| Printer Friendly Format |

Page 1 of 1

Reservoir Capacity Estimation using Semi Automated Package "KSHAMTA" - A Case Study of Jakham Reservoir, Rajasthan, India

Rajashree V Bothale, Vinod M Bothale, J R Sharma and V Jayaraman*
Regional Remote Sensing Service Centre, Jodhpur, India
*NNRMS/ RRSSC, Indian Space Research organization, Bangalore, India

Natural processes like erosion in the catchment area and its deposition in various parts of reservoir gradually reduce the capacity of reservoir. Dead as well as live storages are affected by it. Information about the reduction in capacity is necessary for all the planning and operational purpose and same can be obtained through capacity surveys done at regular interval. Remote Sensing technique to calculate present capacity of reservoir is very useful due to its synoptic and repetitive coverage. Present study aims in updating the stage - area - capacity curve for Jakham reservoir, which was constructed in 1986 over river Jakham, a tributary of river Mahi in Chittaurgarh district of Rajasthan and finding the capacity loss due to sedimentation. For carrying out the analysis IRS LISS III data with 23.5 m resolution has been used. An indigenously developed semi automated package "KSHAMTA" which provides end - to - end methodology for reservoir capacity estimation was used in the study. Satellite data for seven passes falling between MDDL (332.0m) and FRL (359.5m) were used. After present analysis live capacity of reservoir was calculated as 175.451 Mcum, which is much more than the designed capacity of Jakham reservoir. Even the water spread areas were more. To cross check the results WSA was mapped on SOI toposheet. The mapped value (18.133 Mm2) at RL 360m is closer to values calculated using RS techniques.

All our developmental plans have given high priority to water resources projects involving construction of dams and a large number of dams have been constructed since independence. The capacity of reservoirs is gradually reducing due to silting and hence sedimentation of reservoir is of great concern to all the water resources development projects. Silting not only occurs in the dead storage but also encroaches into live storage capacity, which has long and short-range impact on the functioning of the project and economics. Correct assessment of the sedimentation rate is essential for assessing useful life of the reservoir as well as optimum reservoir operation schedule. Since 1958, when it was established that the live storage of reservoir is getting reduced due to siltation, a systematic effort has been made by various departments / organizations to evaluate the capacity of reservoirs. Various techniques like boat echo sounder, etc., were used which are now gradually being replaced by hydrographic data acquisition system (HYDAC). The conventional techniques are either time consuming or costly and require considerable manpower. The completion of survey of one major reservoir may take up to 3 years.

Remote sensing techniques to calculate the present capacity of reservoir are found to be very useful due to synoptic and repetitive coverage. The surveys based on remote sensing data are faster, economical and more reliable. Reservoir sedimentation surveys are essentially based on mapping of water-spread areas at the time of satellite over pass. It uses the fact that water-spread area of the reservoir reduces with the sedimentation at different levels. The water-spread area and the elevation information are used to calculate the volume of water stored between different levels. These capacity values are then compared with the previously calculated capacity values to find out the change in capacity between different levels.

Reservoir capacity estimation are required to be carried out at regular intervals in order to assess the rate of sedimentation and useful life of reservoir. As per International Standard Organization (ISO), the interval between repeat surveys should be between 3 to 5 years or after capacity loss of 5%, whichever is shorter.

The sedimentation surveys of reservoirs in India although dates back to as early as 1870, the systematic surveys started only in 1958 when the Central Board of Irrigation and Power undertook a coordinated scheme of reservoir sedimentation. Under the Eighth Five Year Plan period, Central Water Commission started surveys for 144 reservoirs of the country using conventional techniques(CWC, 2002). In their recommendation CWC recommends that the capacity surveys need to be done on regular basis of once in 5 years for major reservoirs. 'Compendium of silting of reservoirs in India' recommends that owing to limitation of remote sensing based survey being applicable between MDDL and FRL, hydrographic surveys should be conducted at longer intervals (20 years) and remote sensing based sedimentation surveys at shorter intervals (5 years) to make both surveys complementary to each other.

Realising the gravity of sedimentation problem and owing to the need to do regular survey of reservoir using remote sensing techniques, a customized package, which can help the analyst in carrying out capacity estimation in a faster and systematic way has been brought out. Present study aims in estimating the capacity of Jakham reservoir in Rajasthan using "KSHAMTA", the semi automated package for reservoir capacity estimation and comparing it with original values.

