MODELLING THE HYDROLOGY OF WATERSHED BY

USING HEC-HMS

by
MAKKENA JYOTHI
(2014 - 1 8 - 118)

THESIS

Submitted in partial fulfilment of the requirement for the degree of

MASTER OF TECHNOLOGY
IN
AGRICULTURAL ENGINEERING
(Soil and Water Engineering)
Faculty of Agricultural Engineering & Technology
Kerala Agricultural University

DEPARTMENT OF IRRIGATION AND DRAINAGE ENGINEERING

KELAPPAJI COLLEGE OF AGRICULTURAL ENGINEERING AND TECHNOLOGY

TAVANUR- 679573, MALAPPURAM
2016
 D EC LA R A TIO N

I, hereby declare that this thesis entitled “M odelling the hydrology of

w atershed by using H EC -H M S” is a bonafide record o f research done by me
during the course o f research and the thesis has not previously formed the basis
for the award to me o f any degree, diploma, associateship, fellowship or other
similar title, o f any other University or Society.

Place: T a v a n u r M akkena Jyothi.

Date: (2014 -1 8 - 118).
 CERTIFICATE

Certified that this thesis, entitled “M odelling the hydrology o f w atershed by

using H EC -H M S” is a record o f research work done independently by

E r. M akkena Jy othi (2014-18-118) under my guidance and supervision and

that it has not previously formed the basis for the award o f any degree,

diploma, fellowship or associateship to her.

T av an u r, E r. V ishnu, B.
D ate: Associate Professor
Department o f Irrigation and Drainage Engineering
Kelappaji College o f Agricultural Engineering & Technology
Tavanur, Malappuram- 679573
 CERTIFICATE

We the undersigned members o f the Advisory Committee of Er. M akkena Jyothi (2014-18-

118) a candidate for the degree o f M aster of Technology in A gricultural Engineering, agree

that the thesis entitled "Modelling the hydrology of w atershed by using H EC-H M S” may be

submitted by E r. M akkena Jyothi, in partial fulfilment of the requirement for the degree.

E r. Vishnu B. Dr.Sasikala D.
(Chairman, Advisory Committee) (Member, Advisory Committee)
Associate Professor Professor and Head
Department o f IDE Department of IDE
KCAET, Tavanur KCAET, Tavanur

E r. AnuVarughese E r. Shivaji K.P.

(Member, Advisory Committee) (Member, Advisory Committee)
Assistant Professor Assistant Professor
Department o f IDE Department of FPME
KCAET, Tavanur KCAET,Tavanur

EXTERNAL EXAMINER
ACKNOWLEDGEMENT
 A C K N O W LE D G E M E N T

A n d so, comes the time to look back on the path traversed during the
endeavour and to remember the fa ces and spirits behind the action with a sense o f
gratitude. Nothing o f significance can be accomplished without the acts o f
assistance, words o f encouragement and gestures o f help from the other members
o f the society.

It is with immense pleasure I avail this opportunity to express m y deep

sense o f whole hearted gratitude to my major advisor Er. Vishnu, B., Associate
Professor,Department o f IDE, KCAET, Tavanur, fo r his excellent guidance,
caring, patience and providing me with an excellent atmosphere fo r doing
research. I consider it m y greatest fortune in having his guidance fo r m y research
work and my obligation to him lasts forever.

I express my heartfelt thanks to Dr. Hajilal, M .S., Dean, KCAET, Faculty

o f Agricultural Engineering and Technology fo r support that he offered while
carrying out the project work. M y sincere thanks to Dr. M. Sivaswami, Former
Dean, KCAET, Faculty o f Agricultural Engineering and Technology fo r the
unfailing guidance and support that he offered while carrying out the project
work.

I express my deep sense o f gratitude and indebtedness to my advisory

member Er. A n u Varghese, Assistant Professor, KCAET, Tavanur. I am thankful
fo r her valuable advices, creation o f immense interest in the subject, motherly
approach and constant encouragement during every phase o f my thesis work.

I engrave m y deep sense o f gratitude to my advisory committee members

Dr. Sasikala, D., Professor and Head, Department o f Irrigation and Drainage
Engineering, KCAET, Tavanur, fo r her help, valuable suggestions, evaluation and
due encouragement during the entire period o f the project work and
Er. Shivaji, K.P., Assistant Professor department o f FMPE, KCAET, Tavanur fo r
his valuable advice and moral support provided during the project w ork

I have boundless pleasure in recording my profound thanks to

Dr. Sathian K.K, Professor, department o f LWRCE, fo r his expert advice,
inspiring guidance, valuable suggestions, constructive criticisms, affectionate
advice and above all, the extreme patience, understanding and wholehearted co­
operation throughout the project work.

I
 M y heartfelt thanks to my beloved teacher, Dr. A sha Joseph, Assistant
Warden, L H KCAET, Tavanur, fo r granting me permission to stay in hostel
during the course o f my study.

With great pleasure, J express my heartfelt special thanks to my batch

mates fo r their support, encouragement and enthusiastic cooperation fro m time to
time fo r each step o f my thesis work. I whole heartedly thank my loving friends
Anjali, N eetha, J o m o lB h a v y a , Sagarika, Saranya, Seem a, Asw athi, A y is ha,
Laclti, A rju n fo r their love, caring, encouragement and moral support which
helped me get over all odds and tedious circumstances.

1 am in dearth o f words to express my soulful gratitude to my loving

brother Er. H yram fo r his blessings, selfless support, boundless patience,
prayers, inspiration, sacrifices, unflagging interest and eternal love which
sustains peace in my life. I am forever indebted to my loving parents babay, aunt,
cousin and m y entire fa m ily without their blessings and support I would not have
completed this w ork

I express my deep sense o f gratitude to Kerala Agricultural University fo r

financial and technical support fo r persuasion o f my study and research w ork

It would be impossible to list out all those who have helped me in one way
or another in the successful completion o f this w ork I once again express my
heartfelt thanks to all those who helped me in completing this venture on time.

Above all, I thank to “The A lm ighty G od” fo r his blessing and

procurement, which helped me in the completion o f the endeavour.

M akkena Jyothi
Dedicated
to
My mother
 CO NTEN TS

C h ap ter No. Title Page No.

LIST OF TABLES I

LIST OF FIGURES II

SYMBOLS AND ABBREVIATIONS V

1 INTRODUCTION 1

2 REVIEW OF LITERATURE 6

3 MATERIALS AND METHODS 24

4 RESULT AND DISCUSSION 81

5 SUMMARY AND CONCLUSION 102

'
REFERENCES 104

APPENDICES 110

ABSTRACT
 LIST OF TABLES

Table Page.
Title
No. No

Anderson land use and land cover classification codes: First and secondary 29
3.1
level categories

3.2 Hydrological models 57

3.3 HEC-HMS loss rate methods 61

3.4 Surface runoff methods in HEC-HMS 62

3.5 Flood routing in HEC-HMS 63

3.6 Usual rangesof r values 67

4.1 The name and area o f each Grama panchayats with in Thuthapuzha sub basin 83

4.2 Description o f the soil mapping units o f Thuthapuzha sub basin 84

4.3 Calculated Initial and optimized parameters for the watershed 97

4.4 Observed and simulated flow before and after calibration/validation 100
 II

LIST OF FIGURES

M ap No, Title Page No.

3.1 Location o f Thuthapuzha watershed 24

3.2 Flow chart of HEC-GEOHMS 36

3.3 Stream grid 40

3.4 Catchment grid delineation 41

3.5 Drainage line map 43

3.6 Dem for selecting watershed outlet

48

3.7 Selecting watershed outlet 49

3.8 Selecting area of sub basin 49

3.9 Sub basin delineation, merging streams 50

3.10 HEC-HMS model 52

3.11 Legends HEC-HMS 52

3.12 Schematic calibration procedure 70

3.13 Snapshot of dem reconditing, fill sinks operation 73

3.14 Snapshot of creation of flow direction, flow accumulation in HEC-

74
GEOHMS

3.15 Snapshot of creation of stream, stream segmentation map in HEC-

74
GEOHMS

3.16 - Snapshot of creation of catchment grid delineation, catchment

75
polygon processing in HEC-GEOHMS
 Ill

3.17 Snapshot of creation of adjoint catchment processing in HEC-

75
GEOHMS

3.18 Snapshot of creation of drainage point, drainage line processing in

76
HEC-GEOHMS

3.19 Snapshot of creation of slope map 76

3.20 Snapshot o f HEC-GEOHMS project setup 77

3.21 Snapshot o f HMS process 78

3.22 Snapshot o f creating HMS project 79

3.23 Snapshot o f opening the HMS project 80

4.1 Dem o f T huthapuzha 82

4.2 Hypsometric curve of Thuthapuzha sub basin 82

4.3 Soil m ap o f T huthapuzha sub basin 84

4.4 Land use m ap o f T huthapuzha sub basin 87

4.5 Fill sink m ap 88

4.6 Flow direction m ap 88

4.7 Flow accum ulation m ap 89

4.8 D rainage line m ap 89

4.9 C atchm ent extraction m ap 90

4.10 Ad jo in t extraction m ap 91
 IV

4.11 River map 92

4.12 Slope map of Thuthapuzha basin 93

4.13 Longest flow path map 94

4.14 Stream extraction map 95

4.15 Stream linking map 95

4.16 HMS project area 96

4.17 HMS model area 96

4.18 Observed and simulated flow before calibration (Monthly 98

average flow for calibration)

4.19 Observed and simulated flow before calibration (Monthly 99

average flow for validation)

4.20 Observed and simulated flow after calibration (Monthly 99

average flow for calibration)

4.21 Observed and simulated flow after calibration (Monthly 100

average flow for validation)
 V

SYMBOLS AND ABBREVIATIONS

HEC Hydrologic Engineering Center

HMS Hydrological Modeling System

RS Remote Sensing

GIS Geographical Information System

HSU Hydrological Similar Unit

GEOHMS Geospatial Hydrological Modeling System

ARC-GIS Aeronautical Reconnaissance Coverage-Geographical Information System

HEC-RAS Hydrological Engineering Center’s river analysis system

DEM Digital Elevation Model

SCS Soil Conservation Service

PEP Percentage Error in peak

PEV Percent Error in volume

SCS-CN Soil Conservation Service-Curve Number

SRTM Shuttle Radar Topographic Mission

TOPAZ Topographic parameterization

SWAT Soil and Water Assessment Tool

MS-Excel Micro Soft-Excel

USDA United States Department o f Agriculture

WMS Watershed Modeling System

CN Curve Number

NSE Nash Sutcliffe Model Efficiency

R2 Coefficient of Determination

SMA Soil Moisture Accounting

RMSE Root Mean Square Error

 VI

GWI Ground water 1

GW2 Ground water2

FEWS Flood Early Warning Systems

USGS United States Geological Survey

WGS World Geodetic Survey

USACE Unites States Army Corps o f Engineers

UH Unit Hydrograph

SCS-UH Soil Conservation Service

TC Time of Concentration

PET Potential Evapotranspiration

PEPF Percent Error in peak flow

RRMSE Relative Root Mean Square Error

R2 Coefficient o f Correlation

ELE-UP Elevation-up

ELE-DS Elevation-down stream

INTRODUCTION
 CHAPTER I
INTRODUCTION

Water is an essential resource for human life and all other spheres o f his
activity like agriculture, industries etc. W ater is becoming scarcer due to the
increasing consumption o f exploding world population and due to the decreasing
quality o f water by various contaminations. A judicious use o f this scarce resource
requires scientific management by way o f conservation and planning. Water
resources conservation and management plans are made on a watershed basis as it is
the basic unit for the water balance studies. Rivers, around which all the major
civilizations in the world arose, play a major role in the hydrological response o f a
watershed. The ever-growing demand o f water in domestic, agricultural and industrial
sectors calls for better management o f the water available in the rivers. This requires
the study o f precipitation, hydrological response o f the catchment and its relation to
the catchment characteristics.

Kerala, situated on the south-west com er o f India, receiving an average

rainfall o f about 300 cm and blessed with 40 minor and 4 medium rivers, and the
chain o f backwater bodies, ponds, tanks, springs and wells are often considered as a
land o f water. However, Kerala experience severe shortage o f water during the
summer months. This is due to the peculiar topography o f Kerala in which most o f
the rivers flow towards the West in a short length with a steep slope, emptying them
in a short spam At the same time, Kerala has 3.6 times more population per square
km when compared to Indian national level scenario. Hence only the proper
management o f water resource o f Kerala could make the situation comfortable.
Bharathapuzha is one o f the medium rivers which supplies water to a considerable
part o f Kerala and understanding its hydrological response is quite important for the
sustainable management o f water resources in this area. Hence, Thuthapuzha river, a
tributary o f Bharathapuzha river was chosen for this study.
Hydrological phenomena are very complex, highly non-linear and highly variable in
space and time (Patel 2014). A hydrological model is the mathematical
representations o f the response o f a catchments system to hydrologic events during
the time period under deliberation. Hydrological models are the paradigms used for
the analysis and forecast o f stream flow for the purposes o f estimating it for
sustaining not only human life but also to furnish the needs o f industries (Narasayya
2015). Here, a most widely used model suited for varying geographical locations was
selected for this study.

1.1 HYDROLOGIC M ODELLING SYSTEM (HEC-HMS)

Hydrologic Modelling System (HEC -H M S) is developed by the US Army
Corps o f Engineers’ Hydrologic Engineering Centre (HEC). It is calculated to
simulate the precipitation runoff processes o f watershed systems in a large range o f
geographic areas such as large river basins and small built-up or natural watersheds.
The system encompasses losses, runoff transform, open channel routing, and analysis
o f meteorological data, rainfall-runoff model, and parameter estimation. HEC-HMS
uses different models to represent each component o f the runoff process, models
including that compute runoff volume, models o f direct runoff, and models o f base
flow. Each one model run combines a basin model, meteorological model, and
control specifications with run options to obtain results (Kumar 2011, Majidi and
Shahedi, 2012, Halwatura and Najjim, 2013). It can provide information about
current and future runoff from watersheds, including estimates o f runoff volumes, of
peak flow rates, and o f timing o f flows.

1.2 USE OF HEC-HMS

HEC-HMS is a numerical model (computer program) that includes a large set

o f methods to simulate watershed, channel, and water-control structure behaviour,
thus predict flow, stage, and timing. The HEC-HMS simulation methods represent:
1.2.1 Watershed Precipitation and Evaporation

These describe the spatial and temporal distribution o f rainfall on and

evaporation from a watershed.

1.2.2 Runoff Volume

These address questions about the volume o f precipitation that falls on the
watershed: How much infiltrates on pervious surfaces? How much runs off from the
impervious surfaces? When the runoff does occur?

1.2.3 Direct runoff (Including overland flow and interflow)

These methods describe what happens as water that has been stored omot
infiltrated on the watershed and it moves over or just beneath the watershed surface.

1.2.4 Base flow

These simulate the slow subsurface drainage o f water from a hydrologic

system into the watershed’s channels.

1.2.5 Channel flow

These are also called routing methods simulate one-dimensional open channel
flow, thus predicting time series o f downstream flow, stage, or velocity, given
upstream hydrographs. (Ford et.al 2008)

1.3 MODEL CALIBRATION AND VALIDATION

The calibration is a process whereby the model parameter values are adjusted
so as to match the predicted model outputs, as closely as possible, with observations
from the study site. It is usually formulated as an optimization problem to determine
the best set o f model parameter values in order to minimize the differences between
the observed and simulated. The calibration and validation o f the model are essential
and critical steps in any model application. For most o f the watershed models, the
iterative procedure o f parameter evaluation and refinement is calibration, a result o f
comparing predicted and observed values o f interest (Donigian and Rao 2014). The
data from a different time period or from the different locations can be used in the
validation process, using the calibrated parameter values, in which measures o f
goodness-of-fit are used to measure the similarity between the observed and
simulated data

The model validation is actually a part o f the calibration process. Its purpose
is to assure that the calibrated model suitably assesses all the variables and conditions
which can affect model results, and express the ability to predict field observations
for periods separate from the calibration effort. The model performance after the
calibration and validation are evaluated through quantitativeandqualitative measures,
involving both graphical comparisons and statistical tests.

Steps involved in using a hydrological model for a watershed can be summarised as

follows:

> Model setup: Data collection, model input preparation, and parameter
evaluation, i.e. the steps needed to setup a model, characterize the watershed,
and prepare for model executions.
> Model testing: Involves calibration, validation and if possible post audit also.
> Analysis o f alternatives: The step is used for the ultimate use o f the model, as
a decision support tool for management and regulatory purposes
1.4 RESEARCH OBJECTIVES

The main objective is to calibrate and validate a hydrological model, HEC-

HMS for the Thuthapuzha sub-basin o f Bharathapuzha river basin in India.
Thespecific objectives o f thestudyare:

> To calibrate HEC-HMS, a physically based deterministic hydrological

model for Thuthapuzha basin
> To validate HEC-HMS for Thuthapuzha basin
REVIEW OF LITERATURE
 CHAPTER II

REVIEW OF LITERATURE

A review o f previous research works related to modelling o f hydrological

response o f watersheds, calibration, sensitivity, and validation are presented in this
chapter.

A watershed is the area o f land where all o f the water that is under it or drains
off o f it goes into the same place. According to John Wesley Powell a watershed is
that area o f land, a bounded hydrologic system, within which all living things are
inextricably linked by their common water course and where, as humans settled,
simple logic demanded that they become part o f a community. Hence watershed is
adopted as a basic developmental planning or management unit especially for natural
resources. The hydrology as well as the developmental strategy depends on the size
of the watershed. The sizes o f the watersheds vary from a few hectares to thousands
of hectares. Watersheds can be classified on the basis o f area as: micro watershed (0
to 10 ha), small watershed (10 to 40 ha), mini watershed (40 to 200 ha), sub
watershed (200 to400 ha), macro watershed (400 to 1000 ha), river basin (above 1000
ha).

Indian River basins are classified as major, medium and minor river basins
respectively based on the size o f the catchment area being more than 20,000 km2,
between 20,000 km2 and 2,000 km2, and less than 2,000 km2. (Jain et al., 2007)

Watersheds have distinct characteristics and those characteristics influencing

the runoff production are important in hydrologic analyses.

2.1 ROLE OF REMOTE SENSING AND GIS IN HYDROLOGICAL MODELLING

Su et.al (1992) used RS and GIS functions for estimating the parameters for
the distributed hydrological modelling. Here they have used two case studies. In one
 7

study, they were used RS & GIS for integrating and developing some techniques. For
the estimation of the model parameters for the second study they have taken satellite
imagery, digital elevation model as well as digitized thematic data. Since this type o f
information provides a very high resolution in space it was necessary to aggregate
small area elements into so called hydrological similar units (HSUs) in GIS. The most
difficult problem caused by large dimensions o f river basin was to cope with the
enormous amount o f data resulting from the high spatial resolution. So, the
application o f RS and GIS is essential in hydrological modelling. This coupling o f the
two models will allow the running o f the model with future scenarios o f a changed
climate. The impact of hydrological processes can be analysed with the aid o f these
two models.

Geographic InformationSystem (GIS) has emerged over the past decade as

widely used software systems for input, storage, manipulation and output o f the
geographically referenced data. (Stewart and Goodchild 1993). GIS was currently
used to assemble and manage the large databases to manage to perform the spatial
and statistical analysis and to produce effective visual representation o f the model
results.

❖ GIS was used as a preprocessor o f data, making the use o f its functions for
projection change, re sampling, generalization, windowing and other data
extraction tasks, map digitizing and editing. Reformatting was often needed to
meet the requirement o f simulation packages.

❖ GIS was used for storage o f data, database management facilities o f GIS for
keeping track automatically o f many housekeeping functions such as simple
documentation and provide a uniform mode o f access.

