Keywords

19.1 Introduction

Gradual reduction of reservoir capacities due to silting is great concern, and efforts should be made to study its impact on all the water resources development projects globally. Hence, accurate assessment of the sedimentation rate is necessary not only for studying its effect on the reservoir life but also required for planning reservoir operations optimally. Particularly, impact of sedimentation on multipurpose reservoirs is significant as silting affects both dead storage and live storage capacities, which in turn induce long and short-range impacts on reservoir operations. Particularly, sedimentation adversely affects planning for long-term utilization of reservoir capacity for irrigation, power generation, drinking water supply and flood moderation. Sedimentation in the reservoir also affects the river regime, which increases the back water levels in head reaches of reservoir and formation of island deltas. Some reservoirs in the world have been silted up so fast that have become useless. Yasuka reservoir in Japan has lost 85% capacity in less than 13 years (CWC 2002). Several reservoirs spread in India are experiencing capacity losses at the rate of 0.2–1% per annum. Nearly 40,000 minor tanks in Karnataka have lost more than 50% of their capacities. Therefore, it is necessary that the surveys be conducted in all the existing reservoirs for ascertaining siltation rate and their useful life. Thus, periodical surveys will enable selection of appropriate measures for controlling sedimentation along with efficient management and operation of reservoirs thereby deriving maximum benefits for the society.

Knowledge on the characteristics of reservoir sedimentation that include amount, distribution configuration and configuration of reservoir deposits is necessary to understand the sedimentation mechanism. Velocity of water that enters into the reservoir is inversely proportional to channel cross-sectional area. Owing to different densities of clear stored reservoir water and muddy inflows into reservoir, heavy turbid water flows along the channel bottom towards the dam under the influence of gravity (Fig. 19.1) (Varshney 1997). It is believed that sediment deposition takes place at the bottom of reservoir affecting dead storage, but contrary to earlier belief that sediment is deposited throughout the reservoir thereby reducing the incremental capacity at all elevations. Several factors such as sediment loading, size distribution, stream discharge fluctuations, reservoir shape, stream valley slope, land cover at the reservoir head, reservoir size and its location, outlets control the sediment movement and deposition pattern in the reservoir.

Fig. 19.1
figure 1

(Source Varshney 1997)

Conceptual sketch of density currents and sediments in a reservoir

Full size image

Recent past, much attention was paid to issues such as reservoir sedimentation that affects live storage capacity of reservoirs due to incoming sediments (Garde 1995; Varshney 1997; Morris and Fan 1997; Goel et al. 2002; Jain and Goel 2002). Narayan and Babu (1983) reported that soil erosion rate in India is nearly 0.16 t/km2/year, where 10% of sediments contributes to reservoir sedimentation problems and 29% is deposited into the sea. CWC (2002) reported that nearly 20% of the live storage capacity of major and medium reservoirs in India was lost due to sedimentation, which means a loss of irrigation potential of about 4 M ha. Several reservoirs in India are experiencing capacity losses at the rate of 0.2–1.5% per annum (CWC 2001, Goel et al. 2002). Shangle (1991) analysed sedimentation surveys based on the 43 reservoirs, which revealed that sedimentation rates range between 0.34 and 27.85 ha m/100 km2/year, 0.15 and 10.65 ha m/100 km2/year and 1.0 and 2.3 ha m/100 km2/year for major, medium and minor reservoirs, respectively. Analysis of capacity surveys also shows a wide variation in sedimentation rates that may be attributed to various factors like physiography, climate, hydrometeorology, etc. Region-wise average rate of siltation in India is given in Table 19.1.

Table 19.1 Region-wise average rate of siltation
Full size table

19.2 Sedimentation Surveys

Sedimentation surveys can be carried out either through conventional ways or by using remote sensing data.

