water quality i€¦ · i i i i i i i i i i i i i i i i 3 water quality dsm'.2-qual is a water...
TRANSCRIPT
I I I I I I I I I I I I I I I I
3WATER QUALITY
DSM'.2-QUAL is a water quality model based on the Branched Lagrangian Transport Model, a public domain code written by Harvey E. Jobson and his colleagues at the U.S. Geological Survey. The main advantages of DSM2-QUAL over the Department's existing water quality module include:
• DSM'.2-QUAL can simulate the transport of multiple constituents in the Delta. In-ternal kinetic reactions can be incorporated, and the decay or growth of each con-stituent can be numerically accounted for.
• DSM2-QUAL does not need to know the shape of the cross-section. The model needs the flow, flow area. and top width. A I the other parameters are recalculated.
Bodies of water are divided into parcels. The movements of these parcels within the channels are controlled using the flow data provided in the tide file. The size of these parcels usually do not change as Jong as they are entirely within one channel. but once they reach a junction. incoming parcels are merged and/or split to form new parcels. The model continues to update the concentration of each of the simulated constituents during each time step. Dispersion only takes place at the interface between the parcels. However. full mixing at the junctions is as umed.
The following describes the Delta Modeling Section's efforts to improve and add nev.· features to DSM2-QUAL during the past year. The development of subroutines to model nonconservative constituents and a preliminary model evaluation are also described.
Hydro File The original version of BLTM reads hydrodynamic information, provided at every time step, from an ASCII file. This had several disadvantages, as briefly described in Chapter 2. The main disadvantage is that the file can be massive for a long-term model run. The other disadvantage is that the user is forced to use a time step that matched the hydro file. To change the size of the time step, the user must develop a new hydro file with the matching time step. In DSM2-QUAL, the user is now free to choose the size of the time step, and all the values are properly averaged.
Open Water Reservoir An open water reservoir is currently modeled as a tank of water. The concentration of each of the constituents is assumed to be constant throughout the reservoir and is updated every time step. The following describes how the concentration of a conservative constituent is calculated based on different cases. To illustrate this, a reservoir is considered with two connecting junctions. Three different cases can be encountered depending on the direction of the two flow components. Updated concentrations are denoted with primes:
3-1
DWR-1044
Annuz! Prc;fess .(.or.lo tl1e S/a/e ',A/3/c.· Flesources Control Board
Case I: Flow Entering the Rescn·oir from Iloth Junctions
where
Figure 3-1. Channel Schematics for Case 1
I
Cres :;: V res "'Cres+O I *CI* D..t+Q2 *C 2 * t. t Yres
Yres = Water volume in the reservoir
Crcs :::: Constituent concentration in the reservoir
Q 1,Q2 = Flow rate entering the rcservmr
C 1.C2 == Constituent concentration assigned to QI and Q2
t-..1 = time step
Case 2: Flow Entering the Reservoir from One Junction and Leaving from the Other
Figure 3·2. C'hanneU Schematics for Case 2
(3-1)
(3-2)
Ii
I I
I I
I
I I
I I I I
I
Vrcs = v f C + (0r-Q_::)* l
C rc s = Va :s*Cw - 0 1 Cc cs* t+Q 1 *C1*.6t
\Ires
Case 3: Fl-ow Leaving the Reservoir from Both Junctions - - - - - - - - - - -
Figure 3-3. Channel Schematics for Case 3
Y rc s = Y res - (Q1+Q:,)*.6t
(3-3)
(3-4)
(3-5)
(3-6)
(3-7)
(3-8)
(3-9)
l f the constituent is nonconservative, the growth or decay of the constiruent is first determined, and a new Cres is computed. From there, the procedure for the conservative constituent is followed
Agricultural Diversions and Returns The original version of BLTM has the option of assigning flows (either inward or outward) to any grid point within a channel. However, flow estimates for agricultural drainage(Oagl and diversions (Q,h) are currently provided to DSM2 at specified junctions by the Delta Island Consumptive Use (DICU) model. Thus DSM2-QUAL was modified to include this possibility. Future model enhancements will relax this constraint. The following example illustrates the impact of agricultural diversions and returns (see Figure 3-4).
(3-10)
Water Ouo/1;y
3.3
Annual Pro:;,css .Re:>o,1 tc :1e Slale W21 , Resources Control Board
3-4
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
t.
Figure 3-4. Impact of Agricultural Drainage on Water Quality
(3-11)
Equation 3-11 is enforced in DSM2-HYDRO.
Mass Tracking Routine MTR is a very powerful and useful tool that lees the user keep track and account for all simulated water quality constituents. First, the amount (or mass, M) of each of the constituents in the Delta is calculated:
v. ... he re
M = Mres + Mchan (3-12)
M res = The initial constituent mass in the reservoirs at the beginning of the time step
Mchan = The initial constituent mass in the channels ac the beginning of the time step
Then the mass of each of the constituents inJected by all the incoming flows in one time step 1s calculated. This calculation includes all the rim flows, agricultural returns, and the flood tide:
where
M,n = The constituent mass injected by incoming flows
Mnm = The constituent mass injected by rim flows
Mag = The constituent mass injected by agricultural rerums
Mn000 = The constituent mass injected by the flood tide
(3-13)
Next, the mass of each of the constituents released by outgoing flows in one time step is calculated. This calculation includes all the exports. channel diversions. and the ebb tide:
(3-14)
where
I I
I I I I I
I
I I
I I I I I I
I
\!,, .. ;, = The con:-.t1tucnt rna s remo\·ed by ourgomg now,;,
\1c,ppn= Thl' ·011 tit112nt m.l\ rcmon:d by L',;ports
l'v(-11 = The constituent mas removed by channel diversions
:'\1cbh = The con lituent ·rna. s removed by the ebb tide
In addition to tht: physical transport of the consrituents. there can be kinetic reactions among the noncon ervative constituents. resulting in internal growth or decay. For each of the nonconservative constHuents. lhe amount of growth (M,.rowth) or decay (Mdec 3v) is calculated for the entire Delta.
The amount of constituents at the end of the t,ime-stcp can then be determined using the above components·
(3- l 5)
The above value can be compared against the calculated amount of constituents left in the Della. See Equation 3-12.
I
MI I
= i\\c, + Mchar. (3-16)
The results of Equation (3-15) should be exactly equal to Equation (3-16). If all the calculations abo\·e are done corTecrly, and the results differ. the reason can be a11ributed to one of two factors.
• Incorrect formulation. Mass is being lost or gained without being properly ac-counted for -an rnd1cauon that there is probably a naw (which needs 10 be fixed)in the program.
• Round off error. Tens of thousands of calculations are needed to obtain the results of equal ions (3-15) and (3- l 6 ). Some of the parameters in the above equations are the results of DSM2-HYDRO simulations, which in cum require hundreds of thou-sands of calculations. Computers can store numbers with a limited precision. Theloss of precision may be insignificant for a given number but may in fact become significant after numerous accumulations.
One use of the M T R is to check for the problems just described. Many problems have been detected and corrected using this routine. MTR aided in the development of the reservoir routine and accounting for agricultural drainages and channel diversions. Due to the very complex network representing the Delta and other complicated problems such as flow reversals, it would be very easy to overlook all the possible scenarios and events.
MTR detected a flaw in the original BLTM formulation. The problem was in a part of the code where it was combining two small parcels of water into one. Under certain condi-tions, the smallest parcel in a channel simply disappeared. Without the aid of MTR, it would have been almost impossible to detect this problem.
The most imponant application of MTR is probably its use as an investigative tool. In a hypothetical study, a dye can be injected in a certain location. and MTR can monitor how. the dye is spreading. For every time step, MTR reports (in terms of percentages) the amount of the dye in the Delta channels and reservoirs, the amount which has left the Delta from each outlet location, and, in case of nonconservative constituents, the amount of growth or decay. This information can help evaluate the impact of various sources on the
... · - · · · · · - -.... __ .. , ......... _____ , , , , __ .. _ , , _____ , _ , , ____ .. _______ _ - - - - - - - - · - · - - - - - - · · · · " " " · · · · · - -
W2!er Oua!./ ..