❖ GIS was used for statistical analysis o f data and for the model results.
 8

❖ GIS was used for visualization, particularly for presenting the results of
simulations in map form often in combination with other data.

Finally, this study examined the status o f GIS as a technology to support

environmental simulation modeling in the atmospheric, hydrologic and ecological
sciences.

Ravinder Kaur et.al (2002) made a study on GIS based digital delineation of
watershed and its advantage over conventional manual method in Hazaribagh and
Bankura district o f Jharkhand and West Bengal. The study indicated that automatic
digital delineation of watershed boundaries avoids the subjectivity in locating the ridge
lines in the manual method and thus gives more accurate shape and size o f the delineated
watershed.

Gangodagamage (2001) used the RS and GIS techniques to extract the land
surfaces, which exist as a threshold in early days to approach reasonable results in
hydrological modeling.Accurate modeling will require estimation o f the spatial and
temporal distribution o f the water resource parameters. GIS and RS have become
efficient tools to integrate the spatial and non-spatial databases for the hydrologic
modeling. In this study, hydrologic model was developed for the Bata river basin,
which is one o f the tributary of the Yamuna river basin India. Infiltration losses, unit
hydro graph and river routing were the main components o f the model. ILWIS,
ERDAS and AUTOCAD software were used. (SOI) topo maps, field data, IRS LISS
III multi temporal satellite for Rabi and kharif seasons and IRS pan data were used.
SCS CN method, unit hydrograph is used for infiltration losses and synthesis o f unit
hydrographs respectively. Watershed was divided into 10 sub areas. Muskingum
method was used for river routing. Finally, the model was capable o f forecasting
the runoff for given event o f rainfall and deriving hydrographs for the required
duration.
 9

Upadhye et.al(2005) used remote sensing and GIS technique for prioritization of
watershed for development and management. Selvi et.al (2006) studied about the
utilities and limitations o f remote sensing and GIS applications in micro-watershed
planning o f Kuruthukuli watershed in Kundah basin o f the Nilgiris district, Tamil Nadu.
Asis Mazumdar and Sujana Dhar (2007) have conducted a study to quantify the impact
o f climate change on the water resources o f the Jamtara watershed, West Bengal
using a distributed hydrological model HEC-HMS.

The Geographic information systems (GIS) evolution facilitated the use

digital terrain data for topography based hydrological modelling. (Kherde and
Shawant 2013). The use o f spatial data for hydrological modelling emerged from the
great capability o f GIS tools to accumulate and handle the data related to the hydro-
morphology o f the basin. For converting rainfall into surface runoff these models
utilize the spatially variable terrain data. In analysing and designing large scale water
resources projects manual map manipulation has always posed difficulty.

2.2. HEC-HMS

Ford et.al (2008), used the HEC- HMS program for hydrological modelling in
a watershed.

2.2.1. Applications of HEC- HMS

Fleming (2004) described the Hydrologic Engineering Center’s Hydrologic

Modelling System (HEC-HMS) program and its application in the watershed studies.
HMS has the capability to serve as a keystone program with respect to the watershed
point o f approach. It can simulate the rainfall-runoff at any point within a watershed
for the given physical characteristics o f the watershed. The HMS Program is a
valuable tool for forecasting and for managing human impact on the rainfall-runoff
response at points o f interest within a watershed. The study showed that the results
from the program can be used directly or in coincide with other software for studies
 10

of water availability, drainage from urban, forecasting o f flow, future urbanization

impact, reservoir spillway design, flood damage reduction, floodplain regulation,
wetlands hydrology, and systems operation.

Yener et.al (2006) used HEC HMS for floodplain management and flood
damage assessment studies of the Yuvacik basin. They report that missing data and
bad data must be calibrated periodically to avoid errors in flow gages and
precipitation stations.

Bakir and Xingnan (2008) used HEC-HMS for hydrological modelling. The
performance o f HEC-HMS was compared with that o f the Xinanjiang conceptual
model by using historical flood data from the Wanjiabu catchment in China. The
results obtained in this study indicated that HEC-HMS was more convenient for flood
simulation especially in optimizing parameters. But these are not accurate as
compared with Xinanjiang model. HEC-HMS is a very powerful and flexible
software and it provided enough and accurate hydrological data which is needed to
run multiple models inside it. HMS provides precipitation- requirement options, loss
models which can estimate the volume o f runoff, and direct runoff and hydrologic
routing models, and also used for the calibration optimization

Fleming et.al (2010) used HMS to estimate stream flow as a source for
drinking water and the water effluent treatment works, such as shipping, receiving,
navigation. However, analysis o f the water o f a river plays a key role in the
development o f plans for protecting the city. Analysis o f the typical components such
as pump stations adjacent to the holding ponds,and river banks are expensive.

Razi et.al (2010) developed the flood model by using HEC-HMS for the Johar
river in Kotta Tinggi watershed. Calibration and validation procedures were carried
out using different sets o f data. The evaluation and performance o f the model was
carried using HEC-HMS yield a correlation coefficient close to 1. The percentage of
 11

error values were 4% and R was 0.905. Based on these findings, they suggested
HEC-HMS can be used as a tool for estimating the Qpeak.

Martin et.al (2012) performed the hydrological modelling using HEC-HMS

software, by delineating the catchment using HEC-GEOHMS in Arc-GIS
environment, populating the meteorological model with design storm data, and
defining control specifications. The 10, 50, 100, 250, and 500-year design storms data
input into the model generated design floods o f 71.8 m /s, 123.0 m 7s, 138.5 m / s ,
163.9 m3/s, and 183.4 m3/s magnitudes respectively. Hydraulic modelling was
performed using HEC-RAS software. The model output channel flood depths at the
gauging station were 5.21 m, 6.53 m, 6.84 m, 7.31 m, and 7.65 m for simulated 10,
50, 100, 250, and 500-year design floods respectively. Flood hazard maps were
generated by exporting the HEC-RAS model output results to Arc-GIS where they
were processed to identify the flood prone areas. The most flood prone areas were
found around the river middle reach by using flood hazard maps

The flood frequency analysis allows the estimation o f design floods with
hydrological modelling for poorly observed basins considering change and also
taking into account the flood protection measures (Haberlandt and Radtke, 2014)
There are several possible choices related to precipitation input, discharge output and
consequently for the model calibration. The main aim o f their study was to compare
different calibration strategies for a hydrological model by considering various types
o f rainfall input and runoff output data sets to propose the most suitable approach.
One main purpose of that paper was to introduce the idea o f calibrating a
hydrological model based on flood frequency distributions using stochastic rainfall
and to evaluate it against classical strategies in an empirical case study. The results
have shown the suitability o f this approach. Generally, this approach may also be
suitable in climate impact studies where hydrological models could be calibrated
directly using the simulated precipitation from regional climate models against
observed flow statistics.
 12

Kabiri et.al used HEC-HMS and HEC-GEOHMS for the accurate calculations
and predictions for flooding events in the Kuala Lumpur watershed. That is located in
Klang basin o f Malaysia. The main aim o f their research was to estimate the peak
flow in the Klang watershed by using Green-Ampt method as a major component in
runoff and flood modelling. The flow data anddaily surface runoff were compared to
observed values at the outlet o f the watershed. The catchment delineation o f the
Klang watershed to obtain the sub-watershed parameters by using HEC-GEOHMS
extension in ARC-GIS. The results showed that the flow measured at a high accuracy
and the discharge at the outlet o f the tank showed a good fit.

Twumasi et.al (2015) used the hydrological modelling with remote sensing
data and GIS techniques for flood volume computation in Dikpe Catchment using
temperature, rainfall and flow rate values. Many different methodologies have been
tried for using Digital Elevation Model (DEM) for hydrological modelling and
watershed delineation. The magnitude o f flood in the delineated catchment area is
calculated using the HEC- HMS.

2.3 HEC-HMS FOR THE PARAMETER ESTIMATION

Kumar (2011) prepared a hydrological model using remote sensing and GIS
technique. Three methods viz. SCS unit hydrograph transform method, SCS curve
number method, constant monthly method were used to compute the direct surface
runoff hydro graph, runoff volume and base flow separation. Both lumped and
distributed models were simulated and validated by using the rainfall stream flow
data. From the observations, it was found that for estimating infiltration parameters
and also for simulating daily stream flow the HEC-HMS model is consistent.

Arekhi (2012) used the HEC-HMS model and results o f evaluation o f

different methods for runoff losses (Green and Ampt, Initial and constant loss rate
and Deficit and Constant loss) He considered different objective functions (percent
error in peaks and volumes) to classify the methods. The method with less difference
 13

between percent o f observed and simulated discharges got 1st preference and other
methods were given subsequent preferences. Results showed that constant rate and
initial loss had better result for six events, (4 events-PEP, 5 events- PEV), while
Deficit and Constant loss had less changes rather than Green and Ampt method in
(2events-PEP, 3events-PEP). So, Initial and constant loss rate method is the 1st
preferred and for the simulation o f surface runoff and Green & Ampt and Deficit and
Constant loss methods took the next preferences.

Majidi and Shahedi (2012) used the HEC-HMS hydrological model 3.4
version to simulate rainfall-runoff process in Abnama watershed in southern Iran. To
compute infiltration, rainfall excess conversion to runoff and routing o f flow,
methods like Green-Ampt, Unit hydrograph o f SCS and Muskingum routing were
chosen. Rainfall-runoff simulation was con-ducted with five events, and starting
results for the first four events showed differences between simulated and observed
discharges.

Rafi et.al (2012) used HEC-HMS model for comparison and classification of
precipitation loss estimation methods with the hold o f two objective functions for
selection o f the best method. Selected methods for runoff loss evaluation were Initial
and constant, Green & Ampt, and SCS curve number in relation to various functions
(PEP, Peak-Weighted Root Mean Square Error). W ith respect to the Root Mean
Square Error o f the peak weighted and Sum o f Absolute Residuals, Green & Ampt
method had better results than SCS and Initial and constant rate methods.

Asadi and Boustani (2013) used hydrologic model HEC-HMS along with the
geospatial hydrologic model extension, HEC-GEOHMS. By using historical observed
data, the model was carefully calibrated and verified. A local sensitivity analysis was
also adopted for evaluating the event model. The results showed that in both the
lumped and distributed models, the maximum difference between the baseline peak
hydrographand hydrograph was caused by initial abstraction Ia.It was also found that
 14

semi-distributed model captured the runoff o f peak discharges and total volume of
runoff these are better than the lumped model.

Chatterjee et.al (2014) stated that the rate o f infiltration and percent o f
impervious area are the two sensitive parameters for the simulation o f stream flow
using HEC-HMS. In their study Nash-Sutcliffe model efficiency criterion of
percentage error in volume, percentage error in peak and net difference o f observed
and predicted time to peak were used for performance evaluation. The model
established good performance, with above-mentioned performance indices for
simulation o f stream flow
Choudhari e/.a/(2014) selected SCS curve number and SCS unit hydrograph
methods for calculating runoff volume, peak runoff rate and base flow. Mean
absolute relative error and root mean square error were the statistical tests for
calibrating the model. The results indicated values o f MARE o f 0.2 and 0.25 for
runoff depth and peak discharge, respectively. Similarly, the values o f RMSE are
2.30 mm and 0.28m /s for runoff depth and peak discharge respectively. After
parameter optimization, the error functions reduced to 0.10 mm, 0.12m /s, 0.75 mm
and 0.09m3/s in sequence. Also, the model validation was also found to be
satisfactory with low values o f statistical error functions.
Narasayya (2015) used a GIS based hydrological model to study the stream
flow estimation. Based on the statistical results and graphs, it was evident that the
SCS-CN loss approach gives higher stream flow volumes than initial and constant
loss method from both Muskingum and Lag routing methods. Ultimate peak flows
and maximum stream flows in Muskingum and Lag routing methods are possible by
taking into consideration of SCS-CN loss method. Stream flow volume can also be
assessed by taking into consideration the higher spatial resolution o f DEM and
temporal resolution o f satellite imageries is the final conclusion drawn from this
study.

I
 15

Sampath et.al (2015) used a hydrological modelling system HEC-HMS 3.0.1

to estimate runoff in the Deduru Oya river. Long-time daily rainfall data in several
rain gauge stations, land use, and soil data in the river basin were selected. Five-layer
soil moisture accounting loss method, Clark unit hydrograph transformation method,
and base flow o f recession method o f the HEC-HMS model were used. The results
depict the capability of HEC-HM S to repeat stream flows in the basin to a high
accuracy with computed averaged by using Nash Sutcliffe efficiencies o f 0.80. In the
Deduru Oya river basin the model developed is a tool for water management.

2.4 WATERSHED DELINEATION BY USING DEM

Oloche et.al (2010) used the hydrologic model HEC-HMS (Hydrologic

Engineering Center, Hydrological Modelling System), in combination with the
geospatial hydrologic modelling system, (HEC-GEOHMS. HEC-HMS is seldom
applied in chain for flood forecasting. The model is calibrated and verified using
historical data. Flood events had the coefficient o f determination above 0.9, and peak
discharges o f virtual flows were all in acceptable range.

Bhatt et.al (2012) used the HEC-HMS in Bhagirathi basin along with
historically recorded rainfall and flow data. A digital elevation model (DEM) map
was obtained from the SRTM (Shuttle Radar Topography Mission). Topographic
Parameterization (TOPAZ) computer program was used to create the basin boundary
from flow accumulation. Daily rainfall data over the years 1992-2002 at the outlet
point (Tehri Dam) was analysed for the same years, in addition to the daily flow o f
data. Daily stream o f data from the point o f outlet flow hydrographs were used for the
calibration and verification purposes.

N ag et.al (2013) studied rainfall runoff model using HEC-HMS and remote
sensing and GIS techniques in the lower basin o f Woochu river in the Paro
Dzongkhag, Bhutan. Precipitation data for 10 years, topographic map, DEM, and
Landsat images o f the study area were collected. Time series data o f rainfall and
 16

discharge was analysed in MS-Excel. For the automatic delineation o f the watershed
Arc-GIS has been used with the SWAT tool. Calibration parameters have been
validated for the coming 10 years.

Hashemyan et.al (2015) investigated the behaviour o f the rivers flood and
how to use in a reach o f Khoshke Rudan river, located in the Chaharmahal and
Bakhtiari province, Iran by using HEC-HMS and HEC-RAS. It was calibrated by
data from rainfall gauging stations in the neighbouring basin. 10, 20, 50 and 100-
year-long relationship with the precipitations in flood zones were determined using
the extension HEC-GeoRAS using GIS environment. The results obtained from the
study confirmed the ability o f the combination o f GIS and HEC-RAS model and
performance model.

Yasin et.al (2015) created a hydrological model by using HEC-HMS and

HEC-Geo-HMS for Mithawan watershed. DEM and HEC-GEOHMS was used to
extract a watershed model. Frequency analysis o f Mithawan basin rainfall and run-off
during different periods o f time was conducted using the simulation output from
HEC- HMS model. The results o f the study in Darraha Mithawan watershed showed
agreement in the accuracy o f peak discharge o f the model simulation.

Gautam (2014) used the GIS based semi-distributed model named HEC-HMS
in the Narayani river basin. It was analysed by using DEM, evapotranspiration, soil
type, and land use data. The GIS based extension tool HEC-GEOHMS and HEC-
GEORAS were mainly used for the preparation o f inputs for HEC-HMS. The model
provides the best result as a function o f peak flow and time to peak. Hydrologic
model parameters can be derived from historic stream flow, precipitation and GIS
database.
 17

2.5 RUNOFF ESTIMATION BY USING CN METHOD

Al-Jabari et.al (2009) used watershed modelling for the estimation o f the
runoff for the Wadi Su’d Watershed. So Many methods were used to estimate the
runoff from a watershed. The curve number method, also known as the hydrological
soil cover complex method, was a flexible and widely used method for runoff
estimation.This method includes several important properties o f the watershed
specifically, soils permeability, land use and antecedent soil water conditions. The
study estimated the flow o f data on rainfall and land use classification and infiltration
rate o f the soil for the use o f the experimental data. In the present study area (using
Su'd basin), the results showed that the average annual runoff depth o f 36.3 mm. This
approach could be applied in other Palestinian watersheds for planning o f various
conservation measures.

This study reports the approach o f remote sensing and GIS-based

hydrological model for the urbanization and its impact to water resource, i.e. runoff
(Abas and Hashim 2012).The runoff estimation was based on the Soil Conservation
Service Curve Number (SCS-CN) model developed by the USDA-Soil Conservation
Service(SCS) in 1972. The use o f this model in relation to the curve o f the earth
Numbers (CN), soil type, soil moisture, runoff water is counted on the situation. For
the Water resource management,remote sensing technique was widely used to
determine the surface runoff. The main inputsfor this model were rainfall and CN.
The CN represents the runoff potential those were based on the natural factors such as
land use/cover, soil type, antecedent soil moisture and rainfall (amount, timing and
magnitude) may influence runoff. Kuala Lumpur revealed that urbanization due to
human activities could lead to greater runoff and risk o f flood disasters.

For any hydrological studies in an un-gauged watershed, A methodology

has to be selected for the determination o f watershed characteristics and runoff at its
outlet (Al-Jabari et a i, 2012). And a number o f methods were used to determine the
 18

morphological parameters to estimate the runoff in the watershed. The Soil

Conservation Service (SCS) curve number method and Watershed Modeling System
(WMS) are versatile and widely used procedure for watershed hydrological studies.
These methods include several important properties o f watershed namely, soils
permeability, land use and antecedent soil water conditions which were taken into
consideration.Geographical Information System (GIS) was used to estimate the
runoff from a small agricultural watershed as well as the morphological features o f
these watersheds.

Gajbhiye 2015 used the CN method for estimating the surface runoff in the
Kanhaiya Nala watershed located in Satna district o f Madhya Pradesh. In this the
total area o f watershed was 19.53km2. Land Use,Soil map and slope map were
generated in GIS Environment. Since the curve number method was used here as a
distributed model, it is required to get information regarding a large number o f sub­
catchments in the basin.Direct runoff in a catchment depends on type o f soil, land
cover and rainfall. Out O f the available methods (SCS-CN) was the most popular
method for estimating runoff from rainfall. The thematic layers like soil map,
elevation map, rainfall map and land cover maps were created in Arc GIS 9.3. In
present study, the runoff varies from 1196.93 mm to 1551.17 mm o f the study area.

2.6 CALIBRATION AND VALIDATION OF HYDROLOGICAL MODEL

Alaghamand et.al (2011) studied the evaluation o f different sets o f calibrated

and validated data by using HEC-HMS 3.1.0 for the Sungai Kayu Ara river basin,
which was located in the west part o f the Kuala Lumpur in M alaysia was the case
study o f this research. In this research imperviousness, lag time and peaking
coefficient had taken as the calibrated parameters. Average, median and mode were
calculated for selecting best set among these values. The results showed that R 2
values for average median and mode were 0.8922, 0.8954, and 0.8678respectively. It
can be concluded that average and median parameter values for the calibration set
 19

was more accurate and reliable compared to the mode o f parameter values o f
validation.

Ali et.al (2011) was used the HEC- HMS (rainfall runoff model) to quantify
the effects o f land use changes in the Lai Nullah basin.The model was calibrated and
validated for the five storm events and the results showed good consistency between
the simulated and observed hydrograph from 76 to 98% o f the Nash - Sutcliffe model
efficiency at the outlet o f the basin. The future land use scenario was forecasted based
on Islamabad Master Plan and growth pattem.The anticipated results o f the master
plan for land use in the futureincreased runoff between 51.6 and 100% and peak
discharge between 45.4% and 83.3% that the total flow was expected to increase. The
results provided the useful information for land use planning and management and
these methods applied can serve as a useful tool for future land use impact studies.

