Do river deltas in east India retreat? A case of the Krishna Delta
Nilantha Gamage , Vladimir Smakhtin
International Water Management Institute, PO Box 2075, Colombo, Sri Lanka
a b s t r a c ta r t i c l e i n f o
Article history:
Received 7 July 2007
Received in revised form 27 July 2008
Accepted 31 July 2008
Available online 14 August 2008
Keywords:
Coastal erosion
Environmental impact
Reservoir
Sediment load
Krishna
Godavari
The construction of multiple dams and barrages in many Indian River basins over the last few decades
significantly reduced river flow to the sea and affected the sediment regime. More reservoir construction is
planned through the proposed National River Linking Project (NRLP), which will transfer massive amounts of
water from the North to the South of India. The impacts of these developments on fertile and ecologically
sensitive deltaic environments are poorly understood and quantified at present. In this paper an attempt is
made to identify, locate and quantify coastal erosion and deposition processes in one of the major river basins
in India--the Krishna--using a time series of Landsat images for 1977, 1990 and 2001 with a spatial resolution
ranging from 57.0 m to 28.5 m. The dynamics of these processes are analyzed together with the time series of
river flow, sediment discharge and sediment storage in the basin. Comparisons are made with similar
processes identified and quantified earlier in the delta of a neighboring similarly large river basin--the
Godavari. The results suggest that coastal erosion in the Krishna Delta progressed over the last 25 years at the
average rate of 77.6 ha yr-1
, dominating the entire delta coastline and exceeding the deposition rate
threefold. The retreat of the Krishna Delta may be explained primarily by the reduced river inflow to the delta
(which is three times less at present than 50 years ago) and the associated reduction of sediment load. Both
are invariably related to upstream reservoir storage development.
 2008 Elsevier B.V. All rights reserved.
1. Introduction
As in much of the world, the Indian water planning and management
approach was to harness river waters through dams and other
structures to the extent that was technically feasible. Most Indian
rivers, at present, are highly regulated (Agrawal and Chak, 1991).
Hundreds of multi-purpose reservoirs for water supply, irrigation,
hydropower and fisheries have been constructed, as well as numerous
barrages for water diversion. Nevertheless, river basins in the
southern and western states of India are experiencing physical or
economic water scarcity. Basins in the east of the country are, in
contrast, often perceived as having surplus water and encounter
recurrent floods. The National River Linking Project (NRLP) has been
proposed as a solution to water related problems in India. The NRLP
envisages transferring waters of the Ganga and Brahmaputra to water
deficient basins in the south and west, but remains a contentious issue
in Indian society (Jain et al., 2005). Many argue that assessment of
water surplus/deficits in Indian River basins, conducted as part of the
NRLP proposal, has ignored environmental impacts.
Inter-basin water transfers are normally associated with the
construction of new storage reservoirs. A lot has been said and written
about land submergence and population resettlement (upstream),
and impacts of changing flow pattern on fish (downstream) due to the
construction of reservoirs (Revenga et al., 2000; Rosenberg et al.,
2000; Bouwer et. al., 2006; Biggs et. al., 2007). At the same time, instream
storages anywhere in a basin have impacts on sediment
transport and, consequently, on river estuaries and deltas (Chakrapani,
2005; Syvitski et al., 2005; Fan et al., 2006). These issues tend to
be ignored in water resources planning worldwide. At the same time,
depending on the river and the magnitude of upstream reservoir
construction, such impacts may be significant. Although they may not
be immediate and are difficult to quantify unambiguously, they may
be eventually detrimental to the economy and ecology (Stanley and
Warne, 1998; Rosenberg et al., 2000; Vorosmarty et al., 2004; Nilsson
et al., 2005).
This study attempts to identify the recent changes in the delta of
the Krishna River Basin in India and examine their possible causes in
the context of continuing new storage development and planned
inter-basin water transfers. The Krishna River (Fig. 1) is one of the
major basins in Peninsular India with a catchment area of almost
259,000 km2
and a long-term natural mean annual flow of 78 billion
cubic meters (BCM). Its water resources are already significantly developed
(Smakhtin and Anputhas, 2006; Biggs et. al., 2007), but water
supply shortages continue to manifest themselves. The basin is located
en route of the many planned NRLP water transfers.
