ML21035A199

From kanterella
Jump to navigation Jump to search
NRC-2020-000123 - Resp 5 - Final, 2017-01-27 FPL Attachment 1 Gw Flow and Transport Model
ML21035A199
Person / Time
Site: Turkey Point  NextEra Energy icon.png
Issue date: 01/21/2021
From:
NRC/OCIO
To:
Shared Package
ML21035A187 List:
References
NRC-2020-000123
Download: ML21035A199 (39)
v • d • e


Text

Attachment 1 Biscayne Aquifer Groundwater Flow and Transport Model: Heterogeneous Hydraulic Conductivity Analyses, January 2017 (Figures)

Biscayne Aqu ifer Model Heterogeneous Hydraulic January 2017 Conductivity Ana lyses Legend D Active Area 0 4,000 8,000

...,_ i

  • Pilot Points FHI GIS Wo,je fl.PP CJ l.b *bt oin t5.mxd Figure 1. Hydraulic conductivity pilot point locations and the TPGW well name associated with the pilot point locations (where relevant) 18

Biscayne Aquifer Model Heterogeneous Hydraulic Conductivity Analyses January 2017 2 - - - ---1 HHC-J LA Model >--- - -

.. ... 2 ICalibrated HHC-Vl I ~

.1 "'* ~

" t... ,,

1.8

,~. 1.8 f- ,..:{l;.'J; .' I'*

1.8

-p.ciV,*: r'.'.~-\*..,,..

1.6 1.6 1.6 u

_~
  • ~ 1.4 1.4

,_l.~: "

    • _:j

~ ~

I I ~

if'

~ 1.2

1. 2

\ \I

~

~

1 3 0.8 1

~

-0

~

3 0.8 1

... >j

~ !_I_*:' i..

    • *I .... -

)' * °5 0.8 ~~

~

E E **~*[ *  :,; b

~6 0.6 Layers 1 to 3 0.6 Layers 1 to3 LayersS to7 Layers 5 to 7 l..lycrs5to7

~4 r ~- 11 Layer 9 to 11 0.4 Layer9to 11 0.4 Lilyer9 to 11

.
  • *~ . : Layer4 Layer4 Layer4

~2 .:t:*----?-... .~ Layers 0.2 Layers 0.2 Layers

,,,""'-r .. .

0

- - PcrfcctMatch - - PcrfectMatch I I - - PcrfcctMatch 0 0.5 1.5 0.5 1.5 0 0.5 1.5 Observed Re lative Salt Observed Relative Salt Observed Relative Sa lt Figure 2. Observed versus simulated CSEM survey salt concentrations for the July 2016, HHC-JLA, and calibrated HHC-Vl models 19

Biscayne Aquifer Model Heterogeneous Hydraulic Conductivity Analyses January 2017 1

0.9 IG-2 1-18 I 1 0,9 IG 181'

.2 0.8 --, - - . -------+-- ~ QB g 0,7 roz Q7

§ 0.6 § Q6 g C u 0.5 3 Q5 Ij o.4 ~M w

1.~ 0,3 ~ Q3

, ~ 0.2 I"" Q2 CC 0.1 0

~ -===---- ~ .~1111111111111111II II Ql t... ,,,,,,...*~

~ ........... . ......, ...... ,,,,,.

~

¥ 1110111111111111111111 ............. . . 0 J-93 0-95 J-98 A-01 J-04 0-06 J-09 A-12 D-14 5-17 J-93 0-95 J-98 A-01 J-04 0-06 J-09 A-12 D-14 5-17

  • Observed - - July 2016 Model - - HHC-JLA - - Ca librated HHC-Vl
  • Observed - -July 2016 Model - - HHC-JLA - -Calibrated HHC-Vl 1 1 -V~\

G-21-58 0.9 0,9 I ~

,g 0.8 ,gC 0,8 z"" 0.7 ~ 0.7

~ § 0.6

! 0.6 C

0 3 0,5

~ 0,5

~ 0,4 ,';; 0.4 w

I ~ *E o.3

~ ,_::-:.:-=.-::..:..

b;;;;-:.:~;;~~~~~~;; :; -- -~~:::::~v.~~...~.,-  ;;:;;*;..-..~.-

0,3 Cl I

~ 0,2 0.1 ,- 0.2 0 - - *** ,,.. - -- 0.1 0

J-93 0 -95 J-98 A-01 J-04 0-06 J-09 A-12 D-14 5-17 J-93 0-95 J-98 A-01 J-04 0-06 J-09 A-12 D-14 5-17

  • Observe d - -Ju ly 2016 Model - - HHC-JLA - - Calibrated HHC-Vl
  • Observed - -July 2016 Model - - HHC-JLA - -Ca librated HHC-Vl Figure 3. Observed and si mulated saltwater wedge breakthrough s for the July 2016, HHC-JLA, and cal ibrated HHC-V1 model s 20

Biscayne Aquifer Model Heterogeneous Hydraulic January 2017 Conductivity Analyses 0 ft 10000 ft 20000 ft L.

