100 - - 113.t ANAEROBIC
t.
---······ ······ ····· , ································· ........ ................ .
4 ~ 100~ •~141.0
···················· t ··~·····························l~·~················
1.114 1.114 l.oNa
I ..,_ .....100 a.iit. ..,_ .....182.2
POC 19n POC
METHANE NtNONIUM
Sedlmut IMtllane and ammonium budtm, All am>W8 . , . ex,,...... u flux of oxygen
equivalent. (gO m4 d"'). 8otll Ille IMtllane and ammonium have units of mg/L
lS~ The addition of the boundary layer necessitates writing extra equations for that layer which should
be solved dynamically. In addition. oxygen balances need to be written for both the boundary and the
aerobic sediment layer. Although this approach would be necessary for most model applications, a
simpler approach is possible for the present problem. Because we assume that the characteristics of
the two layers are identical. we can approximate them as a single large segment. If this is done, we
rapidly recognize that the increased residence time (because of the added thickness of the boundary
layer) results in negative oxygen concentrations. Thus, to attain results comparable to those of Prob.
2S.4, we must employ much smaller decomposition rates. These can be detennined by bial and error
as kc= 0.0448 and k,, = 0.064 o 1 which result in a total layer thickness of 10.911 mm. Interestingly,
the associated SOD is 0.067 f/m2/d, which is lower than the value of 1.23 calculated in Prob. 2S.4. A
higher value could be induced by using a smaller layer thickness. HOMYer, that would mean that the
boundary layer would have to be lea than 1 cm.
This problem has been intended to illustrate the potential importance of sediment-boundary
layers on the computation of SOD. These should become inc:reasingly important as SOD calculations
are coupled with detailed hydrodynamic computaliona
2S.4
LECTURE 21:
26.1 NOTE TO INSTRUCTOR: There are leftral incorrect parameter values and omitted values
ill this problem statement. TIie vohuael of secmeat 1 and 2 a.Id be Id at ss10' m3 and die
surface area of die sediments sbould be I.do' m2• TIie SOD sbollld be chaapd to 1 rfm2/d and
the nows and bulk clispenioa coeff"teieat units should be m3/d.
Mass balances for BOD can be written for segments I and 2 (using backward differences),
0= WLl +Qo1£o +~1L:z + E;1(½- £i)+E;3(½ -£i)-Q13L1 -k41V1£i

0=WL:z +Qn½-~1L:z +E;1(£i-L:z)+E~(½-L:z)-~1L:z-knV2L:z
Collecting terms gives
(Q13 + Ei3 + E;1 + k,nV1 )Li -(E;, + ~1)½ = Wu + Qo, L<, + Ei3L:J
-e;.ii +(Qi, +E;1 +E~+k42V:z)L:z =Wu +Q.n½+E~½
or in matrix format,
Substituting parameter values for the (A] matrix yields
[A]=[~ri= -J:]
which can be inverted to give
£Ar• = [9.96 x 10-s s.s x 10-1 ]

L78 x l0-8 L59xl0-8
Mass balances for DO can be written for segments 1 and 2 (using backward differences),
0 =Wol + Qo1°o +~10:z + E;1(0:z -oi)+ E;3(0]-oi)-Q130J -kd1V1l1 +ka1V1(0, - oi)

0 = W0 2 +Q»OJ -~10:z + E;1(0J -0:z)+ E~(O] -0:z)-{h 10:z -k,12V2L:z -S~A,:z
These equations can also be expressed in matrix format as,
Substituting parameter values for the [BJ matrix yields
[B] -- [11200000 -3700000]

-1200000 6200000
26.1
which can be inverted to give
The [q matrix can then be generated as
(cr1 =[BJ-1[.sooooo
0
o
.SOOOOO
}Ari = [.s.26 x 10--9
2.46x 10--9
1.1.s x 10--9]
l42 x 10-8
The Excel solver was then used to vary Wu until the resulting concentrations of oxygen were above 4
mg/L. The resulting loading was l . 13xlo' g/d The final steady-state calculation was
{o} = [BJ-I{w;}-[q-I{w~}
{o} = [9.54 x 10-8 .5.69 x 10=:]{w.,1 +Qo,Oo + E;~O) + Ka~V.o.i} -[.5.26 x 10=: 7.l.5x lO~]{Wu +Qo1lo + E~1L1}
l8.5x 10-8 l.72 x 10 W02 +Q32o3 +E23o 3 +SgAs2 2.46 x 10 l42 x 10 Wu +Q32 l:J +£23.l:J
{o} =h.ts}
26.2 NOTE TO INSTRUCTOR: Cbaap the BOD of the waste source to 200 msfL.
The QUAL2E result is shown along with the analytical solution below.
mg/L 20
10
oxygen
km
0
0 20 40 60 80
ANALITICAL SOLUflON:
JC L 0 JC L 0
o..s 19.69 9.02 39.S 10.63 7.27
4 ..S 18.48 8.23 44.S 9.82 7.41
9..S 17.08 7.60 49.5 9.08 7 ..S.S
14.S IS.78 7.24 54.S 8.39 7.70
19.S 14..S8 7.07 59.S 1.1S 7.85
24.S 13.47 7.02 64.5 7.16 7.99
29. .S 12.45 7.06 69. .S 6.62 8.12
34. .S II.SO 7.1.S
26.2
26.J The analytical solution to this problem is displayed below:
X L D 0 o. X L D 0 o.
0 14.00 3.67 6.03 9.69 60 7.42 7.25 2.44 9.69
10 13.18 4.04 S.65 9.69 60 7.85 6.73 2.87 9.59
20 12.40 4.33 5.36 9.69 70 7.26 7.06 2.54 9.59
30 11.68 4.55 5. 15 9.69 80 6.72 7.33 2.26 9.59
40 10.04 S.15 3.95 9.69 90 6.22 1.56 2.03 9.59
so 8.63 6.63 3.06 9.69
16
12 DO saturation
8
4
0
0 100 200
26.4 The QUAL2E result should correspond approximately to the analytical solution listed below.
The actual plot generated with QUAL2! is also shown.
X Q L 0 X Q l 0
-S 345600 2.00 7.50 60 734400 7.45 7. 13
0 734400 27.41 · S.65 70 734400 6.SS 7.44
0.5 734400 26.88 S.56 80 475200 S.65 7.93
I 734400 26.36 5.47 90 475200 4.85 8. 19
s 734400 22.56 5.00 100 475200 4.17 8.36
10 734400 18.56 4.77 110 475200 3.58 8.47
15 734400 15.27 4.80 120 475200 3.08 8.56
20 734400 12.57 4.98 130 475200 2.64 8.64
30 734400 11.03 5.81 140 475200 2.27 8.70
40 734400 9.67 6.38 150 475200 1.95 8.76
so 734400 8.49 6.80
30
20
---
-
0-----------------~--
0 100 1&0
26.3
LECTURE rt:
27.1
ml
(a) Qw = 100,000 capita(O.S m 3 /capi1a/d)=SO,OOO-
d
T = $0000(2$)+ lOOOOO(lS) = 18.333 °C

1$0000
11 3
l.Sx 10 oumber / capi1a/ d m 6 number
(b) N = - -4- - = 3 0 x l 0 - -
w 0.5 m3 /capi1a/d 10 (100mL) 100 mL
6
. S0000(30x 10 )+ 100000(0) 6 number
N =--------=10x10 - -
0 1$0000 100 mL
ml
(c) Q= 100,000+S0,000= lS0,000-
1
Q lS0,000 m
U=-=--=IS,000-
HB 0.5(20) d
(d) Ks 1 = 0.8· L071Ul3-20 = 0.7147 / d
k = 9.339 (1- e --0.5(0..5)) = 8.263 d-1

bl 0.5(0.5)
0.71s+8.263(10000)
(/) N = I Ox 106 e Uxta4 = 2S ISO number
• mL
27.2 (a) The bacterial die-off rate can be calculated•
k' = [0.8+0.006(SO)]L0721.s-20 + 200/24[1-•-0.2(2>]

b 0.2(2)
= 1827 + 6.868 =8.7d"1
(b) Equation 9.S1 can be n,ammged to calculate the allowable loading
27.1
27.3 Basic parameters can bf! esrirnated as
=AH= 5x l0s(7) = 35 x 10s m 3
3
V 3,500,000 m month m3
Q = - = - - - - ---=57,534-y
i-,.. 2 mo 30.42 d d
F = K4m = 0.005(2) - 0.0099

P l + K4m l + 0.005(2)
K6 = 0.8·1.07 22_20 +200/24[ 7 ]

- - - l-e--0.4( ) =0.916+2.795= 3.711 d"
1
0.4(7)
A mass balance can be written as
dN
-=W- QN-v AF N- k VN
dt • P •
At steady state can be solved for
3 106
= (57534+3.711(3500000)+0.3(500000)0.0099) m I ~ ~ = I.305 x 1014 number
d lOOmL m d
27.4 Because there is no resuspension or burial, the bacteria balance for the sediments can be written as
which can be solved, along with the steady-state solution for the water for,
whic~ can be solved for
W=-N_2_~_2_H_2_(Q_+_~_1V,_1_+_v_,F~p_1A
_,_)
v,F,1
Substituting the parameters gives
W = (IOOOOO x 103 x 100)0.916(0.1)(57534+3.711(3500000)+ 0.J(S x IOS)0.0099) = _ x 11 number

4 02 10
0.3(0.0099) d
27.2
27.5 Section 27 .s reports a range vL cfiameten; and densities m
Organlam DlalMW(J&m) Dendy
Giordic s l .07S
(3-7) (1.05-1.1)
CrJpt#pori.11,,,,. 10 1.07S
(7.5-14) (l.05-1.1)
Stabs law is
=L(P•-Pw "L2
V
• II µ r
Merely substituting in the meaa. minimum and maximum values from the table gives the following
means and ranges
Organlam Veloolty (mid)
GUll'IM 0.094
(0.022-0.232)
0.427
(0.lJM.931)
A Monte-Carlo analysis (recall Lee. 18) can be uaed to randomly vary the diameter and the density.
Using a uniform distribution. the 1'91Ulting llatisbcs are the same. In addition. however, the analysis
also provides the distn1Jutiona.
25 a-6,
3J
c- 15
I 10
l 5
0
0.01 0.05 0.09 0.13 0.17 0.21
25
3J
o,,.,.,,,.,.,.
i 15
10
l 5
0
0.15 0.31 0.47 0.83 0.78 0.96
27.6 (a) The cyst balance for the lake can be written as
O=n...
~ID
-Qc-v6 Ac
6
27.3
Cjo = (1 + ~
0.06S7t
'ln.1
= 0.7084 cysts
L
(b) To meet the H>-4 risk level,
~.. =(1+~)7xl0~ =4.96 x l0-.scylll

