Diminishing Returns from Large Hydropower Projects in India

Back to Resources
First published on
  • This is a guest blog by Himanshu Thakker of SANDRP.

As per the latest power generation figures released by the Central Electricity Authority, the hydropower generation during the financial year 2014-15 was 4.25% lower than the previous year’s generation, in spite of the installed capacity having increased. Average generation per MW of hydro capacity in India in 2014-15 was more than 20% less than average generation in 1993-94. More worryingly, the hydropower generated per MW installed continues its downward slide, a continuing trend for the past two decades.

The various governments and power sector establishment in India have been pushing large hydro projects as if they are good in themselves. Here’s a graph that shows the rapid increase in installed capacity of large hydro projects in India. A huge additional capacity of hydropower is under construction and at various stages of approval and planning. However, there has been no attempt at credible performance appraisal of hydropower projects in India.

The author (right) on a river tourThe author (right) on a river tourBharat Lal Seth

SANDRP has been doing performance appraisal of large hydropower projects of India for some years. We have plotted the Million Units (MU, one unit is equal of one kilowatt hour of power) power generated per MW installed capacity of all operating hydropower projects in India. This has been based on the official data from Central Electricity Authority for total annual power generated by all existing operating large hydropower projects vis-avis the total installed capacity of such projects, for the last 20 years starting 1993-94. The total installed capacity of large (projects with installed capacity over 25 MW) hydro projects in India as on March 31, 2015 is 40,885 MW.

The graph of the latest power generation figures shows the power generation trend line of large hydropower projects, which clearly shows diminishing generation from existing projects. This should be a matter of serious concern, but neither the CEA nor the Power Ministry or any other agencies are doing such an analysis. This exercise can also help this fall, and what needs to be done to arrest or reverse this trend. The falling generation cannot be attributed to lower monsoon rainfall, since rainfall has been average or above average in most of the years under consideration.

Monsoon Rainfall 

What we see from monsoon rainfall figures from India Meteorological Department is that in 14 of the 22 years the rainfall was above or near normal (above 98 per cent of normal), so one cannot say that the diminishing generation is due to below normal monsoon.

89% of projects are underperforming

A separate analysis showed that 89% of the projects generate at below the design or promised generation level at 90% dependability. Each large hydro project is given techno economic clearance by CEA based on a promise at appraisal stage that the project will generate certain amount of power in 90 per cent of the years. We compared that figure with the actual generation figure for the last 29 years for each of the operating projects, and found that 89% have been under-performing. And among the under performing projects, half of the projects were generating less than 50% of the promised power generation. Yet, no questions are asked, no accountability fixed, in fact such an analysis is not even done by the official agencies. This means, for example that a lot of projects that are being set up now are unviable projects, or installed capacities are above optimum levels.

The reasons

There are many reasons why the generation per MW is dipping: unviable projects, unviable installed capacities, over-optimistic hydrological assumptions, over development (development beyond the carrying capacity of the basin), catchment degradation, high rates of sedimentation, inadequate repair & maintenance, run of river projects (this phrase is a misnomer, these projects do not generate power from the run of the river but through a dam and a tunnel), changing monsoon patterns due to climate change, etc.

Here it should be added that the destruction of forests, rivers, fisheries, biodiversity and submergence of lands is also making the impact of  climate-related disasters worse for the people and also for the hydropower projects, as could be seen during the June 2013 Uttarakhand disaster. All this needs to be part of our impact assessment and decision making process.

Monthly generation

Monthly power generation from hydropower projects in 2014-15 shows that the maximum power in hydro sector (36.3 per cent of annual generation) happens during three monsoon months of July-Sept. During six months (May-October), 63.7 per cent power is generated. February has the lowest generation.

