th

12 Esri India User Conference 2011

IMPACTS OF LAND USEE AND C

CLIMATE CHANGE SCENARIOS ONN THE WATER
W FLUX
A CASE STUDY
UDY OF SSAL RIVER WATERSHED, CHAMBA
A (H. P.)

Kuldeep Pareta1, Amit K Srivastava2

Dr Kuld
1
Project Manager
ger (RS/G
(RS/GIS & Natural Resource Management), Spatial Decisions
ecisions
2
Remote Sensing & GIS Specialist, Spatial Decisions
B-30
30 K
Kailash Colony, New Delhi - 48 (INDIA)

Abstract: About the Author:

With the flow of the Sal River becoming

ming incr
increasingly erratic
especially in the upper reaches, atten attention has been
directed to land use change as the he majo
major cause of this
problem. The semi-distributed hydrological
rological model Soil and
Water Assessment Tool (SWAT), IRS-P6--LISS-IV-Mx (5.8m)
and IRS-P6-LISS-III (23.5m) satellite imageri
imageries were utilized
in the Sal River watershed in order to (i) mmap existing field
scale land use practices in order to determi
determine their impact Dr Kuldeep Pareta Ph.D.. Geomorphology
Geomo
(ii) determine the impacts of land d use ch
change on water
flux; and (iii) determine the impacts pacts o of rainfall and Author has obtained M.Sc. .Sc. degree
de in Geography
temperature variations based on the Inte Intergovernmental from Dr Hari Singh Gour University (Central
Panel on Climate Change (IPCC) projection
rojections on the water University), Sagar - Madhyaadhya Pradesh in 2001,
flux of the Sal River. subsequently Ph.D. in Geomorphology
Geomor & Hydro-
Geology from same university
versity in 2005. Presently, he
This study found that the differentt scenari
scenarios affected the is working as Project Manager ager (RS/GIS & Natural
water-balance components differently.
ently. LaLanduse changes Resource Management) in SpatialSpa Decisions, New
resulted in a slightly more erratic discha
discharge while rainfall Delhi (INDIA), and has over
ver 10 years
ye of research and
and temperature changes had a more ore pre
predictable impact development experience ce in theth field of National
on the discharge and water-balance ce com
components. These Resource Management, t, Geomorphology,
Geom Hydro-
findings demonstrate that the model el result
results show the flow Geology, Watershed Modelling,
Modelli and National
was more sensitive to the rainfall changes than land use Disaster Management. He has published over 23
changes. It was also shown that land use changes could research papers in various variou National and
reduce dry season flow, which is the m most important International Journals. He was conferred
c Prof. S.M.
problem in the watershed. The model del show
shows deforestation Ali Memorial Gold Medal dal in 2001 and MP Young
in the Binnu, Suri, and Mills forestest incre
increased the peak Scientist award in the year
ear 2004.
2004 He is a member of
flows, which can also lead to high sedimen
sediment loading in the Indian Society of Remote te Sensing,
Sens Madhya Bharti
Sal River. The effect of the land usese and climate-change- Journal, International Journal of Scientific and
scenarios on the sediment and water-qua quality of the river Engineering Research, Internatio
nternational Association for
needs a thorough understanding off the sedsediment transport Environmental Hydrology,gy, International
Intern Association
processes in addition to observed d sedime
sediment and water- of Geomorphologists and nd visited
visite Vietnam for his
quality data for validation of modeling
ing resul
results. research works.

E mail ID: kuldeep.p@spatiald

spatialdecisions.in
Contact No: +91-987192433
71924338
 th
12 Esri India User Conference 2011

1. Introduction
Water is the most precious gift of nature to the man
mankind and has been recognized as one of the most st vital natural
n resources. It is not so
just, because water sustains life but it is a renewa
renewable resource. It is also the most essential inputt for agriculture
agr industry and power
generation. The availability of water with proper
roper qu
quality and quantity at appropriate time and space e are of great
g importance. The water
management is very essential to maintain quality,
uality, qu
quantity and the availability of water due to increase
ase in population,
po rapid urbanization
& industrial growth. Though its availability is limited
limited, yet demand for water is ever increasing. It hass direct impact on human being and
socio-economic development. Hence, the need of proper planning management of the precious resource has become the matter of
utmost urgency.
The specific objectives of this study were to:: map ex
existing land use practices using remote sensing and
nd field observations, determine the
impacts of land use change, rainfall and air temper
temperature variation on the water fluctuation of the Sal River in Chamba. The findings of
this study provided scenarios on the impactss of land use and climate change in the Sal River Watershed
shed therefore
the adding to the existing
literature and knowledge base with a view of prompromoting better land use management practices in Ravi RiverRiv Basin, Himachal Pradesh
and application of the SWAT model in similarr densely populated, highly agricultural watersheds all over
ver the world.
w

