Energy Policy 81 (2015) 226–238

Contents lists available at ScienceDirect

Energy Policy
journal homepage: www.elsevier.com/locate/enpol

Energy infrastructure in India: Profile and risks under climate change

Amit Garg a,n, Prakriti Naswa b, P.R. Shukla c
a
Indian Institute of Management Ahmedabad, Wing 16B, Vastrapur, Ahmedabad 380015, Gujarat, India
b
UNEP-DTU Partnership, Room 3.7.39, UN City, Marmorvej 51, 2100 Copenhagen, Denmark
c
Indian Institute of Management Ahmedabad, Wing 3G, Vastrapur, Ahmedabad 380015, Gujarat, India

H I G H L I G H T S

Climate risks to energy infrastructures adversely impact energy security.

Case studies of a port and a railway show their future climate change vulnerability.

Managing climate-induced risks through preventive adaptation policies.

art ic l e i nf o a b s t r a c t

Article history: India has committed large investments to energy infrastructure assets-power plants, refineries, energy
Received 5 July 2014 ports, pipelines, roads, railways, etc. The coastal infrastructure being developed to meet the rising energy
Received in revised form imports is vulnerable to climate extremes. This paper provides an overview of climate risks to energy
28 November 2014
infrastructures in India and details two case studies – a crude oil importing port and a western coast
Accepted 3 December 2014
Available online 12 December 2014
railway transporting coal. The climate vulnerability of the port has been mapped using an index while
that of the railway has been done through a damage function for RCP 4.5.0 and 8.5 scenarios. Our analysis
Keywords: shows that risk management through adaptation is likely to be very expensive. The system risks can be
Energy infrastructure even greater and might adversely affect energy security and access objectives. Aligning sustainable de-
Reverse impact
velopment and climate adaptation measures can deliver substantial co-benefits. The key policy re-
Vulnerability index
commendations include: i) mandatory vulnerability assessment to future climate risks for energy in-
frastructures; ii) project and systemic risks in the vulnerability index; iii) adaptation funds for un-
mitigated climate risks; iv) continuous monitoring of climatic parameters and implementation of
adaptation measures, and iv) sustainability actions along energy infrastructures that enhance climate
resilience and simultaneously deliver co-benefits to local agents.
& 2014 Elsevier Ltd. All rights reserved.

1. Introduction
Post-industrial revolution era has made energy infrastructure
an essential prerequisite for economic growth and development
(Briceño-Garmendia et al., 2004). Energy infrastructure here de-
notes assets involved in the production, transportation, transfor-
mation, and transmission of energy and those, linking these asset
blocks. Energy producing infrastructure assets are deployed for
primary energy extraction and comprise natural assets of primary
energy such as mines, oil wells, gas reserves etc. (Asif and Muneer,
2007; Brown, 2002). Primary energy sources need to be extracted,
and transformed into a usable secondary energy and therefore,
have to be transported to the location of transformation (Brown,
n more critical for the asset in future, which are either not visible
Corresponding author. Fax: þ91 7966306896.
E-mail addresses: amitgarg@iimahd.ernet.in (A. Garg), prana@dtu.dk (P. Naswa),
shukla@iimahd.ernet.in (P.R. Shukla).
2002). These distances are high in India due to the concentration
of natural energy assets in specific areas, for example, coal re-
sources are located in the eastern states of Bihar, Jharkhand and
Orissa but the transformation asset blocks like generation units are
spread across the country (TERI, 2006; Santhiyavalli and Usharani,
2011). Table 1 summarizes the types of energy infrastructure
blocks and the types of assets in each block.
The changing climate poses many threats for energy infra-
structure assets in India. Firstly, it is creating newer risks. Infra-
structure assets are planned with some visibility of magnitude and
type of potential climate-induced risks (Hallegatte, 2009). With
the changing climate, new dimensions are being added to the risk
profile of these assets thereby changing the conceptual basis for
risks (Kousky and Cooke, 2009). Some specific risks may become
today or do not hold importance in the existing basket of risks
(Stern, 2007). Secondly, climate change appears to enhance the

http://dx.doi.org/10.1016/j.enpol.2014.12.007
0301-4215/& 2014 Elsevier Ltd. All rights reserved.
 A. Garg et al. / Energy Policy 81 (2015) 226–238 227

Table 1 optimum stocks and disbursement to priority demand areas.

