MPRWA DEIR Review 2 Subsurface Intakes

download MPRWA DEIR Review 2 Subsurface Intakes

of 15

Transcript of MPRWA DEIR Review 2 Subsurface Intakes

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    1/36

    Review of Subsurface Seawater Intakes

    DEIR, Monterey Desal Project

    DRAFT Presentation to Monterey

    Peninsula Regional Water Authority

    23 June 2015

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    2/36

    Outline

    Overview of Project and Proposed Subsurface Seawater

    Intakes (SSIs)

    Overview of Hydrogeologic Setting

    Site-Specific Feasibility Evaluation of SSIs

    − Borings

    − Pumping Test

    − Groundwater Models

    Findings

    Discussion of Assumptions and Uncertainties

    Conclusions

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    3/36

    Proposed Production and Required Feedwater

    Option 1 (Proposed Project)Production: 9.6 mgd (10,754 AFY = 6,667 gpm)

    Feedwater: 24.1 mgd (26,997 AFY = 16,736 gpm)

    from 10 Slant Wells at CMEX Site

    Option 2 (Project Variant)

    Production: 6.4 mgd (7,169 AFY = 4,444 gpm)

    Feedwater: 15.5 mgd (17,363 AFY = 10,764 gpm)

    from 7 Slant Wells at CMEX Site

    Intent is to maximize contribution to the intakes from

    the ocean and minimize contribution from inland

    coastal aquifers.

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    4/36

    Schematic Example of a Slant Welleosyntec )

    consultants

    r

    il

    l

    Rrg

    • •

    ·

    F ~ s t ~ ·

    1 •

    ;

    Seawater

    .

    • • •

    ••

    Sub

    sea

    q

    ui

    fer

    .

    / .

    .

    .

    • •

    .

    . .

    .

    • •

    • •

    · SlantWell

    j

    / .

    • • • • • • •

    : .......

    • • • •

    .

    .

    Ceosyntec com

    engineers 1 scientists 1 innovators

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    5/36

    Proposed Slant Well Locations at CMEX Siteeosyntec )

    consultants

    7 Apr

    '__,-r.oa_ l t U i p ~ l l l $ " - t : I U z . - N .

    O l O U . G t C l H l l . N O . ~ s . r . u .

    b I ~ , _ , _ . _

    1

    l CO

    00 4

    5

    i lk

    CasJng

    w lt

    screen

    H..,.

    Tide

    IDOC, NOAA«>L . lOU)

    DRAFT

    Figure47

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    6/36

    Coastal Aquifer Hydrostratigraphic Units

    (Figure 4.4-2 of DEIR)

    Geosyntec

    C

    consultants

    No rth

    Moss landing

    Potrero Road

    Sal inas Valley

    Groundwater asin

    Perched A Aquifer

    ~ = = : : : : : : : : : _ _ ~

    180 Foo

    t Aquifer

    180 FTE Aquifer

    4oo Foot Aquifer

    400-Foot Aqultard

    900-t-oot Aquifer

    1Ox Ver

    tica

    l Exilgge

    mtlon

    South

    F

    ort

    O

    rd Du

    ne omplex

    CEMEX

    180/400-Foot Aquitard

    Geosyntec com

    engineers

    I

    scientists

    1

    innovators

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    7/36

    Groundwater Elevations

    in Salinas Valley 180-ft Aquifer

    (Figure 4.4-5 of DEIR)

    Groundwater levels

    well below sea level

    several miles inland

    due to overdraft of

    aquifers.

    MAKE LABELS

    LARGER

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    8/36

    Extent of Seawater Intrusion

    in Salinas Valley 180-ft Aquifer

    (Figure 4.4-9 of DEIR)

    Chloride > 500 mg/L

    extends 8 miles

    from the coast

    (2013).

    The SVGB is

    hydrologically

    connected to the

    Monterey Bay by

    ocean outcrops of

    the 180-Foot and

    400-Foot Aquifers

    that outcrop a few

    miles offshore.

    .

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    9/36

    Groundwater Elevations

    in Salinas Valley 400-ft Aquifer

    (Figure 4.4-6 of DEIR)

    Groundwater levels

    well below sea level

    several miles inland.

    MAKE LABELS

    LARGER

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    10/36

    Extent of Seawater Intrusion

    in Salinas Valley 400-ft Aquifer

    (Figure 4.4-9 of DEIR)

    Chloride > 500 mg/L

    extends 3.5 miles

    from the coast

    (2013).

    .