TPGW-6 TPGW-7 I I w ~ 01 m ""-l --.j co 01 01 0 01 0 01 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Horiz. Hyd. Cond., ft/day Figure 4. Layer 4 heterogeneous horizontal hydraulic conductivity associated with the calibrated HHC-V2 model 21

Biscayne Aqu ifer Model Heterogeneous Hydraulic January 2017 Conductivity Analyses Oft 10000 ft 20000 ft

...... ...... N N w .i,.. <.n <.n Cl) Cl) ...., ....,

I

0) 0)

I co co ......

0 <.n 0 <.n <.n <.n 0 <.n 0 <.n 0 <.n 0 <.n 0 <.n 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Horiz. Hyd. Cond., ft/day Figure 5. Layer 8 heterogeneous horizontal hydraulic conductivity associated with the calibrated HHC-V2 model 22

Biscayne Aquifer Model Heterogeneous Hydraulic January 2017 Conductivity Analyses Oft 10000 ft 20000 ft TPGW-6 TPGW-1 TPGW-4 Horiz. Hyd. Cond., ft/day Figure 6. Layer 9 heterogeneous horizontal hydraulic conductivity associated with the calibrated HHC-V2 model 23

Biscayne Aquifer Model Heterogeneous Hydraulic January 2017 Conductivity Analyses 0 ft 10000 ft 20000 ft Horiz. Hyd. Cond., ft/day Figure 7. Layer 10 heterogeneous horizontal hydraulic conductivity associated with the calibrated HHC-V2 model 24

Biscayne Aquifer Model Heterogeneous Hydraulic January 2017 Conductivity Analyses 0 ft 10000 ft 20000 ft Horiz. Hyd. Cond., ft/day Figure 8. Layer 11 heterogeneous horizontal hydraulic conductivity associated with the calibrated HHC-V2 model 25

Biscayne Aq uifer Model Heterogeneous Hydraulic Conductivity Analyses January 2017 1

0.9

-1G-21-181 1 0.9 IG-28-181

.g 0.8 r--- .g 0.8

~ 0.7 ~ 0.7

§ 0.6 § 0.6 C

u 6 0.5 8o.5 ~ - -

~ 0.4 ~ 04

<!.I <!.I

  • 5r, E 0.3 03

& 0.2 &02 01 ~ **

0.1

__,.,.._,,, ~ * .......... ** . . .

o .;::;;;;; - $ ,,, ......................~ -- - - - - - - 0 ************~ ........... ._ ..~ *,.4't!.. ___ -- ~ --

J-93 0-95 J-98 A-01 J-04 0-06 J-09 A-12 D-14 S-17 J~ 3 0 -95 J-9 8 A-01 J-04 0-06 J-09 A-12 D- 14 5-17

  • Observed - - Ini tia l HHC-V2 - - Calibrated HHC-V2
  • Obse rved - - Initia l HHC-V2 - - Cali brated HHC-V2 1

IG-21-58 I 1 IG-28-58 I I o.9 09

.g 0.8 I g 0.7

§ 0.6

~-

~ OS g

~u 06 C

0.7 -- - * * * *~ * ...

r--*

I~ o.5 r 8 05

==

I~ 0.4 Sil

<!.I 0.4 1 *5 0.3 . E o.3

& "' 0.2 ..... * -

  • _........ I& "' 0.2
  • 0.1 ' - - ~ , 0.1 0
  • ll11111u**u ~*wo ....... * * * *
  • 0 J-9 3 0-95 J-98 A-01 J-04 0-06 J-09 A- 12 D-14 5-17 J-93 0-95 J-98 A-01 J-04 0-06 J-09 A-12 D-14 5-17
  • Observe d - - Initial HHC-V2 - -Calibr ated HHC-V2
  • Observe d - - Initial HHC-V2 - - Calibrated HHC-V2 Figure 9. Observed and simulated saltwater wedge breakthroughs for the initial (July 2016) and calibrated HHC-V2 models 26

Biscayne Aq uifer Model Hetero geneous Hydraulic Conductivity Analyses January 2017 2 July 2016 Model 2 ~ - - - ~1 Calibrated HHC-V2 f--- ~ - - - -,

1.8

  • I ->> t _X,,.

1.8 . . P:-1:

~

~ .. 'W: :I:

. 1 .. ~-

..x _;(.

-1 i' t ~-~ .;

1.6 x_x x

~

I x

,.' 1.6 1.4 sxJ 1.4

<ti ..

xX ; \

<ti Vl

-~ 1.2

~ ~

'5:

  • x * .~

Vl

-~ 1.2

<ti . .