0.0657 L
which 1epeeent6 a log reduction of
4.96x 10-.S = -4.1

1 5
og 0.7084
Therefore, the cooc:entration would have to be reduced over 4 orders of magnitude.
27.4
LECTURE 28:
21.1 First. the system's flow rate can be calculated as
1 3
lx 10' m (2 m) week m
Q=-------=9524 -
3 'Week 7d d
A phosphorus mass balance can be written as
0= QP. -QP-r,,.,vAa
and solved for
QP. -r,..vAa 9524(5)+ 1(0.5)lx 10'(0.2) mg

P = - - " - - = - - - - - - - - = 3.95-
Q 9524 L
Because the alp ratio is 1, then/a ratio should be the same as the nip ratio
. 16 atomsN 14 gN / atomN gN
r =r = - - - . - - - - - P = 1.23-
- "P 1 atomsP 31 gP / atomP gP
A mass balance for nitrogen can be written and solved for
0= QN. - QN-r""vAa-kVN
9524(5)+ 7.23(0.5)1x l0'(0.2) mg

----------20.69-
9524+0.05(2x 10') L
The NIP ratio can be computed as
N 20.69
-=--=5.24
P 3.95
Therefore, the system would tend to be nitrogen limited.
21.2 The photosynthesis rate can be used to determine a uro-order depletion of nutrient. For
phosphorus, this can be expressed as
Px
p(x) = p -r - -
0 "° HU
where p(x) = phosphorus concentration (g m·3:J as a function of distance x [m), Po = the initial
phosphorus concentration (g m·3:J. r,,, = the amount of phosphorus taken up per oxygen generated [gP
g0·11. P = the photosynthesis rate (g m·2 0 1), H = mean depth [ml, and U = velocity [m a1). The
phosphorus to oxygen value can be determined as (based on stoichiometry of Eq. 28.1),
28.1
31
r = gP = 0.00702 mgP
po 138(32) gO gO
This result. along with the olher model parameters can be ued to compute
10 X ➔
p(x) = 0.4-0.00702 ---=0.4-4.062xl0 x
I 0.2(86400)
A similar formula can be determined for nitrogen as
lO X _,
n(x) = 2- 0.05072 - - - = 2-2.935xl0 x
1 0.2(86400)
The resulU can be plotted versus distance
0 20 40 80
distance (km)
Thus, the nitrogen runs out a little before 70 km.
28.3 (a) The NIP ratio is less than 10 and therefore phosphorus should limit growth. Therefore, the
amount of phytoplanlctoo that could potentially be produced is
a = 10 mgP. 40 mgC . 25 µgChla . mgChla = IO mgChla

m3 mgP mg C 1000 µgChla m3
mgP mgC gC gC
(b) C = 10- -40- -----=--- = 0.4- 3
m3 mgP 1000 mgC m
mgP mgN mgN

(c) N=l0- •72- - =720--
m3 mgP ml
mgN mgO gO gO
0 = 720--- 4.57---- = - - = 0.329- 3
m3 mg N 1000 mgO m
28.2
28.4 Several parameters needed for the evaluation are
Q, = 86400 cmd
Q,., = 20000x0.57 = 11400 cmd
p,= 10 µg/L
n,= 100 µg/L
p.,, = 8000 µg/L
n.,, = 40000 µg/L
NIP ratio above and below the input can then be computed as
100
(N: P)abo
vc
= -10
= 10
(N: P) = 4750.92 = 5_047

below 94 l3497
The results suggest that the river may be phosphoros limited above the input and nitrogen limited
below. Note that if the raw sewage did not have significant detergents (Table 28. 1), the results at the
mixing point would be
(N: P) = 4750.92 = IO
below 475.092
This suggests that phosphorus limitation.
21.J
LECTURE 29:
19.1 The runoff' flow rate
cm m m3
Q., = 25-6,070.000 m1 ---1.Sl7.SOO-
yr 100cm yr
and the groundwater flow rate
Q = 100 m3 365 d = 36.SOO m3

, d yr yr
can be combined to yield a total flow of 1,554,000 ml/yr. The total loading can be calculated as
W,1 ::; 40(6,070,000) = 2.428 X 101 Dig

yr
W• = 24(4,050,000) = 0.972 X 101 mg

yr
The total phosphorus coocentration can be computed as
8
= 2.428 X 1a8 + 0.972 X 10 _ J mg
44
p 1.554,000+ 12(4050000) · m3
19.l NOTE TO INSTRUCTOR: If SCUdeatl wut to employ J:q. 19.27 to estimate Secdai depth,
dtey will require a value for a. A pod appromnatioa is a• 0.035 Ufl&Chl/m.
(a) First. the volume and flow can be calc:ulated as V = lo'(5) = 5xlo' m3 and Q = 5x106/2=2.5xlo'
ml/yr.
8
p-. = 2.Sx 10 _ 100 mg
m 2.S x 106 ml
The phosphorus concentration can be estimaled in a variety ~ ways. For example, Vollenweider
(1976) can be used to estimate
The Cbapra and Tarapcbak (1976) model gives
(h) Sowa-al formulas can be med to eatimlfe ..Jllolapbyll
29.l
p= 16.8 p=41.4
Dillon/Rigler 4.35 16.12
Rast/Lee 4.70 9.33
Barucb/Gakslatter 6.23 12.92
(c) Similarly with Seccbi cleplb. if the Rut IDcl Lee dalonJplyll remits are used
p= 16.8 p=41.4
Rast/Lee 3.06 2.21
Eq. 29.27 (a = 0.035) 6.04 3.99
(d) Finally, several approaches are allo available (plots md equatiom) t-- estimate the allowable lOMing.
Model L,natfing Int1ow Areal loading

Oct/Yr) concentration (R/m2lyr)
(µ&IL)
Pig. 29. 1 10 28 0.070
Pig. 29.2 115 46 0. 115
Eq. 29. 13 60 24 0.061
Eq. 29.8 149 60 0. 149
Thia exercile ii meant to dramatically demomtnte the inhenat uncertainty of the empirical approadi. The
major-. ii that the daoice ofmodel can figure aipificantly in the mapitude ofthe predictiom.
29.3 NOTE TO INSTRUCTOR: Aaume tbat the laitial bypolimaedc OKJPD at the . . .nnia1 of the
smD1Deri1lOllllfL.
(a) Pint. the 'VOiume and flow can be calculated as V = (lx10 + 0.8x10 )
7
7 7
= l.8xl07 m3 and Q =
1.8x10 /3=6xl<f m31yr. The phosphorus concentration can be estimated in a variety al ways. For
example, Vollen•der (1976) can be used to estimate
The Cbapra and Tarapcbak (1976) model gives
(b) Sewnl formulu can be med tp ecimlte chlorophyll
p=28.5 p=42.7
Dillon/Rigler 9.36 16.85
Rut/Lee 1.02 9.55
Barucb/Gakslatter 9.54 13.24
(c) For Seccbi depth. the Rut IDcl Lee model and c:hlorapbyll remits can be used
p = 28.S p=42.1
Rast/Lee 2.53 2.18
(d) Usina the Rut md Lee model,
29.2
AHOD = 0.0851(42.7)0·467 = 0.491
Thm, a simple linear model can be Uled tC' c,atim.., t.he bypoliametic oxy,m.
6
o = l0-0.491 lxlO t
0.8x 107
0 30 IO 90
29.4 The amount of phosphorus buried can be computed as
wbln,d = w., -WO,. =2.03 x 101 (100)-2.03 x 101 (33) = 136x 101 mg
yr
The phosphorus concentration in the sediments can be computed as
Pl= (l-;)/X._%P) = (1 - 0.9)2.S x 106(0.12) g x 1000 mg= 300 000 mg

100°/4 100 m3 g ' m3
The burial velocity is then determined as
v• = Wbln,d = 136x 10
1
m x 1000 mm_ 3_47 mm
~Pl (13.068 x 106 )300,000 yr m yr
The total quantity recycled over the whole year can be determined as
wtel)'C.. = w_.. -Wbwial = 46(13.068x 106 )33- 136 x 101 = 62.4 x 101 mg
yr
Since thls only occurs during the ano,tia, we must determine the period of anoxia To do this, we first
employ
AHOD = 0.086(33)
0
A7I = 0.4574 ~
md
which caa be used to compute when the bypolimnion goes below 1.5 mg/L.
AHOD
DO = D 0 ; - - - t
z,,
29.3
1.5 = 8- 0.4574 1
1.5
1.5
I= (8- 1.5)-- = 107 d
0.4574
Thus, the period of anoxia is
,_xia =(315-135)-107 =73 d

Consequently, the recycle velocity can be calculated as
1
v = 62.37 x 10 (365) x 1000 mm= 1.95 mm
r 13.068 x 106 (300000)73 m yr
Theae values can then be used to simulate the following trends. The results for p. = 45 and 25
m,1'/m3 are also shown.
40
I
30
P1 (ffllP/m1) 20
··½.-.-·~y~---:-:.-·-
10
· · · ·· ··· ·· ··· ·· · · o;.••··
•--------------
1NO 1'70 1t80 1tl0 2000
29.S As shown below, doubling the load approximately doubles the concentrations ill the system,
p so eplllmnlon
(lftfllm')
21
0 IO 120 1IO MO 300 310
,, 90 hypollmnlon
(mglm1 )
21
0 _ _ _ _ _ _ _ _ _ _ _ _ _ _---,._ __
0 90 120 190 2'tO IDO NO
29.4
LECTURE 30:
30.1 The key to solving this problem is to n,mgnia that at equihlmum. the temperablJ'el « the three
phases will be equal. This can be expnaed by the beat budget
where the subscripts, i and/. stand for the initial and final states~ and the sublcripts. s, '· andg stand
for the solid, liquid and gas, respectively. The balana, can be IOlwd for
1800(840)0.003(50)+ 998.2(4182)1(10)+ Ll64(1012)0.997(20)

T1 =- - - - - - - - - - - - - - - - - - - 1 0 . 0 4 6 " ' C
1800(840)0.003+998.2(4182)1+ U64(1012)0.997
The "before and after" beats can be computed by Eq. 30.2. For example, the heat contained in the
brick initially can be computed as
kg J 3 0.2388 cal kcal

T1 = 1800-
3
(840--)0.003 m ( 5 0 0 C ) - - -3 - - S4.2 kcal
m q°C J 10 cal
The JMWoing quantities can be calculated in a similar fashion to yield
Before After
Brick 54.2 kcal 10.9 kcal
Wats 9968.6kcal 10014.5 kcal
Air 5.6kcal 2.8 kcal
30.2 A steady state heat balance can be written as
0= Q,:C,T,,,-Q/C,T+KA(T. - T)
Q/C,T., -QJC,T Q/C, T,,, -T

A--~-,---,----------
- KA(T- T.) - KA T-7'.
substituting the parameter values gives
3 3
7 m /s(l g/cm )1 cal/(g °C) 400C-2S°C 86,400 s 100 cm
A=------ 2
- - - · - - - - - - - - - - - = 4.S36 x 106 m2
40cal/(cm d °C) 2S°C- 200C d m
30.3
30.1
30.4 Finl. the eair can be calculated as
e,. )7.27(2))
= 4,5!)6e237.3+2J = 18.714 mmffg
This value can be divided by the relative humidity to determine the saturation vapor pressure
corresponding to the air temperature, e.J..T.) = 18.714/0.3 = 62.38 mmHg. Tbua, the following
relationship bolds,
J7.27T*
62.38 = 4.596. l37.3+T*
which can be solved for T• = 42.21 °C.