Abood et.al (2012) used the hydrological model (HEC-HMS) to evaluate the
performance o f the Kenyir catchment and Berang catchment in simulating the
rainfall-runoff using two different infiltration methods for estimating infiltration
parameters (abstraction losses, rainfall). These methods include SCS Curve Number
method and the Green and Ampt method. Statistical analysis functions they have used
were coefficient o f determination (R2), the mean square error (MSE) and mean
absolute percentage error (MAPE).And also, they have used HEC- HMS for the
accuracy o f simulations. The results o f the simulation using the SCS Curve Number
method mistakes output was 6.5%, 8.2%, respectively for Berang, Kenyir
catchmentsand also for the errors o f Green and Ampt method was 9.13%, 11.11%. As
a conclusion, it was found that the output o f the program was found to be in
agreement with the historical record o f data and they were suggested that to use the
SCS Curve Number infiltration method for the humid catchment was due to its low
error in simulation output.
 20

Asadi and Boostani (2013) used HEC-HMS for the rainfall runoff process in
small sub basin Delibajak and Kabkian basin in combination with HEC-GEOHMS.
SCS curve number method o f rainfall - runoff modelling and model was considered
to be carefully calibrated with the simultaneous. The review o f the historical data was
verified using the sub-basin. Flood events and deal with all the points for the
determination o f the coefficients were above 0.9, and the highest percentage o f errors
in the flow and volume are in the acceptable range. As well, in these two basins the
hydrologic parameters i.e. curve number and initial abstraction were compared.

Majidi and Vagharfard (2013) analysed surface run-off simulation in Abnama

watershed with Green-Ampt method and Soil Conservation Service (SCS) methods
using HEC-HMS hydrological model. After the input parameters to determine the
methods for selecting the appropriate method for two different managed using the
HEC- HMS model was calibrated and validated for hydrograph and related
hyetograph had been used for four events. Model calibration and validation results o f
the Green Ampt method to estimate the difference in the lower peak discharge and
time to peak was less than the SCS system. Minitab software, as well as their values
and the correlation between simulated and observed hydrographs Green Ampt
method o f comparison based on the results o f the SCS method has been shown to
correlate with more than SCS method. It can be concluded that simulation using
Green-Ampt method was more precise than SCS method

Roy et.al{2013) used the soil moisture accounting algorithm (SMA) method
for the predictionof the response o f the watershed by using HEC-HMS.It has been
calibrated and validated for Subamarekha river basin in Eastern India. The analysis
shows that the soil storage, tension zone storage and groundwater 1 storage
coefficient to be the sensitive parameters for the simulated stream flow. The Nash -
Sutcliffe model efficiency criterion, percentage error in volume, the percentage error
in peak, and net difference o f observed and simulated time to peak, which were used
for performance evaluation, have been found to range from (0.72 to 0.84), (4.39 to
 21

19.47%), (1.9 to 19%) and (0 to lday) respectively, indicating a good performance of

the model for simulation of stream flow and thereby quantification o f available water.

Kabiri (2014) had investigated that the assessment o f the SCS CN loss method
in Klan-g watershed to evaluate the performance o f the SCS CN loss method. They
concluded that SCS-CN loss method can be used for Klan-g watershed due to its
good agreement between observed and modelled in HEC-HMS. The results revealed
that a modified CN o f initial abstraction ratio o f 0.5 gives better fit than 0.2.
Therefore, CN 0.05 can be used for the runoff simulation o f SCS method in Klan-g
watershed.

Yao et.al(2014) used the study to assess the impact o f underlying surface
change on catchment hydrological response using the Hydrologic Modelling System
(HEC-HMS). Twenty-one flood events were selected for calibration and validation of
the model parameters. Using the satisfactory results HEC- HMS model used to
simulate the response o f Dong-wan watershed to show that the runoff in the
catchment rainfall

Legesse (2015) used HEC-HMS3.5 hydrologic model HMS for the calibration
and validation o f the upper Blue Nile River Basin by using soil moisture accounting
algorithm-SMA. The Nash-Sutcliff (E) and Coefficient o f determination (R2) used to
evaluate the performance o f the model. The results obtained are suitable and accepted
for simulation o f runoff, some o f the best fit performed methods o f the hydrological
processes o f direct runoff transformation, infiltration loss and base flow are the
deficit and constant loss method, Snyder unit hydrograph method and exponential
recession methods. By analysing these" results he suggested that HEC-HMS
simulation model was suitable for the upper Blue Nile river Basin.

Rahul et.al (2015) used HEC-HMS by using SMA method to model stream
flow in the Vamsadhara River Basin in India. The SMA algorithm was calibrated
using data from 1984 to 1989 by using HEC-HMS and between 1990 to 1993 at a
l
 22

fixed time period had been validated. Using statistical and visual research
Vamsadhara River basin will be carried out to determine the performance o f HEC-
HMS model. The arrangement o f the sample period, which ranges from a peak
performance o f 0.21 % (R2-0.71, Nash - Sutcliffe Efficiency EFF - 0.701, percentage
error in volume PEV = 2.64% percent error in peak PEP= 0.21% and index of
agreement -0.94) Similarly, as long as the ranges from the validation model, was
good to very good (R2-0.78, EFF - 0.762, PEV -12.33% , PEP- -15. 2% and D-0.93).

Supe et.al{2015) used HEC-HMS model to simulate continuous rainfall-

runoff modelling for Wan river basin, Akola, Maharashtra.Rainfall-runoff
relationship was modelled using soil-moisture accounting method (SMA). The
calibrated model parameters i.e. Groundwater 1, Groundwater 2, GW1 coefficient
and GW2 coefficient were fixed as 72, 10, 387 and 1010, respectively, for the model
performed well in terms o f RMSE, R 2ns, CRM (0.12 m3/s,0.93 and -0.02).
Considering the performance o f model in simulating the runoff, it is suggested that
the calibrated HEC-HMS model could be used to predict runoff for the rainfall events
for Wan river basin.

2.7 WATERSHED MANAGEMENT

Assessment o f watershed management operation is one o f the main subjects

for future planning o f practical projects and natural resources management. Flood
damage was one o f the most important problems in countries same as Iran, which was
mostly effected and caused the hazards (Golrang et a l, 2013). Therefore, the
identification o f the area with high potential risk o f flood occurrence was the main
purpose is to the flood control and reducing its damage.

The purpose o f this study was evaluation o f watershed management activities

in Kushk-Abad Watershed by HEC-HMS (Hydrologic Modelling System). HEC-
HMS is one of the computer models for simulation o f its ability for the short
timeperiod events. HEC - HMS was physically processed using extension and
i
 23

divided into 6 sub-basins. The main aim o f the study was evaluation o f HEC-HMS
model using SCS unit hydrograph method in basins, and results showed that in the
bell form (Normal) hydrographs. These parameters were imported to HEC-HMS to
find out the effects o f watershed practices and then flooding condition was simulated.
For assessment purposes, peak discharge and flood volume were calculated for
“before” and “after” construction conditions. Results showed that check dams as
mechanical measures had low effect on time o f concentration while biological
practices lead to decrease in curve number with an average value o f 4.5. This results
in decrease o f peak flow and flood volume meanly 19% and 14% respectively.
MATERIALS AND METHODS
 24

CHAPTER III
M ATERIALS AND METHODS
3.1 STU D YAREA
The w atersheds from the central part o f Kerala State. India have been used in

this study. A m ong these .one is the sub basin o f a m ajor river ‘B harathapuzha’.

3.1.1 T huthapuzha sub basin

T hutha River is one o f the m ain tributary o f the B harathapuzha (N ila River).

This is the second longest river on the south w est coast o f India. B harathapuzha

originates from the W estern G hats and has a total catchm ent area o f 6400km 2.o f

which about70% spreadin Kerala and rem aining in T am ilnadu. T huthapuzha

w atershed lies betw een 10' 5 0 ' to l l " l 5 ‘ North latitude and 76"5‘ to 76(l40'E ast

longitude. T he w atershed has a total area o f 1018 km 2 and covers 27 panchayats 6

blocks and tw o districts. This river is the m ain tributary w hich supplies w ater to

B harathapuzha river especially during sum m er. C atchm ent area at Pulam anthole river

gauging station (10°53’5 0 ” N .76°11’5 0 ” E) m anned by Centra! W ater C om m ission,

Indiais822 km 2.

Fig 3.1 Location o f Thuthapuzha watershed

 25

3.2 PHYSIOGRAPHIC AND METEOROLOGICAL INFORMATION IN

THUTHAPUZHA WATERSHED

3.2.1 Rainfall

Daily rainfall data and also climatic parameters such as temperature,

humidity, wind speed, solar radiations and sunshine hours have taken from the RARS
Pattambi.

3.2.3 River flow

Daily river flow has been collected from Central Water Commission i.e.
Pulamanthole gauging station and that was managed by the central Water
Commission.

3.2.4 Soil data

Soil map and attribute information have been collected from the Kerala State
Land Use Board (KSLUB), Trivandrum.

3.2.5 Land Use

Land use map and attribute information have been collected from the Kerala
State Land Use Board (KSLUB), Trivandrum. The land cover consists o f the
residential, commercial and services, strips mines quarries and pits etc.

3.2.6 DEM

DEM o f the study area has taken from the Cartosat-1: DEM - Version-3Rl 1
arc sec (~ 32 m) resolution was downloaded from ISRO, National Remote Sensing
Centre’s (NRSC), India Geo-platform Bhuvan website (http://bhuvan.nrsc.gov.inl
 26

3.3 GIS APPLICATIONS IN HYDROLOGY

3.3.1 GIS (Geographic Information System)

Geographic information system can be defined as a computer data base

system for the capture, storage, retrieval analysis and display o f tabular data and
spatial data. GIS has a wide application in a variety o f civil and environmental
applications, including demography, meteorology, transportation, urban planning, and
hydrology. In the past three decades, hydrologic applications have been
revolutionized by GIS, making it an indispensable technology for digital terrain and
hydrologic analysis of watersheds. The role and usage o f GIS technology, software,
data formats, and sources of geospatial data are essential to hydrologic analysis. It
Includes

> Attribute or feature information in a database format

> Location information for these attributes or features
> Analysis functions for deriving new information on spatial relationships
Active involvement in the hydrology by the GIS community is represented
today by the user groups devoted to GIS technology applications to the hydrology,
natural resources management, or urban planning. National hydro graphic and
elevation datasets are available from the U.S geological survey (USGS), digital soils
data from the natural resources conservation service (NRCS). GIS is used for
watershed delineation, runoff estimation, flood plain modelling and hydrologic and
hydraulic modelling in hydrology. Digital representation o f topography, soils, land
use/land cover, and precipitation may be accomplished using GIS data and
methodology. The applications are enhanced through the use o f GIS, because
hydrology is inherently spatial in nature.
3.3.2 Raster and vector data
GIS data are usually formatted in a vector or raster structure. A vector dataset
maps a geographic feature using lines drawn between points or coordinate pairs. An
 27

advantage o f vector formats is that a database containing multiple attributes may be

associated with any given point, line, or polygon. Whereas in the raster format, only a
single attribute is usually associated with a grid cell. Vector data sets are used to
show rain gages as points, streams as lines, and the watershed boundary as a polygon.
A raster o f grid cells represents rainfall in relation to the watershed and other features
in the map
3.3.3 Map scale and spatial detail
The scale and resolution at which the data are collected or measured on a map
is termed as resolution (or)native scale. If the mark or point elevations are surveyed
in the field on a grid o f 100 m, then this is its native resolution. A small-scale map is
one in which features appear small. Conversely, large-scale maps have features that
appear large and show significantly more detail.
E.g.: If a paper USGS quad sheet is digitized, the native scale will be 1: 24000, as a
result performing analysis o f this data at a 1:1000 scale could adversely influence the
accuracy o f the analysis.
3.3.4 Datum and spheroid
Geodesy is the branch o f applied mathematics concerned with determining the
size and shape of the earth and the exact location o f the points on its surface. The
earth is often treated as a sphere to make mathematical calculations easier, but it is
actually a spheroid.
The earth must be treated as a spheroid to maintain accuracy in maps o f large
spatial extent, and numbers of standard spheroids are commonly used to describe its
shape. In North America, these include the Clarke 1866 spheroid and the GRS80
(Geodetic reference system of 1980). A horizontal datum is a reference frame used to
measure locations on the surface o f the earth. It defines the origin and orientation of
the lines o f latitude and longitude. A horizontal datum is always related to a specific
spheroid. There are two types o f datum available - those that are earth centred
 28

(geocentric) and local. Local datum is NAD27 (North American datum). Earth
centred (geocentric datum) is NAD83.
3.3.5 Geographic and Cartesian Coordinate Systems
A geographic coordinate system is based on latitude and longitude. A
transformation or projection from a curved geographic coordinate system to a
Cartesian coordinate system is necessary to view all or part o f the earth as a flat map.
Because it is impossible to represent a three-dimensional figure such as the earth as a
two-dimensional map with minimal distortion, various projections have been
developed to preserve one or more o f the following properties: shape, area, distance,
and direction. No projection maintains all o f the above four characteristics, but some
projections maintain more than one. For Example: The Universal Transverse
Mercator (UTM) projection commonly used in the United States.
3.3.6 Land use/land cover
Land use/cover may be used to develop evapotranspiration rates or estimates
o f hydraulic roughness from a raster array or from polygonal areas o f remotely
sensed substitute measures. A GIS map of the land use and associated cover is an
important descriptor of the hydrology, whether it is vegetative, such as pasture or
forest, or the surface is modified with asphalt or concrete pavement. Land use and
land cover (LULC) GIS data are compiled for many applications and may be useful
for hydrologic modelling.
General purpose LULC maps may or may not contain sufficient details or
classification to be useful for deriving hydrological model parameters. LULC
classification maps can be used to develop hydrologic parameters such as hydraulic
roughness, Aledo or aerodynamic roughness height, which influences
evapotranspiration
Hydraulic roughness is a parameter that controls the rate o f runoff over the land
surface. Classification codes from the Anderson classification scheme are listed
below.
 29

Table 3.1 Anderson Land Use Land Cover Classification Codes: First- and Second-
Level Categories

1. U rban o r Built-Up L and 6. W etland

11 Residential 6 1 Forested Wetlands

12 Commercial Services 6 2 Non forested Wetlands

13 Industrial 7. B arren Land

14 Transportation, Communications 71 Dry Salt Flats

15 Industrial and Commercial 72 Beaches

16 Mixed Urban or Built-Up Land 73 Sandy Areas Other than Beaches

17 Other Urban or Built-Up Land 74 Bare Exposed Rock

2. A gricultural Land 75 Strip Mines, Quarries, and Gravel Pits

21 Cropland and Pasture 76 Transitional Areas

22 Orchards, Groves, Vineyards, Nurseries 77 Mixed Barren Land

23 Confined Feeding Operations 8. T undra

24 Other Agricultural Land 81 Shrub and Brush Tundra

3. Rangeland 82 Herbaceous Tundra

31 Herbaceous Rangeland 83 Bare Ground

32 Shrub and Brush Rangeland 84 Wet Tundra

33 Mixed Rangeland 85 Mixed Tundra

4. Forest Land 9. Perennial Snow and Ice

41 Deciduous forest land 91 Perennial Snow fields

42 Evergreen forest land 92 Glaciers
43 Mixed forest land
5. W ater
51 Streams and canals
52 Lakes
53 Reservoirs
54 Bays and estuaries
 30

3.3.7 Soil mapping unit

Computation o f direct runoff requires an estimate o f infiltration characteristics
for the different soil types in a drainage area. A soil-mapping unit is the smallest unit
on a soil map that can be assigned a set of representative properties. Soil maps and
the associated soil properties form a major source o f data for estimating infiltration.
Some adjustments are needed when estimating the infiltration parameters from
generalized soil databases. Obtaining infiltration parameters from soil properties
requires reclassification o f the soil mapping unit into a parameter meaningful to the
hydrological model. Soil-type data were widely available for the United States and
can be downloaded from the Natural Resources Conservation Service at
(www.nrcs.usda.gov.in).
a) Deriving hydrologic parameters from GIS maps o f land use and soils
Generalized land use/cover maps are now available for most locations. Such
maps may be useful for hydrologic analysis if the mapped categories are
representative o f current conditions or those present during a particular storm event.
To obtain hydraulic or hydrologic parameters from such a map, reclassification is
necessary to convert the dominant land use/cover into hydrologic model parameters.
Interpretation o f the dominant land use category is needed to express the
category in terms o f an appropriate value o f the corresponding hydraulic roughness
parameter. Because a particular land use category may be heavily or lightly
vegetated, the assigned hydraulic roughness should be adjusted to represent this in the
model. Similarly, soil maps can be reclassified based on soil texture into infiltration
parameters for use in the Green-Ampt equation. Hydrologic Parameters from GIS
Maps o f Land Use and Soils may be found on the textbook website along with
sample data.
The steps followed in the presentation are:
• Reclassify land use/soils datasets to-hydrologic parameters defining hydraulic
roughness and infiltration.
 31

• Use multiplicative factor when reclassifying integer-based land use/soils

classes, then divide out multiplicative factor with Raster Calculator (RC).
• Resample roughness/infiltration values to desired model grid resolution.
• Clip the re sampled grid to a specific model domain.
• Export roughness/infiltration grids and load in as ARC-GIS Grids or ASCII
files
The resulting watershed maps demonstrate the reclassification o f GIS maps of
land use/cover and soils into hydraulic and hydrologic model parameters. The
resulting maps are for use in the distributed hydrologic model, alternatively, the same
approach to reclassification can be applied to obtain parameters for input to HEC-
HMS by averaging the distributed values over each sub basin area within the
watershed.
3.3.8 Digital elevation model (DEM)
One o f the most useful types of geospatial data is digital terrain.Hydrologic
quantities can be derived from a DEM. The terrain attributes such as slope and
drainage direction are derived from a DEM. A model o f terrain traditionally relies on
contours of equal elevation that define watershed boundaries, channel slope, and
other features.
Within a GIS, representation o f terrain can be accomplished with contours,
gridded elevations (also known as a raster DEM), and triangular irregular networks
(TIN). Watershed characteristics such as slope and drainage length can be derived
from a digital model o f the terrain. A raster DEM consists o f groupof numbers
indicating the spatial distribution o f elevations.
3.3.8.1 GIS Processing o f digital terrain
A basic application o f GIS in hydrology is the delineation o f watershed
boundaries and corresponding stream network. Several modules are available for
automatic delineation from a DEM. Because a vector map o f the watershed boundary
or stream channel location may have more accuracy than the DEM, the automated
 32

delineation can be improved using the vector maps to constrain the delineated
boundary and stream channel location. Constrained watershed delineation helps
ensure that the resulting maps are consistent with the watershed boundary and stream
channel mapped in vector format. This example provides-an overview o f how to burn
in a DEM with a stream network and use the watershed boundary to constrain the
watershed delineation. The watershed delineation relies on the Arc Hydro tools
extension in Arc-Map.
The steps followed in the presentation are:
• Re project original DEM to the desired projection with horizontal and vertical
units of meters.
• Resample original DEM to a selected resolution.
• Merge stream network and basin/sub basin outline into a DEM to ensure
proper flow direction.
• Use Arc Hydro Tools extension to produce flow direction, flow accumulation,
and a stream grid.
The resulting watershed maps demonstrate the burn-in procedure using a
combination o f raster and vector map information. The flow direction map is for use
in the distributed hydrologic model. An alternative approach is to use the HEC-
GEOHMS processing module to process the DEM and to produce sub basin
boundaries and stream channel map information for input to HEC-HMS.
3.3.9 Triangulated Irregular Network (TIN)
It represents a surface as a set o f non over lapping contiguous triangular facets
o f varying size and shape. Generally, a TIN is produced from a DEM through GIS
software and has several distinct advantages over contour and raster representations
of surfaces. The primary advantage is that the size o f each triangle may be varied
such that broad flat areas are covered with a small number o f large triangles, while
vastly variable or steeply sloping areas are covered with many smaller triangles. This
provides some storage and detail efficiency over raster data structures, since the

I
 33

element may vary in size according to the variability o f the surface. TINs have
become increasingly popular because o f their efficiency in storing data and their
simple data structure for accommodating irregularly spaced elevation data.