2. Study Area
Sal River is a moderate size northern sub-tributary
butary oof the river Ravi, and originating at the eastern part
rt of Nichla
Nic Supeka, Chamba district,
it is flow essentially NE to SW and over 32.3 km to jjoin the Ravi River near Chamba town of Himachal al Pradesh.
Prade Though there is no main
tributaries of the Sal River, there are some small tr tributaries pouring into the river, notable amongstgst there are Gandera Nala, Sanluta
Nala, Saredi Khad on the left bank, and Kundidi Nala, KKiri Nala, Ganji Ka Khad, and Sarotha Nala on the right bank.
ba The Sal River watershed
2
covers about 265.63 km and characterized by differ
different types of land cover and land uses as a resultt of different
differ human activities carried
out by the stakeholders in various parts of the
he water
watershed.

Figure 1: Location Map of the Study Area (Sall River W

Watershed)

3.0 Data Used and Methodology

The methodology for this study involves the followin
following steps.
Table 1: Variables used in the SWAT model and nd data sources
S. No. Variables Data Sources
urces & Methodology
1. Remote Sensing Data - IRS-P6 LISS-IVIV M
Mx data (5.8 m), Dated: April 16, 2011
- IRS-P6-LISS--III Data (23.5 m), Dated: 02 December, 2010
- ASTER (DEM
DEM) (30 m), Dated: 22 August, 2006
2. Topographical Map - Surveyy of India Topographical Map at 1:50,000 Scale - 52 D / 2 & 52 D / 6
3. Landuse / Landcover - Digitally LULC map and forest map have been prepared by usingg Image Classification Tool in
Mapping ArcGIS-10,
10, and IIRS-P6 LISS-IV Mx Data, and also verified through limited field check
4. Soil Map - Soil map
ap of Ch
Chamba district has been collected from National Bureau
ureau of
o Soil Survey and Land Use
Planningg (NBSS&
(NBSS&LUP) and updated through IRS-P6 LISS-IV Mx Data
5. Drainage & Slope Map - Drainage
ge netw
network has been generated in GIS environment usingg ASTER-DEM
ASTER data and
ArcHydroTool
roTool in ESRI ArcGIS-10
 th
12 Esri India User Conference 2011

- Slope map has been created using Spatial-Analyst Extension in ArcGIS-10

ArcGIS and ASTER (DEM) data
6. Climate Data - Daily rainfall, ttemperature, wind speed, humidity and evapotranspirati
anspiration data have been
collectedd from Land Records Department, Chamba & IMD Station

3.1 LULC Classification

Digitally landuse and landcover map and forest m map have been prepared using image classification tion tool
too & supervised classification
(
method in ArcGIS-10, and IRS-P6 LISS-IV Mx (5.8 m) (Fig. 2) and IRS-P6 LISS-III 23.5 m) (Fig. 3) satellite
ite imageries.
image LULC of the study area
(Fig. 4) comprise mostly of forest (Reserve, Protect & open), scrub and agricultural land (Table 2). These LULC maps were then converted
to shapefiles and aggregated to make them easier to input into the SWAT model for use in the hydrological
rological modeling exercise. Land use
and management is an important factor affecti affecting different processes in the watershed such ch as surface
su runoff, erosion and
evapotranspiration. Reclassification of the land use map was done in order to present them in a form that tha is acceptable in the SWAT
model and this is the USGS-LULC classificationn schem
scheme for use with remote sensor data level classification
ation (Anderson
(An et al., 1976).