Jakham reservoir is located at 24010'30" N latitude and 74035'30" E longitude at village Anuppura, tehsil Pratapgarh, district Chittorgarh in Rajasthan. It is constructed in the year 1986 on river Jakham, which is tributary of river Mahi. The project provides irrigation benefits to tribal people of the area consisting of 104 villages. The area near the dam is hilly and consists of waste land, hence a pickup weir at Nagalia, which is 13 km away from dam is constructed from which main canals emerge. Project has a catchment area of 1010 sq km with average yield as 212.376 Mcum. The irrigated area in catchment covers 95.85 sq kms (9.49%) and the un-irrigated cultivable area is 366.32 sq km (36.27%). Around 45% of the catchment area has sufficient soil cover for raising crops and this area may contribute to soil erosion to Jakham reservoir. Forest area covers 23.52% of catchment area. Soils in the area vary considerably. The soils are silty loam to clay loam, occasionally clay, grayish brown to dark grayish brown. Soils are deep and occur o buried pediments and are derived from alluvium of phyllites and limestone.

Top of dam is at RL 373.0m with HFL at RL 371.65m. Dead storage level of reservoir is at RL 332.0m and Full Reservoir Level (FRL) is at 359.5m. Length of dam is 253.0 m with maximum height as 81.0m. Two main canals emerge from pick up weir. The canals are lined with a discharge of 3.533 cumecs (RMC) and 7.92 cumecs (LMC). Figure 1 shows the index plan of study area.

Figure 1 shows the index plan of study area.

To select the desired data set, water level achieved between years 2000 to 2003 were analysed. It was observed that reservoir has gone to FRL on cloud free dates in the year 2001 but data towards MDDL was not available. In the year 2002 the difference with MDDL was only 6.3m in total operating range of 27.5m. Since reservoir has not operated between extremes of MDDL and FRL in the year 2003, it was decided to take study year as 2002 with two data set from 2001. IRS 1D LISS III data with a resolution of 23.5 m has been used for the analysis. Table 1 shows the data set used for the analysis. Figure 2 shows the FCCs used in the analysis.

Table 1: Date of pass for satellite data

Figure 2. Satellite FCCs used in the analysis

Apart from satellite data Survey of India toposheet 45L/12 at 1:50,000 scale was used for geo-referencing of data sets. Field data sets included original stage - area - capacity curve obtained from authorities along with reservoir levels on satellite date of pass. Details about any survey carried out earlier were also collected from authorities.

The basic concept is to find out the water-spread area from satellite data for different water levels between MDDL and FRL. The difference between areal spread of water between current year and earlier years is the areal extent of silting at these levels. The methodology for estimation of capacity of reservoir using remote sensing consists of steps like selection of suitable data, digital database creation, estimation of water-spread area, calculation of capacity and estimation of capacity loss due to sedimentation.

To carry out the task, an indigenously designed and developed package "KSHAMTA" (Figure 3) has been used in the analysis. Developed over ENVI and IDL the package has following functionalities.
  • To provide an end-to-end solution to estimate the reservoir capacity.
  • To compute storage loss from the date of reservoir impoundment.
  • To provide series of processing functions in a predefined sequence.
  • To minimize the user interaction.
  • To automate the data nomenclature at various steps of processing.
  • To load data from different sensors.
  • To do rectification using AUTOFIT option.
  • To plot and save the elevation-area-capacity curves.
  • To calculate the changes in capacity.
  • To generate and save tables for elevation -area-capacity and capacity loss
  • To save the composite images using all date water spread areas.

Figure 3.

For the analysis of all scenes in KSHAMTA, plan process details are generated, which include information on name of sensor, Path / Row, date, month and year of satellite pass along with gauge level on date of pass.

All the scenes, which were entered in plan process details were loaded in the system. Three well defined points (tie points) whose latitude - longitude values are known to us were selected on all the scenes and their line and pixel number were noted. Satellite data is georeferenced using the Autorectify module of KSHAMTA. It computes the GCPs from the information extracted from CD during data loading. These GCPs are fine tuned by using the tie points noted in the previous step. All the scenes are georeferenced in this manner. Water spread area extraction is done using Normalised Difference vegetation Index (NDVI) method. Range of water values for all the scenes was recorded for water body extraction. Table 2 shows the water spread areas estimated from satellite data. Figure 4 shows the water spread areas on different dates.

Table 2: Water spread areas estimated from satellite data

Water levels on the date of pass for selected satellite data is not at regular interval, which is needed for capacity calculation. To get WSA values at regular interval of elevation, area - elevation curve is plotted for the reservoir and a best fit polynomial equation of second order is derived. The best fit equation calculated for Jakham reservoir is A = 1571.16- 9.54514h+ 0.0145098h2, where A is Water spread area in Mm2 and h is elevation in meter.

