Recharge Estimation in Groundwater Modelling - PowerPoint PPT Presentation

Loading...

PPT – Recharge Estimation in Groundwater Modelling PowerPoint presentation | free to download - id: 173275-ZDc1Z



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

Recharge Estimation in Groundwater Modelling

Description:

Recharge Estimation in Groundwater Modelling – PowerPoint PPT presentation

Number of Views:2061
Avg rating:3.0/5.0
Slides: 140
Provided by: mjyr
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Recharge Estimation in Groundwater Modelling


1
Recharge Estimation in Groundwater Modelling
February 10, 2014
Professor Jon Sykes Department of Civil and
Environmental Engineering University of Waterloo
2
Outline
  • Background
  • What is groundwater recharge?
  • Estimation Methods
  • Recharge in Groundwater Modelling
  • Methods and challenges
  • Recharge Estimation Methodology
  • HELP3/ArcView GIS/RDBMS
  • Case Studies
  • Toms River Groundwater Model
  • Grand River Watershed

3
Background
  • What is Groundwater Recharge?

4
Background
  • Hydrologic Cycle
  • Precipitation is the driving force
  • Other processes
  • Interception
  • Stemflow
  • Throughfall
  • Percolation

5
Definitions
  • Recharge Infiltrated water from precipitation,
    irrigation, rivers and lakes, that crosses the
    water table into the saturated zone
  • direct (or natural)
  • localized
  • indirect
  • Discharge Water that exits the groundwater
    system through pumping, or naturally at and near
    streams and lakes

6
Scale Problem
  • Surface processes (e.g. storm events) are
    measured interms of hours and take place over
    large areas
  • Subsurface processes (e.g. groundwater flow)
    evolve very slowly locally and are measured in
    months, or even years
  • Freezing temperatures and snow present additional
    modelling difficulties in northern climates

7
Factors Affecting Recharge
  • Recharge is affected by
  • Climatic conditions
  • precipitation
  • temperature
  • Land surface parameters
  • land use / land cover (LULC)
  • vegetation
  • level of imperviousness
  • Physical properties of the unsaturated zone
  • soil types (heterogeneity)
  • permeabilities

8
Factors Affecting Recharge
  • Precipitation
  • Most important parameter (i.e. driving force)
  • High spatial and temporal variation
  • storm events vary widely in duration, velocity,
    and intensity, while lower temperatures can
    result in snowfall or mixed precipitation
  • Difficult to estimate
  • Measurement is also difficult
  • point measurements are commonly used
  • data is extrapolated using Theissen polygons,
    inverse distance weights, kriging, surface
    fitting...
  • use of weather radar seems promising

9
Factors Affecting Recharge
  • Land Use and Cover
  • Impacts infiltration (e.g. severe salinity
    problems and waterlogging in Australia)

10
Factors Affecting Recharge
  • Land Use and Cover
  • Influence of Vegetation
  • Interferes with passage of precipitation
  • interception
  • throughfall
  • stemflow
  • Difficult to quantify (must account for
    seasonality)
  • Most significant impact through evaporation and
    transpiration (ET)
  • many methods of estimation have been developed
  • Impact of plant roots
  • Leaf Area Index

11
Factors Affecting Recharge
  • Land Use and Cover
  • Influence of Urbanization
  • Increased surface runoff due to imperviousness
  • Leaking sewers and water distribution systems may
    result in increased recharge
  • Over-irrigation of lawns and parks can also
    increase recharge
  • Very difficult to quantify
  • Impacts recharge by altering interception,
    infiltration, surface runoff, and ET

12
Factors Affecting Recharge
  • Overland Flow
  • Rare in humid climates
  • Less intensive rainfall
  • Well developed vegetation
  • Sufficient infiltration capacity of most soils
  • Can occur over short distances during special
    events such as intense snowmelt over frozen
    ground
  • Can also occur in areas of high topographic
    relief with low permeability soils near the
    ground surface

13
Factors Affecting Recharge
  • Infiltration
  • Assumed 1-D (may not be true!)
  • Many methods of estimation
  • Horton (1940)
  • Green and Ampt (1911)
  • Philip (1957)
  • others
  • Subject of many assumptions
  • Only describes the rate of water movement into
    the soil at the ground surface

14
Factors Affecting Recharge
  • Flow in the Unsaturated Zone
  • Governed by soil hydraulic properties and
    moisture conditions (also boundary conditions)
  • Richards equation (1931)
  • Need to know relationship with hydraulic
    conductivity (highly uncertain), pressure head,
    and water content with depth!