30.S As in Example 30.4, the individual terms can be determined. The ones that change are
Copducdon: First, M must compute the wind eft'ect (Eq. 30.22).
f(U .) = 19.0+ 0.95(6)2 = S3.2
This result can be substituted along with other values to compute the conduction loues
-cif(U" XT, - Ta1r) = -0.47{S3.2XT, - 2S) = -2S(T, - 2S)
Jyapontioa; Using the new wind function,
J ~
- f(U .Xe, -eo1,) = -S31.4.596e237.J+T. -14.3 ]
The individual terms can now be CODIOlidaced into the total beat budget.
4
.J 17.27T,
0= 300+ 300+ 642-30T, -113Sx 10-8 (T, + 273) -25(T, -2S)-S31.4.596. 237.J+T. -14.3
]
This nonlinear fJQWltioo can be l0lved numerically for T, = 18°C. Tb.is value can then be substituted
back into the beat budpt in order to 888el8 the relative magnitude of the individual terms. The
RIUlta are shown below
T, =18
30.2
LECTURE 31:
1 I.M t& ISL . Ill • • illllU Hm Ut H Ill l!Ul t . U 111:.t U Ill t • • • a t I Q
31.1 The steady-state balances to be IIOlved are
Note that the flux in the first equation is given by
where the term e. is also a function of T. (according to Eq. 30.16). The vertical transport can be
eairnated from Soodg,ass's model (Eq. 31.9), These 2 nonlinear equations can be aolved
simultaneously for T. = 11.19 and T,,,.. 10.13 °C. (A simple way to do this is merely to integrate the
~ balances (written in differential equatioo form) until they reach steady-state.] At this point the
heat transfers for the system are
J,,,• 60000 J-= 840 Jou,= 60840 cal/cm2/d
31.l (a) The tbennocline diffusicn coefficient can be calculated with Eq. 31.8
v = 4 x 101• ml In ( 19.94-7.77) = 0.02117 m

' (l x l010 rn 3 )93d 19.94-12.5 d
which can be converted to a diffusion coefficient by
m m2 10'4 cm2 d 2
E,=v,H, = 0.02117d(7m) = 0.l48 d m2 0.01718~
864005 s
31.l
This result can be compared with that developed from Snodgrass's model (Eq. 31.9)
•l
E, = 1.01 x 10.... (17.S)l.uo, = 0.01941 an
s
(b) (i) The apparent hypolimnetic oxygen demand can be computed as
D =(10.72-0.6)g/m3 4m=0.43Sgm-2d-1
a 93d
(ii) The diffusion correction can be developed by remgnizing that the oxygen budget for the
bypolimruon can be written as
or, collecting terms.
do-,, v, v, Da
-+-o,.= - o - -
dt H, H, • H,
which can be solved (if o,. =o,.., at t =0) for
which in turn can be aolved for
Substituting parameter values give D. = 0.529 g m·2 <f1.
(iii) Equation 29.30 can be used to determine
(c) As expected, the apparent rate is smaller than the actual depletion rate corrected for diffusion.
This is due to the fact that the real rate has to be higher to yield the same oxygen change given the
added oxygen coming acr06S the thennocline. The empirical result is smaller than both estimates
from the data. However, considering the scatter associated with the data from which the empirical
equation was deYeloped, the result are consistent.
31.2
31.3 NOTE TO INSTRUCTOR: The value of 0.13 cm2/s sugested in the problem statement is not
a 1ood one for a lake this small. Therefore, an interestia1 way to eobaace the question is to
have the students first me this value, and tbeo redo the computation with a diffusion coefflcient
demed from Eq. 31.9. The solution below is based on this eobancement.
This problem merely involves integrating
dT11=EA
- -1[
T~(t)-T1, ]
dt t,v,,
where T.(t) is a table of epilimnion temperatures generated by linearly interpolating between the
points in Table 31.1. Using a value of H = (]1S,000+1S,000)12S,OOO = 10 m, Eq. 31.9 suggests that
a value of7.07xl0~(10)1.1505 = 0.01 cm2 s·1 be used. Employing this value yields,
T 20
(oC)
10
0 60 120 180 240 300 360
31.3
LECTURE 32:
32.1 The Lineweaver-Burk approach can be developed by fitting a straight line to a plot of 1//,, versus
1/S. The fit which is shown below yields a slope of 1.21 and an intercept of 0. 718.
1/k 4 ,
2
0 +--+---t-+---+--+---t-+---+--+----i
0 1 2
1/S
These valuea can be employed to give parameter estimates of ks-= 1/0.718 = 1.392 cf1 and k61 =
1.392(1.21) = 1.684 µg L"1• These estimates can be substituted back into the model and the result
shown 31 the dashed line on the following plot. Note that the residual sum of the squares between the
untransformed data and fit equation is 0.243.
kg 2
- .•
Lineweaver-Burk
1 \
~#
Integral least-squares
0---------------
0 4 8 12
s
(b) The integral least-aquares method can be employed to estimate ks-= 1.874 and k,, 3.628. =
Both the fit shown above (solid line) and the residual sum of the squares (0.129) indicates that thls
method is superior to the Lineweaver-Burk approach.
32.2 A 4th-order RK algorithm with a step si7.e of 0.1 was used to integrate the system of ODEs.
dp = -1.5(l) _La + l.5(0.02S)a+_.!._(50- p)

~ 2+p ~
with initial values ofPo= 20 and a0 = 0.25. The results are shown in the plot below,
32. l
algae
20
10
phosphorus
0 ~ ...................t--+.....-+-+---t....+-+-+-t-+-+--+-+---t....+-+-+-......
0 20 40 80 100
t(d)
(a) A transient occurs that last about 12 days. For all intents and purposes, the nutrient is at a very
low steady state past that time. However, the solution for the alp app1oacbes equilibrium slowly.
After about 32 days it reaches about 95% of its ultimate steady-state value.
(b) The ultimate concentrations of the algae and the phosphorus can be computed as
o.11m8;
m
Q 1
a=-~y_(~P•_-_P_)_ _ = 30(50-0.17) =20.162SµgChla
p 1.5(1) O.l? - 1.5(0.025) L
a_k._ ~--a_k,
,.. •. k,,, + p ,... 2 +0.17
Both these results are confirmed by the numerical simulation.
( c) The wa1hout rate can be estirnated 1'}'
r . = k, + p,_ = 2 + SO = 1.091 d
- (k,.max -k, -v,/H)p,_ -(k, +v,/H)k, (1-0.025-0.l/S)S0-(0.025+0.l/S)2
32.3 Using the approach outlined in Example 32.3, the results (similar to Fig. 32. 7) are shown below.
Using the approach outlined in Example 32.4, the results are 1iai. =7.47 hr and S = 80.77 mg/L.
Thus the steady-state substrate level does not change since it is independent on the inflow substrate
concentration.
32.2
500
250
0
0 50 100 150
500
250
0
0 50 100 150
500
250
0
0 50 100 150
32.3
LECTURE 33:
33.1 The light at the top of the hypolimoion can be computed as
l(z = 4 m)=700e◄.l(4) = 3145
This result can be used to determine the alpha's,
a 0 = 3145 = 1.258 al= 3145 ~◄.2(3) =0.690

250 250
which in turn can be employed to compute the light attenuation factor
; = l .7 l8(05)(1258-0.690)=0.492
L 0.2(3)
The nutrient information can be uaed to compute the nutrient attenuation factor
4
; = - = 0.667
N 2+4
The growth rate is
k, = 2(0.492X0.667) = 0.656 d-1
33.l (a) Equation 33. 7 can be applied as
kg,T =0 TS2.S
k =l T-2.5 2.ss T s 2S
g;I' 25 - 2.S
k = l 36-T 2SS TS 36
g ,T 36-25
kg,T = 0 T >36
or simplifying
kg,T =0 T s 2.S
kg,T = 0.04444444T-0.lllllll 2 .5 S T ~ 25
kg,T = -0.090909097 + 3.272727 25 S T S 36
T > 36
(b) Nonlinear reg,essioo can be used to estimate the parameters ofEq. 33.8. The result is
k = le ~ .005S65(T-25)2 TS2S
g,T
kg,T = le~.02333J(25-T)2 T ~ 25
33. l
Both fits are shown in the following plot
1.2
0.8
OA
0
0 10 20 30 40
33.3 The integral to be evaluated is
which can be integrated to give
33.4 The model (Eqs. 33.31 and 33.32) is modified by merely adding a settling term toEq. 33.31,
The results show that settling lowen the ultimate amount of algae in the reactor (compare with Fig.
33.8),
10
&
0
0 10 20
33.2
33.S The sensitivity analysis was implemented by simple parameter perturbation (±25%). The
results are shown below:
10 10
Oo
I I
0 0
0 10 20 0 10 20 0 10 20
10 10 10
k,,.
I I I
0 0
0 10 20 0 10 20 0 10 20
·:k?s
0
0 10
I
20
:;
Some conclusions:
• Although the algal initial condition influences the time to steady state, it bas no impact on the
final levels
• The phosphorus initial condition stroogly influences the time to steady state. However. the
ultimaae algae and phosphorus concentrations are unaffected (Note that if the simulation time is
increased. the two pertwbed simulation converge oo the base case).
• The phosphorus inflow concentration bas a more pronounced impact on the algae than the
phosphorus level..
• The time to steady state is greatly affected by variations of kr_ r and to a somewhat lesser extent by
k,.,,. The raultl are relatively insensitNe 10 both k,, and I. for this example. The ultimate
phosphorus level will be strongly affected by k6,r, k,, and k,.11•
33.6 (NOTE TO INSTRUCTOR: The residence dme used to 1enerate the results from Eumple
33..J was actually 1000 d, ntller than the JO d line4 iJI tile problem statemmt. TIie residence
time for the present problem should be chanced to JO days to better mimic the effect of fast
flushing on the variable chlorophyll model. The solution below is based on f',., = 30 d).
Using a shorter residence time results in the profiles shown below (comparable to Fig. E33.3. Notice
that the most no&eworthy change is that the algae peak and ultimate concentration is smaller because
of the faster flushing.
33.3
200
100
-
100
0 I 10 11 20 25
0
0 5 10 15 20 25
0.1
0
0 10 20 25
' 15
33.4
LECTURE 34:
bl ~a l t t ~ 1l l t UU~M.U
34.1 The results are shown in the plot below for a time step of 0.4. The results for the first few steps are
also shown.
0 10 20 30 40
t X y k1,1 k1,2 X ki.1 ki,i X ,ks,1 k.s,2 X , k4,1 k4.2 ,