QJ 1 4' ai 1

  • ;x x x x:W-.

c:: c::

"C

~

<ti

  • \ ';.:l'i "C QJ

<ti

~

x *xx*x ~

X X Z x ';it.

i i Xi

i 0.8 :i 0.8 E E  ; 'Jr \ : X vi vi ~-* '3r"Zf X 0.6 0.6
  • Layers 1 to 3 X ~X X x JIC i ; X x- x J?,'1(\

~ -~.: Layers 5 to 7 X

  • Layers 5 to 7 0.4 0.4 ,I' 4\ ***

?~"'. -,.-:." Layer 9 to 11 1"\ .x *x - Layer 9 to 11 0.2 ;4

! xxx ~

. x: l .

X*z x

  • Layer 4 Layer 8 0.2 x i :s:

.. x x:

z*x l

.x Layer 4 Layer 8

  • x Jlr: I Xx X 2':,f: xi - - PerfectMatch - - Perfect Match 0 0 0 0.5 1 1.5 2 0 0.5 1 1.5 2 Observed Relative Salt Observed Relative Salt Figure 10. Observed versus simulated CSEM survey salt concentrations for the initial (July 2016) and calibrated HHC-V2 models 27

Biscayne Aquifer Model Heterogeneous Hydraulic January 2017 Conductivity Analyses 0 ft 10000 ft 20000 ft L.

TPGW-6 TPGW*7 I I

--..J --..J co 0 01 0 0 0 0 0 0 0 Horiz. Hyd. Cond., ft/day Figure 11. Layer 4 heterogeneous horizontal hydraulic conductivity associated with the calibrated HHC-V3 model 28

Biscayne Aquifer Model Heterogeneous Hydraulic January 2017 Conductivity Analyses Oft 10000 ft 20000 ft I

..... ..... N N w 01 01 0) 0) --...I --...I 0) (0 (0 .....

0 01 0 01 01 0 01 0 01 0 01 0 0 01 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Horiz. Hyd. Cond., ft/day Figure 12. Layer 8 heterogeneous horizontal hydraulic conductivity associated with the calibrated HHC-V3 model 29

Biscayne Aquifer Model Heterogeneous Hydraulic January 2017 Conductivity Analyses

""L..

Oft 10000 ft 20000 ft I

TPGW-e TPGW-4

\

Horiz. Hyd. Cond., ft/day Figure 13. Layer 9 heterogeneous horizontal hydraulic conductivity associated with the calibrated HHC-V3 model 30

Biscayne Aquifer Model Heterogeneous Hydraulic January 2017 Conductivity Analyses Oft 10000 ft 20000 ft TPGW-1 Horiz. Hyd. Cond., ft/day Figure 14. Layer 10 heterogeneous horizontal hydraulic conductivity associated with the calibrated HHC-V3 model 31

Biscayne Aquifer Model Heterogeneous Hydraulic January 2017 Conductivity Analyses L-Oft 10000 fl 20000 fl Horiz. Hyd. Cond., ft/day Figure 15. Layer 11 heterogeneous horizontal hydraulic conductivity associated with the calibrated HHC-V3 model 32

Biscayne Aquifer Model Heterogeneous Hydraulic January 2017 Conductivity Analyses Oft 10000 ft 20000 ft J

I CX>

0 Vert. Hyd. Cond., ft/day Figure 16. Layer 9 heterogeneous vertical hydraulic conductivity associated with the calibrated HHC-V3 model 33

Biscayne Aqu ifer Model Heterogeneous Hydraulic January 2017 Conductivity Analyses Oft 10000 ft 20000 ft TPGW-5 TPGW-4 I

(X) 0 Vert. Hyd. Cond., ft/day Figure 17. Layer 10 heterogeneous vertical hydraulic conductivity associated with the calibrated HHC-V3 model 34

Biscayne Aquifer Model Heterogeneous Hydraulic January 2017 Conductivity Analyses Oft 10000 ft 20000 ft TPGW-4 I

,, .I

--*..,_.., 'Ii I

CX) 0 Vert. Hyd. Cond., ft/day Figure 18. Layer 11 heterogeneous vertical hydraulic conductivity associated with the calibrated HHC-V3 model 35

Biscayne Aquifer Model Heterogeneous Hydraulic Conductivity Analyses January 2017 Water Levels (1972 to 2010) Water Levels (2010 to 2015) 5 5 4 4 3 *

  • I 11 3

00 00 co co Cl Cl

~2 ~2 z z g g

~1 QJ

    • ~1 QJ
r:
r:

-0 -0 QJ QJ

-:uo -:u 0

i :i E E vi vi

-1 -1

-2 -2

-3 -3 2 -1 0 1 2 Observed Heads (ft, NAVO 88) 3 4 5 II -3 -2 -1 0 1 2 Observed Heads (ft, NAVO 88) 3 4 5 Figure 19. Observed and simulated water levels between 1972 and 2010 (left) and between 2010 and 2015 (right) for the calibrated HHC-V3 model 36

Biscayne Aquifer Model Heterogeneous Hydraulic Conductivity Analyses January 2017 Relative Salt Concentrations (1972 to 2010) Relative Salt Concentrations (2010 to 2015) 2.5 I

2

~-+-------t----~ I : ,: r-----r--+__j___JL___

I I

... .... I

~1.5

-~

~ 1.5

.5 ai er:

  • ai"'

er:

-0 2l ** -0 2l

="'E 1 ** .!!!

