HYDROLOGY II

CIV4105

HYDROLOGIC SYSTEMS ANALYSIS

By Betty Namugenyi

02/17/2025 1
 CIV 4105: HYDROLOGY II

Contents

[Link] analysis in hydrology

[Link] of models
[Link] models
[Link]-run off models

02/17/2025 27
 CIV 4105: HYDROLOGY II

SYSTEM ANALYSIS IN HYDROLOGY

► A system is set of connected parts that form a whole. A systems analysis is

therefore a simplified means of representing a phenomenon into a set connected
parts in order to understand it well.

► The hydrologic cycle may also be treated as a system whose components are
precipitation, evaporation, runoff and other components of the hydrologic cycle.
These components can be grouped into subsystems of the overall cycle and in
order to analyze the total system.

► Below is the global hydrologic cycle represented as a system.

02/17/2025 27
02/17/2025 4
 A model defined as simplified representation of a complex system. Therefore, a
hydrologic system model is an approximation of the actual system, its outputs
and inputs are measured hydrological variables and its structure is a set of
equations linking the inputs and out puts.

If I(t) = Input into a hydrologic system.

Q (t) = Output

Then, Q (t) =  I(t)

Where  = transformation equations of the system = model structure = process.
Hydrologic systems models convert input variables into output variables using a set
of physical, chemical and biological processes.

02/17/2025 28
2. Classification of models

There are basically two classification of models:

▪ Deterministic or physical models: These models provide a fixed mathematic

relationship between the model inputs and outputs. Here, a given input will
always produce the same output. The models do not represent random processes
that may be present.

▪ Stochastic or statistical models: These models seek to identify statistical

probabilities of hydrological events like rainfall and floods and to predict
probability of a given outcome. They account for variability of the response of a
system. These type of models include regressions, correlations, probabilistic
model

02/17/2025 28
 Deterministic /Physical models

Deterministic models are also grouped into two types: the empirical and conceptual
models

a) Empirical Models:

Empirical models are developed by analyzing a large set of data based on

observation and experiments.

Example 1: The rational method given by Q = C.A.I where Q is the peak

discharge, C is the runoff Coefficient, A is the area Area, I is the Intensity

Example 2: The RUSSLE equation for estimating soil erosion.

02/17/2025 28
b) Conceptual models:

Conceptual models combine component theories on a continuous time and

based on physical mechanisms and laws. Such laws include Newton's law of
Motion.

Sometimes, conceptual models are physical models because their parameters

often have a physical meaning.

Example 1: The rational method given by Q = C.A.I where Q is the peak

discharge, C is the runoff Coefficient, A is the area Area, I is the Intensity

Example 2: The RUSSLE equation for estimating soil erosion.

02/17/2025 29
Other classification of deterministic models:

[Link] models:

Lumped models do not account for spatial distribution of input variables

and the parameter characterizing the processes.

[Link] models:

They describe spatial variability through subdivision of the system into

smaller spatial units.

2.3 Linear Models

Linear models are those models described by a linear differential equation. If

x(t) = input variable and y (t) = output of the system, then a single input/output
system is linear if it is described by:

02/17/2025 29
The system would not be linear if this equation contained terms with products
of x and its differential or powers other than the first power.

For these models, the principle of super position holds. That is to say if x1 and x2
are inputs and y1 and y2 are respective outputs, then y1+y2 will be the output of
the input x1 and x2.

02/17/2025 29
 Rainfall Runoff Models
❖Hydrological cycle has many interconnected components: runoff connects precipitation
to stream flow.

❖Surface runoff → some precipitation does not infiltrate into the soil & runs across the
land surface into surface waters (streams, rivers, lakes or other reservoirs).

❖runoff is important in balancing the hydrological cycle→ returning excess precipitation

& controlling how much water flows into stream systems,

❖Surface runoff is key in monitoring water resources, solving water quality & quantity
problems such as flood forecasting, ecological and biological relationships in the water
environment

❖Runoff→ main driver in contaminant transport, excess nutrients & pesticides from
agricultural lands washed into waterways by rain events.

02/17/2025 11
❖information helps water resource managers account for the pollution in water resources
due to runoff.

❖Surface runoff modeling is used to understand catchment yields & responses, estimate
water availability, changes over time, and forecasting

❖Choosing a rainfall-runoff model is based on

i. modeling purpose such as understanding & answering specific questions about the
hydrological process;

ii. assessing the frequency of runoff events;

iii. or estimating runoff yield for management purposes

❖Identifying the priorities of modeling and the limitations of data availability, time, &
budget for models help to narrow the choices & ensure that the model is the best for the
intended purpose.

02/17/2025 12
❖Modeling runoff helps gain a better understanding of hydrologic phenomena & how
changes affect hydrological cycle

❖Some runoff models & data are not readily available to public, may have missing
information which decreases their usefulness, & can be time consuming or cumbersome to
access.

❖To overcome this problem in accessing water resource data, web services allow
researchers, managers, and the public to become more familiar with the data and better
informed to make improved decisions.

❖Improving model and data access can result in reduced duplication of efforts and save
time and money.

❖Accessing web services from data providers, downloading runoff data, shifting time-series
to local time zones as needed, computing statistics, and flagging missing values are
improvements needed in the modeling community.