0.00 20.00 5.00 10.00 -0.50 22.00 4.90 11.22 -0.29 22.24 4.94 11.25 -0.27 24.50 4.89 12.52 -0.05 11
0.40 24.50 4.89 11.22 -0.29 26.74 4.83 13.83 0.17 27.26 4.92 13.85 0.22 30.04 4.98 15.09 0.50 13
0.80 29.94 4.95 15.11 0.49 32.96 5.05 16.31 0.80 33.20 5.11 16.22 0.84 36.43 5.29 17.16 1.21 16
1.20 36.43 5.29 17.17 1.21 39.86 5.53 17.83 1.64 40.00 5.61 17.54 1.68 43.45 5.96 17.55 2.20 17
34.l The results of the sensitivity analysis are depicted in the following plots.
1
+
0 40 80 120 0 40 80 120
0 40 80 120 0 40 80 120
34.1
0 40 80 120 0 40 80 120
0 40 80 120 0 40 80 120
Some general observations:

&
"•
• Increasing the net g,owth rate of the algae advances and heightens the peaks of both algae and
zooplanktoo.
• Increasing the zooplanktoo grazing rate advances and diminishes the peaks m both algae and
zooplankton.
• Increasing the 7.00plankton grazing efficiency diminishes and advances the peak mthe algae.
The impact on the zooplanktoo is similar but less dramatic.
• Increasing the zooplankton death rate increases and advances the peak mthe algae. The impact
on the zooplanktoo is limilar temporally, but less dramatic in terms of magnitude.
34.3 The results of the simulation are displayed below:
1.5
0.5
0
0 50 100 150
34.4 The results of the simulation are displayed below. Initially, the results proceed as expected. That is,
there is a succession of phytoplankton ➔ herbivores ➔ carnivores. However, after the first cycle, the
carnivores death rate is low enough that they are not significantly reduced by the time the algae peak
again. Consequently, the herbivores do not exert sufficient grazing pressure on the algae which then
grow in an unbounded fashion.
34.2
2
0
0 10
~ 20
:t 0 10
.atl
20 30 40
34.5 The results oftbe simulation are displayed below (this plot is analogous to Fig. E34.2). Notice how.
as time progresses. the cycles seem to be dampening out and c:ooverging on steady values.
1
~
', ........._________
0.6 '
'
----------·-· --------
0 60 100 160
34.6 (a) Nonlinear regression can be used to minimi7.e the sum of the squares of the residuals between the
data and the simulation. The resulting estimates are: a= 0.32823, b = 0.01231, c = 0.22445, andd =
0.00029. The fit is:
1200
•
800
400
0 -+---f---+----t--+----+-----1
1960 1990 2020
34.3
(b} The results in state ,pace are.
300
-I
200
0
I 100
0
0 500 1000 1500
Moose
(c}
1200
800
400
0
1980 1990 2020
300
200
100
0 500 1000 1500

Moose
34.4
(d)
1200
•
800
400
0
1980 1990 2020
300
I• 200
i 100
0
0 500 1000 1500
Moose
34.5
LECTURE 35:
35. 1 If the phosphorus loading is increased a factor of about 2.3, the lake begins to experience nitrogen
limitation. The folowing plot shows the trends of total inorganic nitrogen and soluble reactive
phosphorus.
300
200
100
p
0
0 60 120 180 240 300 360
Notice how the n:p ratio is on the order of 7.2: 1 that indicates the point at which nitrogen tends to
become limiting. This is reinforced by the following plot, which also indicates that light limits the
algae the most.
0.8
0.8
0.4
0.2
0
0 80 120 180 240 300 380
35.l A limited sensitivity analysis (parameter± 25%) is depicted in the following plot. According to
these results,
• The growth rate offers the largest leverage over the algal peak. Although it does not change the
magnitude, it has a dramatic impact on the peak's timing. A higher rate causes the peak to occur
earlier.
• The grazing rate bas a strong influence on how the bloom terminates. A high grazing rate cuts
off the peak abruptly, whereas a low grazing rate allows the algae to persist longer.
35. l
20 20
10 10
0 .t=~~-+--+---+---+--+--+--1--+-I
0 IO 120 180 240 300 380 0 IO 120 180 240 300 380
20 20
10 10
0 -t---+---+----t-+---+--+----t-+---+---+---t-----1 0 +--+--+---t-+--+--+---t-+---t---+---t------1
0 IO 120 180 240 300 380 0 IO 120 180 240 300 380
20
10
0-
0 ---
IO- - - - - -180
120 - - -240
- - -300
----380
35.3 The temperature data can be used to calibrate the vertical diffusion with a resulting value m0.01
cm2/s providing a good fit. Other key model parameters are listed in the following table:
Panmeeer S-.ltol Valae Ualts

Alae:
Growth rate b.20 u cfl
Temoerature factor o. 1.066
Rcsoiration rate k,. 0.025 d•I
Temoerature factor ~ 1.08
Settlina velocitv v. 0.2 md·1
Ootimal lillht I, 350 ly d"'
P half-satwation L LS 112P L"I
Background lillht extinction k,,' m•I
Herbivorous .. :
0.2
Grazing rate c.. 2.25 L ml!C"1 d·1

Tempenture factor 0.. 1.08
Grazing efficiency $ 0.8
Resoiration rate It.,, 0.125 d"I
Temoerature factor 14 1.08
Alaae half-saturation
Camivorous ,nnnlankton: "- 10 W!Chla L•I
Grazing rate c.. 5.5 L ml!C·1 d"1

Temoerature factor o.. 1.08
Grazing efficiency 8. 0.7
Rcsoiration/exett.tion/cleath rate ~ 0.025 d"'
Temperature factor ~ 1.08
35.2
Death rate 4 0.04 .tr'
Temoenture &c:tor e.,. 1.08
Habivorc half-saturation 0.4 11DllCL·•
:
JT--• -:t- L.IC
Settl.iu
p
"""v. 0..2 ma'
Decay rate k. -0!075 d"'
~ fKtor ~ 1.08
These values result in the following fits:
24 T(oC)
18
12
6
0
0 61 122 182 243 304 365
12 Algae (mg/m3)
0
0 11 122 182 2'3 304 386
2' SRP(mg/m3)
18
12
I
0
0 11 122 182 243 304 366
I
Unavall P (mg/m3)
4
•
2
0
0 11 122 182 243 304 386
35.3
0.2 • ■
Herb Zoo (mgC/L)
0.1
0 ~===:C:::.'---1---i--'--+~::::::..,~~==
0 81 122 182 243 304 386
0.2 Carn Zoo (mgC/L)

■
0.1
0 ~:;=~~~~-'--+---'--+--.,___~
0 81 122 182 243 304 386
35.4
LECTURE 38:
36.1 Terminal kinetic loaes might be justified by 11Q1rning that the recycle rate from non-living organic
matter is slow relative to the other model rates. Terminal settling I018e8 might be justified if the
water \Wf'e always oxygenated. This would tend to diminish the feedback of bath nitrogen and
phosphorus. In addition. low flows might also support this 8"'1mpti<m bec:aal• low flow conditions
would tend to diminish scour of bottom sediments.
If the assumptions \Wf'e erroneous. it would mean that the loaes would be fed back into the available
nutrient pools. This would tend to allow g,owth to persist longer.
36.l [NOTE TO INSTRUCfOR: Tbe ndo of carbon to dry Mipt sbauld be cbaapd to 0.4 sCJg-
dry weipt.J
a= Soo 8 •dry 10 mgChla 0_4_L_I_ = 2000 mgChla

m2 gC, g-dry 1m m3
36.3 (a and b) Using mass ba1anc:es at the OUlfall to eslablish initial conditions, Eqs. 36.4-6 can be set as
These can be employed to develop trends as shown below.
11000r-------
10000
IOOO p
I
-IOOO
Thus. the system becomes nitrogen limited at a travel time of about 18 to 19 days when the nitrogen
concentnltioo falls below 5xl5 ... 75 J,&gN/L.
(c) If the N loading is reduced to 5000 µgN/L, nitrogen becomes limiting at a travel time of about 11
to 12days.
36. 1
0 4
• 12
36.4 (NOTE TO INSTRUCTOR: This problem is not u simple u it seems and should not be
aaiped Usbdy. The major problem relates to bandlin1 the numerical soludon at tbe point that
tbe nitrosen coacentradon approaches zero]
The result for part (a) can be calculated as
The result for ( c) is that phosphorus becomes limiting at about a travel time of 11-12 days when the
nitrogen concentration falls to limiting levels.
36.S [NOTE TO INSTRUCTOR: See note on Prob. 36.4.)
The result for parts (a) and (b) can be calculated as
0 10 1,
'
36.2
Notice bow Ditropn l,ecomes limiting IOODel' (14 d) than in Probs. 36.3 and 36.4 (18-19 d) due to
nilrifiallioa IClllel rl ammonium. Similarly. when the nitrogen loading is reduced to 5000. the
limitation occurs sooner (1~11 d) in contrast to Example. 36.3 and 36.4 (11-12 d).
36.6 The QUAL2E result sbouJd comspond approximately to the solution shown below.
0 .. 100 1IO
36.7 The QUAL2E result should correspond approximalely to the solution shown below.
30
20
11
0
0 100 11D
20
15 N..,
10
5 erpalc:N
altrlte
D tnfe
0
0 100 150
36.3
8
4
.......
!::::===========
Pw
0
0 50 100 110
36.8 The result for parts (a ) and (b) can be calculated as
0 10
and for part (C) as
11000
· ··· · ·· ·· · ·· ···· -- ·· "-- \..
10000 ...
..
n
D +-- --- -+----__;~i-,----------i
D
• 4
36.4
LECTURE 37:
um
37. 1 The system is defined by the following equations
(1) (2)
(3) (4)
Solve (3) for [NH3] and substitute into (2) and solve for
This, along with (l) can be substituted into (4) to give
After determining that pKw = 14, this equation can then be solved iteratively for pH= 11.458
37.l The system is defined by the following equations
.,.._7 = [H+][CH 3COO-)

