See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/281409357

Remote sensing and GIS based assessment of water scarcity - a case study
from Hambantota District, Sri Lanka

Conference Paper · October 2013

DOI: 10.13140/RG.2.1.1045.6083

CITATIONS READS

2 850

5 authors, including:

Indishe Senanayake Welivitiyage Don Dimuth Prasad Welivitiya

University of Newcastle University of Newcastle
39 PUBLICATIONS   110 CITATIONS    17 PUBLICATIONS   80 CITATIONS   

SEE PROFILE SEE PROFILE

Nadeeka Parana Manage D.M.D.O.K. Dissanayake

University of Newcastle University of Moratuwa
20 PUBLICATIONS   88 CITATIONS    11 PUBLICATIONS   42 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Disaggregating satellite soil moisture products View project

Remote Sensing and GIS View project

All content following this page was uploaded by Indishe Senanayake on 28 November 2015.

The user has requested enhancement of the downloaded file.

PDF Compressor Pro
Proceedings of ACRS 2013

REMOTE SENSING AND GIS BASED ASSESSMENT OF WATER SCARCITY ± A

CASE STUDY FROM HAMBANTOTA DISTRICT, SRI LANKA

I.P. Senanayake1*, W.D.D.P. Welivitiya2, P.M. Nadeeka2, U.G.A. Puswewala3 and

D.M.D.O.K. Dissanayake3
1
Arthur C Clarke Institute for Modern Technologies, Katubedda, Moratuwa, Sri Lanka,
indishe@accmt.ac.lk
2
School of Engineering, University of Newcastle, Callaghan, NSW, 2308, Australia,
welivitiyagedondimuth.welivitiya@uon.edu.au, nadeeka.paranamanage@uon.edu.au
3
University of Moratuwa, Moratuwa, Sri Lanka,
dean-eng@uom.lk, dmdok@earth.mrt.ac.lk

*Corresponding author: indishe@accmt.ac.lk

ABSTRACT
The dry zone of Sri Lanka periodically faces drought conditions due to uneven distribution of
annual rainfall. Hambantota district in the dry zone can be identified as a severely drought affected
area, on the basis of surface water availability. With the ongoing massive development projects in the
district, the demand for water will increase significantly in the next few years. Therefore,
implementation of a proper water management system is a pre-requisite in facilitating this requirement.
The drought severity of Hambantota district was analysed meteorologically, hydrologically and
physically using remote sensing and GIS techniques. The combined effect of the above drought
components were spatially identified through a matrix overlay operation.
The results illustrate that 20% of Hambantota district faces severe water stress conditions, while
11% of the district faces moderate water stress conditions. The resultant drought severity map can be
used as a reference in a future water management endeavour to enhance the effectiveness of the project.

Keywords: Annual rainfall, drought severity, GIS, matrix overlay, water scarcity

INTRODUCTION
Sri Lanka is located in the Indian Ocean from Northern latitudes 50 ¶WR0 ¶DQG Eastern
longitudes 790 ¶WR0 ¶7KHLVODQGKDVDWURSLFDOFOLPDWHDQGLWVORFDWLRQHQGRZVWKHFRXQWU\ZLWK
a warm climate, moderated by ocean winds and moisture.
The lowlands of the island experience an average annual temperature of 270 C, with 60 C of
mean daily range. In the central highlands the average annual temperature is 150 C with the mean daily
range of 100 C. The average annual rainfall of the island ranges from 900 mm to 6000 mm. The rainfall
pattern of the island is mainly influenced by the two monsoons, namely South-West monsoon from
May to September and North-East monsoon from December to February (Survey Department of Sri
Lanka, 1988). Since the rainfall is not uniformly distributed over the island, Sri Lanka has been
traditionally divided in to 3 climatic zones on the basis of the average annual rainfall, viz; i. Wet zone >
2500 mm of average annual rainfall, ii. Intermediate zone: 1750-2500 mm of average annual rainfall
and iii. Dry zone < 1750 mm of average annual rainfall (Zubair, 2002).
The dry zone covers nearly 65% of the total land area of the country; consequently 65% of the
island receives less than 1750 mm of average annual rainfall. The wet zone periodically faces flood
conditions and flood management is one of the major problems in the wet zone. On the contrary, the
dry zone periodically faces drought conditions.