02/17/2025 13
 Comparison of the basic structure for rainfall-runoff models
Empirical Conceptual Physical
Non-linear relationship
Simplified equations Physical laws and
between inputs and
Method that represent water equations based on real
outputs, black box
storage in catchment hydrologic responses
concept

Small number of
Easy to calibrate, Incorporates spatial and
parameters needed, can
Strengths simple model temporal variability,
be more accurate, fast
structure very fine scale
run time
Large number of
No connection between Does not consider
parameters and
Weaknesses physical catchment, spatial variability
calibration needed, site
input data distortion within catchment
specific

In ungauged When computational Have great data

Best Use watersheds, runoff is the time or data are availability on a small
only output needed limited. scale

Curve Number, HSPF[b], MIKE-SHE[a],

Examples Artificial TOPMODEL[a], KINEROS[c], VIC[a],14
02/17/2025
Neural Networks[a] HBV[a], Stanford[a] PRMS[d
 Spatial Processes
❖The spatial processes in runoff models provide a means of representing the catchment for
modeling.

❖They are based on input data & how runoff is generated & routed over the catchment.
❖The spatial structure of catchment processes in rainfall-runoff models can be categorized as
lumped, semi distributed, & fully distributed.

i. Lumped models do not consider spatial variability within the catchment;

ii. Semi-distributed models reflect some spatial variability;

iii. Fully distributed models process spatial variability by grid cells.

❖Semi-distributed models take spatial variability into consideration at smaller scales than lumped
models, but do not calculate runoff at every grid cell.

❖Spatial interpretation in a lumped model, a semi-distributed model, and a distributed model are
shown in Figure
02/17/2025 15
 Spatial Processes

❖Visualization of the spatial structure in runoff models. A: Lumped model, B: Semi-

Distributed model by sub-catchment, C: Distributed model by grid cell.
❖Runoff is calculated for each sub catchment at the pour point represented by the black dots in
Figure B. Distributed models calculate runoff for each grid cell, while lumped models
calculate one runoff value for the entire catchment at the river outlet point represented by the
black dot in Figure A.
02/17/2025 16
 Comparison of the spatial structures in rainfall-runoff models
Lumped Semi-Distributed Distributed
Spatial variability is
disregarded; entire Series of lumped and Spatial variability is
Method
catchment is modeled as distributed parameters accounted for
one unit
Both averaged and
All averaged data All specific data
Inputs specific data
by catchment by cell
by sub-catchment
Fast computational time, Physically related to
Represents important
Strengths good at simulating hydrological
features in catchment
average conditions processes
A lot of assumptions, loss Averages data into sub
Data intense, long
Weaknesses of spatial resolution, not catchment areas, loss
computational time
ideal for large areas of spatial resolution
Physically
Conceptual & some
distributed
Empirical and conceptual physical models,
Examples models,
models, machine learning TOPMODEL[a],
MIKESHE[c],
SWAT[b]
VELMA[d]
02/17/2025 17
 Lumped models
❖Lumped models treat the catchment area as a single homogenous unit. Spatial variability
of catchment parameters is disregarded in lumped models.

❖By assuming homogeneity over the catchment, lumped models lose spatial resolution of
the input data; for example, using the mean soil storage and uniform precipitation
amounts.

❖A lumped model is designed to simulate total runoff & streamflow at the outlet point, not
specific flows within a catchment

❖For this reason, lumped models adequately simulate average runoff conditions with fast
computational times.

❖Lumped models include a lot of assumptions about hydrological process & tend to over-
or underestimate runoff values

❖Empirical & conceptual models are usually run spatially as lumped

02/17/2025 18
 Semi-distributed models
❖Semi-distributed models are variations of lumped models, with features of distributed models.
❖They can consist of a series of lumped parameters applied in a quasi-spatially distributed
manner.

❖For example, semi-distributed models can have data that are separated within a catchment but
homogenous within a sub-area.

❖Sub-areas represent important features in a catchment including soil type, drainage time, and
vegetation zone, which combines the advantages of lumped and distributed models.

❖Semi-distributed models consider spatial variability & land use characteristics without an
overwhelming model structure

❖The benefits of a semi-distributed model are fast computational time and the ability to use
less data and fewer parameters than a distributed model.

❖Examples of semi-distributed models are SWAT and TOPMODEL

02/17/2025 19
 Distributed runoff models
❖Distributed runoff models are the most complex because they account for spatial
heterogeneity in inputs and parameters.

❖Fully distributed models separate the model process by small elements or grid cells.
❖Each small element (or cell) has a distinct hydrological response and is calculated separately,
but incorporates interactions with bordering cells.

❖By calculating runoff for every grid cell, the model provides detailed runoff information at
various points within the catchment

❖This comprehensive information, helps to understand pollutant & sediment transport within a
watershed along with capturing spatial and temporal variability of the hydrological process

❖Distributed models are data-intensive, with all input data distributed spatially and temporally.
❖Inputs needed for a typical distributed model are Digital Elevation Models (DEM); land use
imagery 02/17/2025 20
 Uncertainty

❖A runoff model is a simplification of a physical process; therefore, all models are uncertain to
some degree.
❖Rainfall-runoff model uncertainty can come from the:
i. observed data,
ii. natural uncertainties,
iii. parameter estimation,
iv. calibration,
v. or model assumptions.
❖Input data are a major source of uncertainty for rainfall-runoff models because they rely
heavily on input data and physically based parameters
❖As a result of the imbalance of model parameters and observed measurements, the
equifinality problem creates different optimal parameter sets that lead to good model
performances without having parameters with physical numerical meaning.