3.4TOOLSANDTECHNIQUESUSEDINTHISSTUDY
3.4.1 HEC-GEOHM S
The Geospatial Hydrologic Modeling Extension (HEC-GEOHMS) has been
developed as a geospatial hydrology tool kit for hydrologists and engineers with
limited GIS experience. HEC-GEOHMS uses Arc-GIS and the Spatial Analyst
extension to develop a number o f hydrologic modelling inputs for the Hydrologic
Engineering Center's Hydrologic Modeling System, HEC-HMS. ARC-GIS and its
Spatial Analyst extension are available from the Environmental Systems Research
Institute, Inc (ESRI). The program allows users to visualize spatial information,
document watershed characteristics, perform spatial analysis, and delineate sub
basins and streams. Working with HEC-GEOHMS through its interfaces, menus,
tools, buttons, and context-sensitive online help allows the user to expediently create
hydrologic inputs for HEC-HMS.
(a) Data Management
The USGS digital elevation model (DEM) contains terrain cells of surface
elevations at a spacing o f 30 meters in both the horizontal and vertical direction. This
DEM data was used to generate the stream network and sub basin areas within the
watershed. All data layers were projected to the WGS 84 Northern hemisphere
Plane coordinate system. '
i
(b) Terrain Pre-processing
Before hydrologic modelling with HEC-HMS was possible, the terrain model
is used to derive eight datasets described in proceeding sections. These processes
were performed using HEC-Geo-HMS in Arc-Map. The steps involved delineation of
the Thutha river, its tributaries, watershed, and watershed properties (i.e. runoff curve
 34

number, time of concentration). With the DEM and GIS tools, the watershed
properties were extracted using automated procedures (USACE 2009).

The spacing of elevations in the DEM was not sufficient for the stream
centreline; therefore the DEM was modified to be consistent with the input vector
stream network. This DEM reconditioning increases the degree o f agreement between
stream networks delineated from the DEM and input vector stream network shape
file. If the reconditioning process was not applied, the path of the Thutha river would
be too crude

Meaning is that the reach lengths generated in Arc-Map would not represent
site conditions. This is mainly due to the precision o f elevation data along the stream
in the initial DEM. By manually detailing, or burning in the stream network, a
distinct stream profile was created and a new DEM was created. If a cell is
surrounded by higher elevations, the water is trapped and cannot flow out of the sink.

After the terrain pre-processing, the reconditioned DEM is the starting point
for delineating sub basins and river reaches. The first five o f the eight datasets are
grid layers that represent:

• Flow Direction, defines the direction of steepest descent for each terrain cell

• Flow Accumulation, determines the number o f cells upstream draining to a

given cell (upstream drainage area can be calculated by multiplying the flow
accumulation value (number o f cells) by the grid cell area (30m x 30m)

• Stream Definition, the generation o f a stream defined by the number o f Flow

Accumulation cells

• Stream Segmentation, divides the grid o f streams into segments, these are
sections o f the stream that connect two joining streams (junction), junction
and an outlet, or junction and the drainage divide.
 35

■ Catchment Delineation, creates a grid layer that delineates a sub basin for
each stream segment

After the terrain processing is completed, the first watershed project was
defined by identifying the downstream outlet area. The catchment polygons were
either merged or subdivided until the area o f the sub basins were comparable to the
drainage area size

Next, it was possible to extract physical characteristics o f the streams and

sub basins. These characteristics included computed length of the river segments,
upstream and downstream elevations o f the reach and the slope o f the river segments.
Basin slope uses the slope grid to determine the average slope for the sub basin.
Longest flow path creates a polyline that stores the upstream and downstream
elevations and slope between endpoints. The Basin Centroid, Centroid Elevation, and
Centroidal Flow Path are hydrologic elements easily performed in GIS. Some
techniques for estimating flood-peak discharges require this data. Although these
characteristics were generated, the rainfall-runoff methodology employed in this
research did not require this attribute information.
 36

3.4.2 H Y D RO LO GICA L M ODELING IN ARCHYDROTOOLS ANDHEC-

GEOHM S

Fig3.2. Flow chart of HEC-GEOHMS

i
 37

3.4.3 WATERSHED AND STREAM NETWORK DELINEATION USING

ARCHYDRO TOOLS
Open Arc Map. Create a new empty map. Right click on the menu bar to see
the context menu showing available tools and then check the Arc hydro tools to add
the toolbar to the map document. We can see the Arc Hydro toolbar added to
ArcMap.

The Spatial Analyst Extension has to be activated, by clicking Customize-

>Extensions..., and then check the box next to Spatial Analyst.

3.4.4 Terrain Pre-processing

3.4.2.1 Hydro DEM\ Flow Direction Grid, Flow Accumulation Grid, Stream
Definition Grid, Stream Segmentation Grid, Catchment Grid Delineation.

Arc Hydro Terrain Pre-processing should be performed in chronological

order. All of the pre-processing must be completed before watershed processing
functions using in the model. DEM reconditioning and filling sinks might not be
required depending on the quality o f the initial DEM. DEM reconditioning involves
modifying the elevation data to be more consistent with the input vector stream
network. This shows, an assumption that the stream network data is more reliable
than the DEM data, so we need to use the knowledge of the accuracy and reliability
of the data sources when deciding whether to do DEM reconditioning. By doing the
DEM reconditioning we can increase the degree of agreement between stream
networks delineated from the DEM and the input vector stream networks.

Click on the Add icon to add the raster data. In the dialog box, we can locate
the data; select the raster file containing the DEM for Thuthapuzha and click on the
Add button. The added file will then be listed in the Arc Map Table o f contents.
Similarly add stream.shp, and save the map document.
 38

a) Hydro DEM
DEM Reconditioning:
The function needs input as raw DEM and a linear feature class (like the river
network) that both have to be present in the map document. On the Arc Hydro
toolbar, select Terrain Pre-processing->DEM Manipulation->DEM Reconditioning.

Select the appropriate Raw DEM (Thuthapuzha_dem) and here AGREE

stream feature (stream). Set the Agree parameters as shown. Then we can reduce the
Sharp drop/raise parameter to 10 from its default value 1000. The output is a
reconditioned Agree DEM (default).
After that click ok and then it will process successfully after that message box
will come. Examine the folder where we are working; we will notice that a folder
named Layers has been created. This is where Arc-Hydro will store its grid results
b) Fill sinks
Fill sinks function fills the sinks in a grid. If the cells with higher elevation
surround a cell in model, the water is trapped in that cell and it cannot flow. This Fill
sink function modifies this elevation value to eliminate these problems.
On the Arc-Hydro Toolbar, select Terrain Pre-processing->Data Manipulation->Fill
Sinks
Confirm that the input for DEM is Agree DEM (or your original DEM if
reconditioning was not implemented). The output is the Hydro DEM layer (Named
default Fil). This Leave the other options unchanged.
Then Press ok. Upon completion o f the successful process, the Fil layer is added to
the map. This process takes a few minutes.
c) Flow Direction Grid
Flow direction computes the flow direction for a given grid. The values in the
cells of the flow direction grid indicate the direction o f the steepest descent from that
cell.

On the Arc-Hydro Toolbar, select Terrain Pre-processing->Flow Direction.

 39

Confirm that the input for Hydro DEM is Fil. The output is the Flow
Direction Grid, (default Fdr). And then Press ok. Upon completion o f the successful
process, the flow direction grid Fdr is added to the map.

d) Flow Accumulation Grid

This function computes the flow accumulation grid that contains the
accumulated number o f cells upstream o f a cell, for each cell in the input grid.

On the Arc-Hydro toolbar, select Terrain Pre-processing -> Flow Accumulation

Confirm that the input o f the Flow Direction Grid is Fdr. The output is the
Flow Accumulation Grid (default name Fac).Press OK. Upon successful completion
of this process, the flow accumulation grid Fac is added to the map. This process may
take several minutes for a large grid.
e) Stream Definition Grid
On the Arc-Hydro toolbar, select Terrain Pre-processing -> Stream Definition.
Confirm that the input for the Flow Accumulation Grid is Fac. The output is the
Stream Grid named Str (default)
Objective methods for the selection o f the stream delineation threshold to
derive the highest resolution network is consistent with geomorphologic river
network properties. It has been developed and implemented in the TauDEM software
(http://www.engineering.usu.edu/dtarb/taudem). For this exercise, we have chosen 25
km as the threshold area, and then click ok.
 40

Fig: 3.3 Stream grid

J) Stream Segmentation Grid

This function creates a grid o f stream segments that have a unique
identification. Here either a segment may be a head segment, or it may be defined as
a segment between two segment junctions. All the cells in a particular segment have
the same grid code that is specific to that segment.
On the Arc-Hydro toolbar. Select Terrain Pre-processing -> Stream
Segmentation.
Confirm that Fdr and Str are the inputs for the Flow Direction Grid and the
Stream Grid respectively. Otherwise if we are using our sinks for inclusion in the
stream network delineation, the sink watershed grid and sink link grid inputs are Null.
The output is the stream link grid (default name StrLnk).
And then Press ok. Upon successful completion of this process, the link grid
StrLnk is added to the map. At this point, we can notice how each link has a separate
value. Then Save the map document.
 41

g) Catchment grid delineation

This function creates a grid in which each cell carries a grid code (value)
indicating that to which catchment the cell belongs. The value corresponds to the
value carried by the stream segment that drains that area, defined in the stream
segment link grid.
On the ArcHydro toolbar, select Terrain Pre-processing -> Catchment Grid
Delineation.

Fig3.4 Catchment grid delineation

Confirm that the input to the Flow Direction Grid and Link Grid are Fdr and Lnk
respectively. The output is the Catchment Grid layer. The default name (Cat).Press
OK. Upon successful completion of this process, the Catchment grid Cat is added to
the map. If we want to recolour the grid with unique values to get a nice display (use
properties->symbology)
 42

3.4.5Catchment Polygon Processing Drainage line, Adjoint catchment, Drainage

point
The Three functions Drainage Line Processing Adjoint Catchment
Processing and Drainage Point Processing convert the raster data developed to the
vector format. The rasters created up to now have all been stored in a folder (i.e.
Layers). The vector data will be stored in a feature dataset also named Layers within
the geo-database associated with the map document. Except that otherwise specified
under AP Utilities-Set Target locations the geo-database inherits the name o f the map
document (Terrain.mdb in this case) and the folder and feature dataset inherit their
names from the active data frame which by default is named Layers.

On the Arc-Hydro toolbar, select Terrain Pre-processing -> Catchment

Polygon Processing. Catchment polygon processing converts catchment grid into
catchment polygon feature.

Confirm that the input to the Catchment Grid is Cat. The output is the
Catchment polygon feature class, having the default name (Catchment).

Press ok. Upon successful completion o f this process, the polygon feature
class Catchment is added to the map. Open the attribute table o f Catchment we can
notice that each catchment has a Hydro ID assigned that is the unique identifier of
each catchment within the Arc Hydro. Each catchment also has its Length and Area
attributes. These quantities are automatically computed when a feature class becomes
part of a geo-database.

a) Drainage Line
This function converts the input Stream Link grid into a Drainage Line feature
class. Each line in the drainage feature class carries the identifier o f the catchment in
which it resides. On the ArcHydro toolbar, select Terrain Pre-processing -> Drainage
Line Processing.

1
 43

Confirm that the input to Link Grid is Lnk and to Flow Direction Grid Fdr.
The output Drainage Line has the default name (Drainage Line).

Fig: 3.5 Drainage line map

Press OK. Upon successful completion o f this process, the linear feature class
Drainage Line is added to the map.
b) Adjoint Catchment
This function generates the aggregated upstream catchments from the
Catchment feature class. For each catchment that is not a head (starting) catchment, a
polygon representing the whole upstream area draining to its inlet point is constructed
and stored in a feature class that has an adjoint catchment tag. This feature class is
used to speed up the point delineation process.
 44

On the Arc Hydro toolbar, select Terrain Pre-processing -> Adjoint

Catchment Processing. Confirm that the inputs to Drainage Line and Catchment are
respectively
Drainage Line and Catchment. The output is adjoint Catchment, with a default
name (Adjoint Catchment).
Press ok. Upon the successful completion of this process, we will see a
message box similar to the one below that will give you a summary o f the number of
catchments that were aggregated to create the adjoint catchments.click ok, and a
polygon feature class named Adjoint Catchment is added to the map.
c) Drainage Point
This function allows generating the drainage points associated to the
catchments.
On the Arc Hydro toolbar, select Terrain Pre-processing -> Drainage Point
Processing. Confirm that the inputs are as below. The output is Drainage Point with
the default name (Drainage Point).
Press ok. Upon the successful completion o f this process, the point feature class
“Drainage point” is added to the map.
3.4.6 Watershed Processing
Arc Hydro toolbar also provides a widespread set o f tools for delineating
watersheds and sub watersheds. These tools rely on the datasets derived in terrain
processing. This part o f the exercise will expose you to some of the Watershed
Processing functionality in Arc Hydro tools.
3.5 TERRAIN PROCESSING AND HEC-GEOHMS MODEL DEVELOPMENT
Open Arc-Map. Create a new empty map. Right click on the menu bar to see
the context menu showing available tools and check the HEC-GEOHMS menu. And
now we can see the HEC-GEOHMS toolbar added to Arc-Map.
Click on the Add icon to add the raster data. In the dialog box, we can
navigate to the location o f the data; select the raster file Thuthapuzha_dem containing
 45

the DEM for Thuthapuzha and click on the Add button. The added file will then be
listed in the Arc Map Table of contents. Similarly add stream.shp, and save the map
document. Save the map document as Thuthapuzha.mxd.
3.5.1 Terrain Pre-processing
Creation o f Slope Grid
Terrain processing involves using the DEM to create a stream network and
catchments.
Towards the end o f the section we will have the following datasets:
Raster Data:
1. Raw DEM (File name: Thuthapuzha_dem)
2. HydroDEM (File name: Fil)
3. Flow Direction Grid (File name: Fdr)
4. Flow Accumulation Grid (File name: Fac)
5. Stream Grid (File name: Str)
6. Stream Link Grid (File name: StrLnk)
7. Catchment Grid (File name: Cat)
Vector Data:
1. Catchment Polygons (File name: Catchment)
2. Drainage Line Polygons (File name: Drainage Line)
3. Adjoint Catchment Polygons (File name: Adjoint Catchment)
In addition to this datasets, we can also get the slope grid by using the Arc
Hydro toolbar.
To create a slope grid using Arc Hydro tools, select Terrain pre-processing -> Slope.
Confirm that the input as Thuthapuzha dem, slope type is percent rise, and the output
will be a slope grid with the default name (Wsh Slope).
This concludes the terrain processing part. What we have produced is a
hydrologic skeleton that now we have used to delineate watersheds or sub-watersheds
for any given point on delineated stream network. The next part o f this tutorial
 46

involves delineation o f a watershed to create a HEC-HMS model using HEC-

GeoHMS. Save your map document.
3.6. HEC-HMS MODEL DEVELOPMENT USING HEC-GEOHMS
Before we continue, please make sure that we have the following datasets in
the map document from the previous part.
Raster Data:
1. Raw DEM (File name: Thuthapuzhadem)
2. HydroDEM (File name: Fil)
3. Flow Direction Grid (File name: Fdr)
4. Flow Accumulation Grid (File name: Fac)
5. Stream Grid (File name: Str)
6. Stream Link Grid (File name: Str Lnk)
7. Catchment Grid (File name: Cat)
8. Slope Grid (File name: Wsh SlopePct)
Vector Data:
1. Catchment Polygons (File name: Catchment)
2. Drainage Line Polygons (File name: Drainage Line)
3. Adjoint Catchment Polygons (File name: Adjoint Catchment)
Save the map document.
3.6.1 HEC-GEOHMS Project Setup
The HEC-GEOHMS project setup menu has tools for defining the outlet for
the watershed, and delineating the watershed for the HEC-HMS project. As multiples
of HMS basin models can be developed by using the same spatial data, these models
are managed by defining two feature classes: Project Pointand Project Area.

Management o f models through Project Point and Project Area. Let users see
areas for which HMS basin models are already created, and also allow users to
rebuild models with different stream network threshold. It is also convenient to delete
projects and associated HMS files through Project Point and Project Area option.
 Dataset Setup: Select HMS Project Setup -> Data Management on the HEC-
GeoHMS Main View toolbar. Confirm/define the corresponding map layers in the
Data Management window as show. Click OK.
To create a new HMS Project, click on Project Setup->Start New Project.
Confirm Project Area for Project Area and Project Point for Project Point, and click
OK.
Note: For some reason, if you get an error message about accuracy/resolution o f the
data, this has to do with tolerances for x, y, m, z coordinates in your spatial
coordinates which you need to fix in Arc-Catalog).
This will create Project Point and Project Area feature classes. In the next
window, provide the following inputs:
If we click on Extraction Method drop-down menu, we will see another
option “A new threshold” that will delineate streams based on this new threshold for
the new project. For now accept the default original stream definition option. We can
write some metadata if we need, and finally choose the outside Main view
Geodatabase for Project data location, and browse to your working directory where
Thoothapuzha.mxd is stored. Click ok on the message regarding successful creation
of the project. We will see that new feature classes Project Area and Project Point are
added to Arc Maps table o f contents. These feature classes are added to the same geo­
database Thoothapuzha.gdb.

Next Zoom-in to downstream section o f the Thuthapuzha to define the

watershed outlet as shown below:
 Fig 3.6 DEM for selecting watershed outlet

Select the Add project Points tool on the HEC-GEOHMS toolbar, and then
click on the downstream outlet area of the Thuthapuzha to define the outlet point as
shown below as red dot:

Then accept the default Point name and description (Outlet), and click ok.
This will add point for the watershed outlet in the Project Point feature class. Save the
map document.

Next, select HMS Project Setup->Generate Project. This will create a

interlock (by delineating watershed for the outlet in Project Point), and then display a
message box. If we want to create a project for this hatched area as shown below:
 49

Fig 3.7 Selecting watershed Outlet

Fig 3.8 Selecting area of sub basin

(Note: This part could be challenging sometimes. If we face any problem in

creating Project Area, just delineate a watershed (using the point delineation tool) in
Arc H\dro for the Project Point feature, and load this watershed poly gon into Project
Area feature class. Make sure the Hydro ID of Project Area is same as Project ID of
Project Point. Also we need to make sure the name and description match with each
other).
 50

Click yes on the message box. And next confirm the layer names for the new
project (leave default names for Sub basin. Project point. River and Basin header),
and then Click ok. This will create a new folder inside your working folder with the
name of the project

3.6.2 Basin Processing

The basin processing menu has features such as revising sub-basin

delineations, dividing basins, and merging streams. For merging basins, we will
follow the process that allow us to merges two or more adjacent basins into one.
Zoom-in to the area marked in the rectangle below:

Select the three adjacent basins (shown above) using the standard select tool.
Click on Basin processings Basin merge.

