12
th Esri India User Conference 2011
GIS, REMOTE SENSING AND
HYDROLOGICAL PAR
Anupam K Singh1
Professor, Department of Civil Engineering, 2
Senior Research Fellow, Department of Civil Engineering, 3
Graduate Student, Department of Civil Engineering,
Abstract:
In India most of the watersheds up to 500km
area can be classified as ungauged catchments. The
estimation of surface runoff from these catchments is not
only inaccurate, but impossible to understand the
catchment behavior and hydrological response to rainfall.
The hydrological response from each catchment helps in
flood routing vis-à-vis in flood modeling and flood
forecasting. In order to understand the hydrological
response a stream gauge network for a pilot catchment has
been established in Vare Khadi catchment. WL
automatic data logger sensors have been installed at three
locations for an area of 437km2 and river length of 48km.
Discharge measurements at Amli, Visdalia and Godsamba
sites have been carried out during monsoon period in 2010
and 2011 at 30sec time interval with an ac
0.0025%. The above river cross-sections and part of
catchment area have been surveyed using Trimble Geo
explorer XT global positioning system. The GPS point
elevation data has been combined with contour lines and
mass points obtained from topo-maps to produce a digital
elevation model (DEM) of 30m cell size in ArcGIS software
Hydrological parameters such as cross-section area, wetted
perimeter, hydraulic radius, flow velocity and longitudinal
slope have been calculated using Manning’s formula. A
hydrological model has been developed for depth
discharge calculation at given cross sections. The empirical
formulas of the form b
L )a(WQ = have been developed
for Visdalia and Godsamba. The depth-discharge empirical
formulas developed for discharge site Godsamba 2.62
Lg )13.25(WQ = and v (W77.21Q =
Visdalia site have been validated for showed good fit.
These depth-discharge relationships can be used for
predicting flood discharge in ungauged catchments.
(wc256)
Page 1 of 8
SENSING AND FIELD MEASUREMENTS FOR ESTIMATING
PARAMETERS IN UNGAUGED CATCHMENT
Anupam K Singh
1, S Sharma
2 U Vakharia
3
fessor, Department of Civil Engineering, Nirma University Ahmedabad
Senior Research Fellow, Department of Civil Engineering, Nirma University Ahmedabad
Graduate Student, Department of Civil Engineering, Nirma University Ahmedabad
ost of the watersheds up to 500km2 geographical
area can be classified as ungauged catchments. The
estimation of surface runoff from these catchments is not
only inaccurate, but impossible to understand the
catchment behavior and hydrological response to rainfall.
from each catchment helps in
vis in flood modeling and flood
forecasting. In order to understand the hydrological
response a stream gauge network for a pilot catchment has
been established in Vare Khadi catchment. WL-16
logger sensors have been installed at three
and river length of 48km.
a and Godsamba
sites have been carried out during monsoon period in 2010
and 2011 at 30sec time interval with an accuracy of
sections and part of
catchment area have been surveyed using Trimble Geo-
explorer XT global positioning system. The GPS point
elevation data has been combined with contour lines and
ps to produce a digital
in ArcGIS software.
section area, wetted
perimeter, hydraulic radius, flow velocity and longitudinal
slope have been calculated using Manning’s formula. A
hydrological model has been developed for depth-area-
discharge calculation at given cross sections. The empirical
have been developed
discharge empirical
discharge site Godsamba 2.32
L )(W for
Visdalia site have been validated for showed good fit.
discharge relationships can be used for
predicting flood discharge in ungauged catchments.
About the Author:
Dr.-Ing. Anupam K Singh, Ph. D
Dr.-Ing. Anupam K Singh is working as Professor in
Civil Engineering Department, Nirma University
Ahmedabad. He holds a PhD and master
Civil Engineering from University of Karlsruhe (TH)
Germany, and has more than 2
professional experience in water, infrastructure and
geomatics. Dr Singh has worked on 7
and consulting projects funded by EU, DFID, GTZ,
Government of India, and USAID.
5-dozens of publications in journals, conferences
and technical notes on hydrological modeling, flood
forecasting, irrigation, climate change, and urban
infrastructure planning.
Dr Singh is Editorial board of
Associate Editor of IJWR&EM, reviewer of almos
dozen journals. He has been awarded with the
engineering college faculty award
(2009), INFAS best poster award (2008), SIWI best
poster award (2002), and numerous fellowships
from Germany, Japan, Poland and India.
E mail ID: [email protected]
Contact No: +91 – 2717- 241912 ext 552
IELD MEASUREMENTS FOR ESTIMATING
AMETERS IN UNGAUGED CATCHMENT
irma University Ahmedabad
irma University Ahmedabad
Ing. Anupam K Singh is working as Professor in
Civil Engineering Department, Nirma University
olds a PhD and master’s degree in
Civil Engineering from University of Karlsruhe (TH)
more than 2-decades of
professional experience in water, infrastructure and
geomatics. Dr Singh has worked on 7-large research
and consulting projects funded by EU, DFID, GTZ,
Government of India, and USAID. He has more than
ons in journals, conferences
and technical notes on hydrological modeling, flood
forecasting, irrigation, climate change, and urban
of few journals incl.
reviewer of almost 1-
dozen journals. He has been awarded with the best
award for Gujarat State
INFAS best poster award (2008), SIWI best
poster award (2002), and numerous fellowships
from Germany, Japan, Poland and India.