19.2.1 Conventional Surveys

Conventional surveys for reservoir capacity estimation conducted so far are time intensive and often require up to three years to complete the survey of a major reservoir. These surveys involve the conventional equipment such as theodolite, plane table, sextant, range finders, sounding rods, echo-sounders and slow moving boat. However, field surveys are difficult to carry out in the dense forests due to presence of wild life and heavy rainfall during monsoon season. This problem is further compounded due to difficult intervisibility land steep foreshores. In general, these surveys are supposed to be conducted at the normal frequency of five years. However, in practice, it is found that the interval between two successive surveys varies between 2 and 15 years. Modern survey techniques such as hydrographic data acquisition systems (HYDAC) were also used for capacity surveys. This system uses ultrasonic waves for depth measurement. A short sound pulse is emitted by a transducer in the form of a beam vertically towards the bottom. Part of sound energy is reflected from the bed and returns as an echo to the same transducer, which operates both as transmitter and receiver. This eliminates the possibility of angular errors in shallow depths. Depth is computed from the time recorded between emission of pulse and return of its echo. With the help of graphic recorder, a continuous curve presenting a true picture of the bottom is obtained. The range of the stem is up to 1500 m (33 kHz) and 250 m (210 kHz), and its accuracy is ±5 cm. The system is also equipped with transducers, which can detect and allow the measurement of low-density sediment deposits. Use of global positioning system (GPS) has also become common in HYDAC system. Simultaneous use of bathymeter and differential GPS (DGPS) gives x, y, z coordinates. Hydrographic surveys based on DGPS allow faster data acquisition with better accuracy than any previous hydrographic techniques. Thus, DGPS-based surveys eliminate the need to establish line of sight between base stations to boat, while achieving centimetre accuracy data. Central water commission carried out surveys of eight reservoirs, viz. Matalia (UP), Konar and Tilaiya (Bihar), Idukki and Kakki (Kerala), Balimela (Odisha), Linganamakki (Karnataka) and Jayakwadi (Maharashtra) using HYDAC system.

19.2.2 Remote Sensing Survey

Remote sensing data by virtue of its synoptic viewing capability and large area coverage at periodic time periods is of great help for reservoir capacity assessment. It also helps in the acquisition of timely, reliable and comprehensive data on various natural resources. Particularly, its capability to map and monitor inaccessible areas helps to acquire data on reservoir waterspread, which is an important parameter for reservoir capacity surveys. Employing geospatial technologies for reservoir capacity surveys has gained importance and popularity in the recent past globally, particularly with the launch of satellite missions such as Landsat, SPOT, and IRS (Goel et al. 2002). Mohanty et al. (1986) reported that the area capacity curve derived from the multidate Landsat imagery for Hirakud reservoir, Orissa, was nearly coincided with the curve obtained through conventional means. Suvit (1988) correlated the Landsat-MSS-derived surface areas of Ubolratana reservoir in Thailand with the corresponding water levels for computing reservoir capacity. Managond et al. (1985) analysed Landsat-MSS data for estimating the waterspread areas of Malaprabhas reservoir as on different dates and compared with the field surveyed figures. Rao et al. (1988) used multidate Landsat-MS imageries for evaluating capacity of Sriramsagar reservoir. Chandrasekhar et al. (2000) calculated capacity loss for Tawa reservoir as 14.2% in 20 years since its impoundment in 1975. Remote sensing-based approach was envisaged by many researchers in the past for reservoir capacity assessment (Managond et al. 1985, Manavalan et al. 1993, Goel and Jain 1996, 1998, Gupta 1999, Jain et al. 2002, Rajashree et al. 2003, 2004). Peng et al. (2006) derived new storage curve for Fengman reservoir, China, using Landsat imagery and reported that the remote sensing-based storage curve estimation is reasonable and has relatively high precision.

Contrary to remote sensing and conventional approaches, Jothiprakash and Garg (2009) used artificial neural network (ANN) model for assessment of sediment volume retained in the Gobindasagar reservoir on the Satluj river, India. In this approach, annual rainfall annual inflows and reservoir capacities were used as inputs. Several studies used gauge water levels and optical or microwave remote sensing data derived waterspread areas for computing reservoir capacities (Bates et al. 2006; Alsdorf et al. 2007; Smith et al. 2008). Recently, satellite altimetry provides an alternative means to monitor water-level variations over large waterbodies from space. Monitoring of rivers and large water bodies was demonstrated through SARAL/AltiKa and Envisat altimetry missions (Arsen et al. 2015; Dubey et al. 2015; Ghosh et al. 2015). Further, waterspread areas derived using SAR images or multispectral images were used in conjunction with altimetry-based water levels for reservoir capacity estimation of large lakes (Frappart et al. 2015; Ding and Li 2011; Duan and Bastiaanssen 2013; Chowdary et al. 2017). Overall, these studies highlighted that remote sensing-based surveys are less tedious and yielded good accuracy relative to traditional approaches. The present study was carried out with a specific objective of updating the stage—area—capacity curves (estimating loss in the live storage capacity) for 30 reservoirs spread across India due to sedimentation through satellite remote sensing.