Figure 2: IRS-P6 LISS-III Data (23.5m) Figure

ure 3: IRS
IRS-P6 LISS-IV Data (5.8m) Figure 4: LULC Map
p 2010-11
2010
Table 2: Land Use / Land Cover Type Reclassification
ification into SWAT LU/LC Classes
Landuse & Landcover Type Settlement Agricultu
Agricultural Reserve Forest Protect Forest Open Forest
For Scrub Land River
SWAT LULC Type Settlement Agricultu
Agricultural Land Generic Forest Evergreen Forest Deciduous Forest Mixed
M Range Grasses Water
Area (in Hectare) 383.82 4586.61 999.60 12574.29 508.69 7376.43 133.62

3.2 Soil Data Classification

Soil data was obtained from the National Bureau
ureau of Soil Survey and Land Use Planning (NBSS&LUP), Nagpur and update through IRS-P6
LISS-IV Mx data (Fig. 6), and fit into SWAT database by using ArcSWAT-Extension in ArcGIS-10. Soil layer wasw obtained and used in the
hydrological model as one of the main inputs ts to the SWAT model which requires soil property data (Fig. 5)such
5) as the texture, chemical
composition, physical properties, available moist
moisture content, hydraulic conductivity, bulk density and organic
org carbon content for the
different layers of each soil type (Setegn, 2008).
Soil Region Soil Taxonomy Soil Texture Soil Depth
Dep Particle Soil Temperature Soil Recaction Area (in
Size Class Regime Class pH Class Sq Kms)
Greater Lithic Udorthents & Mesic, loamy-skeletal soils; Shallow to Medium Loamy Mesic (8 - 15 Moderate Acidic 156.89
Himalayas Dystric Eutrochrepts fine-loamy soils deep (25-50
(25 cm) Skeletal Degree) MAT (pH 4.5 to 5.5)
Greater Typic Udorthents Rock Outcrops, mesic, loamy- Deep Ccoarse Mesic (8 - 15 - 11.12
Himalayas skeletal soils Loamy Cal. Degree) MAT
Greater Lithic Cryorthents Rock outcrops covered with Shallow (25-50 cm) Sandy Mesic (8 - 15 Slightly Alkaline 7.88
Himalayas glaciers, Sandy-skeletal soils Skeletal Degree) MAT (pH 7.5 to 8.5)
Lesser Typic Udorthents & Thermic, coarse-loamy soils; Medium deep to Coarse Thermic (15 - 22 Moderate Acidic 56.45
Himalayas Dystric Eutrochrepts fine-loamy soils deep (50 - 75 cm) Loamy Degree) MAT (pH 4.5 to 5.5)
Lesser Typic Udorthents & Thermic, loamy soils; fine- Shallow; medium
Shallow Loamy Thermic (15 - 22 Slighty Acidic 32.26
Himalayas Dystric Eutrochrepts loamy soils deep to deep (25-50 Degree) MAT (pH 5.5 to 6.5)
Figure 5: Soil Property Data of the Study Area cm)
 th
12 Esri India User Conference 2011

3.3 DEM
The digital elevation model of 30m by 30m resolutio
resolution for the study area obtained from the ASTER was used.
used The DEM (Fig. 7) gives the
elevation of a particular point at a particularr spatial resolution and was used in the delineation of the
e watershed
waters and analysis of the land
surface characteristics and drainage patterns.

Figure 6: Soil Type of the Study Area Figure 7: ASTER - DEM (30m)

3.4 Climate Data

Climate data used in the SWAT model consists
ists of da
daily rainfall, temperature, wind speed, humidity and evapotranspiration
eva data. These
data have been collected for the period of 2006 to 2010 by Directorate of Land Records, H.P., IMD D Station,
Statio and using LocClim V1.10
software. The climate data was able to provide
vide cont
continuous and complete data ranging from the years ars 2006 to 2010, which was used in
the model simulations.
Table 3: Rainfall Data of the Study Area (Sal River Wa
Watershed)
District 2005 2006 2007 2008 2009 2010
Chamba (mm) 1952.7 1583.7 706.4 857.2 xxx 1100.3
Source: Directorate of Land Records, Revenuee Departm
Department H.P.