15
Factors Affecting Recharge
  • Flow in the Unsaturated Zone
  • Soil Properties
  • hydraulic properties are very sensitive to
    changes in moisture content and pressure head
    distributions

16
Factors Affecting Recharge
  • Flow in the Unsaturated Zone
  • Soil Properties
  • Modelled using relationships by
  • Brooks and Corey (1964)
  • Mualem (1976)
  • van Genuchten (1980)
  • Analysis complicated by occurrence of preferred
    pathways due to plant roots and cracks and
    fissures

17
Factors Affecting Recharge
  • Temperature
  • Subzero temperatures (i.e. freezing) complicates
    everything!!!
  • Frost layer
  • constricts or blocks infiltrating water
  • reduces hydraulic conductivity
  • depends on initial water content of soil before
    freezing takes place, snow depth etc.
  • porosity changes due to expansion by ice
    formation
  • Snow
  • spatial and temporal distribution is complex
  • melting, drifting, etc

18
Factors Affecting Recharge
  • Climate Change
  • Impacts both precipitation and temperature,
    therefore, also recharge
  • Difficult to predict the future
  • General predictions include (IPCC, 2001)
  • increase in global average surface temperature
  • increase in intensity and frequency of extreme
    precipitation events
  • possible reduction in incoming solar radiation
    due to increase in greenhouse gases
  • How will these affect recharge???

19
Why Bother?
  • Many rural areas and municipalities depend on
    groundwater for their drinking water needs
  • Recent events have raised questions about the
    safety and sustainability of our drinking water
    resources
  • Factors stressing the existing supply include
  • improper agricultural management practices
  • rapid municipal development and urbanization
  • contamination from various sources

20
Estimation Methods
21
Estimation Methods
  • Direct Measurement
  • lysimeters
  • Empirical Methods
  • regression and stochastic analysis
  • Water Budget Methods
  • calculate soil moisture balance with recharge as
    residual

22
Estimation Methods
  • Darcian Approaches
  • calculate water movement in the unsaturated zone
    based on Darcys equation and conservation of
    mass (e.g. Richards equation)
  • Tracers
  • e.g. 2H, 3H, 36Cl, Cl-, 14C, 18O, CFCs
  • calculate tracer mass balance

23
Estimation Methods
  • Inverse Groundwater Modelling
  • based on the knowledge of the depth of water
    table and the underlying hydraulic conductivity
    distribution

Conduct mass balance for each model cell
using Darcys Law Problems?
(after Stoertz and Bradbury, 1989)
24
Estimation Methods
  • Combined Modelling
  • surface water balance in conjunction with
    unsaturated zone modelling
  • can account for all important hydrologic
    processes, such as precipitation, surface runoff,
    evapotranspiration, and even snowmelt
  • many examples with varying levels of
    sophistication... (HELP3, UNSAT-H, SWAT)

25
Estimation Methods
  • Others
  • Plane of Zero Flux
  • based on water content analysis
  • Temperature and EM Methods
  • screening tools
  • Stream Gaging
  • stream hydrograph analysis (i.e. baseflow
    separation)

Q
time
26
References
  • Lerner, D., A.S. Issar and I. Simmers. 1990.
    Groundwater recharge a guide to understanding
    and estimating natural recharge. Heise, Hannover.
  • Stephens, D.B. 1996. Vadoze zone hydrology. Lewis
    Publishers, Florida.
  • Robins, N.S. (ed.) 1998. Ground-water Pollution,
    Aquifer Recharge and Vulnerability. Geological
    Society, London, Special Publications, 130.

27
References
  • Hydrogeology Journal (Feb. 2002, Vol. 10, Issue
    1) special issue on recharge
  • Groundwater recharge an overview of processes
    and challenges
  • Choosing appropriate techniques for quantifying
    groundwater recharge
  • Interactions between groundwater and surface
    water the state of the science
  • Using groundwater levels to estimate recharge
  • Recharge and groundwater models an overview
  • Identifying and quantifying urban recharge a
    review
  • Groundwater recharge and agricultural
    contamination

28
Recharge in Groundwater Modelling
  • Methods and Challenges

29
Overview
  • Groundwater flow models are used for numerous
    hydrologic investigation purposes, such as
  • vulnerability assessments
  • remediation designs
  • water quality/quantity assessments
  • Applied across various scales
  • Both steady-state and transient methods are used

30
Overview
  • Three types of models
  • Variably saturated (i.e. Richards equation)
  • Fully saturated
  • Fully integrated (both surface and GW)
  • Both finite element and finite difference methods
    are used (e.g. MODFLOW vs. FEMWATER) in the
    solution

31
Paradigm
Groundwater models often simplify surface water
processes
VS.
Surface water models simplify the movement of
groundwater
32
Methods
  • Engineering Judgment
  • Zoning (with calibration)
  • Recharge Spreading Layer
  • Coupled Models
  • Integrated Models

33
Engineering Judgment
  • Recharge is used most commonly purely as a
    calibration parameter
  • Assume constant recharge or a fraction of
    precipitation
  • Rule of Thumb

Recharge 1/3 of Precipitation
34
Zoning
  • Use zoning to reduce the total number of
    calibration parameters
  • Zones with uniform or constant properties (i.e.
    recharge rates) may be loosely based on
  • type of land use and soils
  • topography
  • climate (large scale models)
  • Objective is to have the minimum number of
    recharge zones (fewer degrees of freedom)

35
Zoning
(after Varni and Usunoff, 1999)
36
Zoning
  • The estimated/assigned recharge rates are then
    adjusted during model calibration (along with
    hydraulic conductivity other model parameters)
    until satisfactory results are obtained
  • i.e. compare observed heads to simulated water
    levels at available monitoring wells
  • Problems?