10 (I) (2)
[CH3 COOH]
(3) (4)
The first equation can be solved for
[H•1cCH 3 coo-1
[CH COOH]=
3 10◄.7
and the result substituted into the mass balance (Eq. 2), which can be solved for
- 1 ....
[CH 3 COO )= +/ .,.. 7 10 (S)
l+[H ) 10 .
which can be resubstituted back into the mass balance which can be solved for
(6)
Equations S and 3 can be substituted into the charge balance (Eq. 4) to yield
37.1
This result can be used to determine the other species
10-14
[OH -1 =--=2.75 x l0-10 M
10-4·44
- 1 -4 ~
[CH3COO ]= -4.+t/ -4 7 10 =355xl0 M
1+10 10 ·
10-4·44 / 10-4·7 -4 _,
[CH 3COOH]= -4 44 / -4 7 10 =6.45xl0 M
l+ 10 . 10 .
37.3 The activity coefficients can be computed as
0.5(1)1 ,/o.ooi, 0.5(1)1./o.oiifi

y H• = 10 H0.33(9)J0.0025 = 0.9511 y HCOi- = 10-l+o.33(4)J0.00l, = 0.9474
which can be employed to compute
37.4 The hydrogen ion concentration is simply,
The ioni7.3tion product of water can be calculated as
K = 104717.J/303.1'+7.llll...C303J5)+o.010365(303.15)-ll.lO = 10-13.133 = L47 X 10-14

"'
which can be used to calculate the hydroxyl ion.
[OH-]= L47 X 10-14 4.64 X 10➔

3.16x 10-9
which corresponds to a pOH = 5.333.
37.2
LECTURE 38:
I • Ut:iHJI t .H l l U!ol!UU1L ;_i nna~Ul .v ¥
JI.I Example 38.1 can be repeated for different pHs. The results displayed below show that because of
our choice of parameters. the total coocentration will remain unaffected by the pH change. This is
due to the fact that the volumetric reaction and volatiliminn rates are equal.
k,,V = v.,A.,
0.05(50.000) = 0.1(25.000)
2.500 = 2.500
Therefore. rewprdless of whether the ammonia is ioni7.ed or unionized. the total loss rate will be
equal since Fj + F., = l. As expected (from Eqs. 38.22 and 38.23). the ammonium is primarily in
ioni7.ed form at low pH and unioni7.ed at high pH.
TOTAL
0.0003
,.
NH• 4 _/NH,
0.0002
0.0001
,
..
4 6 8 10
(NOTE TO INSTRUCTOR: This problem can be made more iaterestiac by makin1 the lake's
vohune biger (100,000 m'). The results are shown below.)
0.0003
0.0002
0.0001
0 +---+--+---+-..a.=.+---+--+-~
4 I
• 11
31.l Example 38. l can be implemented in a ~variable mode with a sinusoidal pH. The results
ddplayed below show that because of our choice of parameters. the total -c:oncentration will remain
anaffected by the pH change. This is due to the fact that the volumetric reaction aad volatilization
rates are equal.
38.1
0.0003
.'
,
0.0002 \.
pH
0.0001
••
•• .. . ..
' '' • :NH,
0
.
• , , .
'
0 1 2 3
38.3 (a) The acid dissociation constant is (Eq. 23.28)
K =10-<D.09011-tl'n9.921293.1') =10--9.4 =3.96 X 10-10
F. = l = l =0.556
• l+[H,.]/K l+l0--,·'/3.96 x I0- 10
(b) Assuming that ammonia is primarily gas-film controlled (Pig, 20.4), Eq. 20.14 can be used to
compute
V =-J,37
--X 10-'(200)
. . ,. . .- - 0.114
" 8.206x 10-'(293.15)
The flux can then be computed as
38.2
LECTURE 39:
39,1 Equations 39.24 and 39.25 can be substituted into Eq. 39.26 to give
Combining terms over a common denominator yields the following numerator,
Collecting terms by powers of (W) gives,
3 2
[H+j4 +{K1 +Alk}[H+J +(K1¾ +AlkK1 -K.,. -K1cr)CH+J
39.2 Using an identical approach to Example 39.1, the results are: pH = 7.076, [H:zCQ,•J = 0.000502M,
[HCC>,l =0.002997M, [CO,1 = 1.79xl0°'M.
39.3 [NOTE TO INSTRUCTOR: Tbe alkalinity for tbi1 problem should be 2 meq/L u in Example
39.1.)
At steady state, the caJbonic acid level should be fixed by the Henry's constant and the partial
pressure of C~ in the atmolphere. Thus, the total organic carbon can be computed as
This value can then be substituted into the alkalinity balance (Eq. 39.26) which can be solved for pH
= 8.5.
39.4 The differential equation to be integrated is
dcr = v,, (CH CO •] -(H CO •J)+a R - p

d1 H l l• l l •n
Substituting values gives
dcT = o.s (t.23 x Io·'., - F. CT)+ 0.03125 l.5- P(t)

di 3 ° 3(1000)
where
P(t) = 3sin(2nt) Os t S 0.5

P(t) = 0 0.5s t S I
39.
and F0 is defined by Eq. 39.23.
This equation can be integJated with a numerical method At each time step. the
electroneutrality condition can be used to detenniDe the pH. This is done by determining the root or
The results are shown below (using pH= 7.96 from Example 39.2 for the initial condition)
0 1 2 3
39.5 (a) The alkalinity at the mixing point is
6 6
Alk = 5 x 10 (0.0012)+ 1 x 10 (0.003) 0.0015
6 x 106
( b) The inorganic carbon in the river and the waste are
er.,= Alk,-[OH-], +(H+], = 0.0012-4.6&x 10-1 +6.3lx 10-a = O.OOl04

a 1 +2a1 0.8457+2(0.154)
c = 0.003-l26x 10-1 +7.94x 10-7 _ 0_

00183
T.w 0.36+ 2(0.64)
These values can be flow-weighted awraged to give er= -0.001172.
(c) The pH at the mixing point can theft be detenniaed by finding the toot of
/([H•D= ( 10""'-""(H+] +2 10-6.44 K2 )0001112

(H+]2 + 10-6M(H+)+ 10-6MIQ-l0.4SI (H+]l + l0-6M(H+)+ I0-6.441Q-10ASI '
10-14.435
+ - - - [H ♦ J- 0.0015
[H+)
The root is 1.41"10"7 which con•ponds to a pH af'6.IS.
39.2
LECTURE 40: U l
40.1 first. an inflow concentration can be determined as
m = 11.4 x 1012 = 68.26 mg

• 167x 109 L
Next. Eq. 40.25 can be used to calculate
9
v = 167 x 10 • 6826-05 = 0.0ll l m = l 1.1 mm
6 6
• 3O3OxlO (1-O.9)2.4x1O yr yr
This result can be substituted into Eq. 40.25 to determine
20 m mm
v,. =912.5 6
-0.0111=0.06494-=64.94-
(1- 0.9)2.4 X JO yr yr
40.l (a) The fraction in dissolved form is
1
F = · =0.9989
" 1+ 0.002(0.531)
and F, = 1 - F, = 0.0011. Thus. the steady-state a,ncentration would be
11
c--------------------
_ 161 x 10 2.63 µg L-1
161 10 + 100(5.96 10 )0.9989 + 19182(5.96 1010)0.0011
X
11
X
10
X
Parts (b) and (c) can be evaluated similarly as
11
c= 161 x 10 =2.42 L-1
L61 x 1011 + 100(5.96 x 1010 )0.904+ 19182(5.96x 1010 )0.096 µg
11
c= 161 x 10 _ 180 L-1
161 x 1011 + 100(5.96 x 1010 )0.485+ 19182(5.96x 1010 )0.515 µg
40.3 (a) The fractions in dissolved and particulate form can be computed as
1
F 1 =---=0.8333 F, = 1- F, = 1-0.8333 =0.1667
" I +O.O2(1O)
(i) The COOC81tration in the water can be calculated as
5 µg
Cl = - - - - - - - - 1 0 0 = 37.5-
5 + 0(0.8333) + 50(0.1667) L
where the value of S is q1 =HI i-...
40.1
5
(ii) c1 = 100=23.0Sµg
5+10(0.8333)+50(0.1667) L
5
(iii) c1 = 100= 5.J7 µg
5+100(0.8333)+50(0.1667) L
(b) (i) First, the deposit.ion rate can be used to dEtennine the burial velocity .
~ ~
J• = 01
V& = . gm yr = 2.67 X 10-7 ~
(t-;)p 0.IS(25xl06 gm"3 ) yr
The resuspension velocity can be calculated on the basis of the diff'ereooe between the settling
flux and the burial flux.
m S m
V =V - - - - V =50----- 6
2.67xl0-7 =0.001333-
, • (1-;)p & (1-0.85)25 X 10 yr
The fraaion of c.ontaminant in the sediment po~water can be calculated as
1
F42 = - - - - -- --=0.000133
0.85+0.02(1-0.85)2.5x 106
The final parameter that must be estimated is the diffusive mass-transfer velocity between the
sediments and water (Eq. 40.30)
l
--3 m
v., =69.35;(300) = 13154-
yr
The ratio of the sediment to the water concentration can be calculated as
.R.ii =_Ci= v. F,1 +v.,F.,1 = 50(0.1667) + 1.315(0.8333) =

859
7
c1 ~H2 +v, +v6 +v4 Fn 0+0.00133+2.67 x 10· +1.315(0.000133)
and the water concentration calculated as
C -
•-
~ C
q. +v,.F41 +v.F,1 +v4 F41 -.Ri 1(v, +v4 F42 ) m
S 100 = 34.68 µg
5 + 0(0.8333) + 50(0.1667) + L315(0.8333)- S.658(0.00133+ L315(0.000133)) L
(ii) C1 = S 100 = 21.98 µg

5+ 10(0.8333)+ 50(0.1667)+ l.315(0.8333) - 8.658(0.00133 + 1.315(0.000133)) L
40.2
5
(iii) C1 = 100= 5.12 µg
5+ l~Q8333)+50(Ql667)+ l31S(O.S333)-8.658(Q00133+ L31S(Q000133)) L
40.4' Working in parallel with Example 40.1, the following quantities can be calculated:
12 9
v= Qm.-Qm = 4x10 -50xl0 (0.5) =l37_6 ~
Am 57770x 106 (0.5) yr
W l0x 1012 µg
C· =-=---= 9
200-
• Q 50x 10 m3
1
F. = - - - - = 0.985
d 1+ 0.0301(0.5)
P= 50x 10' = 0.0048 (= 0.48%

50x 109 + 178.4S(0.985)57770 x 106 + 137.(i(0.015)57770x 106 )
C = 0.0048(200) = 0.969 ~
m
9
3(4616xl0 )=l34 r
10.324 X 1012 y
40.S Working in parallel with Example 40.2, the following quantities can be calculated:
4x 1012 g v _ 50x lo' 80-0.5 = _ m

ffl· =---= 9
80- 6 0 000275
m 50x 10 m3 - 57770x 106 (l-0.9)2.5 x 106 yr
v, = 912.5 0.5 0.000275= 0.00155~