19.3 Reservoirs Studied

Analysis of sedimentation was carried out for total 30 reservoirs distributed all over Indian river basins located in different geographical regions (Fig. 19.2a, b).

Fig. 19.2
figure 2

a, b Index map of reservoirs studied

Full size image

List of reservoirs analysed in this study along with basic details was presented in Table 19.2.

Table 19.2 Basic details of reservoirs studied
Full size table

19.4 Satellite Data Used

Satellite data (IRS-1D LISS III and IRS P6 LISS III) at 23.5 spatial resolution data was selected based on their acquisition dates and cloud free data availability. Details of the satellite data used in this study are given in Table 19.3.

Table 19.3 Details of satellite pass used in the study
Full size table

19.5 Methodology

Reservoir sedimentation surveys involve mapping of waterspread areas using satellite data for different water elevations between minimum draw down level (MDDL) and full reservoir level (FRL). In general, decrease of reservoir waterspread area at any elevation due to sediment deposition forms the basis for remote sensing-based surveys. Hence, delineation of waterspread areas corresponding to satellite passes on different dates using remote sensing is prerequisite for remote sensing analysis. Out of all land features, discrimination of land and water features on the image is considered to be easy due to high spectral variability between land and water features in near-infrared (NIR) band. This property enables to delineate waterspread area at different reservoir levels and hereby helps to revise elevation–capacity curve using this approach. Difference in the revised and previously computed reservoir capacity values (previous survey) can be attributed to the siltation problem in the reservoir. Remote sensing-based reservoir capacity assessment approach includes suitable satellite data selection, time series database creation, waterspread area delineation, capacity and capacity loss computations.

19.5.1 Criteria for Satellite Data Selection

Selection of suitable data is imperative for accurate reservoir capacity estimation. Before data procurement, first it is necessary to study the reservoir operation schedule from the reservoir authorities. The dates when the reservoir was at or near MDDL and FRL must be recorded. Satellite passes coinciding with MDDL and FRL (if available) are the best choice for analysis. If overpass does not occur on the same day then nearby dates can be chosen. All the available dates between MDDL and FRL spanning over either one operation year or two operation years can be taken up for analysis. More the number of data, better will be the accuracy of sediment estimation between different water levels. The height interval between consecutive water levels should be around 1–3 m. Hence, optimum dates for acquisition of satellite data were selected based on the information about the minimum and maximum reservoir levels attained during last three years that was collected from reservoir authorities and water management directorate of CWC.

19.5.2 Extraction of Reservoir Spread Area

Researchers envisaged image thresholding, spectral indices, supervised and unsupervised classification-based approaches to maximum extent for monitoring water bodies (Nagarajan et al. 1993; Sharma et al. 1996; Goel et al. 2002; Manju et al. 2005). Modelling approaches based on multiple spectral bands were used to resolve boundary issues between land and water features (Goel et al. 2002). Threshold values can be found out for single band and ratioed image, and subsequently water pixels can be classified based on logical criteria that satisfy in both images. In general, both the supervised and unsupervised classification techniques can be used for demarcation of waterspread area. Conventional maximum likelihood classifier is not always appropriate for surface water mapping. Though the spectral signatures of water are distinguishable from other land features, it may fail to identify water pixels clearly at the land and water interface. Further, it depends on the expert’s interpretation capability. Deep water bodies can be easily identifiable, while shallow water bodies can be falsely classified as soil. Contrary to this, likely chance of saturated soil being misclassified as water is high particularly along the reservoir periphery. Hence, spectral indices based on green and near IR bands can be effectively used for delineation of boundary water pixels. Ratioing of two spectral bands and thresholding water (having ratio higher than threshold) can help in enhancing a particular feature or class from the satellite data. Hence, in this study, thresholding technique is used for extraction of reservoir area.