, 1 E

vi vi I

  • 0.5 --~n.--- I I I 0.5 0 0 0 0.5 1 1.5 2 2.5 0 0.5 1 1.5 2 2.5 Observed Relative Salt Observed Relative Salt Figure 20 . Observed and simulated relative salt concentrations between 1972 and 2010 (left) and between 2010 and 2015 (right) for the calibrated HHC-V3 model 37

Biscayne Aquifer Model Heterogeneous Hydraulic Conductivity Analyses January 2017 2 2 Calibrated HHC-V3 t---- ~ - - - -

Initial HHC-V3

.. r - - - - - ~1 7 I

l'X X
  • ...._:-_.---L..,... I X 1.8 I :1~ *~* xx x x':?,i x 1.8 'i
  • x

~

x :xjc X ~ 1.6

  • 1.6 X X

-x!xxX X X X X x* X X ,!

.,!-x

.~h1 ...

X X 1.4 X X I 1.4 ~ E

~ X ~

V)"' xx x x X

V) X X CJ 1.2 X CJ 1.2 *....--r=*l

-~ X XX.,?1-x x1x x x XX

~ ** . l Qi"' ""1/4

      • 'It ..../

X X X "'

Qi c::: 1 X - ~ X c::: 1 "C

X X "?#- "C

....CJ ....CJ

  • ~

X X  : l:_,;X \

X X X 3 0 .8 X

Xx %

X X *: I 3 0.8 X E X X X X X E X

x
l!,
lx"f  ;;;

0.6  !..c

. Layers 1 to 3 ~ 0.6 X

~ ~ *-=--!' ~*°'4\

X xr.: zXx x

. Layers 5 to 7 X

)l: X

0: ll!

"ir.l

  • X X

Laye rs 1 t o 3 Layers 5 to 7 0.4 711 X

  • - I -- -

X X xX Layer 9 to 11 H I 0.4 X Layer 9 to 11 X X X

  • x x ,l *z 0 .2

~ i

.- X

' l

  • x X

Laye r 4 Laye r 8 H I I 0.2 "x

  • Laye r 4 Layer 8

...... I - - PerfectM atch ,l'*~ a X - - Perfect Match t&;s;:-T [ J:'"'~ .._ '!X X :a:

0 )It:

0 0 0.5 1 1.5 2 II 0 0.5 1 1.5 2 Observed Relative Salt Observed Relative Salt Figure 21. Observed versus simulated CSEM survey salt concentrations for the initial and calibrated HHC-V3 models 38

Biscayne Aquifer Model Heterogeneous Hydraulic January 2017 Conductivity Analyses Relative Concentration 2.05 1.95 1.85 1.75 1.65 1.55 1.45 1.35 1.25 1.15 1.05 1.01 0.95 0.85 0.75 0.65 0,55 0.45 0.35 0.25 0.15 0 .05

-0.05 2.05 1.95 1.85 1.75 1.65 1.55 1.45 1.35 1.25 1.15 1.05 1.01 0 .95 0.85 0.75 0.65 0.55 0.45 0.35 0.25

0. 15 0 .05

-0.05 Figure 22. Initial and 10 year relative salt concentrations under RWS alternative 3D in model layer 8 for the July 2016 model (left) and calibrated HHC-V3 model (right) 39

Biscayne Aquifer Model Heterogeneous Hydraulic January 2017 Conductivity Analyses Relative Concentration 1.25 1.1 5 1.05 1.01 0.95 0.85 0.75 0.65 0.55 0.45 0.35 0 .25 0 .15 0.05

  • 0.05 1.95 1.85 1.75 1.65 1.55 1.45 1.35 1.25 1.15 1.05 1.01 0.95 a.as 0.75 0.65 0.55 0.45 0.35 0.25 0.15 0 .05

-0.05 Figure 23. Initial and 10 year relative salt concentrations under RWS alternative 3D in model layer 9 for the July 2016 model (left) and calibrated HHC-V3 model (right) 40

Biscayne Aquifer Model Heterogeneous Hydraulic January 2017 Conductivity Analyses Relative Concentration 2.05 1.95 1.85 1.75 1.65 1.55 1.45 1.35 1.25 1.15 1.05 1.01 0.95 0.85 0.75 0.65 0.55 0.45 0.35 0.25 0.15 0.05