River basins located to the north of the Krishna (such as the
Godavari, Fig. 1) are often perceived as more water abundant and less
developed at present. Malini and Rao (2004) examined the recent
changes in the Godavari River Delta, called the "rice bowl" of Andhra
Pradesh State, using remote sensing images. They discovered that the
Geomorphology 103 (2009) 533­540
 Corresponding author. Tel.: +94 11 2880000; fax: +94 11 2786854.
E-mail address: n.gamage@cgiar.org (N. Gamage).
0169-555X/$ ­ see front matter  2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.geomorph.2008.07.022
Contents lists available at ScienceDirect
Geomorphology
journal homepage: www.elsevier.com/locate/geomorph
Godavari Delta has regressed landward with the total net land loss
of 1836 ha over the period of 1976­2000 (at a rate of 73.4 ha yr-1
). It
was suggested that reduced inflow of sediments, associated with
upstream reservoir construction in the Godavari Basin is the main
cause of reduced vertical accretion at the delta coast and of the
shoreline retreat.
A number of reservoirs and barrages in the Krishna Basin, both
alreadyconstructed and planned, detrimentallyaffect the sediment flow
regime in the basin and sediment delivery to the delta (Subramanian
et. al., 1985; Ramesh et. al., 1990). It is therefore vital to examine the
impacts of water resources development on the Krishna Delta, although
the issue has not been addressed in detail.
2. Data and methods
The study first illustrates the dynamics of river flow, sediment
loads and reservoir storage growth in both the Krishna and Godavari
basins. The comparative analysis of the two basins allows quantitative
understanding of water and sediment dynamics. The major problem
which adversely affects quantitative assessments in water resources is
the lack of observed data and access to existing data--which is often
the case in India. Some internet resources provide very limited (past)
observed flow data for a few Indian rivers, including the Krishna
(http://webworld.unesco.org/water/ihp/db/shiklomanov/index.
shtml). The monthly flow time series for the terminal stations on both
rivers (Polavaram on the Godavari and Vijayawada on the Krishna,
Fig. 1) for the available periods of record from 1930 to the late 1970s
have been downloaded from this source. More recent observed flow
data for Vijayawada have been provided by the Central Water Commission
(CWC) of India. Long-term observations on sediment loads for
the Krishna at Vijayawada are unavailable; the only available data are
for the period from 1991 to 2000 (CWC, 2006). The available sediment
load data for the Godavari at Polavaram (Fig. 1) covers the period from
1970 to 1997 and have been taken from the published study of Malini
and Rao (2004). The information on existing and planned reservoirs
and their storage is available from the ICOLD register (http://www.
icold-cigb.org) and from published feasibility reports (http://nwda.
gov.in/indexab.asp?langid=1).
To examine the morphological changes in the Krishna Delta itself,
several remote sensing images of the delta area were analyzed. The
images were free products obtained from the Earth Science Data
Interface (ESDI) at the Global Land Cover Facility (GLFC; http://glcfapp.
umiacs.umd.edu:8080/esdi/index.jsp) and were selected from the
period of 1977 to 2000 to form a time series. They include:
* Landsat 2 Multispectral Scanner (MSS) image dated 1 June 1977 with
a spatial resolution of 57 m (Note that GLFC resampled the original
MSS image with a slightly lower resolution),
* Landsat 5 Thematic Mapper (TM) image dated 10 November 1990
with a spatial resolution of 28.5 m, and
* Landsat 7 Enhanced Thematic Mapper (ETM+) image dated 28 October
2000 with a spatial resolution of 28.5 m.
Three basic layers were used to detect morphological changes in
the delta: band 4 (NIR), band 2 (Green) and band 1 (Blue). These bands
have characteristics that are suitable for coastal mapping, differentiation
of vegetation from soil, reflectivity of vegetation vigor, and
delineation of water bodies. The images have been enhanced for
better brightness and clarity using ERDAS 9.0 software. The
Fig. 1. Map of the Krishna and Godavari River basins in India.