3-5
Annu2! Progress Repon 10 111e S;a;e Water Resourees Cor.:roi Boarc'
3-6
\\ ;Her qu:dHy at :1 p::inicular location. M T R 1s .1.lso robust enough o !ha\ it can be e;:\1ly modified to focus on a particular location and extract all the desired information ln CS.\cr1ce. M T R c.111 funcuon like a particle tracking model, but it work!; on ;.i homogeneous ,-o lu ! ion rather than rnd'i \'\Cual particles.
Round-Off Error on Continuity at a Junction As mentioned before. hydrodynamic mformation is averaged over a specified 11rne interval inside DSM2-HYDRO. This model enforces continuity al every junction. that 1s, the sum of the flows entering the junction must equal the sum of the flows lcavrng the junction. Due to the extra calculations performed during the averaging and the limited precision. there can be small imbalances. thus anificially creating or removing mass from the system. The amount of this imbalance may seem infinitesimally small; however, it can add up to a significant amount when accumulated over all the junctions and over time. The amount of this error can po1enrially grow larger with larger time intervals used in the tide file This problem \,\·as detected using :v1TR.
The following describes a possible remedy, which is currently implemented m DS'.\1-Q U A L Figure 3-5 1Hus1rates a Junction with three connecting channels. The nows Q 1, Q2, and Q3 are read from the hydro file.
i a, - - - - - - - - - - - - - - - - - - -
Figure 3-5. Schematics of a Typical Junction Before Flow Corrections
Contmmty requires that:
( 3- I 7)
However. due to possible round off error, in general there is a small imbalance (E). The idea is to distribule this imbalance to the connecting channels. One idea is simply to divide this error equally among the three channels. However, this simple division is not recom-mended since a nonz.ero now can result in a channel with a barrier. A superior approach is ro divide this imbalance in proponion to the absolute magnitude of the flows in the three channels:
I I
I I
I
I
I I
I I
I I
I
= Q 1 + Q - Q ,
= l(.)11 + IQ:1 + !Q:11
where 1he .superscript C indicate\ the corrected value
( _;-1 s·i n-191
n-2oi
< 3-21)
(3-22)
The following example illu<:.1ra1es the applicauon of the correcuon techniqut: dc\cnhcd above, assuming Q 1= I 000. ct\. Q2=2000 cfs. anc.J Q,=2999 99904 cfs.
Error m continuity from Equation 3- \ 8:
E = 1000. + 2000. - 2999.99994 = 0.00006 cfs
Sum of :ill the flows from Equ:ition 3-19:
I Q = 1000. + 2000. + 2999.99994 = 5999 99994 cf\
Thus the corrected values arc: C
Q 1 = 1000. - OJJ0006 "' 1000./5999.99994 = 999.99999 cfs
( :
Q2 = 2000. -0.00006 * 2000./5999.99994 = 1999.99998 cfs
C
Q3 = 2999.99994 + 0.00006 * 3000./5999 99994 = 2999.99997 cfs
After correction:
Therefore, continuity is exactly satisfied. It can be shown that in case of a barrier. the magnitude of the correction would also be zero, thus maintaining the status of the barrier.
Nonconservative Kinetics New subroutines for modeling nonconservative constituents in the Delta have been completed. These routines are structured in modular form so that they are easy to understand and can be extended to simulate additional constituents with minimum change in existing routines, if such needs arise in the future. Flexibility has been built into the model so that any combination (one, a few, or all) of the variable.scan be modeled as suited co the needs of the user. The mode!, in its present form, can simulate the following variables:
Wa1er Duality
- - - - - · -·---------------·---···-··--·---·--··-·· -·--·--·····--·-·-···---·-·-····--·--·-···---·---·-······· .. ···---·-··· -·-·--·-·-·------- ----····--··--··-"-·--
3-7
Annual Progress Reoor! to the S:ate l'ia;e, Resources Cori/rof Boi3{d
3-8
I.
1,
4.
Temperature
B1ochem1cal oxygen demand (BOD)
D1c,,olved oxygen
Organic nitrogen
:'i. Ammonia nitrogen
6. 1'itnte nitrogen
7 Nitrate nitrogen
8. Organic phospliorus
9. Dissolved (inorganic) phosphorus
10. Algae (phytoplankton)
11. Arbitrary conservative or nonconservative constituent.
Refer to the Fifteenth Annual Progress Repoli, l 994, for a description of the mathematical relationships that describe the constituent reactions and interactions. The interactiom among water quality constituents are shown in Figure 3-6. A few important rela1Jonsh1p that were not detailed in the 1994 report are described below. Subroutines are described briefly later 1n this section.
Dissolved Oxygen Saturated Concentration The solubility of dissolved oxygen in water decreases with increasing temperature and salinity. Among about half a dozen or so expressions for DO saturation concentration as a function of temperature reported in the literature, the following algorithm recommended by American Public H alth Association ( 1985) and rated the best by Bowie et al. ( 1985), was coded in the model:
In 0, f = - l 39.344 l l + l .575701*105 - 6.642308* l 07 + 1.24380* 10 10 - 8.621949* 10 1 1 (3-23) T T2 y J T4
where 0, f is the "freshwater" DO saturation concentration in mg/Lat I atm and T i s temperature in °K. The effect of salinity on DO saturation is incorporated as shown below· (APHA 1985):
In 0.-s = ln O sf - S* r .7674* 10-- 2 - 1.0754* 10 + 2.1407* l 03]T T2
where 0 55 is the "saline water" DO saturation concentration in mg/Lat 1 atm and S is salinity in ppt. The effect of barometric pressure on DO saturation is to increase the saturation value and is expressed as:
0, = 0 p [ 1-(P wv! P )] ( J - cpP) ]ss ( 1-Pw v )( I --$)
(3-24)
(3-25)
where 0 5 is the DO saturation at pressure P in mg/L, P i s the nonstandard pressure in aLm. Pw v is the partial pressure of water vapor in atm calculated from:
- - - - - - - - - - - - - - - - - - - · - � - - - - - · - - - · - · · · · - - - - - · - · · - · · - · · -
I I
I I I I I I I I I
I
I
I I
I
I I
I
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - = - : . . ' . ' . Va'..'..'.t e'...'...r ..:::0 .::::u e::..::11:..,
ln Pv.-,· =:c 11.'b57 l - 3840.70 - 21696 I T T2
= 0.000975 - 1,.426*JO--"""t + 6.436*JO-R*t 2
= temperature in ° C = T - 273.150
Atmospheric Reaeration
(3- 25a)
(3-25b) (3-25c)
Reaeratton. a process by which oxygen is exchanged between the atmosphere and a water body. 1s one of the main <;ourcc.s of o-xygen in aquatic systems. The transfer rate of oxygen from air to water may be represented by:
- - - - - - - - - - - - -
Heat exchange
Atmospheric reaeration
Air-water interface I
,, ' /
Organic-N DI
' \ . I B � s
' s ,, i--. BS NH3 L CBOD -
' \ . V E
+ B D ' / -, N02 0 Organic-P .__
' \ . . J y ' -
t B G B ,, ' 'E / '
N Dissolved-P --- N03 '--
R I p BS rG G I A Algae R ;r- , .
' \ .
s s s
TRANSFORt.iiA TION PROCESSES: e -= Bacterial decay G = Growth
P = Photosynthesis S = Settling
BS= Bertthic source
R = Respiration BO= Benthic demand
Note: Rates of mass transfer shown by - - - are functions of temperature.