  • 0,05 2.05 1.95 1.85 1.75 1.65 1.55 1.45 1.35 1.25 1.15 1.05 1.01 0.95 0.85 0.75 0.65 0.55 0.45 0.35 0.25
0. 15 0.05

-0.05 Figure 24. Initial and 10 year relative salt concentrations under RWS alternative 3D in model layer 10 for the July 2016 model (left) and calibrated HHC-V3 model (right) 41

Biscayne Aquifer Model Heterogeneous Hydraulic January 2017 Conductivity Analyses Relative Concen tration 1.95 1.85 1.75 1.65 1.55 1.35 1.25 1.15 1.05 1.01 0.95 0.85 0.75 0.65 0.55 0.45 0 .35 0.25 0.15 0.05

-0.05 1.95 1.85 1.75 1.65 1.55 1.45 1.35 1.25 1.15 1.05 1.01 0.95 0.85 0.75 0 .65 0.55 0.45 0.35 0 .25 0 .15 0.05

-0.05 Figure 25. Initial and 10 year relative salt concentrations under RWS alternative 3D in model layer 11 for the July 2016 model (left) and calibrated HHC-V3 model (right) 42

Attachment 2 Addendum to Regional Biscayne Aquifer Groundwater Model Report Incorporating Comments from SFWMD

Addendum to Regional Biscayne Aquifer Groundwater Model Report Incorporating Comments from SFWMD Introduction This Addendum to the July 2016 Biscayne Aquifer model report (Tetra Tech, 2016) describes revisions made to the Variable Density Regional Biscayne Aquifer Groundwater Flow and Transport Model subsequent to both the distribution of the memorandum to regulatory agencies and an external model review meeting held at Miami Dade County Department of Environmental Resource Management (MDC DERM) on July 21, 2016. The motivation for, execution of, and changes to model results as a result of revisions to model boundary conditions are elaborated upon below.

Regional Model Summary A transient variable density groundwater flow and transport model of the Biscayne Aquifer was developed to assess the impacts of the proposed remediation well extraction on local water resources (Tetra Tech, 2016) . The model domain covers an area of approximately 276 square miles and consists of 11 layers. The 11 model layers represent the Miam i Oolite, Fort Thompson Formation and two distinct high hydraulic conductivity layers (high flow zones) which represent preferential flowpaths detected in lithologic logs. The model simulates interactions between the Biscayne Aquifer, Biscayne Bay and surface water canals, including the CCS. Net recharge (i. e. recharge minus evapotranspiration) and groundwater withdrawals for both municipal and agricultural uses are also simulated using the best available information. The model is executed using the latest version of the SEAWAT code (Version 4).

Water density is assumed to vary as a function of both temperature and salinity.

The regional groundwater model was calibrated to match measured water levels and groundwater salinities over a 47-year period (1968-2015). The first 42 years of the simulation period are simulated seasonally (two seasons per year), whereas the final five years and three months of the calibration period are simulated monthly due to greater data availability during this period. Hydrogeologic properties in the regional model were informed by a local-scale model which was calibrated separately to match drawdowns observed during an aquifer performance test conducted near the site of the proposed UIC wells. Automated calibration was performed on the local-scale and regional models using parameter estimation software (" PEST"). Acceptable statistical measures of calibration quality were achieved for salinities and water levels during calibration of the regional model.

Revisions to the Model During a model review meeting on July 21, 2016, South Florida Water Management District (SFWMD) technical staff recommended revising how two surface water boundary condition_s were simulated . The models used in the simulations presented in this application incorporated both of these suggested revisions, which are described in detail below. The final simulated groundwater heads, salinities and 1

temperatures from the end of the revised calibration model were used to initialize predictive simulations.

Card Sound Canal The first suggested revision pertained to the portion of Card Sound Canal (sometimes referred to as "Card Sound Road Canal") south of where the canal nearly intersects the L-31E canal. The models submitted to Miami-Dade County, SFWMD and EPA prior to the review meeting treated Card Sound Canal as a drain boundary condition . Modeling this canal as a drain allowed the Biscayne Aquifer to discharge to the canal, but did not allow the canal to possibly serve as a source of water to the surrounding aquifer. The suggestion made by SFWMD was to model this section of Card Sound Canal as a saline river boundary condition instead of a drain to allow inland saltwater intrusion to occur through the canal. This revision was made, and relative salinity values of one (i.e. 35 practical salinity units, PSU) were assigned to the river boundaries in this section of Card Sound Canal throughout the calibration (1968-2015) and predictive simulations.

L-31£ Canal The second revision suggested by SFWMD was related to the southern section of the L-31E canal. The L-31E canal was treated as a river boundary condition with a relative salinity of zero in the models submitted for review. SFWMD recommended revising the southern portion of this canal (i.e. the section south of control structure S-20) from a river to a drain boundary condition. This revision was also made to the calibration and predictive simulations.