37
Recharge Spreading Layer
  • First proposed by Therrien and Sudicky (1996)
  • Use a thin, highly conductive surface layer to
    redistribute recharge from areas of low
    permeability to areas of higher permeability
  • Reduces computational problems and imitates
    surface runoff/interflow HOWEVER
  • Not physically based
  • Only used in steady-state analysis

38
Recharge Spreading Layer
(after Beckers and Frind, 2000)
39
Coupled Models
  • Estimate groundwater recharge using a separate
    model and then couple it with a groundwater flow
    model e.g.
  • HYDROLOG/AQUIFEM-N (Chiew et al., 1992)
  • SCS-CN method/NEWSAM (Srinivas et al., 1999)
  • MOZART/NAGROM (Vermulst and De Lange, 1999)
  • SMILE/MODFLOW (Beverly et al., 1999)
  • TOPOG_IRM/MODFLOW (Zhang et al., 1999)
  • SWAT/MODFLOW (Sophocleous and Perkins, 2000)
  • WetSpass/MODFLOW (Batelaan and De Smedt, 2001)
  • and others

40
Coupled Models
  • Various degrees of sophistication
  • from rigorous numerical treatment of many
    hydrologic processes to simple water budget or
    empirical models
  • Main limitations include
  • over-simplification of processes
  • the scaling problem
  • data requirements

41
Integrated Models
  • Avoid problems inherent in coupling
  • Examples include
  • HydroGeoSphere (Therrien et al., 2004)
  • IGSM (Montgomery Watson, 1993)
  • MODHMS (HydroGeoLogic, 2004)
  • MIKE-SHE (Abbott et al., 1986)
  • and others
  • Main problems are parameterization and high
    computational requirements

42
Challenge
  • Fully saturated groundwater models such as
    MODFLOW simulate groundwater flow strictly below
    the water table
  • Recharge must be specified as the top or upper
    boundary condition

What will you do?
43
Challenge
  • Example (Toms River, NJ)

44
Things to keep in mind
  • The spatial scale is constrained by the
    discretization of the groundwater model grid
  • Temporal scale is constrained by the size of the
    time steps
  • Conclusion The most you can do is to assign a
    unique value of recharge for each boundary
    element or grid block during each time step

45
Recharge Estimation Methodology
  • HELP3/ArcView GIS/RDBMS

46
Introduction
  • Study Area Toms River, NJ
  • Highly dynamic groundwater flow field
  • dominant well field
  • physical properties of the aquifer
  • Model was used to investigate the historical
    behaviour and aquifer response over a 30 year
    period
  • Contaminant plume impacting the municipal well
    field

47
Required
  • A physically based and accurate recharge
    boundary condition for the groundwater model
  • Account for high temporal variability in
    recharge
  • Consider freezing temperatures

48
Methodology Overview
Microsoft Access
ArcView GIS
Visual Basic
Groundwater Model
MODFLOW
HELP3
49
Methodology Overview
  • ArcView GIS and MS-Access
  • organize, manipulate, and analyze input and
    output data
  • HELP3 (hydrologic model)
  • core of the methodology
  • calculate recharge rates on a daily basis
  • Visual Basic
  • automation, i.e. pre- and post-processing of
    input and output files

50
HELP3
  • Hydrologic Evaluation of Landfill Performance
    (HELP) Version 3 (Schroeder et al., 1994)
  • 1-D hydrologic model
  • Simulates daily water movement into the ground
    (does not model storm events)
  • Accounts for snowmelt, evapotranspiration,
    vegetative interception, surface runoff, and
    temperature effects
  • Built-in weather generator

51
Why HELP3?
  • Widely accepted, readily available, and easy to
    use many comparative studies
  • Available over the internet at
  • Input data is readily available
  • Fast execution times (can be applied to larger
    areas)
  • 1-D model

http//www.wes.army.mil/el/elmodels/helpinfo.html
52
HELP3 Nuts and Bolts
  • FORTRAN .EXE
  • Uses simple input and output ASCII text files
  • User interface written using BASIC (DOS program)
  • Guidance and default parameter values given on
    screen using interactive menus
  • Simulation period from 1 year, up to 100 years

53
HELP3 Input Data
ET
P
  • Precipitation
  • Temperature
  • Solar Radiation
  • Relative Humidity
  • Average Annual Wind Speed
  • Growing Season Dates
  • Leaf Area Index
  • Evaporative Zone Depth
  • Curve Number
  • Soil Column Information

Interception
Runoff
Snowmelt
Soil Layers
Recharge
54
HELP3 Input Data
  • Weather Data
  • Daily information on
  • precipitation (mm)
  • mean temperature (C)
  • total incoming solar radiation (KJ/m²)
  • Can be generated synthetically using the built in
    weather generator (WGEN) if no actual data
    available
  • Minimum 1 year, up to 100 years

55
HELP3 Input Data
  • Evapotranspiration Data
  • Quarterly relative humidities () (used to
    compute the mean vapour pressure of the
    atmosphere for days without precipitation)
  • Average annual wind speed (km/hr) (at the height
    of 2 m above ground surface)
  • Growing season start and end dates (growth is
    assumed to occur during the first 75 of the
    growing season based on heating units)
  • Default values are provided by the program for
    various locations in the US