We will get a message asking to confirm the merging of selected basins (with
basins hatched in background), then click Yes. Similarly merge two more sub-basins
as shown below:

Fig 3.9 Sub basin delineation. Merging streams

 51

As a result of this merging, we now have 12 sub-basins and 15 river segments

in the project. Save the map document.

3.7. HMS INPUTS/PARAMETERS: RIVER AUTO NAME, BASIN AUTO NAME,

HMS UNITS

The hydrologic parameters menu in HEC-GeoHMS provides tools to estimate

and assign a number o f watershed and stream parameters for use in HMS. These
parameters include SCS curve number, Time o f concentration, channel routing
coefficients, etc.

We can specify the methods in the HMS. It will use for transform method
(rainfall to runoff) and routing (channel routing) using this function. There is a
choice, this can be modified and/or assigned inside o f HMS.

Select Hydrologic parameters->Select HMS processes. Confirm the input

feature classes for Sub basin and River, and then click ok, Choose SCS for Loss
method (attaining excess rainfall from total rainfall), SCS for Transform method (for
converting excess rainfall to direct runoff), none for Base flow type, and Muskingum
for Route method (channel routing). Then click ok.

You can open the attribute table o f sub basin feature class to see that the sub
basin methods are added to Loss method, Transform method, and Base method fields,
respectively. The Muskingum method is added to Route method field in the River
feature class. You can treat these methods as tentative which can be changed in HMS
model. Save the map document.

a) HMS Schematic

This tool creates a GIS representation o f the hydrologic system using a

schematic network with basin elements (links/nodes or junctions/edges) and their
connectivity. Select HMS->HMS Schematic. Confirm the inputs, and click ok.
 52

Fig 3.10 HEC-HMS Model

After the schematic is created, you can get a feel of how this model will look
like in HEC-HMS by toggling /switching between regular and HMS legend. Select
HMS

Fig: 3.11 Legends o f HEC-HMS

 53

>Toggle HMS Legend->HMS Legend, we can keep whatever legend we like.

Save the map document.

b) HMS Project

This function copies all the project specific files that we have created (.met,
.map and .met) to a specified information book, and creates a .hms file that will
contain information on other files for input to HMS. Select HMS->Create HMS
Project.

Provide locations for all files. Even if we did not create a gage file, there will
be a gage file created when the met model file is created. Give some name for the
Run, and leave the default information for time and time interval unchanged. This can
be changed in HEC-HMS based on the event we will simulate. Then Click ok.

If an.hms file already exist in that folder, we will get a message asking that we
have to respond to that message a project file report will be displayed as shown
below:

c) Opening the H M S M odel

This section briefly explains how to interface or open the project files created
by GeoHMS using HMS.

Open HEC-HMS, and select File->Open. Browse to Thuthapuzha.hms file,

and click Open. We will see that two folders: Basin models and Meteorologic models
will be added to the Watershed Explorer (window on top-left) in HEC-HMS. Expand
the Basin models folder and click on Thuthapuzha. This will display the Thuthapuzha
schematic. And then click on View->Background map, and then add the river and
basin shapefiles to see the watershed as shown below.
 54

If you expand the Thuthapuzha basin in watershed explorer, we will see the
list o f junctions, reaches and sub basins. You can click on any reach and see its
associated methods. For example, when you click on a Reach, we will see that
Muskingum routing method is associated with it. Similarly, if we click on a
Watershed, we will see SCS Curve Number (for abstractions) and SCS Unit
Hydrograph (for runoff calculations) are associated with it.

Again, if you click on SCS Curve Number, we will see corresponding

parameters in the Component window as shown below. All this information, which is
now independent o f GIS, which is extracted from attributes that we created in HEC-
GEOHMS.

3.7.1 (c) Hydrologic parameter estimation

After extraction o f the physical characteristics o f the streams and sub basins, a
number o f hydrologic parameters are estimated. These are the model input parameters
used in HEC-HMS. HEC-GEOHMS has the tools to estimate and assign a number of
watershed and stream parameters (i.e.CN, loss rates, reach routing, and time of
concentration). In order to simulate the process o f direct runoff o f excess
precipitation on the watershed, the specifications for this project included a loss and
transform method

The SCS CN loss method was selected to determine the loss o f total
precipitation for the watershed during rainfall events. This loss method equates the
sum of infiltration and precipitation left on the surface equal to the total incoming
precipitation.

A precipitation transform method (converting rainfall to runoff) is selected to

generate actual surface runoff. Several HMS options are available and this research
used the SCS unit hydrograph (UH) method. The basic concept o f the SCS UH is a
dimensionless, single-peaked UH that when watershed lag time is specified, an entire
 55

hydrograph can be generated from precipitation. Lag is the time separation between
the centroid o f the rainfall excess hyetograph and the peak o f the hydrograph. Lag is
empirically related to time o f concentration by: Lag = 0.6(Tc).

Channel characteristics for the reach routing is an estimated parameter in GIS

and performed in the same way the NRCS channel flow regimes; however, the
parameter needs to be entered manually into the HEC-HMS model. The Muskingum
method was selected for reach routing as a placeholder until the simulated parameter
in the channel reaches are established through HEC-HMS optimization trials
(d) HEC-HMS M odel File
Upon completion of the previous steps, HEC-GEOHMS verifies all the data
for consistency. The two data project files were checked for unique names us for river
reaches and sub basins in order to keep data separated and not risk any overwriting or
loss o f information. Additionally it confirms that river reaches and centroids are
contained within each sub basin and that there is connectivity between the stream
segments, sub basins and the outlet point. Once confirmed, the project schematic of
the hydrologic system was generated to show sub basin nodes and reach
links/junctions. Geographic coordinates are tabulated for each hydrologic feature to
maintain the geospatial information after export. Finally a background map to capture
the geographic information o f the sub basin boundaries and stream reaches is
prepared for export. A HEC-HMS basin file was generated containing all the
hydrologic elements, their connectivity, and related geographic information
3.8. WATERSHED DELINEATION
HEC-HMS uses parameters averaged in space and time to simulate the runoff
process. The size o f sub basins, routing reaches, or computation interval is selected
based on the basin physiographic, available rainfall data, available stream flow data,
and required accuracy. A watershed is subdivided into small and relatively
homogeneous sub basins according to drainage. It provides based on mapped or
digital topography. The size o f a sub basin should generally be in the range o f 1-10
 56

mi because of limitations ofU H theory, especially in urban areas. Routing reaches

are identified, and the overall order o f the runoff computation is defined (from
upstream to downstream) for input to HEC. Routing reaches should be long enough
so that a flood wave will not travel faster than the computation time step.

3.8.1 Steps in watershed modelling

With so many hydrologic models available to the hydrologist or civil

engineer, very little new model development is currently being supported. Rather, one
must select one o f the available simulation models based on characteristics o f the
system to be studied, the objectives to be met, and the available budget for data
collection and analysis. Once the model is selected, the steps involved in watershed
simulation analysis generally follow the sequence o f order.
Steps in watershed modelling
1. Select the model based on study objectives and watershed characteristics,
availability o f data, and project budget.
2. Obtain all necessary input data-rainfall data, digital topography, land use and
soils, infiltration, channel characteristics, stream flow data, design floods, and
reservoir data.
3. Evaluate and refine study objectives in terms o f simulations to be performed
under various watershed environments.
4. Choose methods for determining sub basin hydrographs and channel routing
5. Calibrate model using historical rainfall, stream flow and existing watershed
conditions. Verify model using other events under different conditions while
maintaining same calibration parameters.
6. Perform the model simulations using historical or design rainfall ■various
conditions of land use and various control scheme for reservoirs, channels or
diversion to the extent possible.
7. Perform the sensitivity analysis on input rainfall, routing parameters and
hydrograph parameters as necessary.
 57

The selection o f a model is a very difficult and important decision, since the
success o f the analysis hinges on accuracy o f the results. In general, unless digital
watershed data are extensive in space and time, the usual approach to watershed
analysis is to use a deterministic event model with lumped parameter concepts for
developing hydrographs and flood routing
Advances in computer hardware and software since the 1970s combined with
larger and more extensive hydrologic data-monitoring efforts allowed for the
development and application o f a number o f models in hydrology. These computer
models can be used for a variety o f purposes in simulating hydrologic response under
a number of assumptions within a watershed area. Such models incorporate various
equations to describe hydrologic transport processes and storages and to account for
water balances in space and time. Complex rainfall patterns and heterogeneous basins
can be simulated with relative ease if watershed and hydrologic data are sufficient,
and various design and control schemes can be tested with hydrologic mode
Table 3.2 Hydrological Models

Model Type Exam ple of Model

Lumped parameter H EC-HMS, Snyder or Clark UH

Distributed Vflo, MIKE URBAN, MIKE FLOOD

Event H EC-HMS, SWMM, SCS TR-20

Continuous S WMM, HSPF, Vflo

Physically based H EC-HMS, SWMM, HSPF, Vflo

Stochastic Synthetic stream flows

 58

Distributed parameter models attempt to describe physical processes and

mechanisms in space, as evidenced by certain classes o f hydrologic simulation
models. Kinematic wave methods have become very popular to compute both
overland flow and channel routing. This is due to heightened interest in distributed
hydrologic modelling with the advent o f GIS and digital elevation data for watersheds
that has become available since the mid 1990s.

3.8.2 Advantages of simulation models

• A major advantage o f simulation models is that the insight gained by gathering
and organizing data required as input to the mathematical algorithms that
comprise the overall model system.
• This exercise can often guide the collection of additional data or direct the
improvement o f mathematical formulations to better represent watershed
behaviour.
• Another advantage is that many alternative schemes for water supply systems, for
urban development, or flood control options can be quickly tested and compared
with simulation models.
3.8.3 Limitations of simulation models
• The major limitation of simulation models is the inability to properly calibrate
and verify applications in which input data are lacking.
• Current practice assumes that the simplest model that will satisfactorily explain
the system for the given input data should be used.
• Model accuracy is largely determined by available input data and observed input
and output time series at different locations in a watershed.
• Modem radar rainfall, hydrologic, and topographic datasets are now available for
many areas, and model accuracy has increased accordingly.
Despite their limitations, simulation models still provide the most logical and
scientifically advanced approach to understanding the hydrologic behaviour of

1
 59

complex watershed and water resources systems. (Maidment 1993, DeVries and
Hromadka 1993, Hoggan 1997, James and James 1998a, and McCuen 2005).

3.9 HEC-HMS MODEL

3.9.1 HEC-HMS 4.0
The hydrological modeling system (HEC-HMS) is designed to simulate the
precipitation-runoff processes. It is designed to be applicable in a wide range of
ecological areas for solving the widest possible range o f problems. This includes the
large river basin water supply and flood hydrology, and small urban or natural
watershed runoff. Hydrographs produced by the program are used directly or in
conjunction with other software for thestudies of water availability, future
urbanization impact, reservoir spillway design, urban drainage flow forecasting, flood
damage reduction, flood plain regulation system approach. The basin model consists
of mainly two models such as basin model and meteorological model.
3.9.2 Program setup and application
• Create a new project.
• Enter Time-series, paired data, and grid data.
• Create a basin model.
• Create a meteorological model.
• Create control specifications.
• Create and compute a simulation run.
• View results.
• Create other alternatives, compute, and compare results.
• Save the project and exit.
3.9.3 Basin model in HEC-HMS
The physical representation of a watershed is related with a basin model.
Hydrologic elements are connected in a dendrite network to simulate the runoff
process. Available elements are sub-basin, reach, junction, reservoir, diversion,
source and sink. Computation proceeds from upstream elements in a downstream
 60

direction. A classification o f different methods is available to simulate infiltration

losses. Options for event modelling include Initial constant, SCS curve number, and
Gridded SCS curve number, Exponential, Green-Amp, and Smith Parlange.
The one-layer deficit constant method can be used for simple continuous
modelling. The five-layer soil moisture accounting method can be used for simple
continuous modelling o f complex infiltration and evapotranspiration environments.
Gridded methods are available for both the deficit constant and soil moisture
accounting methods. There are seven methods are included for transforming excess
precipitation into surface runoff. Unit hydrograph method includes the Clark, Snyder
and SCS techniques. User-specified unit hydrograph or S-graph ordinates can also be
used. The modified Clark method, Mod-Clark is a linear quasi-distributed unit
hydrograph method that can be used with gridded meteorological data. The
performance o f kinematic wave method with multiple planes and channels is also
included.
Sub basin soil types can be divided into Pervious Surface and Directly-
connected Impervious Surface in HEC-HMS. With Impervious Surface, all the
precipitation transforms to the runoff and is expressed in the percentage o f basin area.
In this research, we used the SCS CN method, which was developed by U.S. Soil
Conservation Service to estimate direct runoff. The SCS curve number method is a
simple, widely used and efficient method for determining the nearly equal amount of
runoff from a rainfall, even in a particular area. Although the methods designed for a
single storm event, it can be scaled to find average yearly runoff values.
3.9.4 Sub basin
It represents the physical areas within the basin and produce a discharge
hydrograph at the outlet o f their respective areas. The hydrograph produced is
calculated from precipitation data minus the losses. The resulting precipitation excess
is transformed using a UH methodology to compute runoff at the outlet, which is then
added to base flow. Each component can be calculated using several methods. The
 61

area, loss rate, runoff, and base flow inputs for each sub basin are entered into the
basin model for each one using the Watershed Explorer.
3.9.5 Loss rates
It can be simulated by one o f several methods. For event modelling,
techniques include initial and constant, SCS curve number, gridded SCS curve
number and Green and Ampt methods.
The one-layer deficit and constant model can be used for simple continuous
modelling. For modelling of complicated infiltration and evapotranspiration
environments, the five-layer soil moisture accounting model can be used. The method
is changed in a simple menu-driven window, and the input data are entered on the
“Loss” tab in the editor.
Table 3.3 HEC-HMS loss rate methods

Method Description

Initial and
Initial loss volume is satisfied and then constant loss rate begins.
constant

Loss function is related to antecedent soil moisture condition and

HEC exponential
is a continuous function o f soil wetness.

SCS curve Initial loss is satisfied before calculating cumulative runoff as a

number function of cumulative rainfall using SCS methods

Infiltration rate is computed as an exponential function of

Holtan method
available soil moisture storage from Holtan’s equation.

Infiltration rate is computed from the Green and Ampt equation

Green and Ampt
as a function o f soil moisture and hydraulic conductivity

First an initial deficit storage is filled, and then infiltration rates

Deficit/Constant
can be specified on a monthly basis.
 62

SMA mo'sture account*n8 assigns a value o f initial storage to all

_________________ layers of the ground using a gridded method.__________________

3.9.6 Transform methods

Transform method which convert rainfall excess into surface runoff, it can
also be simulated using a variety o f tools. HEC-HMS includes the popular Clark TC
& R and SCS UH techniques, as well as the Snyder Method or user specified UH.
Spatially distributed runoff can be computed with the quasi-distributed linear
transform of cell-based precipitation and infiltration.

The Modified Clark method (Mod Clark) is a linear quasi-distributed UH

method that can be used with gridded precipitation data. If the Mod-Clark transform
with gridded rainfall is used, a file that contains characteristics o f sub basin grid cells
is required. HEC-HMS can handle grid cell depiction of the watershed for distributed
runoff computations. The kinematic wave method with multiple planes and channels
is also included.

Table 3.4 Surface runoff methods in HEC-HMS

Unit hydrograph input directly

Clark hydrograph method ( TC&R method Snyder unit hydrograph method )
SCS method (CN method + SCS UH)
Kinematic wave for overland hydrograph
Mod-Clark
User-specified S-graph

3.9.7 Base flow

It takes into account normal flow through a channel or the effects o f ground
water. HEC-HMS offers two methods for base flow calculation: recession and
constant monthly. The recession method is an exponential decay function o f a defined
starting base flow. For the constant monthly method, the user simply enters a constant
 63

base flow value for each month. No base flow is also an option, and in simple
hydrologic models over short time periods or highly urbanized basins with channels,
base flow can usually be neglected.
3.9.8 Flood routing
In HEC-HMS offers a number o f options for the reaches and routing o f flood
hydrographs. The Muskingum method can be used for general routing; routing with
no attenuation can be modelled with the lag method. The most popular and accurate
is the Modified Pul’s method, which can be used to model a reach with a user-
specified storage-outflow relationship.
Table 3.5 Flood Routing in HEC-HMS

M ethod Description

Muskingum Storage coefficient (x) plus travel time ( K) through each reach

Table o f storage vs. outflow for each reach based on HEC-RAS

Modified Pul’s water surface profile information—various storages are plotted
vs. peak flows in the reach

Channel shape, length, slope, and n ; outflow from each reach

Kinematic wave based on depth o f flow in continuity equation and Manning’s
equation

Muskingum— Same equations and technique as the Muskingum- Cunge but

Cunge 8-point channel is described with 8 station-elevation values

Direct lag o f flow through channel with translation accounted for

Lag
but not attenuation

3.9.9 Reservoirs
It stores the inflow from upstream elements and produces an outflow
hydrograph based on a monotonically increasing storage-outflow relationship. A
reservoir can be entered with one o f three possible types o f relationships: storage vs.
 64

outflow, elevation vs. storage vs. outflow, or elevation vs. area vs. outflow. The
outflow structure must be well understood in order to develop an accurate storage-
discharge relation. The inflow entering the reservoir must be contained with the
minimum and maximum values o f the data entered. The user must also select an
initial condition o f storage, elevation, outflow, or select inflow equal to outflow. The
model assumes that elements have a level pool, such as ponds, lakes, or reservoirs.
The effect o f adding a detention pond to a basin can be modelled by using a reservoir.
The input window for a reservoir is used to relate storage to outflow. A reservoir icon
is used to represent storage at any point in the watershed and is then connected
downstream to a junction.
3.9.10. Sources
These are elements that represent a discharge into the basin as an observed
hydrograph or a hydrograph generated by a previous simulation.
They often are used to represent inflow from reservoirs; un modelled
headwater regions, or a watershed outside the region. This may be entered as gage
data or a constant discharge.
3.9.11 Sinks
These are elements that have an inflow and no outflow. The only inputs are
the name and description o f the sink. It may represent the lowest point o f the drainage
area or the outlet.
3.9.12 Diversions
Diversions for hydrologic models use a simple table relating from the inflow
to diverted flow and finally to the routed flow. These relationships can be determined
using geometric calculations and hydraulic models. Diversions will have two
“downstream” connections, one being the routed path and the other the diverted path.
The user specifies the diverted flow through the use o f a table (inflow-diversion), and
whatever flow is not diverted will travel the main path. If the “connected” option is
 65

selected, then the diverted flow will return to the watershed at the downstream
location.
3.9.13. Meteorological model
For the meteorological model in HEC-HMS, we considers only about the
precipitation influence in HEC-HMS. The duration o f the storm events may be too
short, and the simulation can ignore the evapotranspiration influence in this model. In
HEC-HMS there are7 methods are available in the precipitation those are, Frequency
storm, gage weights, Gridded precipitation, SCS storm, inverse distance, Specified
hyetograph, and standard project storm. In this study we have used SCS storm is the
precipitation method.
3.9.14 Control specifications
Control specification in HEC-HMS set up the duration time and the time
interval for simulation. It includes the starting date and time, ending date and time
and Computation time Step.
3.9.15 Time-series data
In Time-series data, we needed to set up the discharge gage and precipitation
gage in the simulation.
The observed runoff data was inputted in the discharge Gage to compare with
the simulated runoff data after the rainfall-runoff model was setup. Thiessen’s
Polygon method was chosen to divide the represented weight for each rainfall station,
we obtained the average rainfall in each sub basin and inputted it in the precipitation
data.
3.10 PENMAN’S EQUATION (EVAPOTRASPIRATION)

Penman’s equation is based on a combination o f the energy-balance and

mass-transfer approach. Penman’s equation, incorporating some o f the modifications
recommended by other investigators is
 66

AHn + E j
PET= — 2-----—
E+y

Where,
PET = Daily potential evapotranspiration in mm per day

A=Slope o f the saturation vapour pressure vs. temperature curve at the mean air
temperature, in mm o f mercury per °C

Hn = Net radiation in mm o f evaporable water per day

Ea=parameter including wind velocity and saturation deficit.

y =psychrometric constant- 0.49 mm o f mercury/ °c

The net radiation is estimated by the following equation

H „=^(l-r)(<i+6^)-<T r;(0.56-0.092^)(0.10+0.9t)^)

Where,

Ha=incident solar radiation outside the atmosphere on a horizontal surface, expressed

in mm of evaporable water per day.

a = a constant depending upon the latitude <f>and is given by a=0.29 costj)

b= a constant with an average value of 0.52

n = actual duration o f bright sunshine in hours

N = maximum possible hours o f bright sunshine

r = reflected coefficient (albedo).