241912 ext 552
12
th Esri India User Conference 2011
Introduction
In India most of the watersheds up to 500km
2 geographical area can be classified as ungauged catchments. These watersheds
have no past data records on depth-discharge or rainfall
discharge estimation for ungauged catchments, which
at el. (2004) for research study in Yangtza River, China have estimated hydrological parameters and developed a stage
rating curve using very high resolution quickbird-2 sa
two sites viz. Hankou (1.488x106 km
2 area) and Luoshan (1.296x10
used satellite based surface water and ocean topograph
such as Brahmaputra. The authors calculated river cross
produce discharge. Manning’s equation has been in continuou
(2011) used Manning’s formula for real-time river stage forecasting. They considered Minjiang and Fuchun rivers for analysis and
found good correction in real time discharge than river sta
The use of remote sensing and GIS for hydrological parameters estimation has increased considerably in recent years. The
method developed by Zhang et al (2004) has been used on quickbird
obtained show an average error of 9% which can be considered significantly low against results from ungauged basins. The
authors attributed these errors as a result of quickbird
Maghrebi at el. (2010) and Rahimpour el al. (2006) has developed stage
measurement in open channel through current meter data. Maghrebi at el. (2010) produced isovel contours for estimation of
discharge using a single point measurement. They worked with acoustic Doppler current profiler for measuring velocity profiles
and then calculating the discharge. The discharge results thus obtained from field measurements have better accuracy as again
satellite data. However, it is not possible to measure all hydrological parameters in field at each and every location. Therefore,
remote sensing and GIS technology overweight field measurements.
In this paper effort has been made to estimate hydrological parameters using
The research objective was to establish depth-area
at ungauged sites. The methodology has been applied for a pilot watershed Vare
Study Area
Tapi basin covers a geographical area of 65145 km2 and is the second largest India’s westward draining Inter
Arabian Sea. Basin covers three states having an area of 51504 km2 in Mahar
km2 in the Gujarat. The Tapi river basin can be classified in three zones, viz. Upper Tapi basin, Middle Tapi Basin, and Lowe
Basin (LTB). The area between Ukai Dam to Arabian Sea has been considered as LT
along with tens of small towns and villages along the river course. The Surat and Hazira twin cities are almost 106km
downstream of Ukai Dam, and have been affected by recurrence floods. One among the major ca
early peak discharge from various tributaries such as Vare khadi, Anjana khadi, Serul khadi, Mau khadi, and Gal khadi. Theref
Vare kadi watershed has been considered as pilot project for establishment of hydro
Vare khadi watershed a tributary of Tapi river located 40km upstream of Surat city near Mandvi has been a pilot project area
(refer Figure 1). The geographic coordinates of the study area are 21
right corners. The main river has 48km length and occupying geographical area of 437km
including an urban centre Zankhwaw, 150 rural villages, two storage reservoirs (Amli as major and Issar as minor).
reservoir storage capacity of Amli dam has been 37.54 million
created through building a dam. The Amli dam is mainly used for irrigation post
season. The right bank canal from Kakrapar weir located 30km upstream passes through vare khadi watershed which has around
76km2 irrigation command area.
The Landsat 7ETM+ satellite derived land cover categories in the watershed has been built
(29%), fellow (14%), water bodies (2%) and waste land (20%) as derived by Singh et al (2011). The field derived soil type and
Page 2 of 8
geographical area can be classified as ungauged catchments. These watersheds
discharge or rainfall-runoff relationship process. Research literature cites few methods on
discharge estimation for ungauged catchments, which required determination of physical characteristics of a catchment. Zhang
at el. (2004) for research study in Yangtza River, China have estimated hydrological parameters and developed a stage
2 satellite images. Empirical equations for depth-discharge were developed for
area) and Luoshan (1.296x106 km
2 area). Woldemichael et al. (2010) in their research work
used satellite based surface water and ocean topography (SWOT) data for estimation of water elevation for large braided river
. The authors calculated river cross-section using bathymetry data and Manning’s roughness coefficient to
produce discharge. Manning’s equation has been in continuous use from estimating river stage-discharge forecasting. Bao et al.
time river stage forecasting. They considered Minjiang and Fuchun rivers for analysis and
found good correction in real time discharge than river stage forecasting.
The use of remote sensing and GIS for hydrological parameters estimation has increased considerably in recent years. The
method developed by Zhang et al (2004) has been used on quickbird-2 satellite data products for large river basins.
obtained show an average error of 9% which can be considered significantly low against results from ungauged basins. The
authors attributed these errors as a result of quickbird-2 derived river channel geometry. In the research work carried by
Maghrebi at el. (2010) and Rahimpour el al. (2006) has developed stage-discharge curve by using fixed
measurement in open channel through current meter data. Maghrebi at el. (2010) produced isovel contours for estimation of
single point measurement. They worked with acoustic Doppler current profiler for measuring velocity profiles
and then calculating the discharge. The discharge results thus obtained from field measurements have better accuracy as again
wever, it is not possible to measure all hydrological parameters in field at each and every location. Therefore,
remote sensing and GIS technology overweight field measurements.