8
Figure 3-6. Interaction Among Water Quality Constituents in the Model
D
3-9
Annual Progress Re;icrt to 1/le S!?.'f Waler Resources Control Board
dlill = h:: (0, - 0) dt n-26i
\\ '1,cr·c 0 :rnd O arc the oxygen concen1rntions at saiuration and oxygen concentrations of the water bcidy, respectively, and k2 is the reaeration coefficient. The oxygen tran,fc·r coel'f1cicnt in natural waters depends on (Thomann and Mueller, 1987 and Gromiec, 1989):
• temperature
• internal mixing and turbulence
• mixing due to wind
• waterfalls, dams, rapids
• surface active reagents
Numerous equations are available for predicting reaeration coefficients, giving a wide range of predicted values for specific hydraulic cond,itions (Rathbun, 1977). Rev1ew of predictive models for reaeration coefficien.1 can be found in Bowie et al. ( 1985). Rathbun (1977) and Gromiec (1989). For tidal rivers and estuaries, one of the most widely applied models is the O'Connor-Dobhins equation. This equation, adapted for the present wacer quality model, is described below. A more derailed discussion of alternative reaeration rate fom1ulations, including a list of references. i provided by Rajbhandari { 1995)
0 ' Conner and Dobbins' ( 1956) equations we re based on their analysis of reaerat ion. mechanisms that considered the rate of surface renewal through internal turbulence. They recommended the following equation for the reaeration coefficient for moderately deep todeep channels (approximately between I ft 10 30 ft. deep) and veloc1ues hetween O.S ft/s and I .6 ft/s:
where D m is the molecular diffusion coefficient. L 2f f The above equation is often expressed as:
k2 (20° C ) = 12.9 u o.s d-1.5
0-27)
(3-28)
where u is the mean velocity (ft/s) and dis the average stream depth (ft). The effect oftemperature on the reaeration coefficient is represented by:
k2 (t ° C ) = k2 (20 ° C ) ( 0 ) (I - 20) (3-29)
The numerical value of e depends on the mixing conditions of the water body. Reponed values range from 1.008 to 1.047 (Bowie et a!, 1985). In practice a value of 1.024 is often used (Thomann and Mueller, 1987, Gromiec, 1989 etc.) which has been adopted for the model.
- - · - - · · · · - - · - - · - · - - - - - · - - - . , · · · - · - - - - - - - - - - - - · - - - - - - - · - - -
3-10
-I I I I I I
I I I I I
I
I I
I I I
Subroutines F igu re . - 7 rs pre�t:nted to h igh l i gh t the main processes i nvo l ved in the d y n a m i c s imula t ion
o f w a t e r q:..iality i:i the Delta. A b n e J descr ip t ion o f the n e w l y deve loped subrout ines is
p r o v i d e d be low. Figures are inc luded f o r a f e w rouunes to i l lus t ra te 1the main processes.
Hydrodynamic. Simulation using an appropriate now model
Solves the continuity and momentum equations to describe the hydrodynamics of the syscem. � files describing the nerworik geometry and che hydrologic infonnaoon serve as
input.
I + Simulatfon of Water QualHy Variables
Discharge and water surface elevation a.'ld wid1h from the hydrodynamic simulation are input to the transporl model. Meteorology data. and the water qu.a1iry flux a1 the ooundaric:s are input to the l'llOdcl. The Lagrangian equations are solvo:i I for all water quality variables. The soui:tt and sink terms arc 1 evaluated. lnreractions among constiruenis are accounted for
'I at aH times.
.
Processes:
l . Advccrion '2. Dispc�ion 13. Decay and growth
+ Model Ou tpu t
Spacial and temporal values o f constituent concentrations
Posrproccssing o f simulaaon �ul ts : longitudinal profi.les
and diurnal ploo L ______ _. Figure 3-7. Flow Chart for the Dynamic Simulation
of Water Quality Variables In the Delta
'Nater Quall/
3-11
Ar: J " ?rog ess Re;;or. to tne Sti''!? Ware: Resources Con;rol Boa.'d
KI:\TT!CS
CALSCSK
HEAT
3-12
The main (unction of this subroutine is to update constituent con-centrations at each time step. It is called by BLTM's parcel tr:icking subroutine (ROUTE) for every parcel at each time step. The flowchart (Figure 3-8) shows the logic of the subroutine. It has also been extended io simulate kinetics in 'reservoirs' in the Delta
This subroutine builds a source/sink matnx for each constituent by calling each constituent subroutine. Individual constituent routines are listed in Figure 3-9. Flowcharts for three principal constituents (DO, algae. temperature) are presented in Figures 3-10 through 3-12 for illustrative purposes.
Subroutine HEAT has been adapted from the USEPA QUAL2E model (Brown and Barnwell, 1987) wtCh some restructuring. This version computes the net shon wave solar radiation
KINETIC
Call LOC Groups c hanne Is ac::on:ling to I ocari on
Sci limit on change in concenrrarion (C) per o.me step and per iceranon
Reaction step (dr-react) = rime remaining for move nt (dt-rem)
Call CALSCSK Conscruccs soora:/sinlr:: matrix
for e.ach consril.ueru
Predicr consrinxnc concc:n rratioru using Euler e:stimate
Call CALSCSK Updates soun::c/sink maou
Update o:mtirueru coocc:mratioo s using M odificd E.u.ler method
V
Figure 3-8. Flow Chart for Subroutine KINETIC
I I I I I I
I I I I I I I I
I I
I
I
I I
I
MET
R A T E
N
and the long wa\'e atmospheric radiation once every hour ( Figure _,-1 3 :i.
A call qa1ernent to thi. subroutine (for every hour) ,vas added lo the main B L T M routine. Meteorology data read by this subroutine are used in computation of heat components as shown in Figure 3-13.
Physic;:ii. chemical and biological rate coefficients arc read 111 this subroutine. Some of these coefficients are constant throughout the system; some vary by loca-tion; and some are temperature-dependent. A list of these coefficients, including their ranges, is provided in Table 3-1 (Brown and I3armvcll, 1987). Tempera-ture coefficients used in the model are listed in Table 3-2.
y
y
Call C A L S C S K
Count number of times !his lcop is emered
Update C again using Modified Euler
y
J1 may be necessary 10 reduce reaction time srcp
Reduce dt -n::ro by dt-react
Goto I Return
Figure 3-8 (continued)
I/later Oua/11
3-13
An, -u" ' ? , o g r e s s Rr:: ;c l o !he S1a:c ..'.:_ c,_R_e_so_u_rce_s_Co_n_lr_ol_B_oa_rd _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
L O C
3-14
This subroutine groups channels by their number or location. The grouping allows the input of spatially varying rate coeffi-cients in the model.
CALSCSK
. lni tialize source/sin I:: array
Call CAl.ARB
Call CALBOD
i . . . - - - - - . Call CALORGN
Call CALNH3
Call CALN02
Call CALN03
1 & - - - - - 1 Call CA LOR GP
Call CALP04
CaU C A l A L G
Call CAL TEMP
RelUl11
Figure 3-9. Flow Chart for Subroutine CALSCSK
...
I I I I
I I I I I I I I I I
I
I
I
I
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _:.•_:.:':,::..:,:.:_'=·:...:· O:::.u ac:.l1 r,
Compute 0 0 sarurati0r1
Include effect of pressure on DO sarurarion
Compute reaerarion rate
Com e benthic oxygen demand rau
I C o m p " ' " ' ' " ' ' ° ITT
y Include effea of saliniry on 0 0
s arura.tion
NOie: 0 0 used in oxidanon of ammonia & 1 nitrite nicrogens and algal
n::sp'irati on and DO produced by
phoms ynthesis arc computed in the
1vc subroutines.
C orrec:r r.u.es for tempe:ra.turc
l ,, l._ ___ e n r r n
- · · · - - - - - .. - - - - -.. ·-·- --·-·- --------.. -·.
Figure 3-1 O. Flow Chart for Subroutine CALDO
3-15
Annu2· Fr:;9rr.ss Re;;o,1 le: ;; S:,,;!J Water F.cso'.lrces Control Board
3-16
CALALG
Ca..lcu late ni1JOgen and phosphorus factor
CaJcu Lau: al gal growth limi tarion f aa or for lighL Account for self shading effects
CompUTe algal gro\Wth using light faaor and limiting num nt concept
Compute loss due to respiration and loss due to senlin,•
Compute rue of change in algal biomass
Compute the effea of algae on so,iro:/rin.k: terms of nitrogen and
phosphorus series and DO
Return
Figure 3-11. Flow Chart for Subroutine CALALG (algae)
I
I
I I
I
I
I I
I
I I I I I I
I I
I
Conven wacer temperarure from Celsius ro English units
Compute Jong wave radiation eminc::d by Wa..!Cf
Compute evaporative hear transfer
Compute conductive heat o-ansfer
Compute net heat flux from all sources including absorbed radiation
Rate of change in tc:mperarurc = net radiarion/(depth *density* heat capacity)
- R - e t u r n
_J Figure 3-12. Flow Chart for Subroutine CALTEMP
Water Ouati!v
3-17
Annual Progress r;,2por: ;c ;nf 5:a:e \\'a:er Resources Centro, Soard
3-18
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - - - - - - - - - - - · -----.. - · - - - - - - - - - - -
HEAT
y
Incrcrnenr the variables th at de fine I he beginning and the end of each hour
Compuie vapor prcssun:s. dew point and dampening effca of clouds
Compu1e amount of clear sky. solar rarliacion, and altitude of the sun
Compute absorpnon and scattering due 10 aonosphaic conditions
Com pu I e rtflecti viry coe. ffi ci en t
Compucc acmosphtric iransmission
Compute hon wave solar radia.ion absorbed in 1 hour
Compu,c long wave aimosphenc rad! a aon a bso rt>ed in I hour
Rerum
Calculate seasonal and d.a..!.ly pos,11on of the sun rdacive to
the location on eanh
Calcul.lle !he siandard time of sunrise and sunset
Mc:t
n:ui meteorology input:
cloud cover atr temperature wetbulb tempcrarure wind speed atmospheric pressure
Rerum
Figure 3-13. Flow Chart for Subroutine HEAT
I I
I I I I I I I I I
I I
I I
I I
I
I
,
I
Water Oua/1!