Figure 8: Climate data of the Study Area

Table 4: Sensitivity Ranking of Parameters towards
wards Water Flow - Sal Watershed Area
Sensitivity Rank 1 2 3 4 5 6 7 8 9 10
Rain Gauge ESCO CN2 ALPHA-BF GWQMN SOL-Z REVAPMN SOL-AWC CH--K2 BLAI CANMX
REF CN2 GWQMN CN2 SOL-Z ALPHA-BF SOL-AWC REVAPMN CANMX GW-REVAP BLAI
Where: ESCO = Soil Evaporation Compensation Factor, CN2 N2 = Initia
Initial SCS runoff Curve Number for Moisture Condition II, GWQMIN
IN = Threshold
Thre depth of water in the shallow
aquifer for return flow to occur (mm H20), SOL-Z = Depthh to botto
bottom of soil layer (mm), ALPHA-BF = Base Flow alpha factor (days),
ys), REVAPMN
REVA = Threshold depth of water in
the shallow aquifer for ”revap” to occur (mm H2O), SOL-AWC
AWC = A Available water capacity of soil layer (mm / mm), CANMX = Maximum canopy storage (mm H2O), BLAI =
Maximum potential leaf area index for land cover / plant, GW--REVAP = Groundwater “revap” coefficient
 th
12 Esri India User Conference 2011

3.5 River Discharge

The discharge of the Sal and its major snowfed
fed tribu
tributaries dwindles down to lowest during winter months, from December to February
and starts increasing from February end due to the mmelting of snow. The bulk contribution is from rainfall
infall in the
t monsoon months.
Table 5: Hydrological Characteristics of the Study
tudy Are
Area
Characteristics Total Snowfed Rainfed River Average Average Average Sediment
Sedimen Yield River Discharge
Catchment Area Area Annual Annual Annual
Area Rainfall Runoff Sediment Load
Units 26563.06 13 26550.06
550.06 Sal Nala 1100 mm 6960 6.8 mcum 0.144 89.67 - 71.74
Hectare Hectare Hectare mcum Ha. M./Sq.
M./S Km/Year Cum/Sec

4.0 Results and Discussion

4.1 Land Use Scenarios
To explore the sensitivity of SWAT outputss to land use and the effect of LULC changes on the discharge
ischarge of the Sal Rivers, land use
scenarios were explored. Attention was paidid to ens
ensure these were realistic scenarios in accordancee to the ongoing trends of land use
change within the study area. The percent coverage
overage and details of the conversions are presented in the tables
table 6.
Table 6: Percent Areal Coverage of Land Use / Land C
Cover 2010-11
Land Use Scenario / Watershed Landuse
use 2010
2010-11 PD 2010 CD 2010 FA 2010
Settlement 383.82 0 0 0
Agricultural Land Generic 4586.61 0 0 0
Forest Evergreen 999.60 723 56 220.6
Forest Deciduous 12574.29 9786 674 2114.29
Forest Mixed 508.69 413 33 62.69
Range Grasses 7376.43 6345 397 634.43
Water 133.62 0 0 0
Where: PD is Partial Deforestation, CD is Completee Deforest
Deforestation, FA is Conversion of Forest to Agricultural

4.2 Land Use Change Scenarios

As a result of SWAT analysis, the partial deforestat
eforestation and forest to agriculture scenarios resulted in high peak flows and lower base
flows while the complete deforestation scenario
cenario was characterized by high peak flows but hass a base flow that appears almost
equivalent to that of the present day scenario
ario (RFE calibrated model). Details on how these different
rent land use scenarios affected the
different water balance components can be seen in table 7.
Table 7: Percent Changes in the Annual Averages
rages of SSal Watershed Water Balance Components for Land
and Use-Climate
Use Change Scenarios
 th
12 Esri India User Conference 2011

Figure 9: Percent Areal Coverage of LULC 2010-11

11 Figure 10: Percent Changes in Water
ter Balance
Balan Component for
Land Use-Climate Change
nge Scenarios
Scen
 th
12 Esri India User Conference 2011

Legend
Settlement
Agricultural Land
Water
Forest Evergreen to Partial Deforestation
Forest Evergreen to Complete Deforestation
Forest Evergreen to Agricultural
Forest Deciduous to Partial Deforestation
Forest Deciduous to Complete Deforestation
Forest Deciduous to Agricultural
Forest Mixed to Partial Deforestation
Forest Mixed to Complete Deforestation
Forest Mixed to Agricultural
Range Grasses to Partial Deforestation
Range Grasses to Complete Deforestation
Range Grasses to Agricultural
Figure 11: Land Use Change Pattern Analysis Map of Sal River Watershed