In this paper effort has been made to estimate hydrological parameters using part field measurements and part GIS technique.
area-discharge relationship for gauged points and apply the parameters prediction
at ungauged sites. The methodology has been applied for a pilot watershed Vare Khadi in lower Tapi basin in western India.
Tapi basin covers a geographical area of 65145 km2 and is the second largest India’s westward draining Inter
Arabian Sea. Basin covers three states having an area of 51504 km2 in Maharashtra, 9804 km2 in Madhya Pradesh, and 3837
km2 in the Gujarat. The Tapi river basin can be classified in three zones, viz. Upper Tapi basin, Middle Tapi Basin, and Lowe
Basin (LTB). The area between Ukai Dam to Arabian Sea has been considered as LTB, mainly occupying Surat and Hazira twin city
along with tens of small towns and villages along the river course. The Surat and Hazira twin cities are almost 106km
downstream of Ukai Dam, and have been affected by recurrence floods. One among the major cause of flood in LTB attributes to
early peak discharge from various tributaries such as Vare khadi, Anjana khadi, Serul khadi, Mau khadi, and Gal khadi. Theref
Vare kadi watershed has been considered as pilot project for establishment of hydro-meteorological network.
Vare khadi watershed a tributary of Tapi river located 40km upstream of Surat city near Mandvi has been a pilot project area
(refer Figure 1). The geographic coordinates of the study area are 21014'N 73
007'E to 21
030'N 73
030'E as lower
right corners. The main river has 48km length and occupying geographical area of 437km2. The watershed has important features
including an urban centre Zankhwaw, 150 rural villages, two storage reservoirs (Amli as major and Issar as minor).
reservoir storage capacity of Amli dam has been 37.54 million-m3 while Issar has very limited capacity. The dam storage has been
created through building a dam. The Amli dam is mainly used for irrigation post-monsoon and flood control during
season. The right bank canal from Kakrapar weir located 30km upstream passes through vare khadi watershed which has around
The Landsat 7ETM+ satellite derived land cover categories in the watershed has been built-up (4%), agriculture (32%), forest
(29%), fellow (14%), water bodies (2%) and waste land (20%) as derived by Singh et al (2011). The field derived soil type and
geographical area can be classified as ungauged catchments. These watersheds
runoff relationship process. Research literature cites few methods on
required determination of physical characteristics of a catchment. Zhang
at el. (2004) for research study in Yangtza River, China have estimated hydrological parameters and developed a stage-discharge
discharge were developed for
area). Woldemichael et al. (2010) in their research work
y (SWOT) data for estimation of water elevation for large braided river
section using bathymetry data and Manning’s roughness coefficient to
discharge forecasting. Bao et al.
time river stage forecasting. They considered Minjiang and Fuchun rivers for analysis and
The use of remote sensing and GIS for hydrological parameters estimation has increased considerably in recent years. The
2 satellite data products for large river basins. The results
obtained show an average error of 9% which can be considered significantly low against results from ungauged basins. The
2 derived river channel geometry. In the research work carried by
discharge curve by using fixed-point velocity
measurement in open channel through current meter data. Maghrebi at el. (2010) produced isovel contours for estimation of
single point measurement. They worked with acoustic Doppler current profiler for measuring velocity profiles
and then calculating the discharge. The discharge results thus obtained from field measurements have better accuracy as against
wever, it is not possible to measure all hydrological parameters in field at each and every location. Therefore,
part field measurements and part GIS technique.
discharge relationship for gauged points and apply the parameters prediction
Khadi in lower Tapi basin in western India.
Tapi basin covers a geographical area of 65145 km2 and is the second largest India’s westward draining Inter-state River in
ashtra, 9804 km2 in Madhya Pradesh, and 3837
km2 in the Gujarat. The Tapi river basin can be classified in three zones, viz. Upper Tapi basin, Middle Tapi Basin, and Lower Tapi
B, mainly occupying Surat and Hazira twin city
along with tens of small towns and villages along the river course. The Surat and Hazira twin cities are almost 106km
use of flood in LTB attributes to
early peak discharge from various tributaries such as Vare khadi, Anjana khadi, Serul khadi, Mau khadi, and Gal khadi. Therefore,
ogical network.
Vare khadi watershed a tributary of Tapi river located 40km upstream of Surat city near Mandvi has been a pilot project area
30'E as lower left and upper
. The watershed has important features
including an urban centre Zankhwaw, 150 rural villages, two storage reservoirs (Amli as major and Issar as minor). The estimated
while Issar has very limited capacity. The dam storage has been
monsoon and flood control during-monsoon
season. The right bank canal from Kakrapar weir located 30km upstream passes through vare khadi watershed which has around
(4%), agriculture (32%), forest
(29%), fellow (14%), water bodies (2%) and waste land (20%) as derived by Singh et al (2011). The field derived soil type and soil
12
th Esri India User Conference 2011
texture analysis for 32 samples analyzed in laboratory classify the watershed into two hydro
al, 2011). The estimated monsoon season annual rainfall has been 1376mm, minimum and maximum temperature of 22°C and
40°C respectively. The relative humidity values in the watershed have been 89% and 32% as maximum and
As described earlier, quick flood response from Vare khadi watershed has been a problem leading to floods in low laying areas
near Wareli village at the confluence of Vare khadi and Tapi River. The August 2006 flood in Surat and Hazira t
attributed to quick response from major sub-watersheds/ tributaries resulting 300 people being killed and US$ 4.5 billion value
property damage (Singh et al, 2009).