Water supply and scarcity in Sri Lanka

,W KDV EHHQ SUHGLFWHG WKDW WZR WKLUGV RI WKH ZRUOG¶V population will face moderate to severe
water scarcity conditions by 2025 based on current trends (Kuylenstierna et al., 1997). According to the
studies carried out by the United Nations; countries that withdraw their available water resources at a
rate of 40% or more will face severe water scarcity conditions by 2025, and therefore prompt measures
have to be implemented for water management. Since the withdrawal rate of available water resources

SC03 - 820
PDF Compressor Pro
Proceedings of ACRS 2013

in South Asia is 48% (Ariyabandu, 1999); Sri Lanka has an increasing threat of water scarcity in the
near future. Therefore, a comprehensive analysis of droughts in Sri Lanka will play an important role in
the future water management endeavours designed to face these predicted water stress conditions.

Water stress conditions in Hambantota District

The long term human interference with the nature as well as climatic changes has made a
significant effect on the seasonal rainfall patterns. Due to these variations, districts in the dry zone such
as Monaragala, Puttalam, Kurunegala, Ampara, Badulla and Hambantota had experienced frequent
drought conditions in the recent past and the situation seems to become worse with passage of time. On
the basis of the surface water availability, Hambantota appears to be a heavily affected area.
Hambantota district is located in the Southern part of the island from Northern latitudes 6 0 to
6.5 and Eastern longitudes 80.60 to 81.70, occupying an area of 2622.82 km2. Just above 1% of the total
0

area of the district, (29.8 km2) is covered by major inland water bodies, while the rest is mainly land.
The average annual rainfall of the district is 650 mm (Ariyananda, 2007; Senanayake et al., 2012).
Although, Kirindi Oya and Malala Oya are the only two river basins of Hambantota; they are
considered as dry river basins due to demand overriding supply (Ariyananda and Ratnayaka, 2006).
Since Hambantota is a rain-fed agricultural area, a small increment in harvesting rainwater can
considerably increase the crop yields and also lower the risk of crop failures in dry areas. Therefore,
identification of drought risk areas is a prerequisite in designing an appropriate water management
system.

METHODOLOGY
Analysis theory
This research methodology was based on the concept that drought severity is a function of
rainfall, hydrology and physical characteristics of land use (Mongkolsawat et al., 2001;
Suwanwerakamtorn et al., 2005, 2006).

Analysis of meteorological drought

Monthly rainfall data from 17 rainfall stations distributed over Hambantota district during the
fifteen years from 1992 to 2006 was obtained from the Meteorological Department of Sri Lanka. The
average monthly rainfall of each rainfall station was calculated to analyse the rainfall distribution
throughout a year and illustrated in Figure 1.

Figure 1: Distribution of average monthly rainfall of the 17 rainfall stations in Hambantota

District, Sri Lanka during the period 1992 to 2006.

SC03 - 821
PDF Compressor Pro
Proceedings of ACRS 2013

Figure 1 clearly shows that the rainfall in Hambantota district is mainly concentrated in two
seasons caused by depressions from Southwest Bay of Bengal and Southeast Arabian Sea during
October to November and convective precipitations during March to May. Further, it shows that
Hambantota district does not receive a considerable amount of rainfall during the two major monsoonal
seasons of the country. In addition, Figure 1 illustrates that the rainfall has dropped down significantly
in the months, January, February, June, July and August. Further analysis of detailed monthly rainfall
data shows that many rainfall stations have periodically faced long periods without any precipitation or
very small precipitation.
The statistical data of the average annual rainfall of the 17 rainfall stations were calculated and
illustrated in Figure 2. Figure 2 shows that, all the rainfall stations in Hambantota, except Kirama
which is located in the Western most part of the district; have received less than 1500 mm of average
annual rainfall over the past fifteen years from 1992 to 2006. Throughout many years the average
annual rainfall of the area has fallen down to less than 1000 mm and occasionally it has fallen down to
less than 500 mm.