In general, thresholds for single band or spectral index are identified through visual interpretation by experts. However, identification of thresholds for each satellite data pass is dynamic and needs to be done for each dataset independently. Further, aquatic vegetation may pose problems for clear demarcation of land and water boundary. Spectral reflectance of both shallow and deep waterbodies in the NIR region is minimum compared to red and green spectra bands, which signifies that the DN values for water features are significantly less than other land cover pixels. Normalized Difference Vegetation Index (NDVI) (Tucker 1979) and Normalized Difference Water Index (NDWI) (McFeeters 1996) are some of the spectral indices, which enhance vegetation and water features, respectively. Spectral indices NDVI and NDWI can be computed as given below:

$$ {\text{NDVI }} = \, \left( {{\text{NIR}} - {\text{R}}} \right)/{\text{NIR}} + {\text{R}}) $$
(19.1)

where NIR and R are radiance values in near-infrared band in red band, respectively.

Normalized difference water index, NDWI, can be given as:

$$ {\text{NDWI }} = \left( {{\text{G }}{-}{\text{ NIR}}} \right)/\left( {{\text{G }} + {\text{NIR}}} \right) $$
(19.2)

where ‘G’ is radiance value in green band.

Density slicing technique is used for delineation of waterspread areas, where image pixel grey values are categorized based on expert’s identified class intervals. Minimum and maximum pixel values were identified through histogram analysis, where water pixels that occupy lower histogram ranges in NDVI image and higher histogram ranges in NDWI image. A pixel is classified as a water pixel if it meets the threshold condition in both NDVI and NDWI ratioed images, else rejected from water class.

19.5.3 Refinement of Extracted Waterspread Area

Removal of isolated water pixels from the classified image is necessary as the area is continuous within the contour. Further, water pixels located at the mouth of the reservoir that do not account to reservoir spread area also need to be eliminated (Goel et al. 2002). In this study, manual editing process was adopted to refine the reservoir waterspread image. Further, the main river and numerous small rivulets that join the reservoir from multiple directions at the reservoir tail end are slightly at higher elevations and needs to be removed for better accuracy (Goel et al. 2002). Thus, waterspread area was quietly demarcated for study reservoirs by manually editing the few extended channels around the reservoir periphery.

19.6 Assessment of Revised Live Storage Capacity of Reservoir

Waterspread areas for each individual optical image for all the study reservoirs were identified. The waterspread area corresponding to each satellite data pass was computed by multiplying the number of water pixels by the pixel area (23 × 23). In general, reservoir capacity or storage volume in the period of satellite survey can be estimated through either trapezoidal, prismoidal or cone formula. In this study, modified trapezoidal formula was adopted for reservoir capacity assessment in this study and is given as below:

$$ V \, = \sum h/3*\left( {A_{1 } + \, A_{2 } + {\text{SQRT}}\left( {A_{1} * A_{2} } \right) \, } \right) $$
(19.3)

where V is the reservoir capacity between two successive elevations h1 and h2, h is the elevation difference (h1 − h2), A1 and A2 are reservoir waterspread areas at elevation h1 and h2, respectively. Reservoir capacity assessment using this approach necessitates to compute areal spread at periodic intervals. Hence, waterspread areas at regular intervals were assessed using a best-fit second-order polynomial equation that was built based on area and elevation values.

The original elevation capacity tables before the impoundment of respective dams spread across India were obtained from the project authorities. Initially, original reservoir elevations and capacities corresponding to satellite pass date were used to construct best-fit second-order polynomial equation and subsequently capacities were computed for intermediate elevations. Difference between the original and revised volumes in each zone is attributed to capacity loss due to sedimentation. It was further assumed that the original and revised reservoir capacities at the lowest observation level are same. Subsequent aggregation of reservoir capacities at consecutive levels between the lowest and the highest observation levels yield cumulative revised reservoir capacity.

19.7 Effect of Erroneous Waterspread Area Estimation on Reservoir Capacity

Remote sensing-based reservoir sedimentation estimate is highly dependent on important factors such as delineated waterspread area accuracy, water elevation and the original elevation area capacity table. Thus, precise revision of elevation area capacity curves is quite possible through exact waterspread delineation. Thus, analysis was carried out to evaluate the change in the reservoir capacity as a result of underestimation or overestimation of reservoir spread area by ±10% through remote sensing.