534 N. Gamage, V. Smakhtin / Geomorphology 103 (2009) 533­540
methodology employed in the study was similar to that applied by
other coastal zone researchers in different parts of the world (e.g. Azab
and Noor, 2003; Malini and Rao, 2004; Turner et al., 2006; Huh et al.,
2007; Kumar et al., 2007). First, land area was demarcated in all
images using same projection system (UTM), unique band combination
with the same magnification level to avoid the imbalance of
spatial resolution from 1977 to 2000. This consideration was to avoid
over or underestimate of the land area due to changing level of the
information with varying spatial resolution from 1977 to 2000. Then
"oldest" image was assumed to be the reference condition against
which changes in the other two images were detected. The entire
Krishna Delta shoreline of 134 km was examined to demarcate the
zones of coastal erosion and accretion (deposition) between 1977 and
2000. The enhanced images were subsequently analyzed using GIS
tools. The delta coastline from each image was presented in a vector
form. These lines were subsequently superimposed to identify zones
of erosion and deposition and calculate the areas of these zones.
3. Results and discussion
3.1. Water and sediment inflows to the Godavari and Krishna deltas
Fig. 2 illustrates the dynamics of flow and sediment load at the
outlet of the Godavari (at Polavaram) and reservoir storage growth in
the entire Godavari Basin since 1970. The flow time series has missing
data during 1980­1990 and neither flow nor sediment data were
available after 1998. Cumulative dam storage increased significantly in
the early 1970s but remained relatively constant until 2006. However,
it will increase abruptly again after the construction of the Polavaram
Barrage and the large Inchampalli Dam. To illustrate this abrupt growth
in Fig. 2, the year of completion for both dams is assumed to be 2010.
While trends in the flow of the Godavari River cannot be ascertained
from the available disrupted flow time series, the decreasing
trend in annual sediment loads are clear from the sediment data
(Fig. 3, also shown by Malini and Rao, 2004). The mean annual sediment
load has decreased from 100 million tons in 1978 (effectively an
ending point in noticeable reservoir growth in the basin so far) to
46 million tons at the end of the 1990s. The current cumulative reservoir
storage in the Godavari Basin remains relatively low (6.3 BCM,
i.e. approximately 6% of the mean annual flow at the outlet--
Amarasinghe et al., 2005; Smakhtin and Anputhas, 2006). The storage
growth is not the only aspect of significance because much of the
water is also diverted from barrages, i.e., structures without storage. A
relatively small storage in the basin (Fig. 2) and still a noticeable
decreasing trend in sediment load (Fig. 3) imply that the basin
sediment regime is very sensitive to reservoir installation. Further
reduction in sediment inflow may therefore be expected after the
construction of the Polavaram and Inchampalli storages, which will
increase the storage/flow ratio in the basin to 19% (Amarasinghe et al.,
2005; Smakhtin and Anputhas, 2006).
A subsequent attempt has been made to examine whether similar
trends exist in the Krishna Basin, where the proportion of storage to
annual flow is much larger than in the Godavari. The comparison of
the two short (1991­2000) time series of sediment loads at Agraharam
(upstream of major reservoirs) and at Vijayawada (downstream
of them) revealed a significant decrease in sediment downstream of
Fig. 2. Time series of annual flows and sediment loads at Polavaram, at the outlet of the Godavari basin, with basin-wide cumulative storage.
Fig. 3. Time series of sediment load at Polavaram with a decreasing trend line (source:
Malini and Rao, 2004).
Fig. 4. The time series of sediment loads at Agraharam and Vijayavada in the Krishna
River basin.
535N. Gamage, V. Smakhtin / Geomorphology 103 (2009) 533­540
Fig. 5. Time series of annual flows and sediment loads at Vijayavada, at the outlet of the Krishna basin, with basin-wide cumulative storage.
Fig. 6. An image of the Krishna River delta showing several areas (numbered) where detailed analyses of erosion and deposition have been made.
536 N. Gamage, V. Smakhtin / Geomorphology 103 (2009) 533­540
the reservoir system (Fig. 4). The differences are particularly noticeable
in high-flow years (1994 and 1999), when more sediment reaches
Agraharam from the relatively unregulated upstream basin but all
sediments are likely being trapped by the existing reservoir system
(Srisailam and Nagarjuna-Sagar) upstream of Vijayawada. The absence
of sediment data prior to 1991 does not allow further consideration on
changes in the sediment regime. However, these changes are most
likely very significant due to marked reduction of river flow at Vijayawada
over the last 70 years (Fig. 5). This reduction is due to various
water diversions, groundwater development (Biggs et al., 2007) and
increased cumulative small farm dams' and large reservoir storage in
the basin, which has grown from almost zero in 1960 to 28.5 BCM at
present (Fig. 5). This present cumulative storage, made up primarily of
large dams, represents 36% and 132% of the natural and present-day
mean annual flow of the Krishna River, respectively (Amarasinghe et
al., 2005; Smakhtin and Anputhas, 2006; Biggs et al., 2007).