56
HELP3 Input Data
  • Leaf Area Index (LAI)
  • The maximum ratio of leaf area to ground
  • Used to compute interception losses and
    transpiration rates for the ET model
  • Also used to empirically adjust the saturated
    hydraulic conductivity of the top-half of the EZD
    to reflect the impact of root channels
  • LAI is adjusted seasonally by the vegetative
    growth model
  • Values range from 0 for bare ground to 5.0 for an
    excellent stand of grass or dense stand of trees
    and shrubbery (HELP3 is insensitive to values
    above 5)

57
HELP3 Input Data
  • Curve Numbers (CN)
  • Surface runoff in HELP3 is computed using the
    Soil Conservation Service (SCS) curve number
    method
  • CN should be estimated based on the average
    antecedent moisture condition (AMC-II)
  • HELP3 can estimate curve numbers based on
  • surface slope and length information
  • description of the soil texture of the top soil
    layer and the vegetation cover
  • CNs are adjusted during simulation to reflect
    changes in soil moisture and temperature

58
HELP3 Input Data
  • Evaporative Zone Depth (EZD)
  • The maximum depth from which water can be removed
    by ET
  • The influence of plant roots generally extends
    below the actual root depth (due to capillary
    suction)
  • EZD influences calculation of many parameters
    and processes
  • storage of water near the surface
  • ET and surface runoff
  • snowmelt
  • vertical percolation
  • EZD must be greater than zero (cm) (but less than
    total column depth)

59
HELP3 Input Data
  • Soil Column Information
  • The soil profile is described by a sequence of
    layers
  • 4 different types in HELP3
  • vertical percolation
  • lateral drainage
  • barrier
  • geomembranes
  • The sequence of layers must follow certain rules
  • For recharge analysis, only need vertical
    percolation layers!

60
HELP3 Input Data
  • Soil Column Information
  • Geometric data
  • total column depth (cm)
  • layer depths (cm)
  • Physical descriptions of
  • total porosity (f)
  • field capacity (FC)
  • wilting point (WP)
  • saturated hydraulic conductivity (KS) (cm/sec)
  • The program provides default values for 42
    different soil/material textures

61
HELP3 Input Data
62
HELP3 Input Data
  • Optional Parameters
  • For the weather generator
  • mean monthly precipitation (mm/mth)
  • normal mean monthly temperature (C)
  • latitude (degrees)
  • For Curve Number estimation
  • surface slope ()
  • slope length (m)
  • Initial volumetric soil moisture content (q)
  • can be specified by user, or computed
    automatically at the start of each simulation

63
HELP3 Solution Method
  • Surface Processes
  • Subsurface Processes

Interception
Surface Runoff
Snow Accumulation
Vegetative Growth
Surface Evaporation
Snowmelt
Transpiration
Soil Evaporation
Frozen Soil
Drainage
64
HELP3 Solution Method
  • Potential ET is modelled by a modified Penman
    method (Ritchie, 1972)
  • Vegetative growth is based on SWRRB model
  • Runoff and infiltration is partitioned using the
    SCS curve number method
  • EZD is modelled using the CREAMS model
  • Snow accumulation and melt is based on the
    SNOW-17 model
  • Moisture movement in the soil is computed using
    Darcys Law and the continuity equation
  • Capillary pressure and moisture content are
    modelled using the Brooks-Corey (1964) equation

65
Obtaining Input Data
How?
  • Precipitation
  • Temperature
  • Solar Radiation
  • Relative Humidity
  • Average Annual Wind Speed
  • Growing Season Dates
  • Leaf Area Index
  • Evaporative Zone Depth
  • Curve Number
  • Soil Column Information

66
Obtaining Input Data
  • The SCS-CN method requires
  • classification of the ground surface type and
    hydrologic condition
  • hydrologic characterization of the soil
  • Therefore, need
  • land use and land cover (LULC) data
  • soil information
  • BONUS!!!
  • LULC data can be used to estimate LAI
  • soil information can be used to characterize the
    soil column
  • both the LULC and soil data can be used to
    estimate EZD

67
Spatial Data
  • Most LULC and soil information is currently
    available in digital format (and free over the
    internet from national and provincial
    agencies!!!)
  • Analysis and manipulation of spatial data is best
    accomplished within the GIS framework (e.g.
    ArcView GIS)
  • Detailed database tables store additional
    information associated with the spatial features
    (e.g. soil database)

68
Other Data Automation
  • MS-Access provides versatile querying and data
    analysis capabilities
  • Can also store other HELP3 input parameters
    (i.e. weather data etc.) in MS-Access
  • Depending on the scale of the input data and the
    size of the study area, the number of unique
    HELP3 input parameter combinations can range from
    a few dozen up to hundreds, or even thousands of
    groupings
  • Use Visual Basic to automate the HELP3 analysis
    of all input parameter combinations