02/17/2025 21
 Uncertainty
❖The challenging aspect of validating models with observed data is that observed discharge
data cannot be directly compared to modeled runoff values because modeled runoff does
consider subsurface interactions and discharge data does not give information about
spatial distribution of runoff within the catchment.

❖Results of rainfall-runoff model comparisons may be contradictory and hard to interpret

because models differ in so many aspects

❖Models are usually created to answer specific questions and thus cannot be compared in a
general way.

02/17/2025 22
 Modelling in Water Resources

Types of Water Models

• The water models are classified according to their specific areas of application

i. Hydrology, Rainfall/Runoff Models (water enters, moves through, leaves a

catchment)

ii. Hydraulic, Biophysical Process Models (depth, speed, pressure)

iii. Planning, Water Resource Systems Models e.g. WEAP, Aquarius

02/17/2025 23
 Systems Model

Critical questions: How should water be allocated to various uses in time of

shortage?

02/17/2025 24
 Systems Model

Critical questions: How can these operations be constrained to protect the

services provided by the river?

02/17/2025 25
 Systems Model

Critical questions: How will allocation, operations and operating constraints

change if new management strategies are introduced into the system?

02/17/2025 26
 Urban Storm Drainage Models

❖The models applicable to design of storm sewer systems can be classified as

1. Design models,

2. Flow prediction models,

3. Planning models.

02/17/2025 27
 Flow prediction models.

❖Simulate the flow of storm water in existing systems of known geometric sizes or in
proposed systems with predetermined geometric sizes.

❖Most flow prediction models simulate the flow for a single rainfall event, but some can
simulate the response to a sequence of events.

❖The simulation might be for historical, real-time, or synthetically-generated storm events.

❖ At least some simple hydraulics is considered in most models. A model may or may not
include water quality simulation.

❖The purpose of a flow simulation may be to check the adequacy & performance of an
existing or proposed system for flood mitigation & water pollution control, to provide
information for storm water management, or to form part of a real-time operational control
system.

02/17/2025 28
 Planning models.
❖Used for broader planning studies of urban stormwater problems, usually for a relatively
large space frame and over a relatively long period of time.

❖The quantity & quality of storm water is treated in a gross manner, considering only the mass
conservation of water & pollutants without considering the dynamics of their motion through
the system.

❖Planning models are employed for such tasks as studies of receiving water quality &
treatment facilities.

❖They do not require detailed geometric information on the

drainage facilities as do the first two groups of models. Typical examples of planning models
are:

(1) STORM (Storage, Treatment, Overflow, Runoff Model),

created by the U. S. Army Corps of Engineers (1976); 02/17/2025 29

 Planning models.
(2) SWMM (Storm Water Management Model), developed by Metcalf and Eddy, Inc., the
University of Florida, & Water Resources Engineers, Inc., (Metcalf and Eddy, 1971; U. S.
Environmental Protection Agency, 1977);

(3) RUNQUAL (Runoff Quality)-includes the hydraulic portion of the SWMM RUNOFF
model and the stream water quality model QUAL-II (Roesner, Giguere, and Davis, 1977);

(4) HSPF (Hydrocomp Simulation Program—Fortran) developed by Johnson, et al. (1980)-a

later version of the Stanford Watershed Model;

(5) MITCAT (MIT catchment model) by Harley, Perkins, and Eagleson (1970)

02/17/2025 30
INPUTS INPUTS OUTPUTS

Model
parameters: ,
, , 

Process data: rain,

temp, sunshine, Simulated data:
wind, humidity etc Hydrologic model Stream flow, soil
moisture, ET,
Observed data: sediment, P, N, etc.
streamflow, soil
moisture
Catchment characteristics:
size, slope, land cover, soil 02/17/2025 31

etc.
• Hydrologic models are planning tools that ask “what if” questions (e.g., What
would happen to the water quality of a lake if the surrounding forest were
partially cleared for an apartment building and a golf course?).

02/17/2025 32
Hydrologic models can be used in two ways:

1. to assess the existing hydrology and water-quality conditions of a water

resource, and

2. to predict future hydrology, which may develop as a result of changes in

land use, climate or any other physical alteration to the environment.

02/17/2025 33
Consider
(1) the ease of running the model and interpreting the results,
(2) availability of input data,
(3) availability of the model and support,
(4) applicability to land-use practices,
(5) applicability to broad geographic areas
(6) accuracy of prediction.

02/17/2025 34
❖Freely available watershed model that
• Is physically-based

• Has distributed parameters

❖Developed in the USA to predict the impact of land management practices on water,
sediment and agricultural chemical yields

❖Can evaluate Best Management Practices such as terracing, river bank protection,
conservation agriculture

02/17/2025 35
02/17/2025 36
02/17/2025 37
• The land phase of the hydrologic cycle is based on the water balance equation

Sw = soil water content

R = Precipitation
Q = Surface runoff
Ea = Evapotranspiration
Wseep = Percolation and bypass
flow exiting the soil profile bottom
Qgw = return flow

38
• SWAT provides two methods for estimating surface runoff
1. the SCS curve number procedure (SCS, 1972)
• CN is a function of the soil’s permeability, land use and antecedent soil water
conditions

2. the Green & Ampt infiltration method (1911)

02/17/2025 39
02/17/2025 40
02/17/2025 41
• Three methods of estimating potential ET have been incorporated into SWAT:

• the Penman-Monteith method (Monteith, 1965; Allen, 1986; Allen et al., 1989),

• the Priestley-Taylor method (Priestley and Taylor, 1972) and

• the Hargreaves method (Hargreaves et al., 1985).