3.2. Erosion and deposition in the Krishna Delta
Fig. 6 shows the image of the Krishna Delta with several selected
areas where erosion and deposition were examined in detail. Figs. 7 and
8 display the sequence of images for years 1977,1990 and 2000 for some
of the selected areas circled in Fig. 6. The black lines in each image
represent the reference position of the land mass--in 1977. Fig. 9 shows
areas of predominant erosion and deposition during the period between
1977 and 2000 for the entire deltashoreline, and Table 1 summarizes the
calculated characteristics of these processes for the entire delta over the
same period.
The results suggest that while areas of predominant erosion and
deposition interchange, the overall tendency is towards the regression
landward with losses of land to the sea--the situation similar to that in
the Godavari Delta. The annual net loss rate of 77.4 ha is almost the same
as that in the Godavari Delta (73.4 ha yr-1
; Malini and Rao, 2004). One
noticeable feature of the Krishna Delta is also its higher ratio of erosion
to deposition (3.05 versus 1.6 in the Godavari) over the same period,
suggesting that coastal erosion is more distinct in the Krishna Delta than
in the Godavari, despite the slightly smaller area (4700 km2
versus
5100 km2
) and shorter shoreline of the former (134 versus 160 km).
Erosion is also a dominant process through most of the coastline, while
the deposition is only limited to certain sections (Fig. 9).
3.3. Possible causes and implications of coastal erosion
The regression of both deltas cannot be explained by the rise in the
sea level. According to Malini and Rao (2004), analysis of the available
sea level data in the region, for the periods overlapping with that of the
present study, did not reveal any significant rising or falling trends.
Therefore, coastal erosion in the Krishna and Godavari deltas can most
likely be explained by the reduced sediment supply, which, in turn, is
due to upstream flow regulation. In addition, human activities in delta
regions (e.g. conversion of cropland and mangrove swamp areas into
aquaculture ponds) may also be responsible for sea transgression
leading to coastal erosion and shoreline retreat in both deltas (e.g.
Sarma et al., 2001).
Analysis of the longer series of sediment load data for the downstream
parts of the Krishna and the use of more recent and more
Fig. 7. Changing morphology of area #2 (Fig. 6) in 1977 (left two images), 1990 (middle two images) and 2000 (right two images). The top and bottom rows of images show the
dynamics of the right and left banks of the river channel, respectively.
537N. Gamage, V. Smakhtin / Geomorphology 103 (2009) 533­540
resolute remote sensing images could result in more detailed quantification
of delta erosion. However, even with the existing limited
data, it is possible to suggest that upstream basin storage development
leads to the said retreat of deltas. The Krishna River is already a "closed
basin" because only occasional high flows with almost no sediment
spill into the delta (Fig. 4). Therefore, the storage that is already
constructed in the Krishna will have a long-lasting detrimental effect
on the delta and its agricultural productivity. The situation in the
Godavari Delta will most likely deteriorate after the construction of
the additional storages (Polavaram and Inchampalli, Fig. 2) planned as
part of the NRLP.
Detailed sedimentation modeling studies would be useful in all
major deltas of India in order to develop a better understanding and
quantification of the links among upstream runoff reduction, sediment
discharge reduction, upstream storage growth and man-induced
changes in deltas. Such studies could allow the specification of
necessary environmental flow releases to be made for the maintenance
of delta sediment regimes.