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    11/36

    Feasibility Evaluation and Design of

    Subsurface Intakes

    Site-Specific Properties and Testing

    6 exploratory borings− water quality samples in 15 aquifer zone locations

    −grain-size analysis and calcs of hydraulic conductivity

    − laboratory testing of Kh and Kv on core samples

    − geophysical logs

    7 monitoring well clusters

    Long-term pumping test of first slant well currently in

    progress.

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    12/36

    Geologic Cross Section of CMEX Area

    (Figure 4.4-3 of DEIR)

    eosyntec

    ()

    consultants

    Station24.000 t

    o1

    O O)S.Sectlon1·

    1

    IWESTl

    ~

    J.OO

    so

    0

    ..so

    -100

    l

    ~

    -*

    i

    ·200

    ·2SO

    300

    ~ ~ o n

    -400

    )

    ~

    -4.1.0

    D •o

    i

    D ,,..

    E:::J ....

    -500

    ll

    :

    D

    or

    -

    sso

    C l ~ ~

    24.000

    25,000

    <

    <

    t

    I

    ~ ~

    n.J0 J (l(t(J

    lo.

    ooo

    Station30 000 ftof

    Cro.s.s 5

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    13/36

    Feasibility Evaluation and Design of

    Subsurface Intakes

    Groundwater flow modeling of slant well Intakes at coastal

    margin provides predictions of

    Portions of contribution to subsurface intakes from the

    ocean and from inland

    Potential influence of the pumping on coastal margin

    aquifers

    drawdown of groundwater levels− change in groundwater flow

    − change sea water intrusion rates

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    14/36

    Groundwater Modeling

    Three groundwater models of different scale

    Geosyntec )

    consultants

    CAUFORN lA M £ ~ 1 C A N WARR

    JuJ

    J4

    it

    0

    0

    NorthMar lna

    Grou

    nd Water

    Mode

    l (NMGWMI

    ' '--

      '·crwe. ~ ~ P t - c u s , r - t v

    CM 1 

    Qt

    :l5C lNa . 1 p p e r t ~ n ,

    N

    AI - ; l l h . . ,

    0 s

    10

    ~ - - -

    MOHTtREY P£N i

    ttSULA

    WATtR SUPPLY PROJECT

    RESU

    LTSO Tt5T SlANTWEU PREDICTJV£ SCENARIOS USING TM£ CIMEX ~ £ A

    MODEL

    CHtet lc U a ..

    ,

     ..

    ..... ~

    -

    ·

     -

    '

    -

    -

    ........ -

    .......

    ,,... .

    DRAFT

    ·-

     

    ..

    GROUND WATER

    MODELS

    .

    GEOS IE

     

    E

    --.... ..--

    C l : : a :

    l i : S : s a a t ~ · '

    v.o

    m a

    Gli

    trn

    llf liQi 0 1 • • R l i { ~

    .po.. .w t>n

    Mo

    ru:crtns Wd

    :.nt . ~ l h u c : l

    Bl

     r

    nk Cu

    w

     

    g

    a ~ n

    Figu

    re 1

    Ceosyntec.com

    engineers 1scientists 1 innovators

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    15/36

    Groundwater Models of Three Scales

    Regional Salinas Valley Integrated Groundwater and

    Surface Water Model (SVIGSM)

    North Marina Groundwater Model (NMGWM, Geoscience2013-2015)

    CMEX Model (CM, Geoscience, 2014)

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    16/36

    Salinas Valley IGSM

    Regional Salinas Valley Integrated Groundwater and

    Surface Water Model (SVIGSM)

    − Entire Salinas Valley Basin Management Tool

    (Montgomery Watson, 1994; WRIME, 2008)− 650 sq mi model domain with ~0.4 sq mile cell size

    − 3 Layers: 180-ft, 400-ft, 900-ft aquifers.

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    17/36

    North Marina GWM

    North Marina Groundwater Model (NMGWM, Geoscience

    2013-2015)

    − MODFLOW, MT3D, SEAWAT More detailed designed as

    tool to evaluate feasibil ity and potential impact of SSIs

    − coverage of 149 sq mi area along coast; Offshore to 5 mi

    and 5+ mi inland; 200 x 200 ft cell size.

    − 8 Layers: Benthic Zone, Dune Sand, 180-ft and 180 ft equiv,

    400-ft, 900-ft aquifers and intervening aquitards.

    −Uses results of SVIGSM as boundary conditions.

    − Transient runs with variable climate conditions for 63 years.