Fig:1 – Varekhadi a tributary of Tapi River along with
estimation
Sub-watershed boundary and discharge gauge locations
Methodology
To understand watershed response for rainfall and route the river discharge, Vare khadi was considered as pilot watershed for
installation of hydro-meteorological network. Three discharge locations have been selected for installation of automatic water
level sensors keeping in mind, the all-weather site accessibility, firm structures in rocky geology, not having transverse slope, on
downstream of bridge peer, and away from wind and wave turbulence. The automatic water level sensors WL
Water USA have been procured and installed at three discharge sites viz.
N21021'), and Ghodsamba (E73
013' N21
016'). The hydrological parameters for three gauged sites can be determined based on
the following methodology. It consists of four steps viz.
surveys and field measurements, and discharge estimation using Manning’s equation. The methodological procedure for
computation of sub-watershed discharges at above locations has been depicted as flow chart in Figure 2.
3.1 Geo database creation
The ArcGIS 9.2 has been used for database creation and GIS analysis.
created using topological maps, remote sensing images and field surveys. Topographical maps at 1:50000 scales were collated
from Survey of India. The topo-maps have been geo
contours, level points, streams, and watershed boundary. Landsat 7ETM+ satellite image of 21 Nov 2001 has been used for
classification of land use and land cover classes in the watershed
system (GPS) has been used for carrying out field surveys along the river channel and data sparse regions in the watershed. A
digital elevation model (DEM) for 15m cell size and 2.5m vertical accuracy for Vare khadi has been generated. The accuracy
assessment carried on DEM for selected locations shows
Page 3 of 8
texture analysis for 32 samples analyzed in laboratory classify the watershed into two hydrological soil groups B and C (Singh et
al, 2011). The estimated monsoon season annual rainfall has been 1376mm, minimum and maximum temperature of 22°C and
40°C respectively. The relative humidity values in the watershed have been 89% and 32% as maximum and
As described earlier, quick flood response from Vare khadi watershed has been a problem leading to floods in low laying areas
near Wareli village at the confluence of Vare khadi and Tapi River. The August 2006 flood in Surat and Hazira t
watersheds/ tributaries resulting 300 people being killed and US$ 4.5 billion value
along with Fig:2 – Research methodology adopted for discharge
boundary and discharge gauge locations in Varekhadi watershed
To understand watershed response for rainfall and route the river discharge, Vare khadi was considered as pilot watershed for
meteorological network. Three discharge locations have been selected for installation of automatic water
weather site accessibility, firm structures in rocky geology, not having transverse slope, on
downstream of bridge peer, and away from wind and wave turbulence. The automatic water level sensors WL
USA have been procured and installed at three discharge sites viz. Amli dam (E73023' N21
023'), Visdaliya (E73
The hydrological parameters for three gauged sites can be determined based on
logy. It consists of four steps viz. geo-database development, establishment of hydrological network, land
surveys and field measurements, and discharge estimation using Manning’s equation. The methodological procedure for
harges at above locations has been depicted as flow chart in Figure 2.
The ArcGIS 9.2 has been used for database creation and GIS analysis. The geo-database for Vare khadi watershed has been
remote sensing images and field surveys. Topographical maps at 1:50000 scales were collated
maps have been geo-reference, digitized and assigned attribute properties for themes such as
ershed boundary. Landsat 7ETM+ satellite image of 21 Nov 2001 has been used for
classification of land use and land cover classes in the watershed in ENVI softaware. Trimble Geo-explorer XT global positioning
ld surveys along the river channel and data sparse regions in the watershed. A
digital elevation model (DEM) for 15m cell size and 2.5m vertical accuracy for Vare khadi has been generated. The accuracy
assessment carried on DEM for selected locations shows a good fit and has been in coherence with actual elevation values. The
logical soil groups B and C (Singh et
al, 2011). The estimated monsoon season annual rainfall has been 1376mm, minimum and maximum temperature of 22°C and
40°C respectively. The relative humidity values in the watershed have been 89% and 32% as maximum and minimum limits.
As described earlier, quick flood response from Vare khadi watershed has been a problem leading to floods in low laying areas
near Wareli village at the confluence of Vare khadi and Tapi River. The August 2006 flood in Surat and Hazira towns has been
watersheds/ tributaries resulting 300 people being killed and US$ 4.5 billion value
Research methodology adopted for discharge
To understand watershed response for rainfall and route the river discharge, Vare khadi was considered as pilot watershed for
meteorological network. Three discharge locations have been selected for installation of automatic water
weather site accessibility, firm structures in rocky geology, not having transverse slope, on
downstream of bridge peer, and away from wind and wave turbulence. The automatic water level sensors WL-16U from Global
23'), Visdaliya (E73019'
The hydrological parameters for three gauged sites can be determined based on
database development, establishment of hydrological network, land
surveys and field measurements, and discharge estimation using Manning’s equation. The methodological procedure for
database for Vare khadi watershed has been
remote sensing images and field surveys. Topographical maps at 1:50000 scales were collated
reference, digitized and assigned attribute properties for themes such as
ershed boundary. Landsat 7ETM+ satellite image of 21 Nov 2001 has been used for
explorer XT global positioning
ld surveys along the river channel and data sparse regions in the watershed. A
digital elevation model (DEM) for 15m cell size and 2.5m vertical accuracy for Vare khadi has been generated. The accuracy
a good fit and has been in coherence with actual elevation values. The
12
th Esri India User Conference 2011
DEM has been considered as basis for delineation of sub
extracted DEM for Varekhadi watershed along with streams has been
Fig. 3 Development of Digital Elevation Model (DEM) at
15m cell size superimposed by stream network
3.2 Installation of Stream Gauge Sensors
Vare khadi has no discharge measurement gauging station and is classified as ungauged watershed. This is being a remote
location it has been proposed to install automatic sensors with data logger capabilities. WL
cable from Global Water USA were procured and installed in field during June 2010. The sensor has 0.1mm measurement
accuracy, and can record 10 reading per second. Three discharge sites viz. Amli, Visdalia and Godsamba have been selected as
marked in Figure 1, and considering the site selection criteria as listed under methodology.