Figure 2: Distribution of average annual rainfall of the 17 rainfall stations in Hambantota

District, Sri Lanka during the period 1992 to 2006.

Annual rainfall data of the 15 years from 1992-2006 were compiled as a GIS point feature class
and the annual average rainfall of each station was calculated using its attribute table. Inverse Distance
Weighted (IDW) interpolation was used to generate the average annual rainfall map of Hambantota
District on a GIS platform. The decile rainfall range has been determined to classify the average annual
rainfall map. Thereafter, 4 levels of drought severity have been determined by using 10 decile rainfall
classes in order to assess meteorological drought (Table 1). Subsequently the meteorological drought
risk map was generated using the defined 4 drought severity classes. This process is illustrated in
Figure 3.

Table 1: Drought severity parameters based on rainfall in Hambantota District, Sri Lanka
Average annual rainfall Drought severity value
class (Decile range)
<989 mm 4
989-1035 mm 3
1035-1126 mm 2
1126-1983 mm 1

SC03 - 822
PDF Compressor Pro
Proceedings of ACRS 2013

Figure 3: Analysis of meteorological drought in Hambantota District, Sri Lanka (modified from
Mongkolsawat et al., 2001)

Analysis of hydrological drought

Hydrological drought was analysed using 3 variables namely, drainage density, distance from
water sources and irrigated area. Catchment areas of the Hambantota district were delineated by using
Advanced Spaceborne Thermal Emission and Reflection Radiometer Global Digital Elevation Model
(ASTER GDEM) with the resolution of 30 m, a data product of METI and NASA, in order to analyse
the drought risk based on drainage density. Terrain preprocessing functions in ArcHydro, were used to
delineate catchment areas of Hambantota District. Subsequently, drainage density (i.e. proportion of
drainage line length to catchment area) of each catchment was calculated. The drainage density layer
was classified into 4 classes of drought severity using quantile classification and converted to a
polygonal feature class (Table 2).

Table 2: Drought severity parameters based on stream density in Hambantota District, Sri
Lanka
Stream density class Drought severity value
0.00082 - 0.15091 km/km2 4
0.15091 - 0.22510 km/km2 3
0.22510 - 0.37103 km/km2 2
0.37103 - 1.70996 km/km2 1

In order to analyse the drought risk based on distance from water source, water sources in
Hambantota district were classified into 3 classes in relation to their surface area and subsequently
multiple buffers were created around water sources to define drought severity as shown in Table 3.
Irrigated areas were then assigned with the lowest drought risk class value (i.e. 1). Thereafter, irrigated
area layer was union overlaid on a GIS platform with the layer representing drought risk based on
distance from surface water sources.
Hydrological drought risk map was generated by a matrix overlay between the resultant layer
with the stream density layer as illustrated in Figure 4.

Analysis of physical drought

Analysis of physical drought was done on the basis of slope and land use. Slope map of
Hambantota District was generated by using ASTER GDEM on a GIS platform. Thereafter, the slope
map was classified into 4 slope classes and the drought severity values were assigned as shown in
Table 4. Afterwards, the land use classes were assigned with drought severity values as shown in Table
5 by using the land use map of Hambantota (1: 50,000 scale) prepared by the Survey Department of Sri
Lanka.