Reservoir capacity calculation was done using following formula

V = (h/3)(A1 + A2 + SQRT(A1 * A2))

Where V is the reservoir capacity between two successive elevation h1 and h2
h is the elevation difference (h1 - h2)
A1 and A2 are area of reservoir water spread at elevation h1 and h2.

Figure 4: WSA for Jakham reservoir

The area and capacity values were compared with original values to know the change in capacity. Figure 5 shows the elevation capacity curves for Jakham reservoir.

Figure 5: Elevation - capacity curves for Jakham reservoir

Water spread area and cumulative live storage capacities were calculated at regular interval of 1 m between MDDL and FRL for Jakham reservoir. Table 3 shows the elevation - area and capacity values.

Table 3: Elevation - area and capacity values

The calculated values were compared with original stage - area - capacity values obtained from reservoir authorities (Table 4).

Table 4: Comparison of live storage capacity of reservoir (Mm3)

Study of Jakham reservoir reveals that the water spread area and capacities calculated through remote sensing technology are more than what is given in the project report. There is no question of loss in this reservoir if we compare with original curve. The present capacity calculated is 175.451 Mcum, which is 132.6% of the original capacity (132.28 Mcum). To cross check the results obtained through remote sensing, help of SOI toposheets was taken. Contour of RL 360m, which could be clearly demarcated was traced out. Water spread area at RL 360m was calculated as 18.133 Mm2 (inclusive of small inner islands). This area is much more than the area given by reservoir authorities. Over the years due to siltation, WSA reduces, hence the water spread area at RL 360 m using RS technique has been calculated as 15.379 Mm2. Figure 6 shows the water spread area using SOI toposheet.

Figure 6: Comparison of water spread areas

From all the analysis it is clear that the original values given by authorities have some discrepancy in capacity and area calculation. The results were shown to authorities and they were also of the opinion that there could be some error in conversion of area values from one unit to another. One source of error could be the map sheet. It might be possible that instead of using 1:50,000 scale map for area calculation, 4 inch : 1 mile map had been used. If the proper conversion factor (~1.6057) is applied, then the new WSAs will be as shown in table 5.

Table 5: Some justifications for the discrepancies in the result

Following conclusions can be drawn from the study:
  • The live storage capacity of Jakham reservoir is 175.451 M cu. m. in year 2002.
  • There is no capacity loss when compared with original values. Original values are much less than presently observed values.
  • There is lot of variation in the capacity and area calculated through remote sensing and given by authorities, but the supporting analysis of toposheet strengthens the results obtained by remote sensing analysis.
  • KSHMATA provides fast tool for processing of remote sensing data.

Authors are grateful to Chiarman, ISRO to provide all support and encouragement during course of study. Authors are thankful to Shri T S Shah, SE, and Shri R L Bansal, EE, jakham Irrigation Division, Udaipur, Rajasthan for providing necessary reservoir inputs and support during ground visit.

  • RRSSCJ (2004), Technical report on 'Sedimentation Assessment through Satellite Remote Sensing for Jakham Reservoir, Rajasthan', RCJ/TR/2004/4
  • Bothale, Rajashree V., Manavalan, P., Sharma, J.R., et.al. (2003). Reservoir Capacity Surveys through Remote Sensing, ISRO-NNRMS-TR-104-2003.
  • Bothale, Rajashree V., Praveen Kumar, Sharma, J.R., et.al. (1997). Sedimentation analysis of Mahi Bajaj Sagar reservoir through satellite remote sensing, Project Report, RRSSCJ, Jodhpur.
  • Bothale, Rajashree V., Sharma, J.R. and Adiga, S. (2000). Analysis of catchment and reservoir for stage-area-capacity estimation using remote sensing and GIS, proceedings of 10th National Symposium on hydrology with focal theme on urban hydrology, New Delhi, July 18-19, pp 533-541.
  • Bothale, Rajashree V., Sharma, J.R., et.al. (1999). Sedimentation analysis of Mahi Bajaj Sagar reservoir through satellite remote sensing, proceedings of workshop on reservoir sedimentation assessment using remote sensing, Roorkee, pp. 6-1 to 6-7, May 07-08.
  • CWC (2001).' Compendium of silting of reservoirs in India', Technical report on silting of reservoirs in India, WP&P wing, Central Water Commission, New Delhi
Page 1 of 1