 67

Table 3.6 Usual ranges o f values r are given below

Surface Range o f r values

Close Ground Crops 0.15-0.25

Bare Lands 0.05-0.45

Water Surface 0.05

Snow 0.45-0.95

3.10.1 Capability of Arc-View 10.3

Arc-View is full-featured geographic information system (GIS) software for

visualizing, managing, creating and analysing geographic data. Application o f GIS in
the geosciences has grown explosively over the past few years as scientists and land-
use specialists have been able to prepare thematic maps and determine spatial
relations among multiple datasets. Although GIS applications hold great potential for
most organizations, their availability and application have largely been limited to
institutions that can afford and assemble the necessary hardware, software, datasets
and hard-to-find technical expertise.

People can understand the geographic context of data and see relationships
and identify pattern in new ways by using Arc-View. A lot o f GIS data can be
integrated by Arc-View for visualization and analysis, and Arc-View can also help to
author maps using simple wizards and an extensive suite o f map elements

3.10.2. Mapping Quality by Arc-View Software

The rich set o f map interaction tools in Arc View allows navigating and
querying a map as well as creating additional information such as hyperlinks that
 68

integrate a map with other information. Arc View allows users to easily build quality
maps in many different styles, from base-maps to thematic classifications, and also
allows using and composing thousands o f symbols and provides a robust labelling
environment, including on-the-fly automatic labelling based on attribute values.

3.10.3 Advanced Spatial Analysis

Arc View includes a set of tools and procedures that can analyse spatial data
and derive answers from data o f a location-dependent nature, utilizing an extensive
set of analysis tool in a comprehensive framework that facilitates the creation, use,
documentation, and sharing o f geo-processing models. We have adjusted the attribute
tables of land use and soil data in Arc-GIS.

3.11HYDROLOGIC SIMULATION

The time span o f a simulation is managed by control specifications, which

include a starting date and time and an ending date and time, and a time interval. A
simulation run is created by combining a basin model, meteorological model and
control specifications. Run options include a precipitation or flow ratio capability to
save all basin state information at a point in time and ability to begin a simulation run
from previously saved state information. Simulation results can be viewed from the
basin map. Global and element summary tables include information on peak flow and
total volume. A time-series tables and graphs are available for the elements. The
results from multiple elements and multiple simulation runs can also be viewed.

3.11.1 Running HEC-HMS and viewing results

The user may specify different data sets for each component within a project
and then run the hydrologic simulation using different combinations o f models.
 69

For example, one can run a 10-year or a 100-year frequency storm model
using the same basin model and control specifications to compare the resulting flows.
Or, one can model the effects of adding diversions and reservoirs to a basin by saving
it under a different name, altering the new basin model, and running the simulation
under the same precipitation and control specifications.

With several basin models saved in the same project, running simulations with
various models is a simple task. In order to create the scenario for a particular run, go
to “Simulation run” under Compute in the main file menu. Select one basin model,
one meteorological model, and one control specification. Then run the scenario by
selecting “Select run,” and subsequently “Compute run,” both o f which are under
Compute in the main file menu. If changes are made to one o f the models after a run
with the same configuration has been calculated, select the correct run to recalculate.
Note that the “Compute” tab beneath the Watershed Explorer allows the user to easily
select different runs.

Results can be viewed by going to the “Results” tab beneath the Watershed
Explorer, clicking on any object (e.g., sub basin or junction) in the basin model, and
selecting from the menu. The program gives the resulting times and flows for each
basin element and its immediate upstream elements in the form o f a graph, summary
table, or time-series table.

3.11.2 Other features

HEC-HMS includes advanced features such as parameter estimation with

optimization, soil moisture accounting, GIS and grid-cell hydrology, snowmelt
simulation, and improved hydraulics. HEC-GEOHMS is a companion program that
allows for the creation o f HEC-HMS projects from GIS sources, including digital
elevation models, land use data, and other electronic sources.
 70

The parameter estimation and optimization function is used to compare

resulting hydrographs to observed hydrographs, so at least one element in the basin
must have observed data.

The program automatically estimates the parameters in order to find the best
fit o f the generated hydrograph to the observed one for one element. It allows the user
to set initial values for the parameters of each element along with maximum and
minimum values so that the parameter values estimated by the program must fall in a
reasonable range.

3.12 MODEL CALIBRATION

The goal of calibration is to identify parameter value adjustments so that the

simulated results match the observed hydrographs. The mathematical search is a trial
and error analysis (optimization trials) that iterates until the simulated measurements:
runoff volume, peak flow, time of peak, and time of centre o f mass, is within an
acceptable error range (less than 5%) o f the observed hydrograph. By comparing
measured discharge from a significant event to the model, the reliability o f the model
is improved.

Fig 3.12 Schematic calibration procedure

3.13 MODEL OPTIMIZATION

It describes how observed stream flow can be used to optimize the model
performance by automatically estimating the parameters. The optimization process
begins with initial parameter estimates and adjusts them so that simulated results
match with the observed stream flow as closely as possible. Two different search
algorithms are provided that move from the initial parameter estimates to the final
best parameter results. A variety o f objective functions are provided to measure the
goodness o f fit between the simulated and observed stream flow in different ways,
while parameter estimation using the optimization does not produce perfect results, it
can be a valuable aid for calibrating the model.

The quantitative measure o f the goodness o f fit between the computed result
from the model and observed flow is called objective function. An objective function
measures the degree o f variation between the computed and observed hydrograph.

Optimization trials are the one o f the component that can compute the results.
Each trial is composed o f basin model, meteorological model and time of
information. The trial includes the selections for the objective function, search
method and parameters to be adjusted in order find an optimal model.

The iterative parameter estimation procedure is used by the program is often

called optimization. Initial values for all the parameters are required at the start o f the
optimization window. A hydrograph is computed at a target element by computing all
the parameters.

3.14 MODEL PERFORMANCE AND EVALUATION METHODS

The Criteria for using and evaluating the performance o f the models are the
overall agreement between predicted and measured runoff discharges, and the
model’s ability to predict time and magnitude o f hydrograph peaks and runoff
volume. The following statistical measures were used to quantify the performance
and accuracy of both models during each simulation periods, and combined over all
periods:

1. Percent error in peak flow (PEPF): The PEPF considers only the magnitude of
computed peak flow and does not account for total volume or timing o f the peak
flow.

PEPF=100 0„ ( p ea k ) - Q s ( p ea k )

( 1)

Where Q 0, (Qs) is the observed (simulated) flow.

2. Percent error in volume (PEV): The PEV function considers only the computed
volume and does not account for the magnitude or timing of the peak flow

PEV=100 V o -V * ( 2)

3. Coefficient of correlation (R), the lag-0 cross correlation coefficient was calculated
as:

Y p - o x s - S )
r = — =— (3)
I Y ( o , - d ) 2x ( s , - s ) 2

Where, O, (5t) is the observed (simulated) flow at time t, and O , S is the average

observed (simulated) flow during the calibration.

4. The relative root mean squared error, RRMSE, were calculated as:

RRM SE=100x J — f j ^ ...(4 )

 73

Where, N is the number of stream flow ordinates and the meaning of the remainin
symbols is the same as in Equation (3).

Screen shots Related to this Studs

DEM reconditioning

D IM R e c o n d i t i o n i n g

R a w DEM fhuthapuzha_dem

A G R E E S tream lhuthapuiha_vtr v !j
A GREE DEM A gteeD E M

S tre a m b i i t e (n u m b er of c e S * |

S m o o th d ro p /ra rte (OEM Z -urat) 10

S h a rp d ro p /ra rte (DEM Z-unit I 10

Oh HHp C a n ce l

Fill Sinks

r --------------1
1 ' F ill S i n k s X ]

DEM -------------------------- I -

D e r a n g e d P o ly g o n N ul

H y d ro D E M
|5
1 I U s e IsSnF. F ield

Fill M e th o d
O « T h re s h o ld ID E M Z U r» r| 10

<*> F * A I

I OK 11 H e lp ] [ C aned 1
L ............................ 1

Fig 3,13. Snapshots o fd e m reconditing. (ill sinks operation

 74

2. Creation Of Flow Direction. Flow Accumulation Map

J ' Flow Direction

Hycfco DEM pj
Flow Diection Gnd
Outer W al Polygon
Flow AccumiJabon Gnd Fac
Flow Direction Grid Fdr

OK Help Cancel
OK Help Cancel

Fig 3.14 Snapshot of creation of How direction, flow accumulation in HEC-

GEOHM S

3. Creation of Stream, Stream segmentation map

t' S tre a m S e g m e n ta tio n m

' Stream Definition

Flow Accurmiabon Gnd Fa:

Flow D fe d o n Gnd Fd V

Stream Gnd Sir Enter stream threshold to initiate a stream

v
S r k W atershed Gnd Number oi cels.
N il V

- Area Isquare km): 25

S r k L rk Gnd NJ V

Stream tank Gnd StrLrk Stream Gnd Sb

OK Help Cancel OK Heb | | Cancel |

Fig 3.15 Snapshot o f creation o f stream , stream segm entation m ap in H EC -G EO H M S

 75

4. Creation of Catchment Grid Delineation, Catchment Polygon Processing

i Catchment Grid Delineation Catchment Polygon Proces... X

Flow D rechon Grid

Catchment Gnd y
L**6 " 1 S tiff
Catchment C^chment
Catchment Gnd Cat

OK Help Cancel
OIC Help Cancel

Fig 3.16. Snapshot of creation of catchment grid delineation, catchment polygon

processing in HEC-GFOHMS

5.Creation of Adjoint catchment Processing

Adjoint Catchment Processing * )■ Adjoint Catchment Processing

n
AccunJate shapes completed successhiy.
Sart: 6/18/2012 11:59:57 AM End: 6/18/201211:59:S8 AM
Diartage Line Dianagebne y
Merfcer o f features processed: 44
Catchment Catchment v Nunber of loops: 2
Traced Layer. Catchment
Soiree LayerCatchment
A<SonlCatehment AdpntCatchffned Target Layer :AdjointCatchinent
Nunber of souce features not agyegated: 0
Total merge processng Tfne: t Os 0-000 hrs.
Oxidate river order tme: 0.1s
TotalProcess**) tme: llsO.OGOhrs,
OK Help Caned

I <* I

Fig 3.17. Snapshot of creation of Adjoint catchment processing in HEC-GEOHMS

 76

6. Creation of Drainage Point. Drainage Line Map

i Drainage Point Processing !!' Drainage Line Processing

Flow AcamJahorv Gnd

Fac Stream bnk. Gnd StiLnk
Catchment Grid
Cat
Flow Drecton Gnd
Fdr
Catchment Catchment
DrarageLine Dramagebne
DramagePof* DiamagePortf

OH Help Cancel
OK Hdp Caned

Fig 3.l8.Snapshot of creation of drainage point, drainage line processing in HEC-

GEOHMS

7. Creation of Slope Map

I Slope n
Raw DEM cedai dem

Slope Type PERCENT RISE

Slope WshStopel

OK Hdp D ax d )

Fig 3.19. Snapshot o f creation o f slope m ap

 77

1. HEC-GEOHMS Project Setup

D d td M d n o g c m c n t .

R a w D EM Thuth^u/fa tknt
H y tfco D E M N
F lo w O r e c l e n G fO

FV>w A c c i m i o f c o n G ntS ta c
G nd |h
S u o a n t U n it Gr«J :b rt
C dtcN oert C a f t >m e * 4
C^cNmrt
PtO K Ct P o r t Ni*
P iq » c * A iM HJ

L o* I ««*■ |

D e fin e a N e w P i o j a c l

- □
hhs M o ld i s C s d s Cm * 1

Cmnoc*ORMrthod Ov>* Mnkm W

Thn [»om < *t wrong MC CCm q NM S lor ArcGlS TO Th* i

noHnnor* W*od**S k«2l

D«i«LbmBbw
O MMM* if s GeoUMi
© 0 J -.* HwWiee bHdafddte i I
T m<p** D VGc TONgoohma

[ OK I I Ha { I Cwd |

Fig 3.20.Snapshot o f H EC-G EO H M S project setup

 78

9. B asin processing

HMS process
r
\ Se le ct H M S P ro c e sse s B E ®
I tr^ut Suttuw i 1
'Sufcbasnll
Input Rrm
jRiver 1 d @
Stirfcasn * Lam MeSf*xl
sc s V
xi)t.«rr transform M«hcd
SCS v
Siifejur. - SaieNtiw Metlmd
hone
V1
firwjr Route McThcd
Husfengun i l

i 1
1 1 Oe; | Cervcet ) |£nvtonpTents . [ 5ho*H®fc)>> ] |
I _________ __ ______ __ ______ _ . .................... 1

HMS Schematic ' 1

n

Protect Pont p!OfeciParir1

Cenlrod Centrodl52
River Rfverl
Sifcbssm SJDb«int
HMSLnH52
H M SH ode 152

C jl Help Cancel

Fig 3.21. Shows that snapshot o f HMS process

 79

10. Create HMS Project

■CrMteHMSProject

DW OpM CtfrGM
□
Satcif.su% 0 ’A;'L'«*ecCeJa£«l LsnJm ta r
a
SaeciMs-r* DV»:'.'oe8hK •>»&«* ^tarStc re
a
[ Vta'^gttte'CeiiiCeti'httlw >a u
W5Rr*aK
ifuawjijiSi*'n [flAUZHDttftOD
MfME«4D*T« <n/tD2HmnoQ
Tee k in d |fcdK|

Create HEC HMS Pro je ct

• -H “tS project He copy report

V CedarO eel..htns successhJhr copied
CedarCree* .met successr j ly copied
C e d a rO e e i < }& }* succew fcAy copied
C e d arO e e K .b e sn sjc c c jsh jtt copted
Cedar Creel. .nx> successfufy copied
R unt _cant10J ccrtroi w c c w jt iiv copied
River 1 .cfcf successful^ copted
River 1 .JSr |Suitcessfuly coped
R iv e rl.sb n 5sjccc5shjt> copied
River l.s b * successfully coped
River 1 iT p successful*- cop e d
River i ,d w s u c c e s s fu l copted
5rJjbasn I dbf su ccessh A y coped
Sub bosn 1 prj successfully copted
S jb b a s n l sbn su cce ssfu l? copted
5ubbovni sb * S u tc e ssfd y copted
SJD b asn 1 shp successfully copted
S jb b a s n t shv succetshJK. copted

OK

Fig 3.22.Snapshot o f creating HMS project

 80

II. Opening the HMS Model

W j'lU
L»- iS*, W200
a- A * w iw
[* A * W 1 8 0
is- A * w i t o
['f~J Mo C4t»py
S B ) >*> Su rfa ce
Fra 1~
SCS _____
C u rv e N um ber
FFV] S C S Unit H ydro^raph
['{~J M o f U B s H w i
». W I6 0
± * Outlet 1
^3s*
*TP JB2
JS S
358
357
372
381
It L***R3D
Muyljrnjurn
E p J N o C h o rm e l t o s s
I* R70

C om pon e n ts ConpU e

Su b b a sin L o ss Transform O p tion s

Bt»sm Nome: L edarl reek

Element Name: W170
I r w tia l A b s t r a c t i o n ( I N )
* C u rv e N u m b e r:

s(«W>)

Fig 3.23. Shows Snapshots o f opening the FIMS project

RESULTS AND DISCUSSION
 81

CHAPTER IV

RESULTS AND DISCUSSIONS

Basin model is the most important input to run the model and simulate the
rainfall runoff based over the entire watershed to create the basin model in the
Thuthapuzha watershed. Here the work was done on the analysis o f watershed by
using observed rainfall and river flow data for estimating the stream flow discharge in
Thuthapuzha watershed. Calibration and validation of watershed was done by using
hydrological modelling system HEC-HMS with that extension HEC-GEOHMS was
used.

4.1 CATCHMENT CHARACTERISTICS

The digital elevation model o f the Thuthapuzha watershed is presented in fig

4.1. The study area falls within three physiographic zones o f Kerala state namely the
high lands elevation(600-2500m)above mean sea level, the midlands elevation (300—
600m) above mean sea level and the low lands elevation (10-300m) above mean sea
level. The Silent valley reserve forestis located at the northeastern comer o f the Sub-
basin(i.e. The core o f the Nilgiri Biosphere reserve). The study area experiences a
humid tropical climate and wide variation in rainfall varies from 2800 mm to more
than 5000mm/year, the rainfall is higher towards the north eastern part. Like this the
other part of Kerala State region also experiences two distinct monsoons, namely the
south-west monsoon (June-September) and north-east monsoon (October-December).
The south west monsoon accounts for about 65% o f the annual rainfall, the north east
monsoon and the summer showers contribute the rest total annual rainfall in the Sub
Basin.

The hypsometric curve o f the watershed which shows the area elevation
relationship is presented in figure 4.3. The curve shows that about 70% of the basin
lies within an elevation band of 20 to 170m and the remaining 30% area is above
 82

170m and has high relief. The digital soil map showing different soil series are shown
in the table 4.2. The land use map developed through supervised classification ot the
satellite imagery shows that there are 29 different land use classes. Major land use
types developed in the watershed are rice, mixed crop, forest, range land and urban

settlement.

-h

Fig. 4.1 Dem of Thuthapuzha

e le v a tio n
N o rm alised

Percentage area above the elevation

Fig4.2. Hypsometric curve of Thuthapuzha sub basin

I

83

Table 4.1. The name and the area o f each Gram Panchayat within Thuthapuzha sub
basin

Name of the
SI. No Name of the block A reafkm 2)
Panchayat
1 Kulukkallur Pattambi 22.8
2 Koppam Pattambi 20.9
3 Nellaya Pattambi 27.4
4 Vilayur Pattambi 17.7
5 Thiruvengapura Pattambi 20.4
6 Cherupulassery Sreekrishnapuram 27.6
7 Kadambazipuram Sreekrishnapuram 39.7
8 Karimpuzha Sreekrishnapuram 47.4
9 Pookkattukadavu Sreekrishnapuram 22
10 Sreekrishnapuram Sreekrishnapuram 29.5
11 Thrikkadivi Sreekrishnapuram 26.3
12 Vellinezi Sreekrishnapuram 26.8
13. Alanellur Manarkkad 58.3
14 Karakurissi Manarkkad 27
15 Karimba Manarkkad 69.2
16 Kottappadam Manarkkad 29.8
17 Kumaraputhur Manarkkad 37.2
18 Manarkkad Manarkkad 63.4
19 Kanhirapuzha Manarkkad 58.4
20 Thachannattukara Manarkkad 35
21 Tachanpara Manarkkad 54
22 Thazhekode Perinthalmanna 45
23 Aliparamba Perinthalmanna 20.2
24 Elamkulam Perinthalmanna 21.3
25 Pulamanthole Perinthalmanna 33.7
26 Moorkannad Perinthalmanna 11.7
27 Irimbilam Kuttipuram 24

4.2 SOIL MAPPING UNITS

Based on depth, texture, slope and drainage characteristics the soils o f the study area
falls into 12 soil mapping units namely K08, K09, K10, K13, K20, K22, K25, K26,
and K28 & K30 (KSLUB,1995). The description o f the soil mapping units is
 84

presented in Table.4.2 The map showing these units (Fig.2) depicts the spatial

distribution of each soil type in the study area.