SC03 - 823
PDF Compressor Pro
Proceedings of ACRS 2013

Table 3: Drought severity parameters based on water source in Hambantota District, Sri Lanka
(Modified from Mongkolsawat et al., 2001)
Surface area of the water body Buffer distance as area beyond water Drought severity
source
0 - 0.5 km2 0 ± 0.25 km 1
0.25 ± 0.5 km 2
0.5 ± 0.75 km 3
> 0.75 km 4
0.5 ± 5 km2 0 - 0.5 km 1
0.5 - 1 km 2
1 ± 1.5 km 3
> 1.5 km 4
> 5 km2 0 ± 0.75 km 1
0.75 ± 1.5 km 2
1.5 ± 2.25 km 3
> 2.25 km 4

Figure 4: Analysis of hydrological drought in Hambantota District, Sri Lanka (modified from
Mongkolsawat et al., 2001)

SC03 - 824
PDF Compressor Pro
Proceedings of ACRS 2013

Table 4: Drought severity parameters based on slope in Hambantota District, Sri Lanka
(Suwanwerakamtorn et al., 2005)
Slope class Drought severity
> 30% 4
17 ± 30 % 3
2 ± 17 % 2
0±2% 1

Table 5: Drought severity parameters based on land use classes in Hambantota District, Sri
Lanka (modified from Mongkolsawat et al., 2001)
Land use class Drought severity value
Sand-beach 4
Home garden 4
Chena 3.5
Other plantations 3.5
Monsoon forest 3.5
Scrub 3
Grassland 3
Coconut 2
Paddy 1.5
Water holes 1
Tanks (working) 1

Subsequently, the physical drought risk map was prepared by using a matrix overlay between
the drought severity layers based on land use and slope as illustrated in Figure 5.

Figure 5: Analysis of physical drought in Hambantota District, Sri Lanka (modified from
Mongkolsawat et al., 2001; Suwanwerakamtorn et al., 2005)

Preparation of drought risk area map

Finally, the drought risk area map of Hambantota district was generated by using matrix overlay
operations between, meteorological, hydrological and physical drought risk map layers as illustrated in
Figure 6.

SC03 - 825
PDF Compressor Pro
Proceedings of ACRS 2013

Figure 6: Analysis of drought risk area in Hambantota District, Sri Lanka (modified from
Mongkolsawat et al., 2001)

RESULTS AND DISCUSSION

The methodology utilized in this study facilitates drought risk identification meteorologically,
physically and hydrologically. Consequently, the major causes affected for severe drought risk areas
can be identified through the results of the research. Table 6 explains the contribution of each factor
relating to the drought conditions of Hambantota district.

Table 6: Percentage areas of meteorological, hydrological and physical drought severity in

Hambantota District, Sri Lanka
Drought type Drought severity (percentage area)
Very Mild Mild Moderate Severe
Meteorological drought 27% 25% 19% 29%
Hydrological drought 33% 29% 23% 15%
Physical drought 6% 54% 38% 2%
Combined drought risk area 43% 26% 11% 20%

Percentage severe drought risk areas were calculated to be 29%, 15% and 2% in meteorological,
hydrological and physical drought respectively. Hence, meteorological drought can be identified as the
major cause for the drought in Hambantota district with 29% of severe drought risk area. The resultant
drought risk map, which is an integration of meteorological, hydrological and physical droughts
explains that, there is a severe drought risk in the Eastern part of the district covered by Yala monsoon
forest and in South-East area of the district around Gonnoruwa Grama Niladhari Division. When
comparing the final drought risk area map with the meteorological drought risk map of Hambantota
district (Figure 6), it can be clearly seen that low annual rainfall is the major contributor for the severe
drought risk in the area. On the contrary, the Western part of the district where a considerable annual
rainfall is received displays low drought risk.
Accuracy of this methodology can be improved by integrating drainage condition, geology,
lithology, groundwater yield and total dissolved solids into the drought severity calculation model.
Further improvements can be made to the drought risk parametric values and to the values defined for