Energy infrastructure blocks.a Infrastructure lock-ins reduce the flexibility to change or re-
place assets before their expected life span. Therefore, there is an
Production Transportation Transformation Transmission
block block block block emerging need to ensure that existing assets are more climate
proof by taking appropriate risk management measures to prevent
Domestic Domestic
Refineries
Power grids damages from climate-induced risks. In this paper, we discuss the

Mines
Pipelines
Power plants
Distribution climate-induced challenges that energy infrastructure faces and
lines

Oil wells
Rail
Pipelines how it subsequently affects the energy infrastructure networks of

Gas
Road
Road the country. The following sections detail the profile of energy
reserves infrastructure assets in India and the various climate-induced risks

IWT
Railways that these assets face. The impact of climate change on two spe-
Imported Imported cific cases of energy infrastructure assets-one a large port (Kandla

Mines
Shipping port) importing considerable volume of crude oil, and another a

Oil wells
Pipelines

Gas rail-road infrastructure (Konkan Railways) along the western coast
reserves transporting coal have been elaborated.
The objective of this paper is to analyze the current situation of
a
Authors' assessment. these energy infrastructure assets and in light of future climatic
changes how these assets can prioritize risk management through
risks that the assets already face. For example, physical risks due to adaptation for better resilience and energy security in India. These
extreme weather will become more severe in certain conditions infrastructure assets constitute important nodes in the Indian
(Hallegatte, 2009). Thirdly, climate change threatens the usable life energy infrastructure set up and hence their uninterrupted func-
span of the assets. Regulatory or product and technology risks tioning is important for the energy network. The paper assesses
could make the asset redundant sooner than the planned lifespan the current and future risks in the form of a vulnerability index
(Peter and Grimm, 2008). Finally, it creates allied risks that arise and a risk function respectively. These cases show the potential
out of disruptions in network of infrastructure such as supply variables that can be improved to increase resilience of the assets
chain risks (Schenker-Wicki et al., 2010). More variations in future and ensure a smooth energy infrastructure. Finally, the paper
weather conditions, increasing economic activity, increasing po- concludes with concerns about risk management through adap-
pulation, etc. are leading to increased energy consumption. tation of climate-induced events for climate proofing assets and a
Increased dependence on energy assets is making them more discussion on key takeaways for policy makers.
susceptible to supply side vagaries. Energy is a constrained re-
source that is needed in all economic sectors and climate-induced
risks have implications for the ‘energy security’ goal in India. Cli- 2. Energy infrastructure assets in India: profile and risks
mate-induced risks challenge this goal by disrupting supplies, af-
fecting the network of energy supplies, making assets redundant, 2.1. Overview
etc. (Kurian and Vinodan, 2013; Schenker-Wicki et al., 2010). These
risks can affect any part of this inter-connected energy infra- India is the second most populous country and the fourth lar-
structure network and when actual events occur, they have the gest consumer of crude oil and natural gas in the world after the
potential to create an energy crisis through ripple effects. In the United States, China and Russia (IEA, 2014; Kumar and Jain, 2010;
past also, extreme weather conditions have damaged transporta- TERI, 2006; Vivoda, 2010). The demand for energy is much higher
tion blocks causing serious supply constraints for India. Temporary than the supply (IEA, 2014) making it important for the energy
supply interruptions induce price rise due to shortage and, to infrastructure network to work efficiently for a better demand-
some extent, can be dealt by maintaining buffer stocks. India supply match across regions. Network disruptions create stresses
maintains strategic ‘crude oil reserves’ for two weeks (ISPRL, for energy availability to millions and can have damaging spill over
2014). Similarly, power plants maintain buffer stocks of coal. impacts on various sectors of the economy.
However, this reserve mechanism sometimes is inadequate to Domestic sources of coal and crude oil are not sufficient to
meet the demands of so many energy dependent sectors of the meet internal demand and have to be supplemented by imports
economy and faces operations management issues of maintaining (Fig. 1). Planning Commission-GoI (2013) estimates suggest that

Fig. 1. Coal demand and supply (million tonnes).