Coastal erosion may be seen as a slow process. However, there are a
few aspects which promote negative environmental impacts associated
with it. One is the saltwater intrusion. Bobba (2002) conducted
a numerical modeling study of the Godavari Delta and inferred that
saline intrusion may become a major factor of reduced agricultural
productivity in the delta due to increased groundwater pumping and
reduced freshwater inflow. Coastal erosion, caused by similar factors,
facilitates saltwater intrusion deeper in the delta adversely affecting
the productivity of land. An additional problem is the potential rise in
sea level in the next 50 years due to climate change, although the
limited available observations so far do not detect it. This rise can lead
to even more coastal erosion and deeper saltwater penetration,
accelerating degradation of the deltas. Research on this aspect was not
the scope of the current study, and needs to be carried out as a
separate detailed project. While quantification of the above impacts
will be developing, even limited environmental flow releases from
existing reservoirs in the Krishna and the Godavari will delay the
adverse environmental processes in both deltas. New storage
reservoirs need to be planned to allow sediments to reach deltas.
Construction of the most downstream reservoirs, particularly as large
as Inchampalli, will definitely not serve this purpose.
4. Conclusions
Reservoir construction in river basins have multiple ecological
impacts such as the significant decrease in sediment supply to downstream
deltas, which are often referred to as the "rice bowls" of India
because of their high land productivity. It has been demonstrated that
the Krishna Delta has been in retreat for the last 25 years, with areas of
erosion dominating over deposition. This retreat is due mainly to
reduced water and sediment inflow to the delta, related to the development
of upstream reservoir storage development and barrages.
Fig. 8. Changing morphology of area #4 (Fig. 6) in 1977 (left two images), 1990 (middle two images) and 2000 (right two images). The top and bottom rows of images show the
dynamics of the southern and northern parts of the area, respectively.
538 N. Gamage, V. Smakhtin / Geomorphology 103 (2009) 533­540
These conclusions, however, have been reached from limited observation
data, which are not easily and freely available in India.
Coastal erosion detected in the Krishna Delta may be seen as a slow
process. However, it gives some negative environmental impacts such
as saltwater intrusion, and the potential rise in sea level caused by
future climate change may significantly speed up the erosion. Detailed
sedimentation modeling studies need to be conducted in major deltas
of India in order to develop a better quantitative understanding of the
relationships among upstream water and sediment flow reduction,
upstream storage growth and changes in deltas. Such studies could
allow the specification of necessary environmental flow releases to be
made for the maintenance of deltas and could be a valuable contribution
to quantitative justification of environmental water allocations
in coastal regions.
Acknowledgements
The study forms part of the research project on the assessment of
the National River Linking Project in India. The authors gratefully
acknowledge the support from the CGIAR Challenge Program on Water
and Food (http://www.waterandfood.org), who funded this project.
References
Agrawal, A., Chak, A., 1991. Floods, flood plains and environmental myths. State of
India's Environment -- A Citizen's Report 3. Centre for Science and Environment,
New Delhi. 167 pp.
Amarasinghe, U., Sharma, B.R., Aloysius, N., Scott, C., Smakhtin, V.U., De Fraiture, C.,
Sinha, A.K., Shukla, A.K., 2005. Spatial variation in water supply and demand across
the river basins of India. IWMI Research Report N 83. International Water Management
Institute, Colombo. 37 pp.
Azab, M.A., Noor, A.M., 2003. Change detection of the North Sinai Coast by using Remote
Sensing and Geographic Information System. Proc. of the Int. Conference
"Environment 2003". Egyptian Geological Survey and Mining Authority (EGSMA),
Cairo, pp. 1­10.
Biggs, T.W., Gaur, A., Scott, C.A., Thenkabail, P., Rao, P.G., Gumma, M.K., Acharya, S.,
Turral, H., 2007. Closing of the Krishna Basin: irrigation, streamflow depletion and
macroscale hydrology. IWMI Research Report 111. International Water Management
Institute, Colombo. 38 pp.
Bobba, A.G., 2002.Numerical modelingof salt-water intrusion due to human activities and
sea-level change in Godavari Delta, India. Hydrological Science Journal 47, S67­S80.
Fig. 9. Map of the Krishna Delta showing areas of erosion and deposition during the period of 1977­2000.