19.8 Customized Software for Reservoir Capacity Estimation

Periodical assessment of reservoir capacity assessment using remote sensing necessitates the development of customized package that can help the user to carry out capacity estimation studies in a timely and reliable way. The analysis was carried out using software ‘KSHAMTA’, a PacKage for JalaSHay CApacity EstiMation and sTorage Loss Analysis. Environment for visualization of Images (ENVI) software along with its development tool Interactive Data Language (IDL) were used for development of semi-automated package ‘KSHAMTA’ for reservoir capacity estimation. KSHAMTA (Fig. 19.3) is a tailor-made application software package for estimation of the reservoir capacity and storage loss analysis. It provides end-to-end facility for multidate processing of satellite images starting from loading of satellite data to the area Vs capacity curve generation.

Fig. 19.3
figure 3

Main screen of KSHAMTA software

Full size image

19.9 Results and Discussion

19.9.1 Extraction of Waterspread Area (WSA)

Waterspread area (WSA) was computed for each satellite date of pass for all the reservoirs. However, WSA needs be computed at regular intervals for capacity estimations, thereby WSA was computed using best-fit second-order polynomial equation using water levels for each satellite data pass and its corresponding WSA. Minimum and maximum WSA at FRL were observed for Kyrdemkulai reservoir as 0.787 M m2 and Narayanpur reservoir as 123.333 M m2. Different reservoir types were considered for analysis. Table 19.4 shows the categorization of reservoirs as per WSA. Analysis for Halali reservoir was carried out beyond FRL due to availability of satellite data. Kallada reservoir was analysed from penstock level, which is the actual operating level for Kallada.

Table 19.4 Categorization of reservoirs as per waterspread area
Full size table

19.9.2 Live Storage Capacity Loss Due to Sedimentation

Cumulative reservoir capacity was compared with the capacity either from original elevation–area–capacity curve generated at the time of construction of reservoir or with the capacities computed from previous surveys. Change in the capacity gives overall loss due to sedimentation. Capacity loss in live capacity for all 30 reservoirs spread across India was carried out (Table 19.5). Sedimentation rates were also computed for these reservoirs between live storage level and MDDL (Table 19.5). Rate of silt deposited over the years was computed using catchment area of reservoirs.

Table 19.5 Live capacity loss and rate of silt deposition
Full size table

Temporal distribution of Mayurakshi reservoir spread, Jharkhand and Sirpur reservoir spread in Maharashtra for different satellite data passes is shown in Figs. 19.4 and 19.5, respectively, as an example. The reservoir waterspread contours extracted from the satellite data along with revised elevation–area and elevation–capacity curves for both Mayurakshi and Sirpur reservoirs are shown as Figs. 19.4 and 19.5, respectively.

Fig. 19.4
figure 4

Mayurakshi reservoir, Jharkhand state a false colour composites acquired on different dates b waterspread areas on different dates c elevation–area–capacity curves

Full size image

Remote sensing-based survey for Mayurakshi reservoir conducted during 2002–2003 revealed that the storage capacity has gone down from 547.6 in 1955 to 485.4 M m3 in 2002–2003 (Table 19.5). The hydrographic survey also proved that the storage capacity has reduced from 547.6 M m3 in 1955 to 474.8 M m3 in 1999–2000, which substantiated the suitability of remote sensing-based approach. Overall remote sensing-based approach revealed that the loss of live storage capacity of reservoir is around 11.4% within a span of 47 years. While hydrographic survey indicated nearly 13.3% of total live capacity loss during the period 1955–2000, remote sensing-based survey (2002–2003) revealed that the rate of siltation is 1.3 M m3/year while the rate of silting is found to be 1.6 M m3/year as per conventional survey conducted during the year 2000. Sirpur reservoir that was constructed across Bagh river in Maharashtra state indicated capacity loss in live storage of nearly 17.325% (Table 19.5).

Original and computed area capacity values for reservoirs and percent change in the capacity are shown in Figs. 19.6 and 19.7, respectively.

Fig. 19.5
figure 5

Sirpur reservoir, Maharashtra state a false colour composites acquired on different dates b waterspread areas on different dates c elevation–area–capacity curves