(1-0.9)2.5x 106 yr
, 0.000275
F i , = - - - - - - - - - - - = 0.137
0.00155+ 0.000275 + l37(i(0.000133)
P=0.0048
Ca = 0.0048(200) = 0.97 ~
m
Cl = _ __912.5(0.0148)+
__ .;....._13757(0.985) µg
_ ____,;;_...;...._097 = 7181-
0.000275+0.00155+ l3757(0.000133) . m3
40.3
LECTURE 41:
..
-41.1 (a) A nonlinear least-squares model can be fit to the data to yield.
. 28.49c.,
v=--~
l32l+c4
V 30
20
10
0
0 20 40
Cd
(b) The partition coefficient can be calculated as K, = v,.(k./k..) = 28.49/13.21 = 2.156.
(c) The fraction in dissolved form in the water can be calculated as
1
F =----=0.044 Fp1 = I- F,11 = I- 0.044 = 0.956
di 1+2156(10)
(d) the Preuncllicb isotherm can be evaluated with a power fit.
V = J. 76cf·'11:l
V 30
20
10
0
0 40
-41.2 (a) The fraction in dissolved form in the water can be calculated as
1
F = 0.5
" l+l x l0--6(106 )02(5)
Note that ifyou u,ed the nc.rounded relatiomhip (Eq. 41.28). the result would be higher(= 0.618).
for the sediment pore water.
41.l
F = l
6
=4x 10- '
,n 0.8+lx 10°"'(106 X0.0S)(l-0.8)2.Sx 10
Note that if you used the non-rounded relationship (Eq. 41.28). the result would be higher(= 6.48 x
10·').
( b) The concentration in partic:ulale form is
v2 = c2 = l0mg/ml 101µg=002µg
(1-;)p (1-0.8)2.Sxl0""' g/m3 mg · g
41.J (a) The uaction in partic:ulale form can be calculated as
1
F
1
= 1 x 10 (0.4)316227.8 = 0.387
1
" I+ Ix 10- (0.4)316227.8
Note that if you used the non-rounded relationship (Eq. 41.28). the result would be lower(= 0.281).
(b) With DOC. the result is
1
F = 1 x 10 (316227.8)0.4 = 0.197
,.. 1+ 1 x 101 (316227.8)0.4+ 1 x 10""'(316227.8)5
Note that if you used the non-rounded relationship (Eq. 41.28), the result would be lower(= 0. 165).
41.4 Using the Banks-Herrera formula (Eq. 20.46)
K,.02 = 0.728(2)°-' - 0.317(2) + 0.0372(2)2 =0.544 m d"1
which can be extrapolaled to DDT by (Eq. 41.64)

0.2,
K, =( .E_) 0.544 = 0.3 m d"1
350
The gas-film exchange coefficient can be detennined as (Eq. 41 .65)
0.2,
K 6 = 16
~ _!!._ )
350
2 = 160 m d"1
These values can then be substituted into Eq. 41.57 to calculate
v = 0.3 198 x 10_, = 0.14 m d - 1

V 3.98 x 1o-' + 8206 x 1o-' (293.I S)(0.3 / 160)
Alternatively. the Wanninkhof formuJa (Eq. 20.47) can be Wied for the li. . .film rate to determine a
value of0. 103 mer•.
41.2
3
41.! The partition coefficient can be calculated as K. = Jo.Koe = O.OS(l) = O.OS m /g. Therefore, the
fractions can be calculated as
F = 0.02S - 0.012 1
F4 =F0 = - - - - - = 0 . 4 9 4
P 1+ O.OS(0.5) + 1(1) l+O.OS(0.5)+1(1)
The rate of volatilwltion can be determined by the two-film model as
The settling and voJatilivufon can be compared. showing that volatiJimtinG dominates.
settling F,v. = 0.012(73) = 0.901 = 0.04

volatilizarion F11v., 0.494(43.9) 2169
41.6 The Koc can be calculated as K0 , = 10-6(105 ) = O.l The fractions in the water can be calculated as
F = O.l(0.5) = 0.0434 1
F41 = - - - - - - 0.8696
pl 1+0.l(0.5)+0.1(1) 1 + 0.1(0.5) + 0.1(1)
The fractions in the sediment can be calculated as
1
F,n =--------...,.----
6
0.003971
0.9+ 0.01(0.lXl-0.9)2.S x 10 + 0.1(0.9)10
Fill = O.l(l O) - 0.003971

0.9+ 0.01(0.1)(1-0.9)2.S X 106 + 0.1(0.9)10
Fv2 = O.Ol(O.l) =3.971 x 10-6

0.9+0.01(0.1)(1-0.9)2.Sx 106 +O.l(0.9)10
A steady-state mass balance around the sediment compartment can be written as
and solved for
The diffusive mass transfer coefficient between the sediments and water can be computed as
365
V4 = 19(0.9X300)-213 = 0.38158~- d __m_= 1.393 .!!!.
d yr 100cm yr
41.3
Substituting values into the equation for c2 give,
c = 0.0435(36.5) + 0.8696(1.393) + 0.08696(1.393) l =26 l.S mg

2
0.0001 + 0.00397 l(l393) + 0.003971(1393) m1
which can be used to calculate
en = (0.003971)26l.S ~ = 1039 ~ Cal = (0.003971)261.S ~ = l.039 ~

m m m m
Val = (3.971 X 10➔)26l.S ~ = 0.001039 mg • l()()O JI& - 1.039 JI&

m g g g
41.7 [NOTE TO INSTRUCTOR: Gradias of this problem will be facilitated if you make the
problem statement more specifac by stipulatinz that.the Banks-Herrera formula should be used
to determine the o ~ transfer rate and Eq. 41.JS to determine K.,J
(a) For TCDD, the following quantities can be calculated
1
F4 = - - - - = 0.224 FP = 1- 0.224= 0.776
l+0.692(S)
0.1,
3
K, = [o.728(2)0-' - 0.317(2) + o.0372(2)2 ( ~ )
= 0.306 m d"
1
3
,,{ 322
K 6 = 365(168)\
)o.2, = 59633 m yr- 1
18
10-l.61
v = l l l 6 - - - - - - - - - - - - - - - 1 0 9 . 2 m y r- 1
" 10-2·61 +8.206x 10-'(293.15}(11L6/59633)
Fpv, = 0.776(100) = 77.574 m yr- •
Fpv, = 0.224(109.2) = 24.49 m yr-•
Finally, we can calculate that log(K.,m) = 0.54. Together with log(H.) = -2.68, this places TCDD in
the sediment 7.0lle of Fig. 41.11. The other contaminants can be analyzed in a similar fashion with
the results shown below.
lotl(K..) lotl(H,) ~ ~ F1 1" K, FJV, lot!(~)

il'CDD 6.84 ·2.68 322.0 6.92xl<T' 0.224
F•
0.776 111.6 59633 1CS.2
"· Fr"•
77.574 24.49 0.54
2alC
,eds
PDD 6.12 -7.66 320.0 1.32 x10· 1 0.603 0.397 111.7 59726 0 .1 39.727 0.CX3 -0.18 ,eds
&ldria aldehyde 3.15 -8.70 381.0 1.41 xl0.. 0.999 0.001 107.0 s11n 0.0 0.071 0.00 •3.15 water
&cbulf111Mfae -1.3 -4.59 422.9 5.01 x10·• 1.000 0.000 104.2 557a, 37.8 0.000 37.88 -7.60 water
TODC1p1me 3.3 -0.68 430.0 2.00xlc, 0.999 0.001 103.8 5547'3 1CX3.8 0.100 1CX3.85 -3.00 air
41.4
(b)
2
■ TCDD
■ DDD
-2
• Andrln aldehY'I• •
Toxapllene
....... •
Endoaulfan
-10 '
I
'
-10 -8 -6 -4 -2 0 2
41.8 The partition coefficient can be estimated using Eq. 41.35. Assuming a suspended solids
concentration of 20 mg/L gives the results in the following table and plot. Thus, because it is
primarily in dissolved fonn and volatilizes, Bugs 'r Toast would be the choice.
loruK_) Kd 102.(K.i,n) 102.<H.)

Sans-a-Roach 7 5x10· 1 I -4 seds
Gnatmare 2 5 x10-<> -4 -8 seds
BuJ?.S 'r Toast 1 5 X 10"7 -5 -1 water
2
■ Sana-a-Roach
-2 ~
•
-6
Onatmare •
lug■ 'rToa■t
-10
-10 -8 -6 ... -2 0 2
41.5
LECTURE 42:
t b.Qlt K •u U . Dl l Q~ AR ;: ma•n,01,umMilKUQ l. I.Gt II!~ It Q _ It ~;: 2 U. U l t _ U U ~IUll.ilHlli:i.0!~11 l ~ n .ii9!V~ll~~\ll~~;: J!Qil,Q ii ~Q·~~~-),)Q,)i
42.1 Using Eq. 42.2.
a0 (J)=.!.1n(
8
500
100
]= 0201 m-•
42.2 From Table 42.2. the values for pyrene are k,., = 24 c:f , / 0 = 2100 ly c:f1, and l.• = 330 nm. The data
1
from Table 42.1 can be evaluated with cubic interpolation to estimate the extinction coefficients for
l• = 330 nm with the results being a,,.= 0.06785, a.,= 60.3125, ~ = 4.046375, and a.= 0.35.
These results can be used to determine the total extinction coefficient. Because the lake has a lower
aolids 00Dlellt. we will assume that D = 1.2. Applying Eq. 42.22 and using the values for Lake
Tahoe from Pig. 42.8 giws
a O (330) =12[0.06785+60.3125(0.001) + 4.046875(0.1) + 0.35(05)] = 0B49 m · 1
Now. substituting this along with the other parameters into Eq. 42.23 yields
500 1 ~.149(lO)
k =24--133 -• =0.8945d"1
P 2100 10(0.849)
This rate can be put into perspective by expressing it as a half life of 0.693/0.8945 = 0. 775 cl
The calculation can repeated for the eutrophic case from Fig. 42.8.
a 0 (330)= l.6(0.06785+60.3125(0.01)+4.046875(05)+0.35(5)] = 7.111 m·•
500 ) -7.lll(lO)
k =24-- 133 -• =0.1069 d" 1
" 2100 10(7.111)
which corresponds to a half life af 0.693/0. 1069 = 6.49 cl
42.J The biodegradatioo reaction rate can be calculaled as
which corresponds to a half life of 0.693/0.0012 =577.6 d (1.6 yrs).
{b) Prom Table 42.3, we will assume a bacterial concentration of 5000 cells mL"1 to be representative
of a eutrophic lake. The first-order value can bedetermiaed as
which cot responds to a half life of 0.693/0.06 = 11.55 cl
42. 1
42.4 The biodegradatioo reaction rate CID be calculated as
The hydrolysis rate for the lower pH CID be calculated as (Eq. 42.31)
10-1•
kL = 2.6 x 10' - - +o + 1.100-1) = 026
" 10-1
whicb CID be combined with the biodegradation rate to give a half life of
0 693
t 0
= · = 0.806 d
' 0.6+0.26
For the pH= 1.S case, the calculation CID be repeated to give
0 693
t 0 = · =0.487 d
' 0.6+0.822
42.5 Using the mean pH for the epilimnioo, the hydrolysis rate CID be •imated as (Eq. 42.31 ),
10- 14
k,, =2.46 x 103 -=.-+3.6x 10-3 +(1.3 x 1<>2)10-a =0.006061 d"1
10
The range can also be determined as 0.00438 to 0.01138 er• which correspond& to half-lives ranging
from 60.9 to 158.2 days. Thus, it is fairly constant over the range of pH's encountered in the
epilimnioo.
For the hypolimnioo, the rate at the mean pH of 7 is 0.00386 and bas a variation betMen 0.00372
and 0.00438 er• which conei;ponds to half-lives ranging from 1S8.2 to 186. Consequently, it seems to
vary much more in the surface waters. A plot of the decay rate versus pH provides an explanation of
these results.
10
1
0.1
0.01
0.001 - - - - - - - - - - - - -
4 6 8 10
42.2
42.6 The lake's surface area can be calculated as 200x106110=20x106 m2 and the settling velocity
expressed in yearly units as 0.25x365=91.25 m/yr. Then. the lake's suspended aolids concentration
can be determined as
6
m= 50 X 10 (20) 0,5))3~
SOx 106 + 9US(20x 106 ) 7m
A sorption coefficient can be eftirnaf«I 21
Therefore. the ftactions can be calculated as
1
F4 =----=0.914 FP = 1- 0.974 = 0.026
l + 0.05(0.5333)
The hydrolysis rate can be determined as
k,. = 1(10-7 )+8.l6x 10◄ +9182.7(10-7 ) = 10-7 +8.16x 10◄ +9.1827 x 10◄ = l7344x 10-3 /d=0.633/yr
The 95% response time can be calculated by first determining a c:ontarnioaot purging rate
Q v SOx 106 9125