4.3 Climate Change Scenarios

According to the Intergovernmental Panel on n Climat
Climate Change (IPCC) 2007, climate change can be definedfined as an identifiable change in the
state of the climate by change in the mean and/or vvariability of its properties and that persists for an
n extended
extend period, typically decades
or longer. Trends from 2006 to 2010 have been o observed in precipitation and have seen a decrease ease in precipitation
p in the Chamba
district, which has been accounted for in thee precip
precipitation reduction scenarios carried out in this study.
udy. Assuming
Ass accurate estimates of
the water balance components, SWAT wass used tto evaluate the impacts of various scenarios off climate climat change on Sal River. The
combined discharge hydrographs for the climate
mate cha
change scenarios shown below help single out the impact a single climate change event
 th
12 Esri India User Conference 2011

however unlikely, would have on the discharge

harge of the Sal river. Table 9 shows the percent changes
nges in the annual averages of Sal
watershed water balance components to precipitatio
ecipitation for these climate change scenarios.
Table 8: Annual Average Water Balance Components
ponents for the Calibrated Sal Watershed Models
Components PRECIP SURQ LATQ GW Q REVAP DA REHG GW-REHG WYLD PERC ET PET TLOSS SEDYLD
Rain Gauge-2006 1583.70 17.89 72.24 422.2
422.26 26.06 26.75 535.19 511.20 535.90 939.57
939.5 1369.82 1.20 0.81
RFE-2010 1100.03 11.53 43.20 482.4
482.47 3.48 25.39 507.93 536.37 510.83 531.36
531.3 1182.04 0.82 0.70
Where: PRECIP (mm) - Average total Precipitation on Sub-basin,
basin, SSURQ (mm) - Surface runoff, LATQ (mm) - Lateral flow contribution
bution to reach, GW Q (mm) - Ground water
discharge into reach or return flow, REVAP (mm) - Waterr in shallow aquifer returning to root zone, DA RCHG (mm) - Amount off water entering
en deep aquifer from root zone,
GW RCHG (mm) - Amount of water entering both aquifers, rs, WYLD ((mm) - Net water yield to reach, PERC (mm) - Amount of water
er percolating
percola out of the root zone, ET (mm) -
Actual evapotranspiration, PET (mm) - Potential evapotranspiratio
anspiration, TLOSS (mm) - Amount of water removed from tributary channels
annels by transmission, and SEDYLD (T/HA) -
Sediment yield

Table 9: Percent Changes in the Annual Averages

rages of SSal Watershed Water Balance Components for Climate
limate Change
C Scenarios

Note: PR–10 = Precipitation reduced by 10%, PR–20 20 = Pre

Precipitation reduced by 20%, TM+5 = Air Temperaturee increase by 5%, PR+10 = Precipitation
increase by 10%, PR+20 = Precipitation increase byy 20%

Figure 12: Water Balance Components for thee Calibr

Calibrated Sal Figure 13: Water Balance Components
nents for
fo Climate Change Scenarios
Watershed Models
Generally, the reduction of precipitation brought
ught ab
about a reduction in available water in the watersheds
heds reducing
red base flows to very low
levels. The increase in temperature also reduces
uces the water availability to some degree by increasing evapotranspiration
evapotra in the watershed
thus reducing amount of water and discharge.arge. Acc
According to Ficklin et al. (2009), temperature is one of the most important factors
governing plant growth and depending on the optim
optimum temperature of the plants; the plant growth th cycle will be shifted also affecting
 th
12 Esri India User Conference 2011

the water balance components. Increases in precip

precipitation by 10 percent and 20 percent increased the discharge
dis and base flow in the
rivers but on the other hand may have negative
tive effe
effects across land such as erosion and in the reach such as increased
i sediment load and
flooding.

5.0 Conclusions
Based on the results obtained in this study, the digi
digital image classification is a suitable methodologyy to visual
visu image interpretation at a
high and produces an accurate map of a highlyghly vari
variable area using far less time and effort than convention
nventional algorithms. The resulting
map obtained from this land cover classification
ation wa
was of a high accuracy (95%) and was suitable for use as an input into SWAT especially
for the simulation of effects of land use change
nge in a sspatially explicit hydrological model.
The model simulations showed that the Sall River fflow would be significantly affected in the face of the climate
c and land use change
scenarios posing difficulties in adaptation to
o the alt
altered flow regimes of the Sal River. The different
nt water
wate balance components were
affected regardless of the type and amountt of chachange that was undergone thus affecting the magnitude
gnitude and timing of the flow. It is
therefore prudent to work towards establishing
shing and maintaining adequate minimum flows that would ould mitigate
mit the effects of reduced
base flows and put in place measures to maintain
aintain adequate sustained river flows to the benefit of the stakeholders
s of the Sal River
Watershed such as proper land and water managem
anagement practices.