• The model can also read in daily PET values

02/17/2025 42
❑Once total potential evapotranspiration is determined, actual
evaporation must be calculated.

❑SWAT first evaporates any rainfall intercepted by the plant canopy.

❑Next, SWAT calculates the maximum amount of transpiration and the
maximum amount of soil evaporation

❑The actual amount of sublimation and evaporation from the soil is then
calculated.

02/17/2025 43
• SWAT utilizes the Modified Universal Soil Loss Equation (MUSLE)
(Williams, 1975) to estimate sheet erosion for each hydrologic response unit
(HRU).

• HRUs are portions of subwatershed with unique soil, land use, and
management attributes.

02/17/2025 44
The equation reads

sed = sediment yield on a given day in metric tonnes

Qsurf = surface runoff volume
Qpeak = peak runoff rate
areaHRU = area of the HRU
KUSLE = USLE soil erodibility factor
CUSLE = USLE cover and management factor
PUSLE = USLE support practice factor
LSUSLE = USLE topographic factor
CFRG = coarse fragment factor

02/17/2025 45
❑A soil type usually becomes less erodible with decrease in silt fraction,
regardless of whether the corresponding increase is in the sand fraction or
clay fraction.
❑Organic matter content and soil structure also influence soil erodibility

❖ Influenced by plant or residue cover

02/17/2025 46
❑The support practice factor is defined as the ratio of soil loss with a specific support
practice to the corresponding loss with up-and-down slope culture.

❑Support practices include contour tillage, strip cropping on the contour, and terrace
systems.

❑Stabilized waterways for the disposal of excess rainfall are a necessary part of each of
these practices.

02/17/2025 47
02/17/2025 48
Vegetated buffer strip Cultivation on river banks

02/17/2025 49

Stabilized waterway
• The topographic factor is the expected ratio of soil loss per unit area from a
field slope to that from a 22.1-m length of uniform 9 percent slope under
otherwise identical conditions.

02/17/2025 50
 02/17/2025 51
With terracing, the slope length LS is the terrace interval
• Calibration
• Hydrology – should come first
• Sediment – should be next
• Water quality
• Nitrogen, Phosphorus, Pesticides, DO, Bacteria
• Validation

02/17/2025 52
❑Water balance (Amount and distribution among components)
❖Annual
❖Monthly
❖Daily
❖Baseflow
❖Surface water
❖Total water yield

02/17/2025 53
i. Visual inspection of hydrographs

ii. Coefficient of determination R2

iii. Mean and Std Deviation of measured and simulated data

iv. Nash-Suttcliffe Efficiency NSE

02/17/2025 54
• CN2 = Curve Number

• SOL-AWC = Soil available water holding capacity

• GWQMN = Threshold depth of water in shallow aquifer

• CH-K1 = Channel hydraulic conductivity

• ALPHA-BF = baseflow alpha factor

• GW-DELAY = groundwater delay time

• CH-N2 = Manning’s roughness coeff of channel

• CH-S2 = Average slope of main channel

02/17/2025 55
❑CN2 = curve number
❑USLE-P = Support practice factor
❑CH-N2 = Manning’s n main channel
❑USLE-C = Cover factor
❑SPCON = Linear coefficient of sediment routing
❑CH-S2 = average slope of main channel
❑SOL-AWC = soil available water capacity
❑SLOPE = average slope steepness
❑SLSUBBSN = average slope length
❑BC4 = rate constant for mineralization of P
❑SOL-ORGN = initial organic P concentration in soil layer
❑CH-N1 = Manning’s n for tributary channels
02/17/2025 56
• Bracmort et. Al (2006)
• Objective of study was to determine the long-term (~20 year)
impact of structural BMPs on sediment and Phosphorus loads
• BMPs considered are
• Grassed waterway – Channel cover, erodibility, roughness
• Terraces – CN, USLE-P, Slope Length
• Field border – filter width
• Grade stabilization – channel erod. & slope steepness

02/17/2025 57
 0.8
0.7
0.6
0.5
0.4 Mean Sy t/ha
0.3 Sy STD dev

0.2
0.1
0
No BMP BMP-good BMP-varying

Operation of BMPs under good condition would reduce the average annual sediment
02/17/2025 58
yield from the Dreisbach watershed by approximately 32%,
02/17/2025 59
02/17/2025 60
02/17/2025 61
OTHER MODELS

02/17/2025 62
 WEAP Model - WEAP21
What is WEAP?
➢ Water Evaluation Allocation and Planning Model (WEAP)
Water Allocation Principles- These are based on:
•Temporal and Spatial distribution of water resources
• Consider different water uses in the system (e.g. a River basin)

02/17/2025 63
 WEAP Model
❖Water Evaluation And Allocation Planning System
❖Integrated watershed hydrology and water planning [Link]
❖ General purpose model building, data management and scenario analysis tools

❖ Integrated analysis across demand and supply

❖ Transparent, flexible and user-friendly with low initial data requirements

❖ Widely used in Governments, Universities, Consulting Companies, Utilities and NGOs: 100s
of users worldwide

❖ Available at no charge to non-profit, academic and governmental institutions based in

developing countries

02/17/2025 64
 Water Evaluation And Planning System
❖ Integrated watershed hydrology and water planning model

❖ GIS-based, graphical drag & drop interface

❖ Physical simulation of water demands and supplies

❖ Additional simulation modelling: user-created variables, modelling equations and links to

spreadsheets, scripts & other models

❖ Scenario management capabilities

❖ In addition, WEAP has the following Modules:

✓ Groundwater
✓ water quality
✓ reservoir,
✓ hydropower
✓ financial modules
02/17/2025 65
The WEAP Interface

02/17/2025 66
 Using WEAP

• Major Modules
i. Schematic
ii. Data
[Link]
[Link] Explorer

02/17/2025 67
 WEAP Schematic
• Drag and drop system node components
✓Demand sites
✓Reservoirs, etc.
• Drag, click, and drop system link
components
✓Rivers
✓Transmission links
✓Return flows
• Add GIS layers to help place components
• Must include all infrastructure you plan to
test in Scenario Explorer
02/17/2025 68
WEAP- River Example Schematic

02/17/2025 69
 WEAP Data Module
Enter data for each schematic component
i. Rivers: Head flows for each month of the simulation
ii. Reaches: Reach gains for each month of the simulation
iii. Diversions: Minimum flow requirements as reach losses
iv. Demand sites: activity levels, use rates, losses,
consumption, demand priority (1=highest; 99=lowest)
v. Transmission links: Max flows, supply preference
vi. Return flows: routing (percent returned)
vii. Reservoirs: storage capacity, initial storage, volume-
elevation curve, evaporation, pool definitions, buffer
coefficients, priority

Enter data or read from input file

02/17/2025 70
 Alternatively, right-click any schematic
component to also get to the Data module

02/17/2025 71
Tree view, Buttons, and Tabs to navigate to desired data

02/17/2025 72
❖Click the Results WEAP Results Module
icon and recalculate
(all scenarios)

❖Choose results
from schematic or
dropdown lists
❖Numerous options
to view, tabulate,
and export

02/17/2025 73
Water demands by Demand Site

02/17/2025 74
 WEAP Scenario Explorer

• Define and manage scenarios from Data module

• Enter input data here too

02/17/2025 75
 WEAP Scenario Explorer (cont.)
• Use Scenario
Explorer icon
to open
scenario
dashboard

02/17/2025 76
WEAP Network Schematic

Linking supply
and demand

5 Main
Views Return flows to
surface or ground
water or
treatment plants

02/17/2025 77
Data View
Data for the
demand sites is
displayed
numerically and
graphically

02/17/2025 78
 Reading from Files

Read in timeseries
data from text files
or Access database

02/17/2025 79
 Results can be
Results View displayed in a
number of
formats and
scales

02/17/2025 80
Results Displayed on the Map

02/17/2025 81
Scenario Explorer

Favorite charts
can be selected to
give quick
overviews

02/17/2025 82
Hydropower

Specify
capacities,
efficiencies, and
other properties
of power
generation
02/17/2025 83
Financial Analysis

Specify variable
and fixed costs
and revenues

02/17/2025 84
Water Quality

Specify pollutant
loadings

02/17/2025 85
 Linking WEAP to Other Software
• Customized/Programmed links
• Groundwater flow model
• MODFLOW
• Surface water quality model
• Qual2K
• User-defined links to dynamic-link libraries
• California Department of Water Resources
• Delta salinity model
• East-Bay Municipal Utilities District
• Reservoir operations model
• Call WEAP using application programming interface (API)
• Scenario analysis
• CARS (RAND Corporation)
• Model calibration
• PEST
• Sensitivity analysis
• VB script 02/17/2025 86
Linking WEAP to other Models e.g. MODFLOW

02/17/2025 87
Linking WEAP to MODFLOW

02/17/2025 88
 WEAP Applications
I. Water Systems Planning
▪ Small Reservoirs Project, Ghana/Brazil
▪ California Water Plan, California, USA
▪ Guadiana River, Spain
II. Transboundary Water Policy
▪ Okavango River, Angola/Namibia/Botswana
▪ Lower Rio Grande, USA/Mexico
▪ Mekong River, Thailand/Cambodia/Vietnam/Laos
▪ Jordan River, Syria/Israel/Jordan
III. Climate Change Studies
▪ Sacramento and San Joaquin River Basins, California, USA
▪ Massachusetts Water Resources Authority, Massachusetts, USA
▪ Yemen Second National Communication
▪ Mali Second National Communication
iv. Ecological Flows
▪ Connecticut Department of Environmental Protection
▪ Town of Scituate, Massachusetts, USA
IV. Water Utility DSS Application
▪ Case studies in Portland, Oregon; Austin, Texas; and Philadelphia,
02/17/2025 89

Pennsylvania.
 Linking Water Decision Processes
Groundwater depletion Limited sectoral water, increased
Water quality energy requirements for pumping.
Unmet ecological flows
Costs Increased energy requirements
for desalination.
Insufficient water for sectors, even
with increased groundwater Water demand in each sector
pumping.
Still insufficient water--further Efficiency and Sustainability
enhance supply with desalination.
Water Water
Supply Demand

DSS Storage
Sectoral policies
Water requirements for sectors
Water conservation
Less water-intensive
processes and technologies
Water requirements
Reduced water demands
02/17/2025
90
 WEAP and Planning
• Provides a common framework for transparently
Organizing water resource data at any scale
desired – local watershed, regional or
transboundary river basin
• Scenarios can be easily developed to explore
possible water uses in the future
• Implications of various policies can be evaluated

02/17/2025 91
i.
Uses
Policy Research
of WEAP
◼ Alternative Allocations
◼ Climate Change