Table 3-1. Typical Ranges for Reaction Coefficients
Global coefficent values. Variable names shown are the same as used in lhe FORTRAN code.
alph(5) alph(6) prcfn
alph(7) alph(l) alph(2) alph(3) alph(4)
oxygen used in conversion of ammonia to nitrite oxygen used in conversion of nitrite to nitrate algal preference factor for ammonia
chlorophyll-a to biomass ratio fraction of algal biomass which is nitrogen fraction of algal biomass which is phosphorus oxygen produced in photosynthesis oxygen consumed with respiration
k.1ght_half half saturation constant for Light knic_half half saturation constant for nitrogen kpho_half half saturation constant for phosphorus
xlamO xlaml x1am2
non-algal light extinction coefficient linear algal self shading coefficient nonlinear algal self shading coefficient
3.0 - 4.0 1.0-1.14 0 - l.O
10 - 100 0.07 - 0.09 0.01 - 0.02 1.4 - 4.8 1.6-2.3
0.02 - 0.10 0.01 - 0.30 0.001 - 0.05
variable 0.002 - 0.02 0.0165
Location dependent coefficient values. Variable names shown below are the same as used in the text of the 1994 Annual Report. These coefficients are described by the array rcoef in the FORTRAN code.
kl BOD decay rate 0.02 - 3.4 k3 BOD settling rate -0.36 - 0.36
k4 benthic source rate for DO variable
karb decay rate for arbitrary non-cons. constituent variable s6 settling rate for arbitrary non-cons. constituent variable s7 benthic source rate for arbitrary non-cons. constituent variable
rnumax maximum algal growth rate 1.0 - 3.0 resp algal respiration rate 0.05 - 0.5 sl algal settling rate 0.5 - 6.0
k.n-org organic nitrogen decay rate 0.02 - 0.4 s4 organic nitrogen settling rate 0.001 - O. l
kn ammonia decay rate 0.1 ._ 1.0 s3 benthic source rate for ammonia variable
kni nitrite decay rate 0.2 - 2.0
kp-org organic phosphorus decay rate 0.01 - 0.7 s5 organic phosphorus settling rate 0.001-0.l s2 benthic source rate for dissolved P variable
Note: rates are in units per day except when specified.
3-19
.t•r;ua Pr :;· 55 ;. .. ::.:>or: to :rie S:are Ware, Resources Conrrol Bozrd
Table 3-2. Temperature Coefficients for Reaction Rates
Consti(Ue:H Reaction type Temperature Variable coefficient (FORTRAN)
--------------------------------------------------------------------------------------------------BOD decay 1.047 thet(l)
settling 1.024 thel(2)
lX) reaerat:ion 1.024 thel(3) benthic demand t.060 thet(4)
ORGANIC-N decay 1.047 thet(5) settling 1.024 thet(6)
AlvfMONIA-N decay 1.083 thet(7) benthic source 1.074 thet(8)
NITRITE-N decay 1.047 thet(9)
ORGANIC-P decay 1.047 thet(lO) settling 1.024 thet(l l)
DISSOLVED-P benthic source 1.074 thet(12)
ALGAE growth 1.047 thet(l3) respiration 1.047 thet(l4) settling 1.024 thec(15)
3-20
I I
I I I I I
I
I I
I I
I I
I I I I I
I I
_ _ _ _ _ ., _ _ _ _ _ _ _ _ - - - - - · - - - - - - - - - - - - - - - - - _ _ _ _ _ _ _ _ Wa:er0:12:.;v
Preliminary Mod,e/ Evaluation 1\ pre li1rnnary e, alu:iti,m of the exienckd BLT ! model wa. conducted by applying the model to the Delia. 1l1i,; section describes the earlier stages of calibration and verification (s1rc--speciftc) in the region including the Stockton Ship Channel. Figure 3-14 shows water quality monitonng stations in the San Joaquin River in the vionity of Stockton, both U[JStream and downstream of the city\ waste water discharge.
The un·ey periods September 20, 1988 and October 12, 1988 were chosen for model calibration and verification because of the unique hydrodynamic conditions prevailing in these periods and availability of water quality data. The unique hydrodynamic conditions were clue in part to placement of a rock barrier at the head of Old River during the period September 22-28. The barrier causes more water to Oow downstream in the San Joaquin R1\'er toward Stockton arid less mto Old River toward Clifton Cour1. Modeling of the tvio
I I 1.
• Stockton MUD stations
Channel numbers closest
to stations are shown
N
l
I i
i
I
- - - - - - - · - .. ·--·· -·---- ··-· - · · - - - - - - - · · - - - - - · ... · - - - - - - - · - · - - - · ···-·-'
Figure 3-14. Monitoring Stations in San Joaquin River Near Stockton
3-21
Ar-.,; i-'rogress Ae · :·· :o ,-.;; S:,;:e V/.;;er Resources Control Board
3-22
,ccl\,Lrin pro, 1dc 1ntcres1ing. altem:iuvec; of local hydrology and hydraulic h.'h..1\ mr :t'>
11cll :h L"iun_::cs ba ed on ,bily flow configur.111ons .:ind water quality
Model Input
Hydrology. A,·erage daily tlows reponed in DWR DayOow Data Summ:iry for 1988 were u)cd for the San Joaquin. Sacramento and Mokelumne River inOows and ,he State Water Project (SWP), Central Valley Project (CVP) and the Contra Costa Water District (CCWD) '"irhdrav.als. Efnuent tlows from the City of Stockton waste water treatment plant were 0btained from the monthly laboratory data file of the Stockton Municipal Utilities District (Huber, pers comm., 1995). Clifton Court intake gate (opening and closing) schedules were obwined from the monthly operation report (DWR, 1988). Agricultural withdrawals and return Oows. which are also input to the model, were based on the Department of Water Resource,· monthly estimates. Hydrologic and hydrodynamic data for both ,irnulauon periods are presented in Table 3-3.
ivater Quality. Smee water quali1y grab sample data were collected only at a frequency of one or two 5amples per month near model boundaries in the San Joaquin and Sacramento Rin:r,. :.ind a1 Martinez. observations closest to the simulated dates were used. These data were extracted from the Water Quality Surveillance Program report (DWR, 1990). Data for Freepon on the Sacramento River and for the Mokelumne River were derived from Water Resource, Data ( USGS. 1989. 1990). Stockton effluent data. available at daily or weekly intcn·als. were al<;o input 10 the model (Huber. pers. comm., 1995). Chlorophyll-a and orthophosphate data for the months of September and October were available from efflucnc monitoring on a monthly bas1 only for 1989: so these values were med for the presern simulations All input data, including some a1 selected locations in the Delta. are listed in Table, -3 through 3-7. Figure 3-15 shows the locations of these monitoring stattons in the Delw
To allow representation of diurnal values, whenever possible, hourly averaged concentra-liom of DO. T D S and temperature were assigned for the boundary at Martinez. Since such detailed data were not available at Vernalis. hourly averaged values of DO, TDS, and temperature available from {he nearby station at Mossdale were used as approximate representa11ons of these quantities for Vernal is. Hourly averaged data provided for the i mu lauon period are shown in Figures 3- 16 and 3-1 7. Figure 3-1 8 is included to show the trend of field data at Rough and Ready Island, near Stockton. No such approximations could be made for the Sacramento River, since no continuously monitored stations were situated near this boundary. Qualities of all eleven constituents included in the model, as with the eight constituents applied at other boundaries, were kept constant for the effective simulation period (25 hours).