69
Summary
Microsoft Access
ArcView GIS
Visual Basic
Groundwater Model
MODFLOW
HELP3
70
Summary
ArcView GIS
MS-Access
Landcover Map
Leaf Area Index

Soil Map
Soil Database

Combination Map
Curve Numbers Evaporative Soil Depths
Pre-Processing
Weather Data
Precipitation
Temperature
Recharge Boundary
Solar Radiation
MODFLOW
Post-Processing
Evapotranspiration Data
71
Summary
MS-Access
ArcView GIS
Landcover Map
Leaf Area Index

Soil Map
Soil Database

Combination Map
Curve Numbers Evaporative Soil Depths
Pre-Processing
Weather Data
Precipitation
Temperature
Recharge Boundary
Solar Radiation
MODFLOW
Post-Processing
Evapotranspiration Data
72
Summary
ArcView GIS
MS-Access
Landcover Map
Leaf Area Index

Soil Map
Soil Database

Combination Map
Curve Numbers Evaporative Soil Depths
Pre-Processing
Weather Data
Precipitation
Temperature
Recharge Boundary
Solar Radiation
MODFLOW
Evapotranspiration Data
Post-Processing
73
Summary
ArcView GIS
MS-Access
Landcover Map
Leaf Area Index

Soil Map
Soil Database

Combination Map
Curve Numbers Evaporative Soil Depths
Pre-Processing
Weather Data
Precipitation
Temperature
Recharge Boundary
Solar Radiation
MODFLOW
Evapotranspiration Data
Post-Processing
74
Summary
ArcView GIS
MS-Access
Landcover Map
Leaf Area Index

Soil Map
Soil Database

Combination Map
Curve Numbers Evaporative Soil Depths
Pre-Processing
Weather Data
Precipitation
Temperature
Recharge Boundary
Solar Radiation
MODFLOW
Evapotranspiration Data
Post-Processing
75
Summary
ArcView GIS
MS-Access
Landcover Map
Leaf Area Index

Soil Map
Soil Database

Combination Map
Curve Numbers Evaporative Soil Depths
Pre-Processing
Weather Data
Precipitation
Temperature
Recharge Boundary
Solar Radiation
MODFLOW
Evapotranspiration Data
Post-Processing
76
Summary
ArcView GIS
MS-Access
Landcover Map
Leaf Area Index

Soil Map
Soil Database

Combination Map
Curve Numbers Evaporative Soil Depths
Pre-Processing
Weather Data
Precipitation
Temperature
Recharge Boundary
Solar Radiation
MODFLOW
Evapotranspiration Data
Post-Processing
77
Parameter Estimation
  • How to estimate
  • Curve Numbers
  • Leaf Area Index
  • Evaporative Zone Depth

78
Curve Numbers
  • NRCS Technical Release 55 (ftp//ftp.wcc.nrcs.usda
    .gov/downloads/ hydrology_hydraulics/tr55/tr55.pd
    f)
  • CN is an empirical design parameter ranging from
    0 to 100
  • The major factors affecting CN are
  • the hydrologic soil group (HSG)
  • cover type
  • treatment
  • hydrologic condition
  • the antecedent moisture condition (AMC)
  • outlet to drainage system (in urban areas)

79
Curve Numbers
  • Hydrologic Soil Groups (HSG)
  • Good hydrologic condition indicates low runoff
    potential (i.e. high recharge/drainage potential)
  • Common to all soil surveys, therefore should be
    readily available in all soil maps (and/or
    databases)

Group Drainage Approximate Potential Drainage
Rate A High gt 0.76cm/hr B Moderate 0.38
0.76 cm/hr C Low 0.13 0.38 cm/hr D Very
Low lt 0.13 cm/hr
80
Curve Numbers
(NRCS TR-55, 1986)
81
Curve Numbers
(NRCS TR-55, 1986)
82
Curve Numbers
  • Higher CN indicates higher runoff potential
    (i.e. lower recharge potential)
  • In urban areas, the CN can be adjusted for the
    level of imperviousness using (assuming that
    the impervious areas are connected to the
    drainage system)
  • Need to match the land cover description in the
    LULC map with the NRCS-TR55 tables!

83
Leaf Area Index
  • Can be measured but may be very difficult (not to
    mention tedious)
  • Values reported in literature are hard to apply
    to other areas due to
  • variability of vegetation types across different
    areas
  • variability of LAI within the same plant
    community
  • seasonal changes in LAI
  • Individual plants are rarely delineated
    separately in LULC maps (e.g. golf courses are
    represented by a single category)
  • Must estimate an average value for each category
  • Estimation is subjective and relies on good
    engineering judgment

84
Evaporative Zone Depth
  • Can be estimated based on maximum root depth
  • But plant root development is affected by many
    factors and can be highly variable even within
    the same plant community
  • Beside vegetation type, root penetration is a
    function of the soil (e.g. roots generally
    penetrate deeper in sandy soils than in clay)
  • NJ Geological Survey Report 32 provides a table
    of maximum rooting depths for various vegetation
    and soil combinations

85
Evaporative Zone Depth
  • Maximum Rooting Depth (cm) (NJGSR-32, 1993)
  • Similar to LAI, must estimate an average EZD for
    each LULC category

