Computer Models in Hydrology

By
Dr. Rizwan Aziz

2nd Semester, M.Phil. Applied Hydrology

CEES, University of the Punjab Lahore
 Two important words: Catchment & Model

What is a Catchment?
river network
An area of land and water bodies
that drains runoff through the
same outlet.

Synonyms: Watershed, drainage

basin, drainage area basin etc.
unsaturated
soil zone
groundwater/aquifer
(saturated)
 What is a Model?

A model is a representation of a real system or process.

A model is a simplification of a real-world system.

“A model should be made as simple as possible, but not

simpler.”
(after Albert Einstein, 1879 - 1955)
A Hydrological model is a simplification of a real-world
watershed/hydrological system (e.g., surface water, soil water,
wetland, groundwater, estuary) that aids in understanding,
predicting, and managing water resources. Both the flow and
quality of water are commonly studied using hydrologic
models.
Process understanding
(measurement, observation)

Modelling
(simulation, prediction)
 Why Model?
The main reason is a result of the limitations of hydrological
measurement techniques.
We are not able to measure everything we would like to know
about hydrological systems. We have, in fact, only a limited
range of measurement techniques and a limited range of
measurements in space and time.

We therefore need a means of extrapolating from those available

measurements in both space and time, particularly to ungauged
catchments (where measurements are not available) and into
the future (where measurements are not possible) to assess the
likely impact of future hydrological change.
provide a means of quantitative extrapolation or prediction that
will hopefully be helpful in decision making.

There is much rainfall–runoff modelling that is carried out purely

for research purposes as a means of formalising knowledge about
hydrological systems. The demonstration of such understanding is
an important way of developing an area of science.
MODEL CLASSIFICATIONS
 Deterministic vs. Stochastic Models

Deterministic model – in a deterministic model, given the input

data and parameter values, the model determines exactly the output,
such that we always get the same result. If the deterministic model
perfectly described the hydrological system under consideration and
there was no uncertainty or source of error, then given the value of
the independent variable (x) and the model parameters, we would be
able to predict the value of the dependent variable (y) exactly (i.e.,
with no uncertainty).
Stochastic model – in a stochastic model, given the input data and
parameter values, the model gives variable output, such that we always
get a different result due to randomness. If there is some uncertainty in
our model parameters, then we would expect for a given value of the
independent variable (x) to generate a different value of the dependent
variable (y) each time, since the model is imperfect. The stochastic
model is simply the error in our ability to predict the outcome
(dependent variable) for a particular input. All statistical models have a
stochastic component.
 Lumped Vs. Distributed Models

Lumped Model
The lumped model considers individual sub-basins as a single unit,
whereas the distributed model sub-divides each sub-basin in
smaller cells. A lumped hydrological model averages spatial
characteristics related to rainfall-runoff response for the entire area
of a sub-basin being analyzed.

Distributed Models (Semi-Distributed or Fully Distributed)

A distributed model explicitly accounts for spatial variability using
spatial datasets describing soils, vegetation, and land use. A
distributed model will subdivide a sub-basin area based on a
particular grid size to capture spatial and temporal variability and
then route flows through the subbasin from cell to cell.
Lumped versus Distributed Basin
A conceptual model is a representation of a system, made of the
composition of concepts which are used to help
people know, understand, or simulate a subject the model
represents. It is also a set of concepts.

Conceptual model may refer to models which are formed after

a conceptualization or generalization process.

A conceptual model allows to represent the main processes of the

rain-flow relationship without describing the physical laws governing
the processes involved. This type of model generally consists in
interconnected reservoirs, in which the level increases and
decreases over time and represents the different hydrological
compartments of the watersheds.
 Physically Based Models
The physically based models are based on our understanding of the
physics of the hydrological processes which control the catchment
response and use physically based equations to describe these
processes.

These models are based on our understanding of the physics of the

hydrological processes which control catchment response and use
physically based equations to describe these processes.

Some typical examples of field applications include study of effect of

catchment changes, prediction of behavior of ungauged catchment, of
spatial variability in catchment inputs and outputs, movement of
pollutants and sediment etc.
In these models transfer of mass, momentum and energy are
calculated directly from the governing partial differential equations
which are solved using numerical methods, for example the St. Venant
equations for surface flow, the Richards equation for unsaturated zone
flow and the Boussinesq equation for ground water flow.

As the input data and computational requirements are enormous, the

use of these models for real-time forecasting has not reached the
‘production stage’ so far, particularly for data availability situations
prevalent in developing countries like Pakistan.
 Black Box Models

As the simplest models, black-box hydrological models are based on

input-output relations and do not describe the
underlying hydrologic processes.

Compared to other types of models, black-box models require the

least input data.

Precipitation and temperature are frequently used as input.