Agricultural return water quality was based on an estimate using l 964 data (DWR, 1967). Flow weighted averages of nitrogen and phosphorous concentrations were computed using data from thirteen sampling stations in the Delta. Estimates for DO, temperature and T D S concentrations were obtained by averaging the values corresponding to three sub-regions in the Delta for the particular months simulated (MWQI Data Request, 1995 ). These subregions were classified according to the distribution of dissolved organic carbon in the Delta. Since data on chlorophyll-a was not available, it was set to a value corresponding to an average for the Sacramento and San Joaquin rivers during the period of simulation.
Hourly meteorological conditions for September 20 and October 18 of 1988 were genera1ed from climatological data at Sacramento Executive Airport (NOAA, 1988)
I I I I
I
I
I I
...
I
I
I
I I
I I
SELECTED MONITORING SITES CJ · s.:,.,,_,., R @O,_,·, Ut.f'O"Q C7 · S6A JGacµn RN0t @ 1.Aoea:!a!, llnd\l• C 1 o • S.a.n .Joal:µI\ Rrvr.- n a a t ' \ / « n.aJ1 t 06 • SuiaJo a.., Bull , , Pt ..,.., Mar,,,,., 07 · Gnzz,y El.-, @ Dc,l)hln .,,., Suiw o SJ 02< · S-•m•>t"O A b..o- R.o V"'a BM9o Pe · S&n " A @ llu::lu•y C . : , , •P10A. 1-idd• [email protected] P'12 • C ) d R@Tracy Rd B,ldoo
Boundary tid6
, I . MajOI inflow at mooel bour.oary
l I" I
j
I ---- ----Figure 3- 5. Selected Monitoring Stations In
the Sacramento-San Joaquin Delta
Scenario 1: Calibration of The Model
Hydrodynamics. This scenario represents conditions corresponding to September 20, l 988, when there was no barrier at the head of Old River. The actual tide at the Martinez boundary was adjusted so as to repeat each 25 hours. With this tide imposed at the seaward boundary and using the freshwater inflows presented in Table 3-3, the D W R D S M hydrodynamics model was run until a state of hydrodynamic equilibrium was achieved.
Upon examination, che computed stages compared reasonably well at locations near Stockton. The flow split (daily mean) at the head of Old River was close to that actually observed dunng <he rime period simulated. Model results showed 83 percent now into Old River with only 17 percent (260 cfs) passing downstream in the San Joaquin River (Figure 3- l 9). Figures 3-20 and 3-2 l are included to show the variation of channel widths and depths in the river.
Wiler Ouai.'!
3-23
,:..nnua1 •01ogress Report t:J the State v'ialer Resources Cor,:rci Board
3-24
Table 3-3. Hydrology Used In Model Callbration and Verification
Inflow/Export
Sacramento River San Joaquin Rjver Mokel umne River State Water Project expon CVP export Contra Costa Canal expon Stockton City Effluent
Consumptive Use
Net Delta Outflow (compULed)
Discharge (cfs) Sep 20,1988 11700 1610 29 2383 4610 221 31.6
2048
4108
Discharge (cfs) Oct 12,1988 11100 1050 30 4388 4407 214 23.6
1260
1934
Water Quality. Site-specific calibration of the water quality model was focused on the region of San Joaquin River near the Stockton Ship Channel where relatively more water quality data were available. Using this region for model calibration also provided some unique opportunities to examine the local effects of Stockton's wastewater effluent and the unique hydrodynamic conditions due to the placement of the Old River barrier. Initially, all rate coefficients were set to intermediate values in the ranges suggested in the QUAL2E manual(Brown and Barnwell, l 987), or they were set based on previous modeling expenences in the Delta and other similar estuarine systems (Rajbhandan and Orlob, 1990. Smith D.1.,1988, Hydroqual, 1985). Detailed discussions on the sources of data. ranges and their reliabilities are given in Bowie et al (1985).
Calibration started with comparison of diurnal variations of computed temperature wirh observed (hourly averaged) data at Rough and Ready Island near Stockton (Channel 20). Adjustments of dusc attenuation and evaporation coefficients were made until a reasonable agreement between the temperature patterns was obtained. These coefficients affect ware.r tcmpera1ures by increasing or decreasing net shortwave solar radiation input and heat energy losses due to evaporation, respectively. The selected values were within the range suggested in the literature (Brown and Barnwell, 1987, TVA, 1972).
Dissolved oxygen is the most important parameter of interest in the current study, consequently it was the primary constituent to be calibrated. DO was generally observed to be at depressed levels, well below saturation, in the Stockton Ship Channel near the wasce water outfall (DWR, 1990).
Calibration for DO involves a trade off between the rates of reaeration and benthic oxygen demand (SOD). Other processes such as photosynthetic oxygen production and chemical-b1ochemical oxidation also affect oxygen balance but have comparatively minor influences on DO balance in this case. Either or both of these dominant processes (reaeration and benihic demand) may be adJusted to achieve acceptable calibration. The O'Connor-Dob-bins reaeration equa11on was coded in the model (Thomann and Mueller, 1987). It computes reaeration rates based on instantaneous channel velocity and the depth of water derived from hydrodynamic simulation. A minimum rate (as a function of depth) of 3
I I I I I I I I I I I
I I I I
Wa:er Ou·a/r/y
Table 3-4. Water Quality (Field Data 1) at the Model Boundary September 988 Simulation
SacramcnlO San Joaquin Suisun Bay Mokc!urnnc Agncultural . Slockton River al River at near River at Return City
Green's L. Verna.Es Martinez Woocfuridgc Effluent
I (C3) (CIO) or (06) or Freeport Mossda.le2
I Date/Time Sep 15/0815 Scpl6/0950 Sep2l/1255 Sep 14/1120 Sep 13/1 l30f
ms 139 f hourly hourly 63 1030 900 (450-480) ( 16k-21 k)
0 0 8.0 f hourly 7.9 8.7 (97% Sat) S. l 5.3 (7-8)
Organic N 0.5 f 0.3 0.3 0.2 I .4 2.5
I NH3-N 0.2 O.Ol 0.02 0.03 0.31 14.0
N02+N03 0.09 1.30 0.37 0.09
N02 0.01 e 0.13 e 0.04 e 0.01 0.02 1.46
N03 0.08 e J.17 C 0.33 e 0.08 1.3 l.19
I Organic P 0.03 b 0.11 • 0.07 b 0.04 0.09 (estimate)
Ortho-P04 0.1 b 0.12• 0.15 b 0.02 0.4 0 0.66 a
Chlorophyll-a 0.3 19.l 0.2 0.3 10.0 14.0 a µg;1
Temperature °C 20 hourly hourly 21.5 19.6 20.0 (18-19) (18.5-19.5)
BOD 3.9 11.0
a = W/20/89 = (J)/01/88 = estimate = Free)Xlrt 09/13/88 = 09/02/88
1 Note: All units are in mg/l except when noted. Organic phosphorus was obtained. by subtracting onho-P04 from total phosphorus. Dissolved nitrite and nitrate were obtained by subdividing thetr known !Dtal into 10 and 90 percent values. Hourly data are also shown by plots (Figures 3-16, 3-17).