k1 =-+F,-L+F4 k,, = 1
+0.026--+0.974(0.633)=0.25+0237+0.617=ll04/yr
V z 20xl0 10
3
19, =- - = 2.718 yr
U04
42.3
LECTURE 43:
43.1 The results for strontium 90 are summari2:ed by the following plots (comparable to Fig. E43. l):
....
10000
0 +-------i=----.-..:::::~~---i
1NO 1NO 1170 1IIO
·~, L. I=
1NO 1NO 1170 1NO
I
-~...____I~~
1NO 1NO 1170 1IIO
43.l The fractions can be calculated as
1
F41 = - - -- = 0.9109 Fpl = 1- 0.9709 = 0.0291 F,n = l = 6.67 x 10- '
1 + 0.06(0.5) O.9+O.06(1-O.9)2.Sx 106
Then the recycle fraction can be determined as (Eq. 43. 11)
F; = 0.000157 + 1(6.67 x 10-') = 0.2279

0.000157 + 0.000756+ 1(6.67 X 10-')
and the net loss rate as (Eq. 43. 10)
vr = (I - O.2279~456.25(O.0291)+ 1(0.9709))= 11 m
yr
The a,nc:eotrations in the water and the sediments can be determined as (Eq. 43.8)
C = 4 X 1010(1) = 0.1537 µg
1 10
4xl0 +11(2xlO10 ) L
3
Vl = 456.25(0.0291) + 1(0.9709) O.1 . 10 µg = _ µg
6
537 8 95
(l- O.9)2_5 x 10 (0.000157 + 0.000756+ )(6.67 X 10-')) mg g
43.1
.. . - .... - .... - .. .
43.J [NOTE TO INSTRUCTOR: TIie problem statement sbouJd be moclif"aed to read: "Repeat
Example 43.2, but determine...")
Because the model is linear, a 50-/4 reduction in the sediment concentration (from 10 to 5 µg g-1) can
be attained by merely halving the inflow concentration from 1 to 0.5 1,1g L"1. The water concentration
would also be halved from 0.055 to 0.0275 µg L·1•
43.4 (NOTE TO INSTRUCTOR: Tile lo&K. for cadmium is-14.1.J
Parameter values can be substituted into Eq. 43.26 which can be solved for FHs = 0.5339. This value
can be substituted into Eq. 43.30 along with other parameters to give
o.012sc1-o.s)2.s x 106 ) 1.94 x 10-"(lo-1 xo.8)2

C2 =( - - - - - - - + 1 - - - - - - - + c •
0.8 0.5339(0.001-c.)
which can be solved numerically for c. = 0.0005 mol L· 1• Thus, almost all of the cadmium
precipitates with sulfur. This result can be substituted into Eq. 43.26 to calculate
[HS- ]= 05339 O.OOl-O.OOOS =3.34 x 10◄

0.8
which can be substituted into Eq. 43.29 to give
7.94 x Io-" co.sx10-1 )

C - -----"-~-....;.. 1.9 X 10"11
d - 3.34 x 10-4
which is negligible.
The calculation can then be repeated for the case where c2 = 0.002. In this instance, c. is fixed at the
level oftheAV.S(= 0.001) and Eq. 43.25 can be solved for
= 0.002-0.00} _ 128 X }0-7

6
l + (I- 0.8)2.5 x 10 0.0125
0.8
43.S [NOTE TO INSTRUCTOR: As mentioned ia Prob. 43.2, tbe Kd for copper is 6slO' L k1·1.)
(a) The fractions can be calculated as
l 1
Fdl = - - - - 0.9109 Fpl =1- 0.9708 =0.0291 F42 = = 2 x 10-5
1 + 0.06(0.5) 0.9 + 0.2(1- 0.9)2.SX l 0 6
Then the recycle fraction can be determined as (Eq. 43.11), F,' = 0. 1893.
F' = 0.000157 + 1(2 x 10-') = 0.1893

r 0.000157 +0.000756+ 1(2 X 10-5 )
and the net loas rate as (Eq. 43.10)
43.2
vT = (1- 0.1893X4562S(0.0291) + 1(0.9709)] = 11 m
yr
The inflow concentration to achieve a water concentration of 0.1 µr/1, can be determined by
rearranging Eq. 43.8 to yield
10 10
c (Q+v1 A) o.{4xl0 +11(2xl0 ] µg
c- = -1 - - - = - - - - " - -10- - ------=- = 0.65-
.. Q 4xl0 L
(b) The resulting sediment concentration is
vl = 456.2S(0.0291) + 1(0.9709) 0.l. l a3 µg =5_825 µg

(1- 0.9)2..5 X 106 [0.000157 +0.000756+ 1(2 X 10-')] mg g
43.3
- .. .
LECTURE 44:
44.1 [Note to Instructor: Tbe distuca for dais problem are in kilometen, not aaeten.)
The resuspension velocity can be estirnated frorn Eq. 44.8
v = "'i(ao) v = 2.S 0.25= l.25x 10➔ md"1

" (1-;)p • (l-0.8)2.Sx 106
This value along with other parameters can be substituted into Eq. 44. 7 to calculate the profiles in the
following plot,
0 50 100
44.2 The mass balaaoe at Flint can be developed as
c = 2.66(1.1)+ 168(55) =26 L-1

IO 2.66+ 168 µg
Then using Eq. 44.28 along with the proper ~ value (take care to convert it to the proper units of
3
m lg). the zinc profile can be simulated down to the Ragone WWTP. At this point. the level will
have fallen to 16.5. A mass balance can then be employed to determine the concentration for the
lower stretch as
c = 4.34(16.5)+0.69(84) = 26.2 L"'

IO 4.34+0.69 µg
The resulting profile is as displayed below:
0 20 40 60
44. 1
44.3 The mass balances for this problem can be written as
The second can be solved for
which can be substituted into the first to give
or collecting terms
where
_ v,, F. v.F,,1 v _ 02 0.25(0.2) 0.000825 _

kn - - , n + - - -6 - - - -08•+ - 001826
.
H1 H 1 v,. +v6 10 10 0.001+0.00082.5
The solution can therefore be developed as
or, in terms of the sediment concentration
(1-0.9)2.5 X 106 (0.001+ 0.000825)1.Sx 1()4 1+ 4(0.0l826)l X lO'

. 2rr 1 = 18.4965
025(02)
44.2
.u., The organic carbon partition coefficient can be computed as
This can be Uled to determine
1
F = - - - = 0.909
" 1+ 0.1(1)
The reaeration rate for oxygen can be computed using the O'Connor-Dobbins formula
This in tum can be employed to compute the volatiliz.atioo rate for the CODtaminant
32 )o.2,
k« =( 300 0.439 =0.251
Therefore, the concentration at KP-8 can be calculated as
50 ( 0.909(0.2'1)IOOO) ffl
c = - - - -- 1-, "400 =O 6955~
60(0.909)0.251 · m3
Then. the concentration at KP-16 can be determined as
_o.909co.2,1>ci6000-1000> m
C =0.695~ "400 =0563 ~
m
44.J
LECTURE 41:
4S.l As in Example 45.2. the upCalce rate can be calculated as (Eq. 45.30)
k = 10' 2.67(0.4)0.014) 0.7 = 3848 L lcg-lipid-ld-1

• 4(0.08)8.5
The total loa rate can be determined as
, 3848
K3 = -,-+ 0.004 = 0.04248
10
and the bioconcentration factor as
3848
N3 = = 90584
"' 0.04248
1
Thus. the concentration due to direct uptake would be "'.I= 90584(1)=90.584 µg lcg-lipicf .
4!.2 As in Example 45.2. using relationships for partitioning developed in Lee. 41 (e.g.. Eq. 41.37). we
can first calculate the concentration in disaolwd form as
c4 = l I = 0.8333 µg
1+101 (106 )0.2(1) L
The solids concentration of the small organic particles (i.e.• phytoplankton. detritus) can be
computed as
As in Example 45.2. we can calculate the moplankton upCalce rate as 31.814 lJ(lcg d). and their total
loa rate as
which can be used to compute the bioconcentration factor as
N = 31,814=558195 L
lw 0.0569 • lqPJlipid)
The food-chain multiplier can be calculated as
= 0.7(0.154) 1894
Klt 0.0569
This value. along with the BCF. can be employed to determine the 7.00plankton concentration as
45. 1
V2 =558,795(0.8333)+ l894(8333.J33) =456,663+ 1,578.532 =2,044,195 kg(~
Results for the rest of the food chain can be computed in a similar fashion. and the results are
summarized in the following table and plot:
Organism N,. v,. g Vfo V N