Source: Ministry of Coal-GoI (2012, 2013).
228 A. Garg et al. / Energy Policy 81 (2015) 226–238
the total domestic energy production in 2016–2017 will be about
669.6 million tonnes of oil equivalent (MTOE) and 844 MTOE by
2021–2022. This will meet around 70 percent of expected energy
consumption, and the balance, projected to be about 267.8 MTOE
by 2016–2017 and 375.6 MTOE by 2021–2022 will have to be met
by exports. In the year 2012–2013 India's coal consumption was
772 million tonnes while the domestic supply was only 580 mil-
lion tonnes (Ministry of Coal-GoI, 2013). The gap of 192 million
tonnes was met through imports. The Twelfth Five Year Plan
(TFYP) projects an annual demand for coal at 980 million tonnes
during 2012–2017, 19 percent of which will be met through im-
ports (Ministry of Coal-GoI, 2011).
Energy imports are a significant component of India's total
foreign exchange spending. The refining of crude oil largely hap-
pens indigenously. The production of crude oil was to the tune of
37.79 million metric tonnes (MMT) (up by 0.2 percent) in 2013–
2014 (Ministry of Petroleum and Natural Gas-GoI, 2014). The
consumption of petroleum products on the other hand was
159.197 MMT during 2013–2014. To meet the demand,
189.24 MMT of crude oil (valued at 8648.75 billion rupees) was
imported. The petroleum imports have a 36.5 percent share in
total imports for 2013–2014 (Ministry of Petroleum and Natural
Gas-GoI, 2014). The demand for petroleum products is expected to
be 847 MMT with an average growth of 5 percent p.a. The share of
coal and petroleum is expected to be about 66.8 percent in total
commercial energy produced and about 56.9 percent in total
commercial energy supply by 2021–2022. 60 percent of the elec-
tricity units being generated in India are coal based (Table 2).
Many new Ultra Mega Power Projects (UMPPs) have now come up
at the coasts that are using imported coal for electricity genera-
tion. The total installed capacity of the power sector in India
228.7 GW. This is not sufficient to meet the internal demand
(Pargal and Banerjee, 2014). Therefore, the peak deficit is close to
10 percent and energy deficit is close to 9 percent (Table 3).
Figs. 2 and 3 show the regional spread of energy infrastructure
for crude oil and coal. They map coal mines vis-à-vis thermal
power plants and crude oil wells vis-à-vis refineries and pipelines.
The network shows the presence of large nodes, that is, nodes
which have concentration of economic activities due the volume
or number of processes being dealt with at the location. The
consumption points are well spread across the country but the
sources of production and processing are confined to specific lo-
cations. This makes the effective functioning of these large nodes
in the energy network crucial. If a node is threatened by a climate-
induced risk, it will break the seamless flow of energy in the
network and create supply issues for the paths/end points that it is
catering to.
Some important points emerge from this section. Firstly, there
is a huge demand for energy resources in India. Indigenous re-
sources are insufficient to meet the growing demand; at times
even after imports the demand is not met. Secondly, infrastructure
resources are yet to develop fully to meet the sub-regional energy
Table 2
Electricity generated based on source (as on October 2014).
Fuel MWa Percentage sharea
Total thermal 176,779 69.58
Coal 152,971 60.21
Gas 22,608 8.90
Oil 1200 0.47
Hydro (renewable) 40,799 16.06
Nuclear 4700 1.85
Renewable energy sources 31,692 12.47
Total 2,28,722 100.00
a
Ministry of Power-GoI (2014).
demand and supply. There are large points in the infrastructure
network that have concentration economic activities. As a result of
resource constraints, (large energy demand, primary energy sup-
ply constraints, limited processing capacity, etc.) these large points
or nodes are already overworked and therefore their unin-
terrupted functioning is critical for seamless energy supplies to
other sectors and regions. Thirdly, because of high economic
concentration, the inherent exposure level of these large points is
high, which in case of an actual climate-induced event can affect
many sectors of the economy and create concerns for energy
security.
2.2. Climate risks for energy infrastructure in India
Climate change, especially extreme events, poses various types
of risks for infrastructure assets that represent a potential for loss.
Kapshe et al. (2003) formulated a reverse impact matrix to high-
light the impact of environmental variables on project variables.
When the extremes cross a threshold, the functional form of da-
mage functions changes as tails become fat and have a sizeable
probability mass (Weitzman, 2007, 2011). Risks are critical for the
“large nodes” in the energy infrastructure network as they have
high exposure levels due to concentration of activities.
Physical risk
The risk of physical damage to an asset due to a climate-in-
duced event is the most common form of risk conceptualization
and is based on the location and nature of the project. Many ultra-
mega power projects running on imported coal in India are located
on the coast to minimize transportation costs. Increased frequency
of cyclonic disturbances in the Arabian Sea (Murakami et al., 2013;
Murakami and Sugi, 2010) poses risks for the new projects like the
Udupi Power Plant (Udupi), Tata Power, Adani Power (Mundra),
etc.
Supply chain
The allied effect of physical risks is supply chain risks. Adani
Power and Tata Power have a combined capacity of 8220 MW
adjoining the Mundra port (Power-Technology, 2013). Cyclone
induced physical damages at Mundra can create electricity supply
shortages in Gujarat, Maharashtra, Rajasthan, Punjab, and Har-
yana, with whom power purchase agreements have been made.
Product-technology risks
Energy infrastructure assets have to keep upgrading with
newer technological developments in their respective domains to
deal with climate change. However, infrastructure lock-ins allow
only incremental changes. Policy regulations do not incentivize
total upgrade as the fines may not be strictly imposed or they are
not enough to make total upgrade viable.
Regulatory risks
New global and national regulations may add additional fi-
nancial burden to the energy infrastructure. For example, building
dykes of specific standards for coastal energy infrastructures like
refineries and thermal power plants could be mandated by the
regulators, and can change the financial model of the asset. In an
extreme situation, the asset owners may quit functioning than to
pursue requirements of newer regulations due to the excessive
financial burden of regulations.
The projections for climate variables in India show that there
is a wide range of future states of climate that infrastructure
assets will face. Changing climate has always been a part of the
earth's systems but problems begin when this pace of change
is accelerated and systems cannot react to abrupt changes.
Representative Concentration Pathways (RCPs) capture four
 A. Garg et al. / Energy Policy 81 (2015) 226–238 229