2 Mossdale data were used as hourly data for the model boundary at Vemalis.
3-25
A'lm;a/ Pro;;;ess Repor. lo !he S:a!e \'\iarer Resources Control Board
3-26
Table 3-5. Water Quality (Field Data 1 ) at the Model Boundary October 1988 Simulation
Sacramento San Joaquin Suisun Bay Agricultural Stockton River at River at near Martinez Return City Green's Vemalis (D6) Effluent
Landing (C3) (ClO)or or Mossdale2
Freepon (C7)
Dare/firne Oct 17/0955 Oct 18/1210 Oct 5/1300 Oct 18/1100(
I D S % f hourly hourly 807 1 0 0 4
(498-536) (J 4500-23800)
8.9 f hourly hourly 5.3 6.2
(96% Sat) (7.2-8) (7 .8-8.3)
Organic N 0.2 0.3 0.3 l.4 3.8
NH3-N 0.51 0.00 0.01 0.31 21.4
N02+N03 0.09 l.10 0.34
N02 0.01 e 0.10 e 0.03 e 0.02 0.26
N03 0.08 C 1.0 C 0.31 e 1.3 0.7
Organic P 0.03 0.11 0.18 0.09
Ortho-P04 0.l8 0.1 0.06 0.40 2.94a
Chlorophyll-a µg/J 0.9 27.6 0.1 10.0 2ga
Temperature ° C 19 hourly hourly 17 .3 19.0
(18.8-19.6) (17.7-18.7)
BOD 6.0 8.0
a = 10/25/89 = estimate = Freeport
1 Note: All uniis are in mg/l except when noted. Organic phosphorus was obi.ained by subtracling ortho-P04 from lot.al phosphorus. Dissolved nitrite and nio-ate were obi.ained by subdividing their known tot.a.I into IO and 90 percent vaJues. Hourly data are also shown by plots (Figures 3-16, 3-17).
2 Mossdale data were used as hourly data for the model boundary at Vemalis.
I I I I I
I I I I I I I I I
I I I I I
I I
Water Qua/it
Table· 3-6. Water Quality (Field Data1) at Selected Delta Locations Sep ember 1988 Slmu latlon
Old R. at Middle R. Sac R. below GrizzJy Bay Tracy Road at Union Rio Vista
Bridge Point Bridge Dolphin (D7) (P12) (PlOA) (D24)
Date/fime Sep 16/1135 Sep 15/1100 Sep 20/1420 Sep 20/1025
Tide LH LH LH LH
IDS 527* 257 b 178 ( +) 13500 (+)
I X ) 8.6 7.7 8.3 8 .1
Organic N 0.3 0.1 0.4 0.3
NH3-N 0.03 0.04 0.1 0.01
N02 (estimate) 0.12 0.02 0.03 0.04
N02+N03 1.20 0.21 0.27 0.40
N03 (estimate) 1.08 0.19 0.24 0.36
Organic P (estimate) 0.10* 0.03 b 0.03(+) 0.10
Onho-P04 0.14* 0.08 b 0.11(+) 0.18 ( +)
Chlorophyll-a mg/l 26 2.4 2.1 1. 2
Temperature ° C 20 21 20 18 °c BOD
b = 09/1/88 data = 09 /2/8 8 data
(+) = 09/6/88 data
1 Note: All units are in mg/I except when noled. Organic phosphorus was obtained by subtracting Onho-P04 from Lota! phosphorus. Dissolved nitrite and nitrate were obtained by subdividing their known tot.al imo 10 and 90 percent values.
3-27
An:1L. ' ::,·:-;ress Aeporr to the Slate Via/er Resou:ces Con:rol Board
3-28
Table 3-7. Water Quality (Field Data 1 ) at Selected Delta Locations October 1988 Simulation
Old R. at Middle R. Sac R. Grizzly Bay Tracy Road at Union below Rio at
Bridge Point Vista Bridge Dolphin (D7) (P12) (PlOA) (D24)
Date/rime Oct 1811435 Oct 17/l240 Oct 19/1325 Oct 19/l015
Tide LH LH LH LH
IDS 59l (+) 270 b 335 ( +) 13200 (+)
I X ) 8.7 7.8 8.1 7.8
Organic N 0.4 0.2 0.1 0.2
I\1-I3-N 0.00 0.07 0.07 0.00
N02 (estimate) 0.09 0.03 0.02 0.04
N02+N03 0.93 0.32 0.22 0.43
N03 (estimate) 0.84 0.29 0.20 0.39
Organic P (estimate) 0.09 0.03 0.06 (+) 0.14 (+)
Ortho-P04 0.l2 O.lO 0.07 (+) 0.12 (+)
Chlorophyll-a, µg/1 39.2 1.4 2.5 0.8
Temperature ° C 21 21 20 19
BOD
b = 10/03/88 = 10/03/88
(+) = 10/04/88
1 Note: All units are in mg/I except when note<l. Organic phosphorus was obtained by subtracting ortho-
I I I I I I I I I I I I
P04 from total phosphorus. Dissolved nitrite and nitrate were obtained by subdividing their known tot.al I into 10 and 90 percent values.
I I I
I
I I
I
I
Waier Quality
MOSSOALE-OO-DwA· 1HOUA , , f-; - - - - - - - - - · - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
21SEP19B8 010CT196& Tmo
MO$.S0"1.E-TEMP-OWR. 1HOUR 2 : 1 - - - - - - - - - - - - - - - - -
1 7 t c . · - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -21!SEP>- 070CTI'""' 1,0C,-1!168 r.,,. 210CT1968
L---·-----·-----···------------ -----·- .. ··--··----· ·····-- _______ _ _ j
Figure 3-�6. Dissolved Oxygen, Total Dissolved Solids, and Temperature at Martinez (September � 4 to October 24, 1988)
ft/depth (in feet) per day was adapted ro account for reaeration during very low velocllies during slack cide conditions.
ln calibration there was generally more latitude for adjustment in DO by varying benthic demands, although these were only varied spatially, not temporally. In areas near the effluent outfall site, SOD is expected to be higher because deposits of settleable organic solids tend to build up over time, especially due to historically poor circulation of water in the area, most notable during extended droughts. During the calibration process various values of the benthic oxygen demand based on the suggested range of 2-10.0 g/m 2 --day (Thomann, 1972) were tried.
After repeated simulacions. adjustments of parameters, and examination of DO profiles from the downstream end of channel IO (approximately 20 miles downstream of Vema\1s) to channel 3 J (approx:ima1ely 38 miles downstream of Vemalis), model results were found to be in good agreement with field data. The calibrated model uses benthic ox:ygen demands of 1.6 g/m 2 -day for channels I through 9 and 4.9 g/m 2--day from channels 10 through 20 (in the vicinily of the outfall). A uniform demand rate of 0.5 g/m 2 -day was assigned to 1he res1 of the Delta.
3-29
Annual Progress RcpOri lo the Stale Waler Resources Con/rol 3oard
l0 -
)00 - ,-
MTZ·DO-DWf\: lHOUR - - - - - - - - - - - - - - - - - -
MTZ-TE"4P-OWR tHOUR 2 1 g - - - - - - - - - - - - - - - - ·- '
11.:- - - - - - - - - - - - - - -070Cl1
r""'
l _________ _ -- ----·-·--···-------------···· - - - - - - - - - - - - - - - - - - - - - - - - - ·
3-30
Figure 3-17. Dissolved Oxygen, Total Dissolved Solids, and Temperature In the San Joaquin River at Mossdale (September 14 to October 24, 1988)
Profiles of calibrated DO and temperature are shown in Figures 3-22a and 3-23a. The time \'ariation of computed discharge at channel 13 rs shown as an inset in the DO plot.
I I
I
I I I I I I I I I I
I I
I I
I
I
STOCKCKAN·OO·OWA; lHOUA , o - - - - - - · - · - - - · · - - - - - -
Figure 3-18. Dissolved Oxygen, Total Dissolved Solids, and Temperature In the San Joaquin River at Stockton (September 14 to October 24, 1988)
Wa1er Qua/11
-
3-31
.",1, / P'c:;'(55 Roporr :c the S:a:e \\'o'.N Resources Control Boa.-c
2000
1000
0
- 1 00 O·
- 2 0 00 l 0:
0 I I l l
I -0 <ll
0 "' , I I
- - - - - - - - - - - - - - - - · · - - - - - - · · · · · · · - - - - - - - - - - -
� 9/20/88 No barrier at Head of Old R.
0 I O/l 2/88 With barrier at Head of Old R. lJ
· J11]Negatov, flow md,oa<e, flow up" ,eam towa,d Mossdale
1
:;;: :, 0 (.) I
(/")
'i Cl a, :5 C
I I I ' I
6 7 8 9 10 1, 12 13 14 15 16 17 18 19 20 21 22 23 24 25 30 31
Channel Number
Figure 3-rn. Computed Flows (tidally averaged) in the San Joaquin River
I I I I I
I
I I
I
I
I I I I I I I I I I I I I
r-····-- ---···I l
C:
1000
800
600
400
200
0
1 o " 12 13 " 1 s 16 11 , e 1; 20 21 22 23 2• 2s 30 . ,
Channel Number
Figure 3-20. Channel Widths along the San Joaquin River
40
20
0
10 \ l 1 2 1 3 H I 5 1 t? I 9 19 20 21 22 23 2 • 2S JO 31
Channel Number
Figure 3--21. Channel Depths along the- San Joaquin River
! _J
Waler Ouah/
3-33
Annua, Fro9ress Re,Jor110 the S!ale Waler ResourD:!s Control Board
3-34
0 , E
0 0
C c,,
1 D
s
0
, 0
5
0
r !