It was decided to collect discharge data from all the stations at a time interval of 30sec. The data logger at other end of s
cable can store up to 80,000 values, which can be import
output data has standard spread sheet format as *.csv (
window-XP compatible water level logger software. Figure 3 de
installed sensor on bridge peer. The output file can be imported in Microsoft excel or ASCII format for further analysis. The
output can be further imported in major hydrological models for ra
Fig. 4 WL-16U SENSOR, Data Import process on
differential level Laptop and SENSOR installed on bridge peer
Page 4 of 8
DEM has been considered as basis for delineation of sub-watersheds boundary, sub-watershed areas and river slopes. The
rekhadi watershed along with streams has been shown in Figure–3.
evelopment of Digital Elevation Model (DEM) at
15m cell size superimposed by stream network in ArcGIS
Vare khadi has no discharge measurement gauging station and is classified as ungauged watershed. This is being a remote
location it has been proposed to install automatic sensors with data logger capabilities. WL-16U stream gauge sensors of 25m
Global Water USA were procured and installed in field during June 2010. The sensor has 0.1mm measurement
accuracy, and can record 10 reading per second. Three discharge sites viz. Amli, Visdalia and Godsamba have been selected as
nsidering the site selection criteria as listed under methodology.
It was decided to collect discharge data from all the stations at a time interval of 30sec. The data logger at other end of s
cable can store up to 80,000 values, which can be imported in laptop computer by connecting through a USB type
output data has standard spread sheet format as *.csv (comma separated by values) and output is acquired in excel format using
XP compatible water level logger software. Figure 3 depicts the WL-16U sensor, data import through computer, and
installed sensor on bridge peer. The output file can be imported in Microsoft excel or ASCII format for further analysis. The
output can be further imported in major hydrological models for rainfall-runoff modeling.
16U SENSOR, Data Import process on Fig. 5 Field survey measurement using
Laptop and SENSOR installed on bridge peer and GPS in Varekhadi.
watershed areas and river slopes. The
Vare khadi has no discharge measurement gauging station and is classified as ungauged watershed. This is being a remote
16U stream gauge sensors of 25m
Global Water USA were procured and installed in field during June 2010. The sensor has 0.1mm measurement
accuracy, and can record 10 reading per second. Three discharge sites viz. Amli, Visdalia and Godsamba have been selected as
It was decided to collect discharge data from all the stations at a time interval of 30sec. The data logger at other end of sensor
ed in laptop computer by connecting through a USB type-B cable. The
) and output is acquired in excel format using
16U sensor, data import through computer, and
installed sensor on bridge peer. The output file can be imported in Microsoft excel or ASCII format for further analysis. The data
Fig. 5 Field survey measurement using
and GPS in Varekhadi.
12
th Esri India User Conference 2011
3.3 Field survey and measurements over cross section of the river
Global positioning system (GPS) having sub-meter accuracy was used for carrying out field surveys. The field surveys were
conducted along river course and across the selected river cross
surveyed using GPS having 2.5m vertical accuracy. The river cross sections were surveyed using differential
vertical accuracy. The differential leveling has been aimed to achieve the enhanced accuracy in
levels obtained from differential-leveling were integrated into levels for generation of DEM.
The survey findings from river cross-sections have been employed for development of stage
stage-discharge curves. It was possible to analyze the longitudinal slope between discharge stations.
hydraulic radius have been calculated for each section to determine flow velocity based on Manning’s equation. The glimpse of
field surveys using differential-level and GPS are depicted in Figure 5.
The field data collection survey has been carried out for
reference levels (RL’s) have been plotted for both stations; they are plotted at 3m horizontal interval with chainage on X
Both the river cross-sections and chainage are depicted in Fi
Visdalia and 48 points for Godsamba stations were drawn. The river width at Visdalia and Godsamba cross
as 35m and 45m respectively.
Fig. 6 (a) River Cross – section at gauge site Visdaliya
Godsamba
3.4 Discharge Estimation
Literature cites several flow estimation methods for ungauged catchments viz. rational method, SCS
cook’s method, Manning’s method, and unit hydrograph method. Given the limited field measurements and data availability
Manning’s method has been found to be accurate and reliable for discharge estimation. Based on the watershed prope
physical conditions, it was decided to use Manning’s method.