Acknowledgement
The author is grateful to Mr. Kapil Chaudhery,
ry, Direct
Director, Spatial Decisions, Ha Noi (Viet Nam) for providing
iding the necessary facilities to carry
out this work. I am also thankful to my Guru Ji Prof. JJ. L. Jain for the motivation of this work.
Reference
1. Anderson, J.R., Hardy, E.E., Roach, J.T., and Witmer, R.E. (1976). A Land Use and Land Cover Classification
Classific System for Use with
Remote Sensor Data. US Geologicall Survey P Professional Paper, 964.
2. Battles, J. J., and Robards, T. (2008). Climate Change Impacts on Forest Growth and Tree Mortality:
rtality: A Data-Driven Modeling
Study in the Mixed-Conifer Forest off the Sie
Sierra Nevada, California. Climatic Change 87. pp. 193-213
213.
3. Ficklin, D.L., Luo, Y. Z. (2009). Climate
te Chang
Change Sensitivity Assessment of a Highly Agriculturall Watershed
Watersh using SWAT. Journal of
Hydrology. Vol. 374(1-2). pp. 16-29.
4. Gommes, R., Grieser, J., and Bernardi,
rdi, M. (2
(2004). FAO Agroclimatic Database and Mapping Tool. Food
Foo and Agricultural
Organization of the UN, European Society
ciety fo
for Agronomy, Newsletter. No. 26.
5. IPCC - Intergovernmental Panel on Climate
ate C
Change (2007). Climate Change Synthesis Report. Cambridge
Cambr Press, Cambridge.
6. Pareta, K. (2004). Hydro-Geomorphology
ology of Sagar District (M.P.): A Study through Remote Sensing
ensing Technique.
T Proceeding in XIX
M. P. Young Scientist Congress, Madhya
dhya Pra
Pradesh Council of Science & Technology (MAPCOST), T), Bhopal.
Bhop
7. Pareta, K. (2011). Geo-Environmental
tal and G
Geo-Hydrological Study of Rajghat Dam, Sagar (Madhyaadhya Pradesh)
P using Remote
Sensing Techniques. International Journal
ournal oof Scientific & Engineering Research. Vol. 2-8. pp. 1-8.
8. Pareta, K. and Koshta, Upasana (2009). Soil Erosion Modeling using Remote Sensing and GIS:: A Case Study of Mohand
Watershed, Haridwar. Madhya Bhartirti Journ
Journal, Dr. Hari Singh Gour University, Sagar (M.P.). Vol. No. 55.
5 pp. 23-33.
9. Peterson, D.H., and Smith, R.E. (2000). An O
Organized Signal in Snowmelt Runoff over the Western
stern United States. Journal of the
American Water Resources Association.. Vol
Vol. 36(2), pp. 421-432.
10. Setegn, S.G., Srinivasan, R., and Dargahi,
rgahi, B. (2008). Hydrological Modelling in the Lake Tana Basin, Ethiopia
Et using SWAT model.
The Open Hydrology Journal. Vol 2. pp. 2424-40.
11. Wentworth, C.K. (1930). A Simplified
ed Metho
Method of Determining the Average Slope of Land Surfaces.
rfaces. American Journal of Science.
Vol. 21. pp. 184-194.
12. Winchell, M., Srinivasan, R., Luzio, M.D., and Arnold, J. (2007). ArcSWAT User Guide - ArcSWAT
AT Interface
Inter for SWAT2005.
Blackland Research Center, Texas Agricultur
gricultural Experiment Station, Texas.
13. Yuan, F.S., Linsley, B.K. (2004). A 1200
00 Year Record of Hydrologic Variability in the Sierra Nevada
vada from Sediments in Walker Lake,
Nevada. Geochemistry Geophysics Geosyste
Geosystems. Vol 5.