◼ Land Use Change

◼ Infrastructure Planning

ii. Capacity Building

iii. Negotiation
iv. Stakeholder Engagement

02/17/2025 92
 WEAP Capabilities
Can do Cannot do
❖High level planning at ◼Sub-daily operations
local and regional ◼Optimization of supply
scales and demand (e.g. cost
minimizations or social
❖Demand welfare maximization)
management
❖Water allocation
❖Infrastructure
evaluation
02/17/2025 93
A Simple System with WEAP21

40

60

02/17/2025 94
An Infrastructure Constraint

10 Unmet
30

70

02/17/2025 95
A Regulatory Constraint

10 Unmet
30

70
IFR Met

02/17/2025 96
 Different Priorities

•For example, the demands

40 of large farmers (70 units)
might be Priority 1 in one
scenario whereas the
60 demands of smallholders
(40 units) may be Priority 1 in
0 another
10 unmet
 Different Preferences

30

10

0 •For example, a center

pivot operator may prefer
90 to take water from a
tributary because of lower
pumping costs
A Simple System

02/17/2025 99
 What are we assuming?

▪ That we know how much water is flowing at

the top of each river
▪ That we know how much water is flowing
into or out of the river as it moves
downstream
▪ That we know what the water demands are
with certainty
02/17/2025 100
 WEAP is Scenario-driven

◼ The scenario editor readily accommodates

analysis of:
◼ Climate change scenarios and

assumptions
◼ Future demand assumptions

◼ Future watershed development

assumptions

02/17/2025 101
 Futures and Scenarios: Why?

❑Scenarios: a systematic way of thinking about the

future
❑To gain a better understanding of the possible
implications of decisions (or non-decisions across
scales and time
❑To support decision-making

02/17/2025 102
 Driving Forces
Demographic
Technological
•More people
•Computer and information
•Urbanization technology
•Older •Biotechnology
Economic •Miniaturization
•Growing integration of global Environmental/Climatic
economy •Increasing global stress
Social •Local degradation
•Increasing inequality •Some remediation in richer
•Persistent poverty countries
Cultural Governance
•Spread of values of •Global institutions
consumerism & individualism •Democratic government
•Nationalist & religious •Role for civil society in decision-
reaction making 02/17/2025 103
 Who are the Actors?
◼Government ◼Rich farmers
◼Private sector ◼Poor farmers
◼Civil society ◼Urban users
◼Public ◼Environmentalists
◼Or?

02/17/2 10
025 4
 WEAP Highlights

➢WEAP is comprehensive, straightforward and

easy-to-use, and attempts to assist rather than
substitute for the skilled planner
➢Integrated water resources planning system
➢GIS-based, graphical drag and drop interface
➢Model-building tool – user-created variables
and modeling equations
➢Basic Methodology: physical simulation of
demands and supplies
➢Scenario management capabilities
➢Dynamic links to spreadsheets & other models
02/17/2025 105
 WEAP Highlights cont.

➢As a database, WEAP provides a system for

maintaining water demand and supply
information
➢ As a forecasting tool, WEAP simulates water
demand, supply, flows, and storage, and
pollution generation, treatment and discharge
➢ As a policy analysis tool, WEAP evaluates a full
range of water development and management
options, and takes account of multiple and
competing uses of water systems
02/17/2025 106
 Conclusions

❖ WEAP can simulate priority-based water allocations

❖ Drag and drop interface to draw system schematic
❖ Entered head flow data for one river
❖ View results in numerous formats

02/17/2025 107
 Calculate Allocations
• Step 1: Draw the schematic
• Step 2: Determine delivery targets for demand sites
(demands)
• Step 3: Assign priorities to demand sites (delivery
preferences)
• Step 4: Determine water availability
✓ Sources
✓ Return flows
• Step 5: Allocate remaining available water to meet
delivery target of highest priority demand site
• Repeat Steps 4 and 5 for next highest priority site

02/17/2025 108
 Aquarius Model

❖Aquarius is a state-of-the-art computer model

devoted to the temporal and spatial allocation of
water among competing uses in a river basin
❖ The model is driven by an economic efficiency
operational criterion requiring the reallocation of
stream flows until the net marginal return in all water
uses is equal
❖ Aquarius is an analysis framework rather than a
single dedicated model for water allocation
02/17/2025 109
QUESTIONS?

02/17/2025 110
HEC -HMS

02/17/2025 111
 Data4Water Summer school on Intelligent Metering and Big Data in Hydroinformatics,
Bucharest, June 20 – July 8, 2016

DEVELOPMENT OF A SENSITIVITY
ANALYSIS APPLICATION FOR AN EVENT-
BASED HEC‒HMS RAINFALL-RUNOFF
MODEL

A. BUILDING AND RUNNING THE

HYDROLOGICAL MODEL

02/17/2025 112
 Data4Water Summer school on Intelligent Metering and Big Data in Hydroinformatics,
Bucharest, June 20 – July 8, 2016
HYDROLOGIC
AL MODELING
Hydrological modeling – simulate the precipitation – runoff of
MODEL watershed
COMPONENTS
Simulation options:
SIRON • Event modeling (Initial constant, SCS Curve Number, Exponential,
BASIN MODEL Green & Ampt, Smith Parlange)
• Continuous modeling (one-layer deficit constant method, three-
SIRON layer soil moisture accounting method)
METEO
MODEL Objective:
• Building a model to simulate a storm event on Siron river basin
SIRON
CONTROL
Model requirements:
SIRON • State variables
TIME SERIES • Parameters
• Boundary conditions
• Initial conditions
Siron basin
02/17/2025 113
 Data4Water Summer school on Intelligent Metering and Big Data in Hydroinformatics,
Bucharest, June 20 – July 8, 2016
HYDROLOGICA
L MODELLING Basin Model
• Uses time series data and specific sets of parameters to represent the
MODEL hydrologic system
COMPONENTS • Contains multiple methods to simulate each hydrological processes:
runoff volume = loss (infiltration-saturation-runoff)
SIRON direct runoff = transform (effective rainfall-runoff)
BASIN MODEL baseflow