Let us assume that hydraulic radius Rh [m] and longitudinal slope
calculated using Manning’s equation;
2132 .1
SRn
Vh
=
Where, the term n is manning’s roughness coefficient. The equivalent hydraulic radius for a trapezoidal cross
calculated using equation (2) as given below;
P
ARh=
Page 5 of 8
over cross section of the river
meter accuracy was used for carrying out field surveys. The field surveys were
conducted along river course and across the selected river cross-sections. The regions having sparse elevations have been
surveyed using GPS having 2.5m vertical accuracy. The river cross sections were surveyed using differential
vertical accuracy. The differential leveling has been aimed to achieve the enhanced accuracy in cross-section. The cross
leveling were integrated into levels for generation of DEM.
sections have been employed for development of stage-area, stage
discharge curves. It was possible to analyze the longitudinal slope between discharge stations. The longitudinal slope and
lic radius have been calculated for each section to determine flow velocity based on Manning’s equation. The glimpse of
level and GPS are depicted in Figure 5.
The field data collection survey has been carried out for Visdalia and Godsamba using differential level. The cross
reference levels (RL’s) have been plotted for both stations; they are plotted at 3m horizontal interval with chainage on X
sections and chainage are depicted in Figure 6 (a) and (b) respectively. In total 40 chainage points for
Visdalia and 48 points for Godsamba stations were drawn. The river width at Visdalia and Godsamba cross
section at gauge site Visdaliya Fig. 6 (b) River Cross – section at gauge site
Literature cites several flow estimation methods for ungauged catchments viz. rational method, SCS-curve number method,
cook’s method, Manning’s method, and unit hydrograph method. Given the limited field measurements and data availability
Manning’s method has been found to be accurate and reliable for discharge estimation. Based on the watershed prope
physical conditions, it was decided to use Manning’s method.
[m] and longitudinal slope S [1:n] are know, the mean flow velocity
is manning’s roughness coefficient. The equivalent hydraulic radius for a trapezoidal cross
1
2
meter accuracy was used for carrying out field surveys. The field surveys were
aving sparse elevations have been
surveyed using GPS having 2.5m vertical accuracy. The river cross sections were surveyed using differential-level having 0.5mm
section. The cross-section
area, stage-area-perimeter and
The longitudinal slope and
lic radius have been calculated for each section to determine flow velocity based on Manning’s equation. The glimpse of
Visdalia and Godsamba using differential level. The cross-section
reference levels (RL’s) have been plotted for both stations; they are plotted at 3m horizontal interval with chainage on X-axis.
gure 6 (a) and (b) respectively. In total 40 chainage points for
Visdalia and 48 points for Godsamba stations were drawn. The river width at Visdalia and Godsamba cross-section was observed
section at gauge site
curve number method,
cook’s method, Manning’s method, and unit hydrograph method. Given the limited field measurements and data availability
Manning’s method has been found to be accurate and reliable for discharge estimation. Based on the watershed properties and
[1:n] are know, the mean flow velocity V [m/s] can be
is manning’s roughness coefficient. The equivalent hydraulic radius for a trapezoidal cross-section can be
12
th Esri India User Conference 2011
Where, the terms cross-section area A [m2] and wetted perimeter
discharge at an open channel having known cross
product of both. Thus, discharge Q at given cross-
follows;
A.V=Q
Therefore, several parameters for discharge calculation s
using field surveys and measurements for both sites.
Results
The water level data at 2-gauging sites viz. Visdalia and Godsamba have been collected during June
sections at both locations were surveyed using GPS and differential
stations has been established and equations been developed. The polynomial trend line for both stations shows very good fit
having root mean square error (R2) in the range of 0.9953 and 0.9989 for Visdalia and Godsamba respectively. The depth
curves and associated equations for both stations are depicted in Figure 7(a) and (b).
Manning’s equation as given in equation (1) has been used to calculate the flow velocity at each station. The flow velocity f
to be varying and has been in the range of 1.32
velocity has been in upper range as compared to normal river velocity. It has to be noted that flow velocity increases with
increase in river depth, resulting in increased discharge
velocity, and has been found in the range of 29.50
and (b)). The root mean square error (R2) for both stations
equations developed for each discharge stations are given below as equation (4)
Godsamba;
2.32
Lv )(W77.21Q =
2.62
Lg )13.25(WQ =
Fig. 7(a) Depth – Area Curve for Visdaliya
Page 6 of 8
] and wetted perimeter P [m] are for equivalent trapezoidal cross
discharge at an open channel having known cross-section area A [m2] and mean flow velocity V [m/s] can be expressed as
-section [m3/s] for Visdalia or Godsamba can be calculated from equation (3) as
Therefore, several parameters for discharge calculation such as A, P, Rh, and S are derived from the database developed in GIS
using field surveys and measurements for both sites.
gauging sites viz. Visdalia and Godsamba have been collected during June-Sept 2010. The river cross
sections at both locations were surveyed using GPS and differential-leveling. Later the depth-area relationship at both discharge
stations has been established and equations been developed. The polynomial trend line for both stations shows very good fit
) in the range of 0.9953 and 0.9989 for Visdalia and Godsamba respectively. The depth
nd associated equations for both stations are depicted in Figure 7(a) and (b).