N

Legend
Soil
TYPE
KQ8
^^K 09

^ H K1°
^H K 13
K1G

| ^ | K22
|m K25
^ ■ 1 K26

K28
K30

10 20 30 40
I Kilometers

Fig. 4.3 Soil map o f Thuthapuzha sub basin

Table.4.2. Description of soil mapping units o f Thuthapuzha sub basin

Soil
mapping Texture Description
unit
Very deep, moderately well drained, clayey soils with
moderately shallow water table in nearly level narrow valleys,
K08 Clayey with slight erosion associated with imperfectly drained, very
deep, clayey soils with moderately shallow water table on nearly
level lands.
Gravelly Very deep, well drained, gravelly clayey soils with moderate
K09 surface gravelliness on moderately steeply sloping laterite
clay
 85

mounds, with moderate erosion; associated with deep, well

drained, gravelly clay soils on gentle slopes.
Very deep, well drained, gravelly clay soils on gently sloping
midland laterites with valleys of northern Kerala, with moderate
Gravelly
K10 erosion, associated with deep, well drained, gravelly clay soils
clay with moderate surface gravelliness and ironstone layer at 100-
150 cm on nearly level lands, slightly eroded.
Deep, well drained, gravelly clay soils with moderate surface
Gravelly gravelliest on ironstone layer at 100-150 cm on gently sloping
K13
clay midland laterites, with moderate erosion; associated with laterite
outcrops.
Very deep imperfectly drained loamy soils with moderately
shallow water table in nearly level board valleya of palghat gap,
K16 Fine loamy with slight erosion associated with moderately deep, moderately
well drained, gravelly loamy soils with coherent material at 75-
lOOcmon gentle slopes, moderately eroded
Deep, somewhat excessively drained, gravelly clay soils with
Gravelly moderate surface gravelliness on steeply sloping high hills with
K20
clay thick vegetation, with moderate erosion; associated with very
deep, well drained, clayey soils on gentle slopes.
Very deep, well drained, clayey soils on gently sloping low hills
with isolated hillocks, with moderate erosion; associated with
K22 Clayey
deep, well drained, gravelly clayey soils on moderately steep
slopes
Very deep, well drained, gravelly clayey soils with strong
Gravelly
K25 surface gravelliness on moderately sloping medium hills with
clay moderate erosion, thin vegetation, associated with rock outcrops.
Very deep well drained clayey soils with loamy soils on gently
K26 Clayey sloping medium hills with thick vegetation, with moderate
erosion, associated with rock outcrops.
Moderately deep, well drained, moderately calcareous, gravelly
loam soils with moderate surface gravelliness on gently sloping
Gravelly foot hills and valleys, with moderate erosion; associated with
K28
loam moderately shallow, somewhat excessively drained, gravelly
clay soils with strong surface gravelliness and coherent materials
at 50 to 75 cm on moderate slopes, severely eroded.
Very deep, well drained, loamy soils on gently sloping uplands
K29 Loam with valleys, associated with very deep, with moderate erosion,
well drained, loamy soils on gentle slopes
Very deep, well drained, clayey soils on moderately sloping high
hills with thick vegetation, with moderate erosion; associated
K30 Clayey
with deep, well drained, gravelly clayey soils with moderate
surface gravelliness on moderately steep slopes.

(Source: Land Resources Map o f Kerala State, Kerala State Land Use Board, 1995)
 86

4.3 PREPARATION OF SOIL MAP

The soil map o f the watershed (indicating soil type and Series Name) was
traced and the boundaries o f different hydrologic soil groups viz., A, B, C, and D
were delineated. The soils o f group A were o f shallow, well drained, the soils of
group B were o f moderate infiltration rate, moderately well drained. The soils of
group C of the moderately fine to moderately coarse textures, water transmission of
moderate rate and the soils o f group D were o f very deep ,slow infiltration and high
runoff potential

4.4 LAND USE PATTERN

The land use pattern o f the study area varied according to the local
physiography. The high land areas are mainly occupied by forests. The forest area
contributes 30% of the land area of Thuthapuzha Sub Basin. The Silent Valley
Reserve Forest which forms the core o f Nilgiri Biosphere Reserve and is one of the
best representative evergreen forests in the country is located towards the northern
part of the study area. The region has an exceptional diversity ofbiotic communities,
forest ecosystem and species o f flora and fauna. A large percentage o f the species
recorded from this forest especially plants and lower animals, are endemic species
occurring only at a few locations in the Western Ghats. There is no recorded history
of human settlements within the Silent Valley Reserve Forests, but there are nine
tribal settlements along the fringes. Rubber plantation also occupies the high land
region and contributes about 10% o f the study area.
The low land and mid land areas are utilized for agricultural purpose and
settlement. Mixed crops o f coconuts, banana, tapioca, seasonal vegetables etc. are
cultivated in the midland region and accounts to approximately 65% o f the study
area. Only 5% of the study area is utilized for single crop cultivation.
 87

Fig. 4.4. I.and use map of Thuthapuzha sub basin

4.5 TERRAIN PREPROCESSING

Terrain processing involves using the DEM to create network ol stream and
catchments. The Processing menu (shown below) in HEC-GEOHMS is used tor

terrain processing.

Raster Data

1. Raw DEM
2. Hydro DEM (DEM after reconditioning and tilling sinks)
3. Flow Direction Grid
4. Flow Accumulation Grid
5. Stream Grid
6. Stream Link Grid
7. Catchment Grid
 88

Vector Data

1. Catchment Polygons
2. Drainage Line Polygons
3. Adjoint Catchment Polygons

4 -

Fig. 4.5. Fill sink map

Lag end

nz*

Fig. 4.6. Flow direction map

 89

Fig. 4.7. Flow accumulation map

Fig. 4.8. Drainage line map

 90

4.6 BASIN PR OCESSING

The basin processing menu lias features such as revising sub-basin

delineations, dividing basins, and merging streams. For merging basins, we will
follow the processthat allow us to merges two or more adjacent basins into one.
Zoom-in to the area marked in the rectangle below:

Select the three adjacent basins (shown above) using the standard select tool. Click on
BasinprOcessing-> Basin merge. After that we will get a message asking to confirm
the merging o f selected basins, then click Yes. Similarly merge two more sub-basins
as shown below. I hen the resulting pictuie is like this.

Fig. 4.9 C atchm ent extraction map

 Fig. 4.10 Adjoint catchments map

4.7 EXTRACTING BASIN CHARACTERISTICS

The basin characteristics menu in the HEC-Geo HMS Project vision provide
tools for extracting physical characteristics of streams and sub-hasins into attribute

tables.

a) River Length

The river length tool computes the length of river segments and stores them in
River length field. Select river length. Confirm the input River name, and click OK -
Characteristics. You can check the River length field in the input River) (or whatever
name we have mputtedlor river) feature class is populated. Save the map document.
 92

b) River Slope

Ri\ er slope tool computes the slope o f the river segments and stores them in
Slope field. Select- characteristics- Basin River Slope. Confirm inputs for Raw DEM
and River, and click OK. You can check the Sip field in the input River 1 (or whatever
name you have for your input river) feature class is populated. Fields ElevUP and
ElevDS are also populated during this process. Slope = (Elev UP - Elev DS) River

Length.

Fig. 4.11. River map

c) Basin Slope

The basin slope tool computes average slope for the sub-basin using the slope
grid and sub-basin polygon. Add (percent slope for watershed) grid to the map
 93

document. Select - characteristics- Basin Slope. Confirm the inputs for Suh basin and

Slope Grid, and then click OK.

o is r u 2i a

Fig. 4.12. Slope map of Thuhapuzha basin

d) Longest Flow Path

This will create a feature class with poly line features that will store the
longest flow path for Longest flow path. Confirm the inputs, and leave theeach sub­
basin. Select_ Characteristics_ default output (Longest Flow Path) unchanged. Click

OK.
 94

e) Basin Centroid

This will create a Centroid point feature class to store the centroid of each sub-basin.
Select input Sub basin Characteristics Basin Centroid. Choose the default. Center of
Gravity Method, leave the default name (Centroid). Computes the method Click OK.

f) Basin Centroid Elevation

This will compute the elevation for the each centroid point using the underlying
DEM. Select Characteristics Centroid Elevation update. Confirm the input DEM and

centroid feature class, and click OK.

g) Centroidal Longest Flow Path

Select-Characteristics-Centroidal Longest Flow Path. Confirm the inputs, and

leave theSelect Characteristics default name for output Centroidal Longest Flow' Path,

and Click OK.

Fig. 4.13 Longest flow path map

 Fig. 4.14 Stream extraction map

14 r t* Ji »_

Fig. 4.15 Stream linking

4.7.1 HMS

The HMS menu has tools for creating input files fromHEC-GEOHMS.

Confirm the input feature classes. C lick ok on HMS process

 96

Fig. 4.16. FIMS Project Area

4.7.2 HMS Schematic

This tool creates a G IS representation of the system using (hydrologic) a

schematic network with basin elements (nodes/links or junctions/edges) and their
connectivity. Select FIMS- HMS Schematic. Confirm the inputs, and then dick

ok.

Fig. 4.17. HMS model area

 97

4.8 PARAMETER SENSITIVITY

The sensitivity analysis has carried out and the sensitive parameters shown in
Table4.3. Here sensitive parameters related to study are Initial abstraction, followed
by CN, and followed by lag time and remaining parameters are given in sequence
Initial discharge, recession constant, and Muskingum X and K values.
Table 4.3Calculated initial and optimized parameters for the watershed
No parameters Initial value Optimized value
1 Initial abstraction la mm 17.23 13.10
2 Curve number CN 63.15 65.90
3 Lag time flag min 18.51 17.90
4 Initial discharge Q m3/sec 0.0774 0.069
5 Recession constant 0.746 0.627
6 Threshold flow Q m7sec 0.0564 0.055
7 Muskingum K hr 0.378 0.292
8 Muskingum X 0.287 0.301

4.9. MODEL CALIBRATION AND VALIDATION

The calibration o f the model has been carried out by suitably modifying the
sensitive parameters, within the range suggested by the uncertainty analysis. It is
emphasized here that the calibration effort was very much reduced when the
optimum parameter search was limited to the parameters suggested by the sensitivity
analysis and their ranges suggested by the uncertainty analysis as described in
sections respectively. First the calibration was attempted to annual time series and
then it was extended to monthly basis. The summary statistics is given in table 4.4.

The comparison of monthly average flow for calibration and validation period
is shown in figures 4.18 & 4.20 to 4.19 & 4.21. Very high NSE o f 0.77 and 0.82 and
R2 of 0.88 to 0.91 have been obtained for the calibration period. However, the
simulations under estimate the peak values and this under estimation has been
reported by other researchers (Chatterjeeef.a/ 2014, Royet.al 2013)). Similarly the
 98

corresponding NSE values were increased from0.86 to 0.91 and R2 were 0.86 & 0.93
for the validation periods. So, this flow prediction information can be utilized for the
reservoir planning at any section o f the stream along its course within the river basin.
The above said calibration and validation is applicable for the total river
flow. Since, HEC-HMS predicts the hydrologic processes in component way, it may
be better, if different hydrologic processes are compared separately with their
measured or alternatively computed counterparts. Hence, to validate the model
output in a more detailed manner, the simulated base flow was compared with the
observed counterpart. As such there is no observed base flow. But the summer river
flow is considered as the soul contribution o f base flow and that has been matched
with the simulated one. Very close similarity was observed for all the years under
study as shown in previously mentioned figs. The corresponding descriptive
statistics is presented in table 4.6. NSE o f maximum 0.91 and an R of maximum

Fig4.18. Observed and simulated flow before calibration (monthly average flow

for calibration)
 99

Fig4.19. Observed and simulated flow before calibration (monthly average flow for

validation)

Fig 4.20. Observed and simulated flow after calibration (monthly average flow for

calibration)
 100

Fig4.21. Observed and simulated flow after calibration (monthly average flow for

validation)
Table4.4. Observed and simulated flow before and after calibration/validation

Statistics Calibration period(before) Validation period

Observed flow Simulated flow Observed flow Simulated flow

Flow(m3/sec) (m3/s) (m3/s) (m3/s) (m3/s)

Mean 64.46 53,16 46.05 44.63

SD 71.50 46.97 47.84 36.42

Nash
Efficiency 0.77 0.83

Coefficient of
determination 0.88 0.86

Statistics Calibration period(After) Validation period

 Observed flow Simulated flow Observed flow Simulated flow
Flow(m3/sec) (m3/s) (m3/s) (m3/s) (m3/s)

Mean 64.46 56.34 46.05 48.04

SD 71.50 54.76 47.84 41.42

Nash
Efficiency 0.82 0.91

Coefficient of
determination 0.91 0.93

HMS model has been successfully set up for an important river basin,
Thuthapuzha o f Kerala state in India by overcoming all odds regarding the
availability o f basic data and information in the ready to use format. With 7 years of
daily rainfall and river flow, the model has been calibrated and validated. Parameter
sensitivity analysis was carried out to identify the most sensitive parameters towards
river flow. It was followed by an uncertainty analysis to determine the feasible range
of parameter variation. Using the results o f sensitivity and uncertainty analysis, the
model was calibrated using 5 years of observed river flow. Validation o f the
calibrated model was attempted with separate data set of two years not used in
calibration
SUMMARYAND CONCLUSION
 102

CHAPTER V

SUMMARY AND CONCLUSION

The U.S. Army Corps o f Engineers’ Hydrologic Engineering Centre -

Hydrologic Modeling System (HEC-HMS) is designed to simulate for the complete
hydrologic processes o f dendritic watershed systems. HEC-HMS is widely used
under various widely varying geographic conditions and includes both traditional
hydrologic analysis procedures such as event infiltration, unit hydrographs, and
hydrologic routing as well as continuous simulation procedures including
evapotranspiration, snowmelt, and soil moisture accounting. It can be used in
conjunction with other software for studies o f water availability, urban drainage, flow
forecasting, future urbanization impact, reservoir spillway design, flood damage
reduction, floodplain regulation, and systems operation.

The hydrologic and hydraulic model provide updated conditions for the
Thuthapuzha river reach. HEC-HMS uses separate mathematical models to represent
each component o f the runoff process, like models to compute runoff volume, models
to compute direct runoff and base flow. Hydrology was modelled with a rainfall
runoff analysis in order to generate direct runoff from the land use condition. The
Soil Conservation Service Curve Number method developed by the U.S. Soil
Conservation Service was used for the loss method in this study.

In the present study Hydrologic Modeling System (HEC-HMS) is calibrated

and validated for Thuthathapuzha subbasin of Bharathapuzha river basin in Kerala.
The input data required for the model like precipitation, meteorological parameters,
river discharge, soil characteristics, land use characteristics and topographical
characteristics of the study area were collected from various agencies like Central
Water Commission (CWC), Kerala State Land Use Board (KSLUB), RARS Pattambi
and the Bhuvan geo-data portal o f National Remote Sensing Centre (NRSC).
 103

The watershed was divided into homogeneous sub-basins using WMS and HEC-
GEOHMS to get the sub-basin geometric data and then hydrological model was
developed in HEC-HMS.The analysis showed that CN, lag time are the sensitive
parameters for the simulation o f stream flow. Then optimization trials in HEC-HMS
was conducted to generate a calibrated model.

The model performance o f the calibrated HEC-HMS model for Thuthapuzha

watershed was evaluated using the statistics - Nash Sutcliffe- model efficiency
criterion, coefficient of determination and simulated time to peak. The analysis
showed that CN, and lag time are the most sensitive parameters for the simulation of
stream flow. The Nash-Sutcliffe model efficiency (E) was (0.77-0.8) and (0.86-0.88)
and the coefficient o f determination was (0.82-0.91), and (0.91-0.93) before and after
the calibration respectively, indicating the good performance o f the model.
REFERENCES
 104

REFERENCES

Abas, A.A. and Hashim, M. 2012. Estimation of impervious surface using multi temporal
land sat imagery and its impact to the runoff.

Abood, M.M., Mohammed, T.A., Ghazali, A.H., Mohamud, A.R., and Sidek, L.M. 2012.
Impact o f infiltration methods on the accuracy o f rainfall-runoff simulation. Res. J.
Appl. Sci. Eng. Tech. 4(12): 1708-1713.

Alaghamand, S., Abullah, R., and Abustan, I. 2011. Selecting the best set value in calibration
process for the validation o f hydrological modelling (Kayu Ara river basin,
Malaysia.i?es. J. Environ. Sci. 5(4): 354-365.

Alaghamand, S., Abullah, R., and Abustan,I. 2010. GIS-based river flood hazard mapping in
urban area (a case study in Kayu Ara river basin, Malaysia). I. J. Eng. Technol. 2(6):
488-500.

Al-Jabari, S.,4Sharkh, M.A., and Al-Mimi, Z. 2009. Estimation of runoff for agricultural
watershed using SCS curve number and GIS. Int. Water. Technol. conf. 1213-1229

Al-Jabari, S., Sharkh, M.A., and Al-Mimi, Z.2012. Management o f agricultural watershed
using GIS in Palestine. Int. water. Technol.conf. 1-10.

Ali, M., Jamal, S., Aslam, I., and Khan, Z. 2011. Simulation o f the impacts o f land-use
change on surface runoff o f Lai Nullah Basin in Islamabad, Pakistan. Natl. Uni Sci.
Technol. 102:271-279.

Asadi A.,and Boostani F. 2013. Application o f HEC-HMS for flood forecasting in Kabkian
basin and Delibajak sub basin in Iran. IOSR J. Eng. 3 (9): 2250-3021.

Asadi, A. and Boustani, F. 2013. Performance evaluation o f the HEC-HMS hydrological

model for lumped & semi-distributed storm flow simulation. Am. J. Eng. Res.
2(11):115-121.
 105

Asis Mazumdar and Sujana Dhar (2007) have conducted a study to quantify the impact of
climate change on the water resources o f the Jamtara watershed, West Bengal using a
distributed hydrological model HEC-HMS.

Arekhi, S. 2012. Runoff modelling by HEC-HMS model (Case study: Kan watershed, Iran).
Int. j. agric. crop sci. 4(23): 1807-1811.

Bakir, M and Xingnan, Z. 2008. GIS-based hydrological modelling: A comparative study of

HEC-HMS And the Xinanjiang model. Int. water Tech. conf.IWTC12 Alexandria,
Egypt.

Bhatt, A., Yadav, H.V., Kumar, D. 2012. Estimation o f the infiltration parameter for Tehri
Garhwal catchment. Int. J. Eng. Res. & Tech. 1(7).

Chatterjee, M.,Rumpa, D., Roy, D., Das, S. and Mazumdar, A. 2014. Hydrological modelling
studies with HEC-HMS for Damodar basin, India, J. word appl. sci. 31(12): 2148-
2154.

Choudhari, K., Panigrahi, B., and Paul, J. C. 2014. Simulation of rainfall-runoff process
using HEC-HMS model for Balijore Nala watershed, Odisha, India Int J. Geomatics
and Geosciences. 5(2).