Layer 1 Rewetting Subsequent to these boundary condition revisions, the model files were imported into the model pre-processor Groundwater Vistas (GWV) to facilitate external model review. Whereas the model was successfully executed from the DOS command prompt, stable and successful model execution within GWV required that layer 1 model cells in the western areas of the model domain be specified as not rewettable. As explained below, this did not affect calibration quality of the model, and was only necessary to be able to execute the model within GWV.

Model Results Calibration Model In order to assess the impact of the boundary condition revisions upon the calibrated model quality, simulated results for both models (pre- and post-revised boundary conditions) were compared to each other. First, a comparison of the relative salt concentration versus time at a monitoring well near to the revised boundary conditions simulated by both models provides a visual assessment of the local changes to simulated conditions attributable to the boundary condition revision. The simulated relative salt concentration versus time at monitoring well G-28 (18 ft and 58 ft screens) is provided in Figure 1. At the shallow (18 ft) screen, the impact to simulated salt concentrations due to the revision is negligible. At the deeper (58 ft) screen, the model revision produces slightly greater concentrations at the monitoring well. This result should be expected, as both boundary condition changes would likely produce greater salt concentrations at nearby monitoring wells.

2

Results at G-28 illustrate impacts to simulated salt concentrations near to the boundary conditions. In order to confirm that these changes are relatively localized, summary statistics for flow and transport model residuals were tabulated for targets throughout the entire model domain prior to and after the boundary conditions revision. The mean absolute error (MAE) was calculated for water level observation calibration targets and observed salt concentration calibration targets at monitoring wells, as well as CSEM survey-based salt concentrations. The MAE for the target categories were compared between the pre- and post-revision models to assess the degree to which the boundary conditions impacted model quality. This comparison is provided in Table 1, below. Inspection of this table reveals that the MAEs for both models are very similar for all calibration target types, which confirms that the boundary condition revision did not significantly affect model results. This is not surprising given the disparate locations of the revised boundary conditions that were revised relative to calibration target data. In fact, for most of the calibration target types where a change to the residuals occurred, the model improved (MAE reduced) as a result of the boundary condition revision.

Prediction Model Simulation of Recovery Well Systems (RWS) Alternative 3D with the revised boundary condition model yielded similar, yet slightly different results in terms wetland drawdown. Inspection of Figure 2 reveals that the revised model simulates a slightly greater extent of drawdown in wetlands . The slight increase in the extent of wetland impacts simulated by the revised model is consistent with the conversion of L-31E (south of S-20) from a RIV boundary condition (potential source of water to the model) to a DRN boundary condition (not a source of water to the model).

Figure 2 shows the average simulated salt concentration of the water extracted by the five sourthernmost extraction wells that operate as a part of RWS Alternative 3D . These wells were selected because they are nearest to the revised boundary conditions. Note that the simulated average salt concentration of the extracted water is nearly equal in the two versions of the RWS model.

Conclusions Based on the recommendations provided by SFWMD, revisions to portions of the L-31E and Card Sound Canal boundary conditions were made to the calibrated Biscayne Aquifer groundwater flow model presented to MDC DERM in July 2016 . As a result of these recommendations, the representation of these boundary conditions in the revised model is likely improved over the prior model. Based on an evaluation of calibration and prediction models' results, the revisions have an overall minor impact to the historical and future simulated hydro logic and water quality conditions. As of this model improvement and minimal change to model results, these revisions will be preserved in future versions of the calibration and predictions models.

3

1 0 .9 IG-28 (18 ft) I C

0.8

-~roO b 0.7 C

~ 0.6 C

8 0.5

~ 0.4

<I)

.=::

..... 0.3 ro a:i 0.2 0::::

0.1 o b~~~~~~~~~~~~~--~~--_j J-93 J-98 J-04 J-09 D-14 J-20 x Observed - Original - - Revised 1

-1G-28 (58 ft), - -

0.9 C

0.8

-~0ro X C

0.7

<I) u 0 .6 C

0 u 0.5

.:!:=

ro V) 0.4

<I)

-~ro> 0.3 X

<I) 0.2 0::::

0.1 0

J-93 J-98 J-04 J-09 D-14 J-20 x Observed - Original - Revised Figure 1. Observed and simulated relative salt concentrations at monitoring well G-28 (shallow and deep screens) for the original calibrated model and calibrated model with the revise boundary conditions 4

Legend UIC Ptoduction \\'ells D Active Area

- - Original PEST CSEM Calibfation Model (July 21)

Moda l with Revise d Rivet/Drain Boundary Conditio ns o

Figure 2. Simulated wetland (Model Layer 1) drawdown after 9 years of Alternative 3D extraction well operation using the original (pre-revision) models and revised boundary condition models 5