Manning’s equation as given in equation (1) has been used to calculate the flow velocity at each station. The flow velocity f
2-1.71m/s for Visdalia and 1.63-2.32m/s for Godsamba. This shows that flow
velocity has been in upper range as compared to normal river velocity. It has to be noted that flow velocity increases with
increase in river depth, resulting in increased discharge. The river discharge at both locations has been calculated for every
velocity, and has been found in the range of 29.50-67.47m3/s for Visdalia and 65.52-162.65m
3/s for Godsamba (refer Figure 8(a)
) for both stations has been in the same range like depth-area relationship. The empirical
equations developed for each discharge stations are given below as equation (4) Qv for Visdalia and equation (5)
Area Curve for Visdaliya Fig. 7(b) Depth – Area Curve for Godsamba
3
4
5
[m] are for equivalent trapezoidal cross-section. The
[m/s] can be expressed as
/s] for Visdalia or Godsamba can be calculated from equation (3) as
uch as A, P, Rh, and S are derived from the database developed in GIS
Sept 2010. The river cross-
area relationship at both discharge
stations has been established and equations been developed. The polynomial trend line for both stations shows very good fit
) in the range of 0.9953 and 0.9989 for Visdalia and Godsamba respectively. The depth-area
Manning’s equation as given in equation (1) has been used to calculate the flow velocity at each station. The flow velocity found
2.32m/s for Godsamba. This shows that flow
velocity has been in upper range as compared to normal river velocity. It has to be noted that flow velocity increases with
. The river discharge at both locations has been calculated for every
/s for Godsamba (refer Figure 8(a)
area relationship. The empirical
for Visdalia and equation (5) Qg for
Area Curve for Godsamba
12
th Esri India User Conference 2011
Fig. 8(a) Stage – Discharge Curve for Visdaliya
The inter-gauge relationship between Visdalia an upstream station at 85m RL and Godsamba a downstream station at 35m RL
has been studied. Both the stations are located 17km apart where discharge measurements are being carried out since June
2010. It has been estimated that time to flow for various stage levels have been in the range of 1.30hours to 2.32hours
depending upon flow velocity. The actual measurements are depicted in Figure 9 below as light dotted line (Visdalia) and dark
line (Godsamba).
Fig. 9 Gauge inter
Godsamba measurement stations (x
Time and y
Conclusion
The methodology presented above provides possibility of discharge
limited field measurements. This research paper demonstrates an approach to generate reliable stage
perimeter and stage-discharge curves to understand the hydrological behavior of su
estimating the flood response, flood routing time and hydrological process between two stations. Therefore, the future resear
scope will be to provide reliable and accurate flood forecasts at selected stations in
can be very well justified in estimating flood discharge and flood routing. It is believed that approach offers capability to
reproduce complex stage-discharge relationship for small rural sub
Page 7 of 8
Discharge Curve for Visdaliya Fig. 8(b) Stage – Discharge
gauge relationship between Visdalia an upstream station at 85m RL and Godsamba a downstream station at 35m RL
has been studied. Both the stations are located 17km apart where discharge measurements are being carried out since June
stimated that time to flow for various stage levels have been in the range of 1.30hours to 2.32hours
depending upon flow velocity. The actual measurements are depicted in Figure 9 below as light dotted line (Visdalia) and dark
Fig. 9 Gauge inter-relationship between Visdaliya and
Godsamba measurement stations (x- axis represents
Time and y – axis represents stage)
The methodology presented above provides possibility of discharge estimation in ungauged or poorly gauged watersheds under
limited field measurements. This research paper demonstrates an approach to generate reliable stage
discharge curves to understand the hydrological behavior of sub-watersheds. This approach has helped in
estimating the flood response, flood routing time and hydrological process between two stations. Therefore, the future resear
scope will be to provide reliable and accurate flood forecasts at selected stations in Vare khadi. The use of Manning’s equation
can be very well justified in estimating flood discharge and flood routing. It is believed that approach offers capability to
discharge relationship for small rural sub-watersheds.
Discharge Curve for Godsamba
gauge relationship between Visdalia an upstream station at 85m RL and Godsamba a downstream station at 35m RL
has been studied. Both the stations are located 17km apart where discharge measurements are being carried out since June
stimated that time to flow for various stage levels have been in the range of 1.30hours to 2.32hours
depending upon flow velocity. The actual measurements are depicted in Figure 9 below as light dotted line (Visdalia) and dark
estimation in ungauged or poorly gauged watersheds under
limited field measurements. This research paper demonstrates an approach to generate reliable stage-area, stage-area-
watersheds. This approach has helped in
estimating the flood response, flood routing time and hydrological process between two stations. Therefore, the future research
Vare khadi. The use of Manning’s equation
can be very well justified in estimating flood discharge and flood routing. It is believed that approach offers capability to
12
th Esri India User Conference 2011
References:
1. Garth A. Lindner, and Andrew J. Miller, 2011, Numerical Modeling of Stage
Journal of Hydrologic Engineering. Submitted August 31, 2010; accepted June 10, 2011.
2. Singh A K, Sharma S, Jain A O, 2011, Mapping and Prediction of surface Run
ISG2011 Ajmer, 3-5 Feb 2011.
3. Singh A K, Sharma S, 2010, Assessment of SLC
India , Proceeding GRID2010, Lonavala, 1
4. W. C. Sun, H. Ishidaira, and S. Bastola (2010) Towards improving river discharge estimation in ungauged
basins:calibration of rainfall-runoff models based on satellite observations of river flow width at basin outlet, Hydrol.