(BCF) (total) (BAF)
PbytolDetritus 833333 833333 - - 833333 1000000

Zooolankton 558795 465663 1.894 1578532 2044195 2453034
Foraefish 497354 414461 1.581 5726034 6140496 7368595
Piscivorous fish 283559 236299 1.153 13687688 13923987 16708784
V
(101 µg/kg)
10.00
food
1.00 chain
o.oo L~=~;;~==~wa~ter~=::L=~
phyto ZIOO small fish large fish
FIGURE W.1.1
45.3 for simplicity. we will assume that the phytoplankton are the only small particles in this system (not
a good assumption. but in the absence of additional information the only one we can adopt). Using
the new concentration factor for the phytoplankton gives:
1E+10
1E+08
1E+06
1E+04
1E+02
1E+OO
2 4 6 8
45.2
45.4 As in Example 45.3, the following relationship can be u,ed to determine the inflow concentration,
Substituting the parameter values yields
C· = 105 [1 + 100(0.16667)] = 0.0128 µg

• 16,708,784(0.8333) 21.2 L
45.3
APPENDIX A.
(1) PROBLEM ERRATA AND NOTES
p. 83 (Prob. 4.4): The following parameters should be used: spill mass = S mttric tonnes. Vy = 0.1 mid
and outflow= lxlo' m 3/d.
p. 191 (Prob. 10.s): Assume an initial condition mm,= 1 g/cm 2

•
p. 110 (Prob. 11.7): The point source has units m8/d
p. 275 (Prob. 15.2): The x-values in the table are in kilometers. not miles.
p. 275 (Prob. 15.8): The abscissa af'the figure is incorrect. The figure should look like the following in
order to conform to the table:
clllortll•
-:t • •
I
111
• •
I
17N
• •
I
1711
• •
I
ffll
•
I
471
. . . . . .Ill)
p. 387 (Prob. 20.6): The time units for Station 2 should be hours.
p. 403 (Prob. 21.2): Part (c) requires that the depth of the reactor be specified. The solutions manual
assumes an H = l m.
p. 417 (Prob 22.1): Requires the cross-sectional area The solutions manual assumes an A .. = 10 m2 .
p. 418 (Prob. 22.s): Requires the per-capita generation rate of BOD from Table 19.2. That table (p. 358)
has ioc:orrect units. The units mCBOD should be g/capita/d. not m3/capita/d.
p. 418 (Prob. 22.6): This problem requires a base flow. The solution manual employs a value of IO ems.
I'w also made the distributed inflow smaller. 0.00001 m2/s.
p. '32 (Prob. 23.2): The units of U should be mis. not mid. as in the problem statement. This is not a
critical error. as the lltUdents should rec:ogni:ze that U is not required to obtain the
IOlutioo.
p. "8 (Prob. 14.2): The next to last equation in Example 24.2 should be:
A.I
p. 480 (Prob. lS.l): The diffusive mass-transfer coefficient is erroneously labeled as a bu1k diffusion
coefficient.
p. 500 (Prob. 26.1): There are several incorrect parameter values and omitted values on this problem
6 3
statement. The volumes of segment 1 and 2 should be aet at Sx 10 m and the surface area al the
sediments (i.e., segment 2) should be lx10 m • The SOD should be changed to 1 p/m1/d and the
6 1
flows and bulk dispersion coefficient uruts should be m 3/d
p. sot (Prob. 26.l): Change the BOD al the waste source to 200 m&'L.
p. !!9 (Prob. 29.l): If students w.mt to employ Eq. 29.27 to estimate Secchi depth. they will require a
value for a. A good approximation is a= 0.03S Uµgchl/m.
p. !!9 (Prob. 29.3): Assume that the initial hypolimoetic oxygen at the beginning al the summer is l 0
m&'L.
p. !89 (Prob. 31.3) The value al 0. 13 cm2/s suggested in the problem statement is not a good one for a
lake this small. Therefore, an interesting way to enhance the question is to have the students first use
this value, and then redo the computation with a diffusion coefficient derived from Eq. 31.9. The
solution below is based on this enhancement.
p. 621 (Prob. 33.6) The residence time used to generate the results from Example 33.3 was actually 1000
d, rather than the 30 d listed in the problem statement. The residence time for the present problem
should be changed to 30 days to better mimic the effect al fast flushing on the variable chlorophyll
model.
p. 663 (Prob. 36.l) The ratio al carbon to dry weight should be changed to 0.4 gC/g-dry weight.
p. 663 (Prob. 36.4) This problem is not as simple as it seems and should Dot be assigned lightly. The
major problem relates to handling the numerical solution at the point that the nitrogen concentration
plunges to zero.
p. 682 (Prob. 38.1) This problem can be made more interesting by making the lake's volume bigger
(100,000 m1) . See solution for more details.
p. 691 (Prob. 39.3) The alkalinity for this problem should be 2 meq/L as in Example 39. 1.
p. 738 (Prob 41.7) Grading al this problem will be facilitated if you make the problem statement more
specific by stipulating that the Banks-Herrera formula should be used to determine the oxygen
transfer rate and Eq. 41.35 to determine N -
p. 768 (Prob. 43.3) The problem statement should be modified to read: "Repeat Example 43.2, but
determine..."
p. 768 (Prob. 43.4) The logK} for cadmium is-14. 1.
p. 768 (Prob. 43.!) As mentioned in Prob. 43.2, the N for copper is 6xl04 L kg"1.
p. 782 (Prob. 44.1) The distances for this problem are in kilometers, not meters.
A.2
(2) TEXT ERRATA
p. 75 (l:q. 4.31) The equation should be
p. 183 Out after 1:q. 10.M): " ...the effect ar c1ecay reduces the area. .."
p. 186 Oust after l:q. 10.29): "A simulation using Eq. 10.29..."
p. 186 Oust after l:q. 10.41): "Prom Eq. 10.40. it abould. .."
p. 191 (Fi&. Ell.l): Numbers on ordinate are incorrect. Each division represents 0.1, so axis goes from
Oto 1.S.
p. 255 (uuumbered equation at top or pap): 2.s 1-o.06• should be 2.51 -a,_
p. 262 (l:q. 15.3): The second integral should be modified to read:
r r,
q.
m{,c(t-0)] +
cT,+TJ
q,
.
,r+
,c(_t-0-T.)
s r. s+r. r
Q. = - - - - - - - - - - - - = - - - - - -1= d t
T.+T1
p. 291 (£sample 16.7; last equation OD pap): Note the x bu been dropped from the last tenn.
p. 31' (Eumple 17.S; last equation OD paae): Note the first (20Y bas been changed to (20).
p. 358 (Table 19.l): The units afCBOD should be g/capita/d, not m3/capita/d.
p. 439 (Eumple 2'.l): The next to last equation in Example 24.2 should be:
p. '70 (Eumple 25.3): The initial guess in the secooc1 equation should be soo_. not csoo_.
A.3
p. 619 (Eumple 33.3): The residence time used to generate the results from Example 33.3 was actually
I 000 d. rather than the 30 d listed in the problem statement.
p. 686 (Eumple 39.1): The value for [C0/1 should be 4x10·1 M.
p. 704 (Eumple 40.1): v = 191. 72 m yr·•
p. 711 (Eumple 40.l): remove superscript "6" from denominator of equation for v,. Also note that the
quantity F&' = 1 - F,'.
A4

Solutions Manual For Surface Water - Quality Modeling

Uploaded by

Copyright:

Available Formats

Solutions Manual For Surface Water - Quality Modeling

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Solutions Manual For Surface Water - Quality Modeling

Uploaded by

Copyright:

Available Formats

Solutions Manual for

The McGraw -Hill Companies, Inc.

Solutions Manual for

Copyright© 1997 by The McGraw-Hill Companies, Inc. Al rights reserved.

In addition, a number of errors have come to my attention during classroom testing. I

Steve Chapra November 3, 1996

Civil, Environmental and Architectural Engineering

Phone: (303) 492-7573

(a) a = W = 6,950 mta . 1«>9 mg . kml = 868.751cml

(c) % reduction= 6,950-4,343.75 . l()(J9/4 = 37_5%

The waste flow can be converted to the proper units,

and the total flow calculated as

Q=Q, +Qw = 2.831+0.438= 3.269 ems

(b) c= Q,c, +Qwcw = 2.831 cm,(0.2 mg/L)+0.438 CDll(2 mg/L) = _ mg

c= V1c1 +V2 c2 = 3.78S4 L(0.2S0 mg/L)+2 L(2 mg/L) =O.SSSmg

m= cV = 0.85S mg(S.78S4 L) 8 = 4.9464x 10-3 g

Q,.(O.S g/L)+Q8 (0.0S g/L)=4 CIDl(O. l g/L)

Q,. +Q8 = 4 ems

Q,.(O.S g/ L)+ (4-Q,. )CO.OS g/ L) = 4 cms(O.l g/L)

which can be solved for ·

QA = 0.2 =0.444 ems

Q8 = 4 - 0. 444 = 3. SS6 ems

(a) c= 10 ml.(100 mg/L) =3.33 mg

(b) m= 3.33 mg·300 mL• lL · g = Ix 10- 3 g

1.10 The following balance should hold

which can be solwd for

Q,. = 1(100)-5.S -17.182 L . ml . min - 2.864 x 10◄ ml

(b) U = J = 500 g/ m2yr 200 m

5 m{lOO- 20) mg3

The following mass balance should bold

= Q,. (170) + Q 8 (820)

These two equations can be combined to yield,

which can be solved for

Q = 0.494(639-820) = 0.1 3756 ems

Q8 = 0.494- 0.13756 = 0.356 ems

(a) J= 20g = 2-g-

(b) v=J =2g/ml/d=2m

2.1 Applying equal-area differentiation

Linear regression yields a least-squares fit of

1o{- ~;) = - l6213+2.333logc

The results can be summarized as

integral integral/least-squares differential

The constant can be evaluated by applying the initial condition: c = c0 at t = 0,

Therefore. the solution is

Inc = 1. 609 - 0. 0351 ( r2 = 0.99993)

Therefore, k = 0.035 min•1.

l.4 Taking the logarithm ofEq. 2.44 yields

log(k(t)) =log(k(20))+ (T - 20) logO

log(k(t)) = -0.60474 + Q020722(T- 20)

klO =10--0.604"74 =0.248

c = 10 mL(lOO mgC/L) 6(32)mg0 _ 8_ mg0

lnp = -290.86+ 0.149985(1995) = 8.35525

p(l 995) =e 1·rn25 =4252

l.9 The rate of growth can be computed as

k = ln(4x 109)- ln(O..S x 109) 0.00693 yr·•

Thus. in another hundred years (I= 400). the iq,dati,m would be

p = 0.5 x 109 e 0·00693(400> = 8 x Irl

which can be manipulated algebraically to give.

2.11 Aa:ordi.ng to Eq. 2.46.