60

- - Original Model 55

? - - Revised Model

~

....,~ 50 cu s

"O 45 Q) u

....,... 40 cu X

w 4-0 u -- - -

C 35 0

u Q)

?i...i° 30 Q)

<t 25 20 13 25 37 49 61 73 85 97 109 121 Months of 10-Year RWS Figure 3. Average simulated (original and revised models) concentration of water extracted by the five southernmost extraction wells that comprise RWS Alternative 3D 6

Table 1. MAE for the pre- and post-boundary condition revision models for all calibration target types Target Type Pre-Revision MAE Post-Revision MAE  % Difference Seasonal Water Levels (ft) 0.459 0.459 0.00%

Seasonal Relative Salt Cones 0.198 0.191 -3 .54%

Monthly Water Levels (ft) 0.329 0.329 0.00%

Monthly Relative Salt Cones 0.145 0.145 0.00%

CSEM Relative Salt (Layers 1 to 3) 0.116 0.111 -4.31%

CSEM Relative Salt (Layer 4) 0.227 0.225 -0.88%

CSEM Relative Salt (Layers 5 to 7) 0.216 0.213 -1.39%

CSEM Relative Salt (Layer 8) 0.267 0.259 -3.00%

CSEM Relative Salt (Layers 9 to 11) 0.284 0.287 1.06%

7

Attachment 3 FPL Responses to FKAA Model Review Comments

Response to FKAA Groundwater Model Review Introduction Florida Power & Light (FPL) appreciates the review of the Regional Biscayne Aquifer model provided by Florida Keys Aqueduct Authority (FKAA) and its consultant, Water Science Association. With its consultant, Tetra Tech, FPL has evaluated FKAA's comments and concerns associated with the groundwater model. Overall, while FPL believes that incorporating some of FKAA's suggested model revisions will generally improve the caliber of the model, FPL finds that none of these revisions will significantly impact the results of predictive modeling nor will they likely change the optimality of the selected remedial alternative (Alternative 3D).

It is worth noting that the Consent Agreement (CA) between FPL and Miami Dade County (MDC)

Department of Environmental Resource Management (DERM) required FPL to develop the variable density groundwater flow and transport model referenced herein within a 180 day timeframe. Given this relatively short timeframe, the model will evolve over time with the introduction of additional data and improved understanding of regional hydrology, hydrogeology, and geology. As such, it is possible that some of the suggested revisions and recommendations proffered by FKAA may be incorporated in the course of the model's evolutionary development In the Conclusions and Recommendations of the review memorandum, FKAA's consultant elucidated concerns that need to be addressed prior to additional modeling. Responses to these concerns are provided below. Additional and relatively minor concerns, discussed elsewhere in the review memorandum, are addressed at the conclusion of this document FKAA Concerns Requiring Resolution FKAA Comment: Assigning spatially variable hydraulic parameters to model layers should be considered since it could affect the flow and transport significantly.

FPL Response: FPL agrees that adding additional and appropriate complexity to the groundwater model has the potential to improve the calibration of the model and perhaps its predictive capability.

Uniform hydraulic conductivities were informed by a local aquifer performance test (APT) (Enercon, 2016); and, due to the relatively short period of time allotted for model construction and calibration, the incorporation of greater complexity in the form of heterogeneous hydrologic properties was not possible. T_hrough the course of continued model improvement, Tetra Tech is currently conducting a calibration of the groundwater model wherein the definition of heterogeneous hydraulic conductivities is being evaluated.

1

FKAA Comment : The varying rates of net recharge part way through the calibration period are not clearly tied to calibration efforts. Some explanation for these changes is requ ired .

FPL Response: Starting in 1996, NEXRAD-based estimates of precipitation and potential evapotranspiration provided a relatively detailed assessment of precipitation, and were employed to define seasonal recharge for the latter part of the seasonal model (1996 to 2010). Prior to 1996 in the seasonal model, wet and dry seasonal recharge was based on average seasonal precipitation and evapotranspiration, as calculated from available NEXRAD data. A more detailed discussion of precipitation data and the associated estimates of recharge is provided in the modeling documentation (Tetra Tech, June, 2016).

FKAA Comment: The occurrence of "dry cells" and "flooded cells" over large portions of the model domain raise concerns about the appropriateness of model assumptions and/or inputs and could be an issues for overall model accuracy and reliability for predictive appl ication.

FPL Response: Layer 1 of the model represents surficial sediment and muck, which constitute a relatively thin hydrogeologic formation. As such, it is reasonable for this layer to be desaturated in western areas of the model domain. During the continued evolution of the groundwater model the occurrence of dry and flooded cells will be evaluated by the modeling team in order to bolster numerical stability in the model's solution. In the course of this evaluation, modelers will investigate reasonable model revisions that minimize the occurrence of dry cells (e.g. revisions to the representation of layer 1) and flooded cells (e.g. increasing the storage or hydraulic conductivity of shallow model layers).