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    18/36

    CMEX MODEL

    CMEX Model (CM, Geoscience, 2014)− SEWAT. More detailed tool to evaluate feasibility of SSI in

    CMEX area.

    − 4 sq mile domain. 20 x 20 ft cell size. Offshore to 4800 ft

    and ~6000 ft inland− 12 layers:

    Benthic Zone (constant sea level),

    Dune Sand, 180-ft and 180 ft equiv,

    400-ft, 900-ft aquifers and intervening aquitards.

    − Uses results of NMGWM for boundary conditions.

    − Will be calibrated to long-term slant well pumping test

    − Then NMGWM updated

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    19/36

    Model Drawdown in Dune Sand Aquifer

    Model-calculated

    maximum inland

    extent of 1 ft

    lowering

    (drawdown) ofgroundwater due

    to project

    pumping is ~5

    miles.

    (1 ft contours of

    drawdown)

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    20/36

    Model Drawdown in 180-ft Aquifer

    Model-calculatedinland extent of

    1 ft lowering

    (drawdown) of

    groundwater dueto project pumping

    is ~7 miles.

    (1 ft contours ofdrawdown)

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    21/36

    Model Drawdown in 180-ft Aquifer

    and Locations of Wells

    (Figure 4.4-15 of DEIR)

    eosyntec )

    consultants

    -   ' llquoft<

    r

    ,oo

    . ,

    wtll

    M

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    22/36

    Drawdown Impact is Less Than Signif icant

    No local wells within the area of influence would beadversely impacted by the drawdown caused by project

    pumping.

    − Pumps and screens are deeper than the predicted drawdown,

    Shallow wells no longer used, or− Wells are screened in a deeper aquifer with limited hydraulic

    connection to the Dune Sands or the 180-Foot Equivalent

     Aquifers.

    − The nearest municipal water supply wells (Marina Wells 10,

    11, and 12) are more than 2 miles to the SE and screened inthe 900-Foot Aquifer.

    Consequently, the DIER concludes that impact of the

    project on neighboring, local groundwater wells is less than

    significant.

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    23/36

    Proportion of Ocean Water and Freshwater  

    Proportion of Ocean Water (X) and Inland Fresh Water

    (1-X) calculated from model chloride content of intake

    water (Feedwater Salinity FS).

    − Ocean Water Salinity (OWS) Chloride Conc = 33,500 mg/L

    − Inland Groundwater Salinity (IS) Chloride Conc = 440 mg/L

    For example Intake Salinity of

    32,000 mg/L consists of 95%

    Ocean Water and 5% InlandGroundwater.

    X * OWS + (1-X) IS = FS

    X * OWS + IS - X*IS = FS

    X*OWS - X*IS = FS - IS

    X(OWS-IS) = FS - IS

    X = (FS- IS)/(OWS-IS)

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    24/36

     Inland Groundwater Take is Replenished 

    Based on NMGWM pumping simulations, the intake

    water consists of

    − 94.5% ocean water, and

    − 5.5% inland groundwater which is 1,458 AFY.

    The project will return fresh water, which is equal to the

    portion of inland groundwater pumped, to the Salinas

    Groundwater Basin via the Castroville Seawater Intrusion

    Project (CSIP) ponds.

    Thus, the DEIR concludes the project will result in no

    net depletion of inland groundwater.

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    25/36

    Potential Influence on Contaminant Plumes

    at Fort Ord 

    DEIR Fig 4.4-18 Drawdown in180 and Ft Ord Plumes

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    26/36

    Potential Influence on Contaminant Plumes

    at Fort Ord 

    OU1 TCE A-Aquifer Plume (TCE plume 2.25 mi SE):

    Less than significant because drawdown of project

    pumping much less than local remedial pumping

    OUCTP A-Aquifer Plume (carbon tetrachloride plume 2

    mi SE): bioremediation in progress. Cal Am monitoring

    and mitigation if needed.

    OUCTP Upper 180 ft Aquifer Plume (3 mi SE). Less

    than significant because drawdown of project pumping

    much less than local remedial pumping

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    27/36

    The Project Reduces Inland Extent of

    Sea Water Intrusion 

    Project pumping accelerates seawater intrusion to the

    SSIs in the CEMEX area, but reduces the rate of sea

    water intrusion further inland.

    Project coastal margin pumping locally reverses existinginland flow of groundwater and draws some inland

    groundwater toward the coast.

    Thus the project decreases seawater intrusion to

    inland aquifers.