SIRON Meteorological Model

METEO • Defines the gage data input that is transformed during the simulation
MODEL into effective precipitation, cumulated precipitation per time step,
precipitation excess etc.
SIRON • Can use various meteorological parameters (air temperatures, air
CONTROL humidity etc.) to simulate snowmelt and evapotranspiration
Control specifications
SIRON • Sets the time interval of the simulation
TIME SERIES • Includes the time step for the computation
Time series data
• Manually or read from file time series data; gridded data; paired data
• entered manually or automatically (HEC-DSS). 02/17/2025 114
 Data4Water Summer school on Intelligent Metering and Big Data in Hydroinformatics,
Bucharest, June 20 – July 8, 2016
HYDROLOGICA
L MODELLING

MODEL Loss → SCS Curve Number Method - implements curve number

COMPONENTS methodology for incremental losses (NRCS,2007)

SIRON Parameters needed:

BASIN MODEL
• Initial Abstraction (m3/s): amount of precipitation that must infiltrate
SIRON before surface excess results
METEO
MODEL • Curve number: - expresses the runoff potential and it is dependent on
soil permeability, land coverage, basin area
SIRON - the range for CN values is 0 – 100 without ever reaching
CONTROL to the extreme values
CN = 0 → the entire amount of precipitation infiltrates = low runoff potential
SIRON CN = 100 → impervious soils = high runoff potential
TIME SERIES
• Impervious (%): the share of the basin area that is impervious
- all precipitation becomes direct runoff

02/17/2025 115
 Data4Water Summer school on Intelligent Metering and Big Data in Hydroinformatics,
Bucharest, June 20 – July 8, 2016
HYDROLOGICA
L MODELLING
Transform → Snyder Unit Hydrograph – calculates the unit
MODEL hydrograph through Snyder’s method
COMPONENTS

SIRON Parameters needed:

BASIN MODEL • Standard Lag (hr): time between the centroid of excess rainfall and the
centroid of the resulting hydrograph
SIRON depends on: basin area, slope, river network length,
METEO curve number
MODEL
0.8 1000 0,7
𝐿∙3,28∙103 ∙ 𝐶𝑁 −9
SIRON 𝑡𝑙𝑎𝑔 = , where:
1900∙𝑌 0,5
CONTROL
𝑡𝑙𝑎𝑔 = 𝑙𝑎𝑔 𝑡𝑖𝑚𝑒, 𝐶𝑁 = 𝑐𝑢𝑟𝑣𝑒 𝑛𝑢𝑚𝑏𝑒𝑟, 𝑌 = 𝑠𝑙𝑜𝑝𝑒, 𝐿 = 𝑟𝑖𝑣𝑒𝑟 𝑛𝑒𝑡𝑤𝑜𝑟𝑘 𝑙𝑒𝑛𝑔𝑡ℎ
SIRON
TIME SERIES GIS → basin area, slope, river length network

• Peaking Coefficient: steepness of the hydrograph that results from a

unit of precipitation 02/17/2025 116
 Data4Water Summer school on Intelligent Metering and Big Data in Hydroinformatics,
Bucharest, June 20 – July 8, 2016
HYDROLOGICA
L MODELLING

MODEL Baseflow → Exponential recession

COMPONENTS

SIRON Parameters needed:

BASIN MODEL • Initial Discharge (m3/s): the initial value from which the baseflow for
each time step is calculated
SIRON
METEO • Recession Constant: defined as the ratio of present baseflow value
MODEL at previous time step baseflow
SIRON • Threshold Type: discharge value at which all runoff is considered
CONTROL baseflow; if Ratio to Peak is chosen, the threshold is dependent on
the peak value (not a fixed discharge value)
SIRON
TIME SERIES

02/17/2025 117
 Data4Water Summer school on Intelligent Metering and Big Data in Hydroinformatics,
Bucharest, June 20 – July 8, 2016
HYDROLOGICA
L MODELLING
Meteorological model:
MODEL
COMPONENTS Gage Weights → specify the weights
to defined precipitation gages (in our
SIRON study case just one precipitation gage was used)
BASIN MODEL

SIRON Control data:

METEO START DATE and TIME: [Link] hh:mm
MODEL END DATE and TIME: [Link] hh:mm
COMPUTATION TIME STEP: 1 minute → 1 Day (1 hour in our case
SIRON study)
CONTROL
Time series data:
SIRON
Precipitation (mm)
TIME SERIES
Discharge (m3/s)

02/17/2025 118
 Hydrological Modeling for Upper Chao
Phraya Basin Using HEC-HMS

UNDP/ADAPT Asia-Pacific First Regional Training Workshop

Assessing Costs and Benefits of Adaptation: Methods and Data
March 11-14, 2013

Dr. Dilip K. Gautam

RIMES, AIT Campus, Bangkok
02/17/2025 119
 Hydrological Model
 A model is a simplified representation of reality.