 Event Based Vs Continuous Models

An event based model simulates a catchment response to a single storm

or a number of selected storm events (e.g., quantity of surface runoff,
peak, timing of the peak, detention). The duration of the storm may
range from few hours to few days. Event-based hydrologic models are
frequently used for flood design and assessment.

A continuous model simulates a longer period, predicting watershed

response both during and between the storms. Thus continuous
hydrologic modeling synthesizes hydrologic processes and phenomena
(i.e., synthetic responses of the basin to a number of rain events and
their cumulative effects) over a longer time period that includes both wet
and dry conditions.
Fine-scale event hydrologic modeling is particularly useful for
understanding detailed hydrologic processes and identifying the
relevant parameters that can be further used for coarse-scale
continuous modeling, especially when long-term intensive monitoring
data are not available or the data are incomplete.
DATA REQUIRMENTS
Typical Input
• Topography (DEM)
• Soil Characteristics
• Land cover
• Land use
• Meteorological data

Typical Output
• Streamflow
• Subsurface Flow
• Depth to water table
 Usual Steps to Hydrologic Modeling

1. Delineate watershed
2. Obtain geographic data (DEM, Land use/Land
Cover etc.)
3. Obtain hydrologic/meteorological data
4. Select modeling approach
5. Calibrate/Verify model
6. Use model for assessment/prediction/design
All watershed delineation means is
that you're drawing lines on a map to
identify a watershed's boundaries.
These are typically drawn on
topographic maps using information
from contour lines. Contour lines are
lines of equal elevation, so any point
along a given contour line is the
same elevation.
Digital Elevation Model (DEM) are digital file that stores the elevation
of the land surface a specified grid cell size (e.g., 30 meters, 90
meters)
 Land Cover Land Use Data
Land Cover
Forest
Oak Woodlands
Mesquite Woodlands
Grasslands
Desertscrub
Riparian
Agriculture
Urban
Water
Barren / Clouds

0 5 10 km

1992 NALC Hillshade DEM STATSGO

• Meteorological Data Hydrologic Data
• Temperature Streamflow
• Precipitation Peak discharge
• Wind speed Daily flow volume
• Humidity Annual flow volume
• Extrapolation of point Soil moisture
measurements Groundwater level
• Theisen Polygons
• Inverse distance weighting
 Modeling Approaches (examples)

TIME SCALE
Event-based Continuous Simulation
(minute to day) (days – years)
Empirical
Rational Method
Regression equ’s
SCS Unit Hydrograph Simple Model
Transfer Functions
Simple models
Physically-based KINEROS
Based on physical HEC-HMS SWAT
processes Stanford Watershed VIC-3L
Complicated Model TOPMODEL
Many parameters TOPMODEL HBV
CONSTITUENTS OF MODELS

State Variables
Model Parameters
Boundary Conditions
Initial Conditions
 State Variables

State Variables represents the state of hydrological system at a

particular time and location in terms of equations.

For example, the deficit and constant rate loss model tracks the
mean volume of water in natural storage in the watershed at a given
time and location.

Similarly, the detention models are used as state variables to express

pond storage at any given time.
 Model Parameters

These are the numerical approximations of the properties of real-

world system. They control the relations of the system input to the
system outputs.
They are considered as the “tuning knobs” of the model. The
parameter values are adjusted (during calibration) so that model
accurately predict the physical system response.

For example, Snyder UH model has two parameters, basin lag (tp)
and the peaking coefficient Cp. The values of these parameters can
be adjusted to “fit” the model to a particular watershed so that the
observed and simulated streamflows are as close as possible.
Model Parameters may have obvious physical significance or
maybe purely empirical.

For example, the Muskingum-Cunge channel model includes the

channel slope as a parameter which has a physical significance
and is a measurable parameter.

However, on the other hand, Snyder’s UH model has a peaking

coefficient Cp, which has no direct relationship to any physical
property, and can only be estimated by calibration.
 Boundary Conditions

Typically, a model domain is defined by Boundary Conditions such as

watershed boundaries that are no‐flow conditions or the coastal tidal
boundary that has a known head time series.

The other boundary conditions are the values of system input-the forces
that act on any hydrological system and cause it to change.

For Example, upstream flows (inflow) hydrograph to a channel reach

which is a boundary condition used in routing models.
 Initial Conditions

In order to execute a simulation on a hydrologic model, it is

necessary to specify the initial conditions of the simulation, which
represent the variables states at the first-time step of the simulation.

The initial hydrologic conditions have a strong impact on the

prediction of cumulative runoff and soil moisture.

Errors on the initial states have as much influence on the quality of

the flow prediction than those related to weather forecast. For
example, When the soil is saturated, the watershed tends to respond
rapidly and intensively to rainfall, while when it is dry, the watershed
absorbs most of the rainfall. Hence, antecedent soil moisture is an
important initial condition.