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    28/36

    The Project Reduces Inland Extent of

    Sea Water Intrusion 

    DEIR Fig 4.4-16

    Exist ing Condit ions (No Project) 24.1 MGD Pumping

    Inland flow direction in Coastal Aquifers Local Reversal of Flow toward Ocean

    eosyntec t

    consultants

    engineers scicnlists

    I

    innov tors

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    29/36

    Conservative Model Assumptions 

    Offshore No-Flow Boundary Conditions 

    Offshore portion of Model Layer 1 (“Benthic Layer”) is

    constant sea level elevation and salinity, but the offshore

    boundary conditions for all the other layers are no flow

    boundaries rather than constant sea level elevation.

    The large offshore extent of the model layers provides a

    “reservoir” of “sea-water groundwater” beneath the sea

    floor, but no flow boundaries result in conservatively large

    model contribution of inland flow to pumping beneath thecoastal margin.

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    30/36

    Conservative Model Assumptions 

    Conservatively Low Kh between Dune sand and 180-FTE Aquifers at Coastline 

    Borings at the CMEX site show that low permeability clay

    layers between Dune Sand and 180-FTE Aquifers, which

    are present inland, do not extent offshore.However, the model includes a relatively low permeability

    layer (Model Layer 3, Kh = 5 ft/d) between the Dune Sand

    and 180 FTE Aquifers extending offshore.

    Thus, the model may underestimate the hydraulicconnection between the 180-FTE Aquifer with the the

    Dune Sand Aquifer and the Ocean.

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    31/36

    Conservative Model Assumptions 

    ConservativelyLow Kh (5 ft/d)

    between Dune

    sand and 180-FTE

     Aquifers atCoastline

    (Model Layer 3) 

    P t ti ll N C ti H d li

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    32/36

    Potentially Non-Conservative Hydraulic

    Conductivity Values in Models 

    Horizontal Hydraulic Conductivity (Kh) values (340 and 114ft/d) assigned to the Dune Sand and 180 FTE Aquifers at

    coastal margin and offshore may be optimistically high.

    Vertical Hydraulic Conductivity (Kv) value (10 ft/d) assigned

    to some of the Dune Sand aquifer may be optimistically high.

    Thus, the models may overestimate hydraulic connection

    between the 180-FTE and Dune Sand Aquifers with the

    Ocean.

    May be appropriate to update the NMGWM properties

    based on calibration of CMEX model to long-term

    pumping test of Slant Well.

    P t ti ll N C ti H d li

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    33/36

    Potentially Non-Conservative Hydraulic

    Conductivity Values in Models 

    Hi Kv (10 ft/d) for

    Dune Sand Aquifer (Model

    Layer 2) may

    overestimate

    hydraulicconnection to the

    Ocean 

    P t ti ll N C ti L ti f

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    34/36

    Potentially Non-Conservative Location of

    Slant Wells in Models

    The models represent the slant wells extending wellbeneath the sea floor, but the actual slant wells may not

    reach that far.

    Test slant well barely reaches the ocean margin.

     At the ocean margin the well screen is more the 200 ft

    beneath sea level.

    Thus, the models may overestimate hydraulic connection

    between the slant well intakes and the Ocean.May be appropriate to update the models with the

    actual Slant Well locations, or conduct sensitivity

    model runs.

    Substantial Separation between Slant Well

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    35/36

    Substantial Separation between Slant Well

    Intakes and the Ocean Floor  

     Actual geometry of installed slant well.

    Geosyntec )

    consultants

    w e . \ ~ d Beva Yon

    SJJOIE A i Je

    2

    n

    am9

    l

    5 6 5 t f ; . .

    ------y_ -------------- -------------------  \ i ·

    .-.. -. . .--.----

    I _..

    ..

    ,

    ___ .--

    -----

    -

     

    _______

    o

    \

    eanSea le vel

    Beach

    Ceosyntec.com

    engineers 1 scientists 1 innovators

  • 8/21/2019 MPRWA DEIR Review 2 Subsurface Intakes

    36/36

    Conclusions Regarding Project Intakes

    The models provide reasonable simulations of pumpingfrom the subsurface intakes.

    The contribution of inland fresh groundwater to the

    proposed pumping beneath the coastal margin is minor and

    can realistically be returned to the Salinas Basin.

    The potential impact to inland wells is not significant.

    The project pumping would decrease sea water intrusion to

    inland aquifers.Updates to the model predictions can be made based on

    the long-term pump testing currently in progress at the

    slant well.