 A mathematical model consists of series of equations defining the system

we are dealing with. The function of model is to convert the given input
into an output.

 A hydrological model is the mathematical representation of the response

of a catchment system to hydrologic events during the time period under
consideration.

 Hydrological phenomena are extremely complex, highly non-linear and

highly variable in space and time.

 A model is needed to predict the watershed runoff for the design and
management of water resources utilization and flood control projects.

02/17/2025 120
 Classification of Hydrological Models

02/17/2025 121
 Hydrological Model Outputs for
Climate Change Impact Assessment
 Simulated flow peaks, volumes and hydrographs
at the outlets of subbasins and the points of
special interest such as reservoirs, weirs or other
hydraulic structures
 Simulated long flow sequences for water budget
and drought analyses
 Simulated extent of flooded areas for different
precipitation events and various antecedent basin
conditions

02/17/2025 122
 Hydrologic processes that need to
be captured by the model
 Single-event precipitation-runoff transformation
 Continuous precipitation-runoff transformation
 Snow accumulation and melt
 Interception, infiltration, soil moisture accounting
 Evapotranspiration
 Regulated reservoir operation

02/17/2025 123
 HEC-HMS
• US Army Corps of Engineers, Hydrologic Engineering
Center’s Hydrologic Modeling System software
• Designed to simulate both single event and continuous
rainfall-runoff process
• Simulates precipitation-runoff and routing processes, both
natural and controlled
• HEC-HMS uses a separate model to represent each
component of the runoff process including:
➢ runoff volume;
➢ direct runoff (overland flow and interflow);
➢ baseflow;
➢ channel routing.

02/17/2025 124
 HEC-HMS representation of watershed runoff

02/17/2025 125
Key Components of Model
 Runoff Volume models: separate infiltration from
pervious surface, runoff from impervious surface,
compute the direct runoff volume
 Direct Runoff models: transform direct runoff volume
from excess precipitation into fast component of flow
 Base Flow models: compute slow subsurface drainage
component
 Routing models: compute flow attenuation and
translation over channel
 Reservoir models: flow regulation
02/17/2025 126
 Data Required
 Digital Elevation Model (DEM), land use, soil
types and other physiographic data
 Precipitation, temperature data
 Evaporation/evapotranspiration data
 Discharge, Water level and Rating curve data
 Channel and reservoir hydraulic data
 Generated sequence of meteorological data
representing various scenarios of future climate

02/17/2025 127
 Upper Chao
Phraya Basin,
Thailand
Catchment Area
= 105553 sq. km.

02/17/2025 128
 Dams and Reservoirs
 Bhumibol dam in the Ping River (Storage 13462 MCM)
 Sirikit dam in the Nan River (Storage 9510 MCM)
 Kwae Noi dam in Kwae Noi River (Storage 766 MCM)
 Kiew Kor Mha dam in Wang River (Storage 171 MCM)
 Kiew Lom dam in Wang River (Storage 112 MCM)

02/17/2025 129
 Data preparation using HEC-GeoHMS
 Delineate catchment and river network
 Obtain catchment characteristics data (area, slope etc)
 Make Thiessen polygon
 Obtain Thiessen weights
 Prepare basin file

02/17/2025 130
02/17/2025 131
 Data preparation using HEC-DSSVue
 Time series data (rainfall, discharge etc.)
 Pair data (elevation-storage)

02/17/2025 132
 Model Setup
 Basin model
 Meteorological model
 Time series data
 Pair data
 Control specification

02/17/2025 133
 Meteorologic model
 Precipitation
 Evapotranspiration
 Snowmelt : not applicable for upper Chao Phraya

02/17/2025 134
 Precipitation methods
 Gauge weights : selected for upper Chao Phraya
 Inverse distance
 Gridded precipitation
 Frequency storm
 SCS storm
 Specified Hyetograph
 Standard project storm

02/17/2025 135
 Evapotranspiration methods
 Monthly Average : selected for upper Chao Phraya
 Priestley-Taylor
 Gridded Priestley-Taylor

02/17/2025 136
 Snowmelt methods
 Gridded temperature index
 Temperature index

02/17/2025 137
 Control Specifications
 Simulation start date/time
 Simulation end date/time
 Time interval

02/17/2025 138
 Model Calibration
 Finding optimal parameter values
 Minimizing difference between simulated flow
and observed flow
 Objective functions
 Peak weighted RMS error
 Percent error peak
 Percent error volume
 RMS log error
 Sum of absolute residuals
 Sum of squared residuals
 Time weighted error

02/17/2025 139
 Search Algorithms
 Nelder Mead
 Univariate Gradient

02/17/2025 140
 The Basin Model

02/17/2025 141
 Simulated Hydrograph at Basin Outlet
R2 = 0.71 BIAS = 6.7 % NS = 0.71

02/17/2025 142
 Conclusions
 Semi-distributed physically based deterministic
hydrological models are powerful tools for assessing
climate change impact on water resources.
 Continuous modeling approach could be taken to
assess the impact on flow volume.
 Care should be taken to interpret the results as there
are lots of uncertainties in the model inputs,
parameters and structure of the model. Uncertainties
associated with climate models will also be carried
over.

02/17/2025 143
 CIV 4105: HYDROLOGY II

READING ASSIGMENT

Conduct a literature review and identify the models used to simulate:

▪ Soil erosion

▪ Surface runoff from single storm events

▪ Ground water flow

▪ Water quality

Identify whether these models are determined or physical or stochastic models.

02/17/2025 30
 END
02/17/2025 145