86
Scale Issues
  • Because HELP3 is 1-D it is independent of the
    scale of the input data
  • i.e. the size or scale of the LULC and soil maps
    does not matter (only changes the total number of
    input parameter combinations)
  • For very large areas, where there is significant
    evidence of weather variability across the site,
    the domain must be divided into sub-areas of
    uniform meteorology
  • results in a potentially large increase in the
    total number of input parameter combinations
  • Discontinuities or overlaps may exists between
    adjacent map sheets due to differences in scales
    and mapping methods by different agencies
  • construction of continuous LULC and soil
    coverages for large study areas may be
    potentially difficult and time consuming

87
Case Studies
  • Toms River Groundwater Model
  • Grand River Watershed Recharge Analysis

88
Toms River GW Model
  • Reich Farm Superfund Site
  • Model was used as a forensic tool to investigate
    the historical origin of a contaminant plume and
    its impact on a municipal well field
  • Highly transient groundwater flow field
  • Challenge Require model calibration to spatial
    and temporal distribution of measured groundwater
    heads over 30 years

89
Study Area
Reich Farm
Municipal Wellfield
BARNEGAT BAY
Toms River
Active Domain
Active Domain
Image courtesy of Visible Earth (NASA)
90
Groundwater Model Grid
  • MODFLOW
  • 208 rows
  • 200 columns
  • 4 vertical layers
  • 166400 grid blocks

Reich Farm
Municipal Wellfield
BARNEGAT BAY
Toms River
  • Simulation period from Jan. 1970 Sept. 2002
  • Monthly time steps

Active Domain
91
Boundary Conditions
Legend
Inactive cells Active cells General Head Boundary
(GHB) Rivers Surface Water
Reich Farm
Municipal Wellfield
BARNEGAT BAY
Toms River
Recharge Boundary Condition
92
Recharge Boundary
  • Use the methodology

ArcView GIS
MS-Access
Landcover Map
Leaf Area Index

Soil Map
Soil Database

Combination Map
Curve Numbers Evaporative Soil Depths
Pre-Processing
Weather Data
Precipitation
Temperature
Recharge Boundary
Solar Radiation
MODFLOW
Evapotranspiration Data
Post-Processing
93
LULC Data
  • Obtained from the NJDEP
  • Map based on 1995/97 land classification
  • Also contains level of imperviousness (IS)
    information
  • Mapping based on the modified Anderson et al.
    (1976) classification system
  • a hierarchical framework based on four digits
  • each digit represents the level of classification
    i.e. Level I general to most detailed Level IV
  • eight different land classification categories

94
LULC Data
  • Example
  • The categories were delineated through
    photointerpretation of color infrared (CIR)
    aerial photos
  • The classification was matched to the NRCS-TR55
    categories resulting in 11 unique LULC groups

4 3 2 2
Level IV Level III Level II Level I
4000 Forestland 4300 Mixed Deciduous/Coniferou
s 4320 Mixed with Deciduous Prevalent (gt50
Deciduous) 4322 Mixed with Deciduos Prevalent
(gt50 Crown Closure)
95
LULC Data
LULC Map
96
LULC Data
LULC Map
97
Soil Data
  • The surface soil information was obtained from
    the USDA
  • Soil database contains information on
  • soil type
  • number of layers
  • layer depths
  • soil texture classifications
  • A default HELP3 soil code was assigned for each
    soil using the soil texture classification
  • The total depth of soil layering was assumed to
    be 3 m

98
Soil Data
Soil Map (total of 27 unique soils)
99
Combination Map
ArcView GIS
(11 groups)


LULC Group Map
- LULC group codes - imperviousness (IS)
100
Curve Numbers
  • Assigned based on the LULC categories and
    hydrologic soil groups (adjusted for the
    level of imperviousness)

101
Leaf Area Indexes
  • Assigned based on each of the 11 LULC categories

102
Evaporative Zone Depths
  • Based on both the LULC categories and soil
    texture classification

103
Input Data
Leaf Area Index
Average Vertical Saturated K
Curve Numbers
Evaporative Zone Depths
104
Combinations
  • Assign unique identifier for each LULC/Soil/IS
    combination (684)
  • Link (query) data tables in MS-Access to get
    HELP3 input parameter values

Ats-04-010
Level of imperviousness (IS) LULC Group Soil Type
105
Linking
Ats-04-010
Soils Database
Leaf Area Index
Curve Numbers
HSG
HSG Soil Texture Layer Depth Layer Thickness
Soil Type
(adjust CN)
LULC Group
LULC Group
HELP3 Default Soils
Evaporative Zone Depth
Soil Texture
Porosity Field Capacity Wilting Point Hydraulic
Cond.
Soil Texture
LULC Group
106
Weather Data
  • Actual daily precipitation and temperature
    records were obtained from the NCDC
  • The measurement location was within the study
    area, in the southwest corner of the domain
  • Solar radiation was generated synthetically in
    HELP3 for Edison, New Jersey (with a latitude
    adjustment to 4000 to correspond with the center
    of the model domain)
  • Precipitation, temperature, and solar radiation
    were assumed to be spatially constant across the
    study area