                                                              Underperformers

Dams have a massive social and ecological cost. River beds are left dry for large sections due to so-called run-of-river projects.Dams have a massive social and ecological cost. River beds are left dry for large stretches due to so-called run-of-river projects.Bharat Lal Seth

During 2014-15, central sector hydro capacity of 13,576 MW generated 1 per cent more power than 2013-14, and private sector capacity of 2,728 MW generated 7.32 per cent more power. It was the state sector capacity of 24,582 MW that generated 8.1 per cent less power than the previous year. The biggest dip in generation was seen in the power generation at Sardar Sarovar Project’s 1,450 MW capacity, which was a reduction of 50.31 per cent compared to 2013-14. Among the central sector hydro organisations, NHDC (1,520 MW capacity on Narmada river) saw the biggest dip of 35.9 per cent compared to previous year generation. One of the reasons for the reduced power generation at SSP and upstream Narmada projects of NHDC was the unjustified depletion of the reservoir before the 2014 parliamentary elections.

There is no doubt that more detailed state wise, basin wise, type wise, age wise, etc analysis would help, some of which are available in “Hydropower Performance” section of SANDRP website. However, to continue to push large hydro without such informed analysis would only lead to bad decisions, leading to avoidable social, environmental, economic and opportunity costs.

There are options

Considering the diminishing returns from existing hydropower capacity and serious underperformance from 89 per cent of projects, the first priority should be to optimise generation from existing projects instead of pushing for new large hydro projects. We are also not doing advance comprehensive project level and basin wide cumulative social and environmental impact assessments, nor getting public consent in upstream and downstream areas.

Even for ensuring power to those that do not have access to power now, there is also a huge potential of sub MW (less than 1 MW or KW scale) capacity hydro projects in most of the Himalayan and other mountain areas. These projects would have least costs, less impacts, and could be implemented in short time with full involvement of the benefiting local communities. Such projects should be a priority before taking up larger projects.

As global trends show, more solar and wind power based installed capacity is being added than in any other sector. This is also likely to be the trend in India in future since cost of solar and wind based generation is going decreasing. One of the obstacles in this regard is that the solar and wind power is only available during day time or wind hours, and storage of surplus (over and above what the grid can consume) power generated during such time is costly. The pump storage hydro projects can provide an option in this regard even for existing dams. Thus surplus power during day time and wind hours can be used to pump water from lower (smaller) reservoir to upper (larger existing) reservoir of such projects. During night and off wind hours, this water than be released from upper to lower reservoir to power the same turbines. This option needs to be explored.

No one is assessing peaking power from hydro projects

One of the unique selling propositions of hydropower projects is that they can provide peak hour power, which coal based or nuclear power projects cannot. However, we do not have the necessary data to analyse what % of hydropower generation is providing peaking power. What is surprising however is that there is no agency in India that is doing such an analysis. Without such an analysis it is not even possible even start understanding and optimizing the peak hour power generation from hydro projects. Needless to add, such an analysis should also involve assessing the impacts of peaking power generation. Depending on site specific conditions, the currently under-utilised peaking power potential of existing hydropower projects can be substantially better utilised before going for new peaking projects. The question is, if we are neither monitoring nor trying to optimize peaking power generation, is there a case for pushing more hydro in the name of peaking power demands?

What about existing large dams?

Lastly, it should be added that less than 3 per cent of India’s large dams have hydropower component and most of the remaining large dams without hydropower component are irrigation projects. However, in these projects, the social, environmental and economic costs have already been paid. We need to at least assess the possibility of adding hydro component to some of these projects. But such an assessment is not happening at all. The United States of America, to whom we look up for many things, has been doing exactly that, assessing the possibility of adding hydro component at existing dams, rather going for new hydro projects.

What all this means is that before we go for more large hydro projects at huge social, ecological and economic costs, we have other options. We hope the questions raised in this analysis lead to changes in some of the directions indicated here.

Himanshu Thakker is the Coordinator of South Asia Network on Dams, Rivers and People (SANDRP). This blog post first appeared on the SANDRP Blog, here.