Devries, J.J and Hromdaka, T.V 1993, computer models for surface water. In handbookof
hydrology D.R Maidment (Ed), Me Graw-Hill, New York

Donigian., A.S Watershed model calibration and validation. The HSPF experience
AQUATERRA Consultants 2685 Marine Way, Suite 1314Mountain View, CA
94043

Fleming, M. 2004. Description o f the hydrologic engineering center’s hydrologic modelling

system (HEC-HMS) and application o f watershed studies. Smart Tech. Notes
Collection, ERDC/TN SMART: 4-3.
 106

Ford, D., Pingel, N., Devries,J.J. 2008. Hydrologic modelling system (HEC-HMS)
Application guide. US Army Corps o f Engineers Hydrologic Center.

Gangodagamage. C 2001. Hydrological modeling using remote sensing and GIS. Asian Conf.
Remote Sensing.5-9.

Gautham, N.P. Flow routing with semi-distributed hydrological model HEC- HMS in case of
Narayani river basin. J, Inst.Eng. 10(1): pp. 45-58.

Gajbhiye, S. 2015. Estimation of surface runoff using Remote Sensing and Geographic
Information System. Int. J. U and eService, Sci. Tech. 8(4): pp. 113-122.

Golrang, B.M:, Lai, F.S., Sadeghi, S.H.R., Khamurdin, M.N., Abdkudus, K Mashayekhi. K.
Bagherian, R. 2013. Assessment o f the watershed management implemented on
sprinkle peak flood discharge and flood volume, using HEC-HMS model (Case
study: KushkAbad sub basin in Iran). B M Golrang etc. Sci. and nature. 2(2):59-64.

Hoggan, D.H., 1997, computer assisted floodplain hydrology and hydraulics, 2nd ed., Mc­
Graw-Hill, New York

Kabiri. 2014. Simulation o f the runoff using modified SCS-CN method using GIS
system. Res. J. Environ. Sci. 8(4): 178-192.

Kabiri using Green-Ampt loss method in surface runoff simulation (Klang watershed in west
Malaysia). Adv. Environ. Computational Chemistry Bio sci. 217-221.

Kaur, R., Kumar,S and Gurung, H. P.2002. A pedo-transfer function (PTF) for estimating
soil bulk density from basic soil data and its comparison with existing
PTFs Australian J. o f Soil Res. 40(5) :847 - 858.

Kherde, R. V. and Sawant., P.H. 2013. Integrating geographical information systems (GIS)
with hydrological modeling- Applicability and limitations. Int. J. Eng. and Tech.
5(4).

I
 107

Kumar, D. 2011. Distributed rainfall- runoff modelling. Int. J. o f Earth Sci. Eng.4(6):270-
275.

Haberlandt, U.and Radtke I. 2014. Hydrological model calibration for derived flood
frequency analysis using stochastic rainfall and probability distributions o f peak
flows.Hydrol. Earth Syst. Sci. 18: 353-365.

https://www.westernwatersheds.org/issues/issuesprotecting-watersheds.

Legesse. S. 2015. Application of the HEC-HMS model for runoff simulation o f Upper Blue
Nile River Basin. Hydrol. Curr. Res.6(2).

Maiden, D.R., 1993, Handbook o f hydrology, Mc-GRAW-Hill, New York

McCuen, R.H., 2005, Hydrologic analysis and design, 3rd ed., prentice Hall, Upper Saddle
River, NJ

Majidi, A. and Shahedi, K. 2012. Simulation O f rainfall-runoff process using Green- Ampt
method and HEC-HMS model (A case study: Abnama Watershed, Iran). IntJ.Hydrol.
Eng. 1(1): 5-9.

Majidi, A., and Vagharfard, H. 2013. Surface runoff simulation with two methods using
HEC-HMS model. Curr. Adv. in Environ. Sci. 1(1): 7-11.

Martin, O., Rugumayo, A. and Ovcharovichova, J. 2012. Application o f HEC-HMS/RAS

and GIS tools in flood modelling: A case study for river Sironko-Uganda. Glo. J.
Engi. Des& Tech. 1(2): 19-31.

Narasayya, K. 2015. A Geo-Informatic approach to estimate stream flow o f an un developed

catchment-a hypothetical research. Int. J. Advmtin Remote Sensing GIS and
Geography. 3(1): 22-31.

Oloche, O.J, Zhi-jia LI. 2010. Application o f HEC-HMS for the flood forecasting in Misai
and Wan a catchment in China. Water Sci. Eng. 3(1): 14-22.
 108

Patel M.M.,Gandhi H.M., and Shrimali N.J. 2014. Literature study on application o f HEC-
HMS for the event and continuous based watershed modeling. I. J. Sci. Res. Dev. 1:
2321-0613.

Rafiei, E., Rostami,N., Khalighi, S., and Tahei, S. 2012. Calibration o f loss estimation
methods in HEC-HMS for simulation o f surface runoff. Adv. in Environ. Biol. 6(1):
343-348.

Rahul, W.S. and Jain., M J. 2015. Continuous hydrological modeling using soil moisture
accountingalgorithm in Vamsadhara river basin, India. J. water. Resour. Hydraulic.
Eng. 4(4): PP. 398-408.

Razi M.A.M., Ariffin. J., Tahir. W., and Arish, M.A.M. 2010. Flood estimation studies using
hydrological modelling system (HEC-HMS) in Johor river, Malaysia. J. Appl. Sci.
10(11): 930-939.

Roy, D., Begam, S., Ghosh, S., and Jana, S. 2013. Calibration and Validation o f HEC-HMS
model for a river basin. J. Eng and Appl. Sci. 8(1).

Sampath, D.S., Weerakoon, S.B., and Herath, S. 2015. HEC-HMS model for runoff
simulation in a tropical catchment with Intra-Basin diversions - Case study o f the
Deduru Oya river basin, Sri Lanka. The Inst. Eng. 1 pp (1-9).

Scharffenberg, W.A 2013 US army corps of the engineers. Hydrological engineering centre
Hydrologic modelling system (HEC-HMS): user manual 4.0.

Sinha, R., Kumar, R., Sinha, S. Tandon, S.K. and Gibling, M.R. (2007) Late Cenozoic fluvial
successions in India: an overview and synthesis. Quaternary Science Reviews, 26,
2801-2822.

Stewart, L.T., and Godchild, M. F.1993. Integrating geographic information systems and
environmental simulation models. Environ. Modeling With GIS: 16-30.
 109

Supe, M.S., Taley, S.M., and Kale,M.U. 2015. Rainfall-runoff modelling using HEC-HMS
for Wan river basin. Int. J. Res. Eng. Sci.Tech. 1(8): 21-28.

Su, Z., Neumann, P., Schumann F.A., and Schultz G.A. 1992. Application o f remote sensing
and geographic information system in hydrological modeling. Earsel. Adv.remote
sensing. 1: 3-7.

Twumasi, Y. D., John Ayer, R., and Mathew, E. 2015. Computing flood volume O f Dikpe
Catchment using HEC-HMS. Glob J. O f Hum-Social Sci: B Geography, Geo-
Sciences, Environmental Science &Disaster Management. 15 (2).

Upadhyay, T.P., B. Solberg and P.L. Sankhayan, 2005. A review of carbon sequestration
dynamics in the Himalayan region as a function of land-use change and forest/soil
degradation with special reference to Nepal. Agric. Ecosyst. Environ., 105: 449-
465. DOI: 10.1016/j.agee.2004.09.007

Yasin, Z., Nabi, G., and Randhawa, S. M. 201. Modeling O f hill torrent using HEC Geo-
HMS and HEC- HMS Models: A Case Study o f Mithawan Watershed. Pakistan. J.
Meteorol. 11(22).

Yener, M.K., §orman, A.U., §orman, A.A., §ensoy, A., and Gezgin, T. 2006. Modeling
Studies With HEC-HMS and runoff scenarios in Yuvacik basin, Turkiye. Int. Congr.
River Basin Manag. 623-634.
APPENDICES
 110

APPENDIX-I

Anderson L and Use L and Cover Classification Codes:

First- and Second-Level Categories

X. U rban or Built-Up Land 6. W etland

11 Residential 6 1 Forested Wetlands

12 Commercial Services 6 2 Non-forested Wetlands

13 Industrial 7. B arren Land

14 Transportation, Communications 71 Dry Salt Flats

15 Industrial and Commercial 72 Beaches

16 Mixed Urban or Built-Up Land 73 Sandy Areas Other than Beaches

17 Other Urban or Built-Up Land 74 Bare Exposed Rock

2. A gricultural Land 75 Strip Mines, Quarries, and Gravel Pits

21 Cropland and Pasture 76 Transitional Areas

22 Orchards, Groves, Vineyards, Nurseries 77 Mixed Barren Land

23 Confined Feeding Operations 8. T undra

24 Other Agricultural Land 81 Shrub and Brush Tundra

3. Rangeland 82 Herbaceous Tundra

 Ill

31 Herbaceous Rangeland 83 Bare Ground

32 Shrub and Brush Rangeland 84 Wet Tundra

33 Mixed Rangeland 85 Mixed Tundra

4. Forest Land 9. Perennial Snow and Ice

41 Deciduous forest land 91 Perennial Snow fields

42 Evergreen forest land 92 Glaciers

43 Mixed forest land

5. W ater :

51 Streams and canals

52 Lakes

53 Reservoirs

54 Bays and estuaries

 112

APPENDIX-II

Hydrological models

Model Type Example of Model

Lumped parameter H EC-HMS, Snyder or Clark UH

Distributed Vflo, MIKE URBAN, MIKE FLOOD

Event H EC-HMS, SWMM, SCS TR-20

Continuous SWMM, HSPF, Vflo

Physically based H EC-HMS, SWMM, HSPF, Vflo

Stochastic Synthetic stream flows

 113

APPENDIX-III

HEC-HMS loss rate methods

M ethod Description

Initial and constant Initial loss volume is satisfied and then constant loss rate
begins.

Loss function is related to antecedent soil moisture condition

HEC exponential
and is a continuous function of soil wetness.

Initial loss is satisfied before calculating cumulative runoff as a

SCS curve number
function of cumulative rainfall using SCS methods

Infiltration rate is computed as an exponential function of

Holtan method
available soil moisture storage from Holtan’s equation.

Infiltration rate is computed from the Green and Ampt equation

Green and Ampt
as a function o f soil moisture and hydraulic conductivity

First an initial deficit storage is filled, and then infiltration rates

Deficit/Constant
can be specified on a monthly basis.

Soil moisture accounting assigns a value of initial storage to all

SMA
layers of the ground using a gridded method.
 114

APPENDIX-IV

Surface runoff methods in HEC-HMS

Unit hydrograph input directly

• Clark hydrograph method (TC&i? method Snydes unit hydrograph method

• SCS method (CN method + SCS UH)
■ Kinematic wave for overland hydrograph
• Mod-Clark
• User-specified S-graph
 115

APPENDIX-V

Flood Routing in HEC-HM S

M ethod Description

Muskingum Storage coefficient (x) plus travel time (K) through each reach

Table of storage vs. outflow for each reach based on HEC-RAS water
Modified Pul’s surface profile information—various storages are plotted vs. peak
flows in the reach

Channel shape, length, slope, and n; outflow from each reach based on
Kinematic wave
depth o f flow in continuity equation and Manning’s equation

Muskingum- Same equations and technique as the Muskingum- Cunge but channel
Cunge 8-point is described with 8 station-elevation values

Direct lag of flow through channel with translation accounted for but
Lag
not attenuation
 116

APPENDIX-VI

The name and the area of each G ram aPanchayat within T huthapuzha sub basin

Name of the
SI. No Name of the block A rea(km 2)
Panchayat

1 Kulukkallur Pattambi 22.8

2 Koppam Pattambi 20.9

3 Nellaya Pattambi 27.4

4 Vilayur Pattambi 17.7

5 Thiruvengapura Pattambi 20.4

6 Cherupulassery Sreekrishnapuram 27.6

7 Kadambazipuram Sreekrishnapuram 39.7

8 Karimpuzha Sreekrishnapuram 47.4

9 Pookkattukadavu Sreekrishnapuram 22

10 Sreekrishnapuram Sreekrishnapuram 29.5

11 Thrikkadivi Sreekrishnapuram 26.3

12' Vellinezi Sreekrishnapuram 26.8

13 Alanellur Manarkkad 58.3

14 Karakurissi Manarkkad 27
 117

15 Karimba Manarkkad 69.2

16 Kottappadam Manarkkad 29.8

17 Kumaraputhur Manarkkad 37.2

18 Manarkkad Manarkkad 63.4

19 Kanhirapuzha Manarkkad 58.4

20 Thachann attukara Manarkkad 35

21 Tachanpara Manarkkad 54

22 Thazhekode Perinthalmanna 45

23 Aliparamba Perinthalmanna 20.2

24 Elamkulam Perinthalmanna 21.3

25 Pulamanthole Perinthalmanna 33.7

26 Moorkannad Perinthalmanna 11.7

27 Irimbilam Kuttipuram 24
 118

APPENDIX-VII

Description of soil mapping units of Thuthapuzha sub basin

Soil mapping
Texture Description
unit

Very deep, moderately well drained, clayey soils with

moderately shallow water table in nearly level narrow
K08 Clayey valleys, with slight erosion: associated with very deep,
imperfectly drained, clayey soils with moderately
shallow water table on nearly level lands.

Very deep, well drained, gravelly clayey soils with

moderate surface gravelliness on moderately steeply
Gravelly
K09 sloping laterite mounds, with moderate erosion;
clay
associated with deep, well drained, gravelly clay soils
on gentle slopes.

Very deep, well drained, gravelly clay soils on gently

sloping midland Iaterites with valleys of northern
Gravelly Kerala, with moderate erosion, associated with deep,
K10
clay well drained, gravelly clay soils with moderate surface
gravelliness and ironstone layer at 100-150 cm on
nearly level lands, slightly eroded.

Deep, well drained, gravelly clay soils with moderate

Gravelly surface gravelliest on iron stone layer at 100-150 cm
K13
clay on gently sloping midland Iaterites, with moderate
erosion; associated with laterite outcrops.

Very deep imperfectly drained loamy soils with

moderately shallow water table in nearly level board
valleya of palghat gap, with slight erosion associated
K16 Fine loamy
with moderately deep, moderately well drained,
gravelly loamy soils with coherent material at 75-100
cm on gentle slopes, moderately eroded

K20 Gravelly Deep, somewhat excessively drained, gravelly clay

 119

clay soils with moderate surface gravelliness on steeply

sloping high hills with thick vegetation, with moderate
erosion; associated with very deep, well drained,
clayey soils on gentle slopes.

Very deep, well drained, clayey soils on gently

sloping lands having low hills with isolated hillocks,
K22 Clayey with moderate erosion; associated with deep, well
drained, gravelly clayey soils on moderately steep
slopes

Very deep, well drained, gravelly clayey soils with

Gravelly strong surface gravelliness on moderately sloping
K25
clay medium hills with moderate erosion, thin vegetation
associated with rock outcrops.

Very deep well drained clayey soils with loamy soils

K26 Clayey on gently sloping medium hills with thick vegetation,
with moderate erosion, associated with rock outcrops.

Moderately deep, well drained, moderately calcareous,

gravelly loam soils with moderate surface gravelliness
on gently sloping foot hills and valleys, with moderate
Gravelly erosion; associated with moderately shallow,
K28
loam somewhat excessively drained, gravelly clay soils
with strong surface gravelliness and coherent
materials at 50 to 75 cm on moderate slopes, severely
eroded.

Very deep, well drained, loamy soils on gently sloping

K29 Loam uplands with valleys, associated with very deep, with
moderate erosion, well drained, loamy soils on gentle
slopes

Very deep, well drained, clayey soils on moderately

sloping high hills with thick vegetation, with moderate
K30 Clayey erosion; associated with deep, well drained, gravelly
clayey soils with moderate surface gravelliness on
moderately steep slopes.
I
I
I
I
I

120

APPENDIX-VIII

Calculated initial and optimized parameters for the watershed

No Parameters Initial value Optimized value

1 Initial abstraction la mm 17.23 13.10

2 Curve number CN 63.15 65.90

3 Lag time T lag min 18.51 17.90

4 Initial discharge Q m3/sec 0.0774 0.069

5 Recession constant 0.746 0.627

6 Threshold flow Q m3/sec 0.0564 0.055

7 Muskingum K hr 0.378 0.292

8 Muskingum X 0.287 0.301

 121

APPENDIX-IX
Observed and sim ulated flow before and after calibration/validation

Statistics C alibration period(before) Validation period

■3 Observed flow Simulated flow Observed flow Simulated flow

Flow(m /sec)
(m3/s) (m3/s) (m3/s) (m3/s)

Mean 64.46 53.16 46.05 44.63

SD 71.50 46.97 47.84 36.42

N ash-
Sutcliffe 0.77 0.83
Efficiency

Coefficient of
0.88 0.86
determination

Statistics Calibration period(After) Validation period

Observed flow Simulated flow Observed flow Simulated flow

Flow(m3/sec) (m3/s) (m3/s) (m3/s) (m3/s)

Mean 64.46 56.34 46.05 48.04

SD 71.50 54.76 47.84 41.42

N ash-
Sutcliffe
0.82 0.91
Efficiency

0.93
Coefficient of 0.91
determination
 MODELLING THE HYDROLOGY OF WATERSHED BY
USING HEC-HMS

By
MAKKENA JYOTHI
(2014 -18 -118)

ABSTRACT OF THE THESTS

Submitted in partial fulfilment of the requirement forthe degree of

MASTER OF TECHNOLOGY
IN
AGRICULTURAL ENGINEERING
(Soil and Water Engineering)
Faculty of Agricultural Engineering & Technology
Kerala Agricultural University

DEPARTMENT IRRIGATION AND DRAINAGE ENGINEERING

KELAPPAJ1 COLLEGE OF AGRICULTURAL ENGINEERING AND TECHNOLOGY
TAVANUR- 679573, MALAPPURAM
KERALA, INDIA
3016
 A B STR A C T

A hydrological model is a commonly used tool to estimate the

hydrological response o f a watershed to precipitation. Hydrologic Modeling
System (HEC-HMS) is a physically based semi-distributed hydrologic modelling
software developed by the Hydrologic Engineering Center (HEC) o f the U.S.
Army Corps o f Engineers. It is designed to simulate the complete hydrologic
processes o f dendritic watershed systems under various widely varying
geographic conditions. HEC-HMS is widely used and includes both traditional
hydrologic analysis procedures such as event infiltration, unit hydrographs, and
hydrologic routing as well as continuous simulation procedures including
evapotranspiration, snowmelt, and soil moisture accounting. It can be used in
conjunction with other software for studies o f water availability, urban drainage,
flow forecasting, future urbanization impact, reservoir spillway design, flood
damage reduction, floodplain regulation, and systems operation.

In the present study, Hydrologic Modeling System (HEC-HMS) is

calibrated and validated for Thuthapuzha sub basin o f Bharathapuzha river basin
in Kerala. The input data required for the model like precipitation, meteorological
parameters, river discharge, soil characteristics, land use characteristics and
topographical characteristics o f the study area were collected from various
agencies like Central Water Commission (CWC), Kerala State Land Use Board
(KSLUB), RARS Pattambi and the Bhuvan geo-data portal o f National Remote
Sensing Centre (NRSC). The model performance o f the calibrated HEC-HMS
model for Thuthapuzha watershed was evaluated using the statistics -Nash
Sutcliffe- model efficiency criterion, coefficient o f determination and simulated
time to peak. The analysis showed that CN, and lag time are the most sensitive
parameters for the simulation o f stream flow. The Nash-Sutcliffe model efficiency
(E) was (0.77-0.8) and (0.86-0.88) and the coefficient o f determination was (0.82-
0.91) and (0.91-0.93) before and after the calibration respectively, indicating the
good performance o f the model.