107
Weather Data
Average Monthly Temperature
Average Monthly Precipitation
108
Evapotranspiration Data
  • Use default values given by HELP3 for Edison,
    New Jersey (also assumed to be spatially
    constant)

109
Run HELP3
  • Daily water budget analysis
  • for all unique LULC/Soil/IS combinations

684
Jan 1970
Sep 2002
Each 33 year simulation took approximately 2.6
seconds on a 1.8 MHz P4 computer with 2GM of RAM
(total time 30 min)
110
Results
Average Annual Recharge
Recharge (mm/yr)
45 136 311 439 518 572 610 643 781
111
Monthly Recharge
  • Recharge varies in both space and time

Typical Dry Month May 1981
Typical Wet Month May 1989
Recharge (mm/mth)
0 30 60 100 125 135 145 160 182
112
Average Monthly Results
113
Average Monthly Results
114
Impact of Averaging
Average Monthly Precipitation vs. Recharge
Average Annual Precipitation vs. Recharge
115
Recharge Boundary
  • Calculate block recharge using daily recharge
    rates for each LULC/Soil/IS combination (areal
    averages)

Physically Based Recharge
Recharge (mm/yr)
45 136 311 439 518 572 610 643 781
116
Impact on GW Model
  • Simulated and observed water levels at a
    selected well

117
Impact on GW Model
  • Calibration comparisons

Constant Recharge
Variable Recharge (methodology)
118
Grand River Watershed
  • ? 80 of the people in the watershed rely on
    groundwater as a drinking water resource
  • Need to quantify input to the groundwater system
    i.e. recharge
  • Test the methodology at a much larger scale (i.e.
    watershed scale)
  • There was no attempt to calibrate the results in
    this stage of the study

119
Study Area
Grand River Watershed
O N T A R I O
LAKE HURON
Toronto
LAKE ONTARIO
Grand River Watershed
Buffalo
Detroit
N E W Y O R K
LAKE ERIE
P E N N S Y L V A N I A
O H I O
(image courtesy of NASA Visible Earth)
120
Landcover Map
GRCA Landcover Map
  • - Landsat TM Imagery
  • 25m grid (raster)
  • 15 unique groups

Legend
Built-up (commercial/industrial) Built-up
(residential) Dense forest (conifer) Dense forest
(deciduous) Dense forest (mixed) Pasture/sparse
forest Golf courses Extraction/bedrock/roads Forag
e Marsh Bare agricultural fields Plantation
(mature) Row crops Small grains Open water
121
Soil Data
Assembled from Ontario Soil Surveys
Survey County Scale Year ON13 Perth
63360 1975 ON17 Grey 63360 1981 ON28 Oxford
63360 1961 ON32 Wentworth 63360 1965 ON35 Wellin
gton 63360 1963 ON38 Dufferin 63360 1963 ON43 H
alton 63360 1971 ON44 Waterloo 20000 1971 ON55
Brant 25000 1989 ON57 Haldimand-Norfolk 25000 1
984
ON17
ON38
ON35
ON43
ON44
ON13
ON32
ON28
ON55
obtained from UW Map Design Library Others
obtained from CANSIS website
ON57
PROBLEMS overlaps and gaps between map sheets
122
Soil Map Assembly
Example of gap filling
Waterloo County
Oxford County
Trim and extend based on scale and year (e.g. use
newer and finer scale maps to trim older and
larger scale maps)
123
Final Soil Map
  • Total of 723 unique soil types

124
Combination Map
(15 groups)
(723 soil types)


47,229 unique combinations
Assign - curve numbers - evaporative zone
depths - maximum leaf area indicies
Landcover Map
Soil Map
Combination Map
125
Input Data
Evaporative Zone Depths
Average Vertical Saturated K
Leaf Area Index
Curve Numbers
126
Weather Data
1
  • 13 Zones of Uniform Meteorology (ZUM)
  • delineated by the GRCA
  • Daily values of
  • precipitation
  • minimum and maximum temperatures
  • Continuous records from Jan 1960 to Dec 1999
  • Estimate solar radiation for each ZUM using WGEN
    in HELP3

2
4
3
7
5
9
6
8
10
11
13
12
127
Weather Data
Average Monthly Temperature
Average Monthly Precipitation
128
Weather Data
  • Solar Radiation
  • generated synthetically in HELP3
  • choose Grand Rapids, Michigan (similar climate
    and latitude)
  • solar radiation is affected by latitude,
    precipitation and temperature
  • adjust latitudes for each ZUM during generation

Locations for synthetic solar radiation in HELP3
129
Weather Data Interpolation
  • 293 GAWSER Sub-Basins
  • adjust/estimate weather parameters for each
    Sub-Basin (to avoid discontinuities along ZUM
    edges)
  • use Inverse Distance Squared (IDS) weighting
    scheme