SC03 - 826
PDF Compressor Pro
Proceedings of ACRS 2013

matrix overlay functions. Remote sensing and GIS techniques provide cost and time effective platform
to make updates and to improve the accuracy.
Since Hambantota is an agricultural area, proper water management system can increase the
crop yield considerably, consequently improving the economy and the livelihoods of the inhabitants.
Currently, irrigation and water supply of the district is mainly provided by the rainwater harvesting
reservoirs and projects such as Udawalawa water supply scheme. However, many abandoned tanks
were distributed over the district due to improper planning (Senanayake et al., 2012). Results of this
study provide insight to the decision makers about the spatial distribution of the water stress conditions
in the district; hence effectiveness of the future water supply projects can be improved by precise
planning and optimization.

CONCLUSION
The drought severity in Hambantota district of Sri Lanka was analysed on the basis of
meteorological, hydrological and physical parameters of drought using remote sensing and GIS
techniques in this study. The resultant drought risk area map of the district was developed by
integrating meteorological, hydrological and physical drought maps using matrix overlay functions.
The results display that Eastern and the South-Eastern area of the district face severe drought
conditions. The availability and the amount of annual rainfall were identified as the major cause for the
drought in the district. The findings of this study can be used as a reference for improving the
effectiveness of the future water management projects, since the results provide an insight for proper
planning which leads towards optimization of water supply.

REFERENCES
Ariyananda, T.and Ratnayaka, N. A., 2006, Improving Water Security in the Southern Province
Affected by Tsunami Through Domestic RWH Systems, International Workshop on RWH,
Kandy, Sri Lanka.
Ariyananda, T., 2007, Improving Water Security in Tsunami Effected Areas in Sri Lanka through
Domestic Rainwater Harvesting, Proceedings of 13th IRCSA conference on Rainwater for Urban
Design, Sydney, Australia.
Ariyabandu, Rajindra de Silva, 1999, Problems and Prospects of Rainwater Catchment for the 21st
Century in Sri Lanka, 9th International Rainwater Catchment Systems Conference, Petrolina,
Brazil, 6-9.
ESRI, 2004, Arc Hydro Tools Overview: Verson 1. Beta 6. ESRI, CA, USA.
Kuylenstierna, J.L., Bjorklund, G., and Najlis, P., 1997. Natural Resources Forum, 21 (3), 181-190.
Mongkolsawat, C., Thirangoon, P., Suwanwerakamtorn, R., Karladee, N., Paiboonsak, S.,
Champathet, P., 2001, An Evaluation of Drought Risk Area in Northeast Thailand using
Remotely Sensed Data and GIS. Asian Journal of Geoinformatics, 1 (4), 33- 44.
Senanayake, I.P., Dissanayake, D.M.D.O.K., Puswewala, U.G.A., 2012. Analysis of the abundance of
abandoned tanks in Hambantota District, Sri Lanka using GIS techniques, The Egyptian Journal
of Remote Sensing and Space Science, 15 (2) 143-150. DOI: 10.1016/j.ejrs.2012.07.001
Survey Department of Sri Lanka, 1988, The National Atlas of Sri Lanka. Survey Department of Sri
Lanka, 22-36.
Suwanwerakamtorn R., Mongkolsawat C., Srisuk K. and Ratanasermpong S., 2005, Drought
Assessment Using GIS Technology in the Nam Choen Watershed, NE Thailand, Proceedings of
the 26th Asian Conference on Remote Sensing, Hanoi, Vietnam.
Suwanwerakamtorn R., Mongkolsawat C., Srisuk K. and Ratanasermpong S., 2006, Matrix Overlay
for Drought Assessment in the Nam Choen Watershed, NE Thailand, Proceedings of the 27th
Asian Conference on Remote Sensing, Ulaanbaatar, Mongolia.
Zubair, L., 2002, El Nino-Southern oscillation influences on rice production in Sri Lanka,
International Journal of Climatology, (22) 249-260.

SC03 - 827

View publication stats