Table 1
Areal extent of erosion and deposition in the Krishna Delta over 23 years (1977­2000)
Point number in Fig. 6 Erosion (ha) Deposition (ha) Net loss (ha) Rate of loss/gain
(ha yr- 1
)
1 598 483 115 5.0
2 478 178 299 13.0
3 275 31 243 10.6
4 326 74 251 10.9
5 79 98 -19 -0.8
6 894 3 890 38.7
Total (23 years) 2 650 867 1 779 77.4
539N. Gamage, V. Smakhtin / Geomorphology 103 (2009) 533­540
Bouwer, L.M., Aerts, J.C.J.H., Droogers, P., Dolman, A.J., 2006. Detecting the long-term
impacts from climate variability and increasing water consumption on runoff
in the Krishna river basin (India). Hydrology and Earth System Science 10,
703­713.
Chakrapani, G.J., 2005. Factors controlling variations in river sediment loads. Current
Science 88, 569­575.
CWC (Central Water Commission), 2006. Integrated Hydrological Data Book (Nonclassified
River Basins). Central Water Commission, New Delhi, 383 pp.
Fan, H., Huang, H., Zeng, T.Q., Wang, K., 2006. River mouth bar formation, riverbed
aggradation and channel migration in the modern Huanghe (Yellow) River delta,
China. Geomorphology 74, 124­136.
Huh, O.K., Coleman, J.M., Braud, D., 2007. The Colorado River Delta: regional geomorphology
and sedimentary processes. Journal of Coastal Research 23, 1355­1358.
Jain, S.K., Sharma, K.D., Singh, V.P., 2005. Interlinking of Indian rivers -- issues and analytical
framework. Proceedings of the IWRA XII World Water Congress, New Delhi,
vol. 1, pp. 1.25­1.34.
Kumar, P.D., Gopinath, G., Laluraj, C., Seralathan, P., Mitra, D., 2007. Change detection
studies of Sagar Island, India, using Indian Remote Sensing Satellite 1C Linear
Imaging Self-Scan Sensor III Data. Journal of Coastal Research 23, 1498­1502.
Malini, B.H., Rao, K.N., 2004. Coastal erosion and habitat loss along the Godavari delta
front -- a fallout of dam construction(?). Current Science 87, 1232­1236.
Nilsson, C., Reidy, C.A., Dynesius, M., Revenga, C., 2005. Fragmentation and flow
regulation of the world's largest river systems. Science 308, 405­408.
Ramesh, R., Subramanian, V., Van Grieken, R.,1990. Heavy metal distribution in sediments
of Krishna river basin, India. Environmental Geology and Water Science 15, 207­216.
Revenga, C., Brunner, J., Henninger, N., Kassem, K., Payne, R., 2000. Pilot Analysis of
Global Ecosystems. World Resources Institute, Washington DC. 83 pp.
Rosenberg, D.M., McCully, P., Pringle, C.M., 2000. Global-scale environmental effects of
hydrological alterations: introduction. BioScience 50, 746­750.
Sarma, V.V.L.N., Murali Krishna, G., Hema Malini, B., Nageswara Rao, K., 2001. Landuse /
land cover change detection through remote sensing and its climatic implications
in the Godavari delta region. Journal of Indian Society of Remote Sensing 29, 85­91.
Smakhtin, V.U., Anputhas, M., 2006. An assessment of environmental flow requirements
of Indian river basins. IWMI Research Report 107. International Water Management
Institute, Colombo. 36 pp.
Stanley, D.J., Warne, A.G., 1998. Nile Delta in its destruction phase. Journal of Coastal
Research 14, 794­825.
Subramanian, V., Van, t Dack, L., Van Grieken, R., 1985. Chemical composition of river
sediments from the Indian sub-continent. Chemical Geology 48, 271­279.
Syvitski, J.P.M., Vorosmarty, C.J., Kettner, A.J., Green, P., 2005. Impact of humans on the
flux of terrestrial sediment to the global coastal ocean. Science 308, 376­380.
Turner, I.L., Aarninkhof, S.G., Holman, R.A., 2006. Coastal imaging applications and
research in Australia. Journal of Coastal Research 22, 37­48.
Vorosmarty, C., Lettenmaier, D., Leveque, C., Meybeck, M., Pahl-Wostl, C., Alcamo, J.,
Cosgrove, W., Grassl, H., Hoff, H., Kabat, P., Lansigan, F., Lawford, R., Naiman, R.,
2004. Humans transforming the global water system. EOS 85, 509­520.
540 N. Gamage, V. Smakhtin / Geomorphology 103 (2009) 533­540