[, o a
Cal ibra t ion 9/20/88 No barner al head of Old River
Field - - m o d e l I
9 '.:! :: .c .c .c u <.J u
I 2 0 1 ' 0
Downs1,eam distance from
i·":: _('_ ··-!·:::: \_J
2.C C U 0
160
Vemal,s.
ol V> N
.c .c u u
\80
1000 ti (0.3 km)
;:; .c u
200
Figure 3-22a. Comparison of Observed and Simulated Dissolved Oxygen Profiles
!' Veril1cat10 10/12/88
Barner at head of Old River
\ F ie ld - model I
.... , ... i .,.,l\ DCU
i 'f \_,, I
'l
I 00
,:. u .c u
.... !I .... . ...
,20 140 160 180 200
Downstream distance from Vernafis. 1000 ti (0.3 km)
Figure 3-22b. Comparison of Observed and Simulated Dissolved Oxygen Profiles
J i-I
j I
1 i
I
I I I
I I I I
I I
I I I I I
I
I I
r-·-II I 25
: ,
(D a . 20 E
(I)
3: 1 5
, 0
9/20/88 I - M o d e l f i e l d 1
t\_
;: - 0 N "' ;:; <'J N N ,:: 0 0 .c .c .c .c .c u u u L) u
100 120 140 160 180 200
Downstream distance trom Vernalis, 1000 ft (0.3 km)
Figure 3-23a. Comparison of Observed and Slmulated Temperature Profiles
, - - - - - - - - - - - - · - - - · · - - - -
( . )
25
.2 (I) a.. E 20
' -
1 5
10
10/12/88 Field - - m o d e l J
i-I r r
0 N U'I ;:; N N N . c . c ..c . c J:: , : ..c: J: (.) (.)
(.) (.) (.) (.) (.) u
1.00 120 140 160 180 200
Downstream distance from Vema!is, 1000 ft (0.3 km)
Figure 3-23b. Comparison of Observed and Simulated Temperature Profiles
l-'.'2te1 Oua!1r
I
J J
, 1 j I
' I
J j I l
J
1
3-35
Annual Pro re. s Re;xr: :o :re S:a:e V.'.'?lf' Resources Ccr:·. Board
3-36
Scenario 2: Ve·.rification of The Model flydrodynamics. h\llo" 111g calibr:1t1on the model wa et up lO mntlare \Valer quality for Cktol1 ·r 1 . l 9S.S. Pnor t0 t Im date, a rock barn er wa put in place at the head of Old R l \ t T , \ in the cal'ibration nin the actual tide at the Martinez boundary for this date 1v:1s .ldju,ted ,o as Lo repeat after 25 hours. With this tide imposed a1 the seaward boundary and u,ing the .frc,hwak'r inflows presented_ in Table 3-3, the DWRDSM hydrodynamics module was run until a state· of dynamic equilibrium ·was achieved.
Th,: flow split at the Old River Head computed by the model was found to be close to that actually obser,·ed during the ame time period. Model results showed the flow split (daily mean) into Old River at its head to be 27 percent (280 cfs). with 73 percent (74 l cfs) of rhe flow passing downstream m the San Joaquin River. This represents a nearly three-fold irn:rease 1n che n t downstream now over that of September 20, 1988 simula{ion period ( ,cc F1 ure 3-19)
Hlater Quality. In this simulation calibration coefficients previously established remain unaltered. This allows us 10 examine how wc:11 the model can simulate hydrodynamic and w:1ter qualiry conditions different from chose of the calibration period, a measure of the model' r lrabil1ty a a simulation 1001.
Figure 3-22b represents the comparison of computed D O and field data for the verification run. Mode! results agree very closely with field observations over the l 8-mile reach of the ri,·er. reproducing most features of the "oxygen sag". E pecially notable is the displace-ment of rhe sag downstream due to the increase in net downstream flow (compared to rhe calihr.111on case) and the increa e in the DO sag minimum, due apparently to improved reaer;.1t1on and dispersion along the channel. A comparison of temperature profiles which look reasonable, especially considering the fact that one sec of climate data was used for the entire system, is shown in Figure 3-23b. (This study used data from Sacramento Airport Sta{ion for lack of detailed data at Stockton Station, although the latter would ha,e been more appropnare because of the focus in this region for the present evaluation).
Figure 3-24 shows the profile of nitrate-nitrogen during the two simulation periods. Ranges of observed data for the months of September and October 1988 (taken at weekly intervals) are shown for comparison. Nitrate is somewhat underesrirnated by the model, due. most likely, to uncertainties in boundary values. Future modeling efforts are expected to improve nnrate simulation capability.
Figures 3-25 and 3-26 present computed profiles of chlorophyll-a and orthophosphate of the reach for the two simularion periods. In these plots, field data at only one station (P8) were available for comparison.
Diurnal Variation in Water Quality. The diurnal variation of DO is important because i1 informs us how low DO levels can get during certain hours of the daily cycle. This variation can be significant when larger populations of algae are present in the water. Figures 3-27 through 3-33 show diurnal variations of selected quality variables at a station near Stockton (Channel 17) for the September 1988 scenario. The chlorophyll pattern shown in Figure 3-27 is typical of what would be expected on a dear day. The sinusoidal pattern seen for afternoon hours shows the influence of solar radiation that peaks during mid-afternoon. DO concentrations (shown in Figure 3-28) exhibit a similar trend in concentration variations with elevated levels during the afternoon. This pa1tem reflects the effect of the increase in photosynthetic oxygen production during afternoon hours. Increased chlorophyll-a levels in the system also deplete nutrients in the system rapidly which, in tum. reduce the demand on oxidation. This process also contributes to the rise of D O levels during the afternoon.
.. I
I I
I I
I I I
I I I
I I I I I I
Il I
I Cl E
(") 0 z
- · - - · ·
3
2
0
f
->-
tt 100
9/20/88
Field - - - model
0, 0 - "' .c .c (.) u
"' N
,::. u
120 140 160 180 200 Downstream distance from Vernalis. 1000 ti (0.3 l<m)
Figure 3-24a. Comparison of Obsen1ed and Simulated Nitrate Profiles
- - - - - - · - - · - - - - - - - - - - - - - - - - -
2
E
0z
L.. 10/12/88
t tl r
I Field Ir 1 - - m o d e l I
e 0, 0 N "' -"' "' N ,: .c .c ,: ,: ,: r.
0 (.J (.)0 <..J (.)
100 1 20 140 160 180 Downstream distance from Vemalis, 1000 It (0.3 km)
Figure 3-24b. Comparison of Observed and Simulated Ni1rate Profiles
;:.:; u
200
Waler Ouaiity
-
j
3-37
100 r--------- l 09/8 8 I -Mode l + Field I
� I� � :,
'?.
I0 L-
� I '.; 0 ;; I � '.:: � N "'
N N N
.c ,:: .c .t::. .c. .t::. .c .c r u u u u (.) 0 (.) 0
100 120 140 160 180 200 I Downstream distance lrom Vernali's, 1000 ft (0.3 km)
I Figure 3�2Sa. Simulated Chlorophyll-a Profile I
-- 00 I
I J1 0 /88 I ------ Model + Field j Ii � I
' .... --;
,d ! J I>..
�
&.
Iu 0
� � '.:: � 0 N "' ;:; IN N "' .c. .c .c .c .t::. .t::. .c. .t::.
0 (.) 0 (.) 0 (.) (.) (.)