FKAA Comment: The change of river conductance at the CCS is a major concern. The changes are significant, late in the simulation period. The issues is identified at locations most critical for the performance eva luation of the various remedial alternatives. The change of river conductance may require the model to be recalibrated or the proposed remediation scenarios be reevaluated if the change is not supported by actual field data.

FPL Response: The reduction of CCS RIVER cell conductance late in the monthly simulation is consistent with the siltation that occurred in the CCS during this timeframe. This siltation is believed to have decreased the conductance of the CCS canal beds and the thermal efficiency of the CCS.

Several cores collected in the CCS bottom sediments are evidence of such siltation. Additionally, the reduction in conductance due to siltation is apparent in the observed salinities directly below the CCS (TPGW-13S and TPGW-13D; Figure 18 of review memorandum and figures copied below). Despite the significant reduction in conductance, the model still over-predicts the groundwater response to the increase in CCS salinity that occurred in late 2013. This modeled response suggests that, rather than being too large, the reduction in bottom conductance possibly may not be large enough. Discussion of this model characteristic was inadvertently omitted from the model documentation (Tetra Tech, June, 2016).

2

TPGW-13S TPGW-13D 2.0 f1 .5 f1.5 2.0 I

ro ro w 1.0 w 1.0

~

Q) ro

~0.5 ro

~0.5 --Simulated I

  • Observed 1
  • Observed 0.0 0.0 L _ - - - - - - --======='-1 10/1/10 10/1/11 9/30/ 12 9/30/13 10/1/14 10/1/15 10/1/10 10/1/11 9/30/12 9/30/13 10/1/14 10/1/15 Date Date As the reviewers point out, riverbed conductance is rarely measured in the field, and, as such, is often a calibration parameter. During a calibration analysis conducted in June 2016, the conductance of the CCS (pre- and post-siltation) was adjusted to improve the calibration quality of the model. These adjustments were relatively minor and the post-calibration conductances were on the same order of magnitude at those in the model reviewed.

The decision to include the reduced CCS conductance during model predictions was twofold. First, Model predictions focused on identifying impacts to remediation-based changes. Maintaining the conductance obtained from the calibration ensures that simulated changes are due solely to the simulated remedial alternatives. Second, reduced CCS conductance reflects a conservative model characteristic insofar as the benefits of CCS salinity abatement (Alternatives 2 through 8) would be less pronounced than those resulting from a higher CCS conductance.

FKAA Comment: It is recommended to consider practical well capacities for the proposed extraction wells in the remediation scenarios. To optimize the remediation designs, the performance of individual extraction wells may be assessed by checking the mass removal rates of particle tracking methods.

FPL Response: FPL assumes that this first part of this comment refers to the single extraction well that is assumed to extract 15 MGD from the Biscayne Aquifer beneath the CCS. This single extraction well has since been revised to four extraction wells (each pumping at a rate of 3.75 MGD), located within the vicinity of the Underground Injection Control (UIC). The reduced extraction rates (15 MGD distributed across four wells) are lower than the rate of pumping during the near-CCS aquifer performance test, early in 2016.

FPL concurs that the continued analysis of mass removal is warranted as the model evolves and extracted salt mass will be monitored and reported as described in the agreements and orders with the agencies.

FKAA Comment : Using the MOD FLOW Drain package to simulate Card Sound canal should be re considered.

FPL Response: This boundary condition has been revised to reflect its potential to contribute saline water to the aquifer. This revision did not have notable impacts on model calibration statistics (less 3

than 1% change in mean absolute error for heads and salinity in both the seasonal and monthly models) or predictive simulation results.

Other Review Findings Comments to the groundwater model were discussed elsewhere in the review document. These comments were not identified as requiring resolution prior to additional modeling. Nevertheless, they are addressed below.

CCS Salinities - The review document notes that the simulated CCS salinity does not match the maximum CCS salinity observed during the simulated timeframe . This is due to the fact that simulated conditions at boundary conditions (e .g. the CCS) for each model stress period reflect the average condition that occurred during that stress period. Peak salinities in the CCS have been relatively brief, and the timescale of the model simulation is more consistent with capturing long term trends in such variables as salinity, rather than short term changes.

Flow from GHB Cells - FPL and its consultant concur that the flow through GHB cells should be evaluated to confirm the amount of water entering the model through this boundary condition is realistic.

Net Recharge Approach -As the review notes, the net recharge approach is generally used when evapotranspiration parameters and/or surface runoff are difficult to quantify. As this is the case for the modeled area, the net recharge approach was employed in the groundwater model.

4

Retrieved from "https://kanterella.com/w/index.php?title=ML21035A199&oldid=3417275"