Full size image
Fig. 19.6
figure 6

Original capacity and calculated capacity using RS

Full size image

Umiam reservoir in Maghalaya state has minimum capacity loss of 1.19% in the span of 37 years (Table 19.5). Gudha reservoir has maximum capacity loss of 26.92% in a life span of 49 years whereas Sondur reservoir has capacity loss of 24.95% in life span of 15 years only, which is quite alarming (Fig. 19.7). Large capacity reduction is observed in Halali, Gumti and Sirpur reservoirs also (Fig. 19.7). Catchment characteristics play an important role and are responsible for erosion processes in the catchment and subsequent capacity loss in the live capacity of reservoirs. Capacity loss was observed to be low for Narayanpur as this reservoir is in succession to other reservoir, thereby getting less silt. Most of the reservoirs followed normal decreasing trend of capacity between different time periods (Fig. 19.7). Reservoirs in Gujarat that were analysed by one agency show more capacity loss when compared to the capacity obtained using remote sensing. Jakham reservoir shows absolute reverse trend, where present capacity is much higher than the original capacity itself. To cross check such results, original capacity was computed using original area values and the capacity formula. Most of the discrepancies could be explained with this approach.

Rate of sedimentation was compared with number of years of reservoir operation (Fig. 19.8). Rate of sedimentation is more for reservoirs, which are relatively young with less operating years. This may be attributed to the fact that during the initial stages of reservoir, sedimentation from catchment is generally more. As the reservoir life increases, stability of the regime changes with time and the rate of sedimentation gradually reduces.

Fig. 19.7
figure 7

Percent change in the capacity of reservoirs

Full size image
Fig. 19.8
figure 8

Rate of sedimentation versus life of reservoir

Full size image

Remote sensing-based method is promising as satellite-derived reservoir spreads are more realistic, whereas conventional hydrographic surveys suffer from certain inherent limitations. Particularly, these surveys face difficulties in fixing survey ranges, movement of boat along survey lines, dynamic and complex nature of silt deposits at reservoir bottom which effects on reflectance of sound waves during echo sounding. It was reported by Rajeev et al. (1992) that remote sensing storages are relatively underestimated. Higher rate of siltation than the assumed value at planning stage can be attributed to severe soil erosion in the reservoir catchments due to inadequate forest cover, denudation of forest, uncontrolled grazing and negligence of available pasture land. Conservation measures, viz. construction of retention terraces, bench terraces, gull control works and silt detention dams were adopted in some of the catchment areas, still the catchments are experiencing soil degradation.

The remote sensing-based approach highlights the need to adopt appropriate measures for controlling soil erosion in the reservoir catchment areas thereby increasing the life of reservoir. Further, effect of over- or underestimation of reservoir spread area on reservoir capacity was also studied for Mayurakshi reservoir (Table 19.6). It may be observed that the change in the reservoir live capacity is less than 4% of our actual estimate, when the error in waterspread area estimation is not more than ±2.5%, while changes in the reservoir spread area beyond 2.5% has brought noticeable changes in reservoir capacity (Table 19.6). It signifies that any possible error associated with border pixels while delineation of waterspread will significantly affect the reservoir capacity estimation, which can be overcome by using higher spatial resolution satellite. Further, remote sensing images can be chosen at closer time intervals so that the revised waterspread area can be computed for smaller elevation intervals, which may lead to increased accuracy in sedimentation assessment.

Table 19.6 Sensitivity analysis of waterspread area and reservoir capacity for Mayurakshi reservoir
Full size table

Although remote sensing-based approach is promising, it is limited to assessment of reservoir live capacity only. However, this limitation can be insignificant as reservoir levels rarely fall below the MDDL during normal years. Further, sediment deposition in the live storage zone is of great significance considering the reservoir operations between MDDL and FRL. Remote sensing approach coupled with hydrographic survey can be envisaged for reservoir capacity assessment both in live and dead storage zones.

19.10 Conclusions

Multispectral satellite data available at higher spatial and temporal resolutions promises fast and reliable estimation of waterspread areas required for capacity estimation of reservoirs as remote sensing images can be obtained at frequent interval for better representation of ground profile. This study signifies that any possible error associated with border pixels while delineation of waterspread will significantly affect the reservoir capacity estimation, which can be overcome by using higher spatial resolution satellite for waterspread area delineation. One of the major limitations with remote sensing technique is that it provides accurate estimates for fan-shaped reservoir where there is a considerable change in the waterspread area with small change in the water level. For U- or V-shaped reservoirs where steep banks exist, there may not be remarkable change in waterspread area with fluctuation in water level. In such conditions, capacity estimation may not provide accurate results. Remote sensing-based survey being applicable between MDDL and FRL, remote sensing surveys can be carried out in conjunction with hydrographic surveys at shorter and longer intervals, respectively, for assessment of sedimentation for the entire reservoir zone.