P
S
i
2
d
sub
i
P
i

S
1
2
d
i
i
130
Impact of Interpolation
Average Annual Precipitation
Legend
849 869 889 901 914 928 934 942 947 952 957 964 97
6 995 1022 1045 mm
Sub-Basins
ZUMs
131
Evapotranspiration Data
  • Quarterly Relative Humidities
  • use data from 6 weather stations around/in the
    Grand River Watershed
  • obtained from Environment Canada - Canadian
    Climate and Surface Water Information, and the
    University of Waterloo
  • use Inverse Distance Squared weighting scheme

Mount Forest
Toronto Pearson Intl Airport
UW Weather Station
Hamilton Airport
London Airport
Simcoe
132
Evapotranspiration Data
  • Average Annual Wind Speed
  • use data from 10 weather stations around/in the
    Grand River Watershed
  • obtained from Environment Canada - Canadian
    Climate and Surface Water Information
  • use Inverse Distance Squared weighting scheme

Mount Forest
Elora Research Station
Toronto Pearson Intl Airport
Fergus Shand Dam
Waterloo/Wellington Airport
Hamilton Marine Police
Hamilton RBG
Hamilton Airport
London Airport
Simcoe
133
Evapotranspiration Data
May 5 Oct 7
  • Growing Season Dates
  • estimated for each ZUM based on HELP3 values
  • use ZUM values for each Sub-Basin

May 5 Oct 7
May 4 Oct 8
May 4 Oct 8
May 3 Oct 9
Station Start End
May 3 Oct 9
La Crosse, WI Apr 29 Oct 9 Madison, WI May 4 Oct
6 Milwaukee, WI May 10 Oct 10 Chicago, IL Apr
27 Oct 17 Fort Wayne, IN Apr 26 Oct 16 Grand
Rapids, MI May 3 Oct 10 Lansing, MI May 3 Oct
10 Detroit, MI May 1 Oct 13 Toledo, OH Apr 29 Oct
13 Cleveland, OH Apr 30 Oct 17 Buffalo, NY May
6 Oct 12 Syracuse, NY May 4 Oct 11
May 3 Oct 10
May 3 Oct 10
May 3 Oct 10
May 2 Oct 11
May 2 Oct 11
May 5 Oct 12
May 6 Oct 12
134
Run HELP3
  • Daily water budget analysis
  • for all unique Landcover/Soil/Weather combinations

47,229
Jan 1960
Dec 1999
(Total computation time 37 hours)
135
Results
Recharge Histogram
Legend
0 98 123 170 191 209 229 253 290 345 421 mm
136
Results
Legend
0 98 123 170 191 209 229 253 290 345 421 mm
137
Monthly Recharge
December 1992
December 1998
Legend
0 1 30 58 71 81 88 94 100 107 119 mm/month
138
Average Monthly Recharge
Time (years)
139
Average Monthly Recharge
140
Runoff and ET
Average Annual Evapotranspiration
Average Annual Runoff
Legend
208 378 447 490 505 521 536 551 571 597 683 mm
Legend
62 116 139 163 186 208 230 257 299 383 642 mm
141
Climate Change
  • Investigate the impact of potential future
    climate change on groundwater resources in the
    Grand River Watershed
  • Perturb HELP3 input parameters based on general
    observations by the IPCC (2001)

Case Description Base Actual data (Jan 1960 Dec
1999) 1 Precipitation 5 in winter (Dec, Jan,
Feb) 2 Precipitation 20 in winter (Dec, Jan,
Feb) 3 Precipitation 20 for all months
4 Temperature 0.016C/year 5 Temperature
0.070C/year 6 Solar radiation 2 for all
months 7 Combination of Cases 1, 5, and 6
8 Combination of Cases 3, 5, and 6
142
Results
  • Surface Runoff (difference from Base Case)

143
Results
  • Evapotranspiration (difference from Base Case)

144
Results
  • Recharge (difference from Base Case)

145
Results
  • Average Monthly Differences (Case 8)

146
Results
  • Change in Recharge
  • Average Base Case Watershed Recharge 189
    mm/year
  • Average Case 8 Watershed Recharge 289
    mm/year
  • The average annual increase in watershed
    recharge due to climate change over the 40 year
    study period is approximately 100 mm/year

147
Climate Change Summary
  • Increasing precipitation will increase runoff,
    ET, and recharge rates
  • Global warming will reduce surface runoff by
    reducing ground frost and shifting springmelt
    toward winter months
  • Decrease in incoming solar radiation will have a
    minimal impact
  • Climate change will increase recharge from 189
    mm/yr to 289 mm/yr (on average over 40 years)

148
Other Applications
  • Can use the recharge methodology to study the
    impact of urbanization (i.e. land use change) on
    groundwater recharge rates

149
Sensitivity Analysis
  • Determine the HELP3 input parameters that have
    the most impact on the estimated recharge rates
  • Use normalized sensitivity analysis
  • Adjust volumetrically to account for relative
    contribution across the domain
  • Use the Toms River Study Area for the analysis

150
Average Monthly Sensitivity
151
Average Monthly Sensitivity
152
Average Monthly Sensitivity
153
Average Monthly Sensitivity
154
Cumulative Sensitivity
155
Cumulative Sensitivity
156
Cumulative Sensitivity
157
Cumulative Sensitivity
About PowerShow.com