100 120 140 160 180 200
D0wns1ream d,stance lrom Vemalis. 1000 ft (0.3 km) I Figure 3-25b. Simulated Chlorophyll-a Profile
I
3-38 I
I '§, E
,q
0 0
0 I I I I I
<:: a,
I E
. ; 0 o 0
r
f 09/88
. ·-·--,--
E. Model
- - , - - -+ Field\
:: .c [:,
e o- "' "- ,:0 0
N "' N N ,: ,, 0 0
,oo 1 2 0 , 40 160 > 80
Downstream distance from Vernatis. 1000 It (0.3 km)
Figure 3-26a. Simulated Orthophosphate Profile
i 1 0188 1 - - M o d e l Fie ld\
t r l l f r
" '.'.'.! :: 0 N "' N "' "'
.c ,, [:, "- "- .c. 6<.) u <.J u 0
100 1 ? 0 l 40 160 180
Downstream distance from Vernalis, 1000 11 (0.3 k.m)
Figure 3-26b. Simulated Orthophosphate Profile
WalerOuaiic
l 1j j
;; 1 200
j i
lj
;:; c5
200
3-39
Ar'lnuaf Pro;tt1!s . f.::0 1: :o /he S a;c W ::er Hesi::..1'C"eS Cc--: rol Board
3-40
I I I
· - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Hourly \':tr::it1ons of ammonia and nitrate concentrations (sec Figures 3-29 and 3-3 l ) show gradual depression during the hours that correspond to rising chlorophyll-a levels, most likely rc: ultins from the incrc;:iscd uptake rates of nitrogen co maintain cell growth. A s1milo.r trend 1::, exh1bi:ed hy the diurnal pattern of inorganic phosphorus (orthophosphate) as shown in Figure 3-32. The fall in phosphate concentrations results from the increased phosphorus uptake during growth.
hgure 3-30 .<:hows computed organic nHrogcn concentrations over a 24-hour period. The rise and fali pattern follow that of chlorophyll-a resulting from the increased release of algal cells during respiration. The hourly variation ·of computed temperature is included [or reference (see Figure 3-33). Note that all of the above transformation processes are dependent on temperature.
Further work on model evaluauon is in progress. Future modeling efforts will involve more deiailed calibration of the model in the Delta.
::::::
to 100
so
0 0 4 8 12 16 20 24
hO<Jf
L _ _ _ _ - - - - · · .. -· .. · - - - - - - - - - - - - - - · - - - · · - - - - -Figure 3·27. Computer Hourly Variation of Chlorophyll·a at Channel 17
I I I I I I
I I I I I
I I I
I I I I
I
I
I
I
- - - - - · - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - . . . : . : W'...'::.a .'..:::/e:_:r O L'...'.:'.'a=/1,
r - · · · · · -
I I I I !
12
�
./
::::::: 8
0 4
0 0 4 8 12 16 20 24
Hour
Figure 3-28. Computer Hourly Variation of Dissolved Oxygen at Channel 17
::::::: 01 E ,..;
0.8 -
0.4 -
0 I I
0 4 8
' 12 16 20
Hour
- ---·····-·--·,
-
-
24
I I
Figure 3-29. Computer Hourly Variation of Ammonia at Channel 17
3-41
Annual ,Proqri:ss Reper: !C ?,t? State Water Resources Con;rol 8_c-ard - - - - - - - - - - - - - - - - - - - - - - - -
3-42
0.8
·2: 0.4
0
t 0 4 8 12
Hour 16 20 24
Figure 3-30. Computer Hourly Variation of Organ le Nitrogen at Channel 17
( ' " ) 0z
4 8
- - - - - - - · - - - · · - - · - - - - · - - - ...
12 Hour
16 20 24
Figure 3-31. Computer Hourly Variation of Nitrate at Channel 17
•
I I I I I I
I I I I I
I I I
I I I I
I
- - - - · · · · · - - - · - - - - - - - - - - - - - - - - - - - - - - - - - - - _ ! _ V � l a : ' . . ' : t e : ' . . . . r � O u � 8 � 1 i ,
0 8 1
J'<T 0.4 j 6 _j .c
J 0
0 4 8 12 Hour
16 20 24
Figure 3-32. Computer Hourly Varlatlo.n of Orth phosphate at Channel 17
u
I
30
'
20
10 r0 4 8
I f
12 Hour
l16 20 24
· - - - - - - - - - - - - - · · - - · · · · · · - · - · · - - - - .. · · - - -
Figure 3�33. Computer Hourly Variation of Temperature at Channel 17
3.43
Arm ·a/ Progress ReoOli to '",' s·ate Wa!er Resources Control Board
References ,\ Pl J..\ r .-\ mcric:111 Puh!ic I kJlth A ,,ocialiPn } .. 1985 Standard Mer hods (or 1/re 1:· rn111111011011 o f \\c11er and \\<me H:.,o. ! 6th ed .. \\'3,hingw:1. D C r :-i7-I .
B0"-·1c. (i I., c't JI. l 985 Ratti. Cnnst(mts nnd Ki11et1cs Fnnnulat,ons in Surfa,e \ nlcr Qua/in· A1odelinp. Second FJ11:1,n. L S. Environmental Prtitccl\on Agency. Athens. Georgia. EPA 60013-85/40
Brown.LC. and T.0 Barnwell. Jr.. 1987. The £11hanced S1rea111 Wilier Qiwli1y ,',,fodels QUAL2f_· and Q l .. 'AL7.f' .. l._.'.\'CAS D0cume111a1im1 and 1../sers Manual. l.]_S_ Environmental Pro1ec11on Agency, Athens. Georgia, EPA 600/3-87/007.
Dcpanmcnt of Water Resources. 1990 \Vata Quality Corui11ions in 1hc Sacramento-San Joaquin Delia D11nnfi 1988. D1vi 10n of Local Assistance
Depan:ncnt nt' Wa1cr Re,ources. 1967. Delw a11d Summ Marsh Water Qriality /11vernia11on. B11llen11 I 23
Dcp:mment of Water Resources. 1988 S1a1e Water Projcc1 Operations Data Monthly repons.
Department of Water Resources. 1990 Sacramento-San Joaquin Delta Water Qua/try Sun-e11/ance Pro?J()III -· I 988· Vo/w,,,: I Di,,,ion of Local Ass1s1:mcc.
Gromiec. M J. Reacration Ch.3 in. Mmhemm1ca! Submode ls Iii Water Qunlin S n terns S E Jorgcn\en :ind \ ! J Grom1ec led,. I. Elsevier.
Hu her. L , ! 995 Personal communication and fax Department of Munic1pal U11 l1ties. City of Stockton
HydroQual. \ 985 Central Della F:111rophica1ion Model-Mu/1ip!e Algal 5(? ecie:i Prepared for the Dcp:mmcnt of Water Resources. California HydroQual. Inc .. Mahwah. NJ 07430.
M WQI Dat:i Request. l 995. I nlcmal DWR memorandums 10 P Hu11on from R Woodard. February l Jnd 15
National Occ:inic and Atmosphcnc Adm1m tralion. NOAA O 9b8) Local cl1matolog1Cal d,11-a.
O'Connor. D.J Jnd W.E DobhinL \ 956 Mecha111sm o f Reaeratwn ,n Nawral Streams. J. of the Samtary Engineenng Division. ASCE. 82. SA6. pp 1-30
RaJbhandar1. H L. and G. T. Orlob .. 1990 £,.a/un1io11/ De1·1·/opmen1 o f Water Qualiry Models for Wastelood Allocation in r.,11,anne Envzronme111s. Prepared for California State Water Resources Control Board and Cahfom1a Rcg1011;il Water Quali1y Control Board. San francisco Bay Region. Department of Civil and Env1ronmcntal Engineenng. University of California. Davis.
Rajbhandan. H L.. 1995. Dmam11: S11nu/o1wn o f Water Qunliry Van.ables 111 Surface Water Sntems U11ki11g a Lagmngwn Reference Frame. PhD. D1ssenation-in-prcpara11on Department of Civil and E1wironmcnwl Engineering. l;nl\,erSHy of Cahfom1a. Davis.
Rathbun. R E . 1977. Reaeratwn Coefficients o f Streams. Stare-of-the-Art. ASCE. J Hydraulics Division. Vol. 103. No. HY4. pp. 409-424.
Smith, D J . ( 1988) Effects o f 1/1e S1ockton Slup Chnnnel Deepening on the Dissolved O x ygen o f Wmer Near 1he Port o f S1oc/.:1011. Cahjornia (Phase II). Resource Management Associa(es. Inc .. Lafayette. California
_ , , ___________ _ 3-44
•
I I I I I I
I I I I I I I I I