Earth Syst. Sci., 14, 2011–2022, 2010.
5. A T Woldemichael, A M Degu, A. H. M. Siddique
Manning’s Roughness Parameter in Space
Brahmaputra River in Bangladesh Role of Land
Borne Estimation of Discharge for Braided Rivers: A Case Study of the Brahmaputra River in Bangladesh, Applied Earth
Observations And Remote Sensing, Vol. 3, No. 3, September 2010.
6. Mahmoud F. Maghrebi and Mohammad Givehchi, Discharge Estimation in
Journal of Waterway, Port, Coastal, and Ocean Engineering © ASCE / September/October 2010.
7. Singh A K, Sharma A K, (2009), GIS and a remote sensing based approach for urban flood
catchment, India. IAHS Publ. 331 (2009) 389
8. GUOLIANGYU, SIOW-YONG LIM, (2003) Modified Manning formula for flow in alluvial channels with sand
of Hydraulic Research Vol. 41, No. 6 (2003), pp. 597
9. D. Ganora, P. Claps, F. Laio and A. Vigli
ungauged basins, WATER RESOURCES RESEARCH, VOL.45, W10418,doi:10.1029/2008WR007472, 2009
10. J. J. V´elez1, F. L´opez Unzu, M. Puricelli, and F. Franc´es. (2001). Parameter extrapolatio
hydrological distributed model in a regional framework. Hydrol. Earth Syst. Sci. Discuss., 4, 909
11. M Rahimpour, M. F. Maghrebi, (2006) Prediction of stage
measurement. Flow Measurement and Instrumentation 17 (2006) 276
Page 8 of 8
Garth A. Lindner, and Andrew J. Miller, 2011, Numerical Modeling of Stage-Discharge Relationships in Urban Streams,
Journal of Hydrologic Engineering. Submitted August 31, 2010; accepted June 10, 2011.
O, 2011, Mapping and Prediction of surface Run-off using SCS CN Method, Proceedings
Singh A K, Sharma S, 2010, Assessment of SLC-off Landsat 7ETM+ images for Land use mapping in Lower Tapi Basin,
onavala, 1-3 Dec 2010.
W. C. Sun, H. Ishidaira, and S. Bastola (2010) Towards improving river discharge estimation in ungauged
runoff models based on satellite observations of river flow width at basin outlet, Hydrol.
A T Woldemichael, A M Degu, A. H. M. Siddique-E-Akbor, and F Hossain (2010). Role of Land–Water Classification and
Manning’s Roughness Parameter in Space-Borne Estimation of Discharge for Braided Rivers: A Case Study of the
Brahmaputra River in Bangladesh Role of Land–Water Classification and Manning’s Roughness Parameter
Borne Estimation of Discharge for Braided Rivers: A Case Study of the Brahmaputra River in Bangladesh, Applied Earth
Observations And Remote Sensing, Vol. 3, No. 3, September 2010.
Mahmoud F. Maghrebi and Mohammad Givehchi, Discharge Estimation in a Tidal River with Partially Reverse Flow.
Journal of Waterway, Port, Coastal, and Ocean Engineering © ASCE / September/October 2010.
Singh A K, Sharma A K, (2009), GIS and a remote sensing based approach for urban flood-plain mapping for the Tapi
nt, India. IAHS Publ. 331 (2009) 389-394.
YONG LIM, (2003) Modified Manning formula for flow in alluvial channels with sand
of Hydraulic Research Vol. 41, No. 6 (2003), pp. 597–608.
D. Ganora, P. Claps, F. Laio and A. Viglione (2009), An approach to estimate nonparametric flow duration curves in
ungauged basins, WATER RESOURCES RESEARCH, VOL.45, W10418,doi:10.1029/2008WR007472, 2009
J. J. V´elez1, F. L´opez Unzu, M. Puricelli, and F. Franc´es. (2001). Parameter extrapolation to ungauged basins with a
hydrological distributed model in a regional framework. Hydrol. Earth Syst. Sci. Discuss., 4, 909–956, 2007.
M Rahimpour, M. F. Maghrebi, (2006) Prediction of stage–discharge curves in open-channels using a fixed
measurement. Flow Measurement and Instrumentation 17 (2006) 276–281.
Discharge Relationships in Urban Streams,
off using SCS CN Method, Proceedings
off Landsat 7ETM+ images for Land use mapping in Lower Tapi Basin,
W. C. Sun, H. Ishidaira, and S. Bastola (2010) Towards improving river discharge estimation in ungauged
runoff models based on satellite observations of river flow width at basin outlet, Hydrol.
Water Classification and
Borne Estimation of Discharge for Braided Rivers: A Case Study of the
Water Classification and Manning’s Roughness Parameter in Space-
Borne Estimation of Discharge for Braided Rivers: A Case Study of the Brahmaputra River in Bangladesh, Applied Earth
a Tidal River with Partially Reverse Flow.
plain mapping for the Tapi
YONG LIM, (2003) Modified Manning formula for flow in alluvial channels with sand-beds. Journal
one (2009), An approach to estimate nonparametric flow duration curves in
ungauged basins, WATER RESOURCES RESEARCH, VOL.45, W10418,doi:10.1029/2008WR007472, 2009
n to ungauged basins with a
956, 2007.
channels using a fixed-point velocity