Table 3
Electricity generated (gross) in BKWH.

Categorya 2005–2006 2006–2007 2007–2008 2008–2009 2009–2010 2010–2011 2011–2012

Hydel 102 114 120 110 104 114 131

Thermal 505 538 585 616 677 665 709
Nuclear 17 19 17 14 19 26 32
Sub-total (utilities) 624 671 723 741 800 806 871
Non-utilities 74 82 91 100 106 121 128
Grand total 697 753 813 841 906 926 1000

Deficitsa,b
Peak deficit (MW) 11,463 13,897 18,073 13,024 15,157 12,031 13,815
Peak deficit (%) 12.3 13.8 16.6 11.9 12.7 9.8 10.6

a
Central Electricity Authority-GoI (2011).
b
Ministry of Power-GoI (2013).

pathways for global warming potential of future greenhouse gas
emission projections in the world (IPCC, 2013a). Projections for the
world and India show an increase in the frequency and intensity of
climate-induced events. In India, the annual mean air temperature
is projected to increase by around 1.7–2 °C by the 2030s (MoEF,
2010; Chaturvedi et al., 2012) for the Business as Usual (BAU)
scenarios, and could be as high as 3.3–4.8 °C by the 2080s com-
pared to the pre-industrial phase for India (MoEF, 2010). Coupled
with an increase in the average temperatures, more hot extremes
as compared to cold extremes are likely to occur (MoEF, 2010).
This will have direct impact on energy requirement for heating
and cooling and will further put a stress on the already con-
strained energy infrastructure assets.
The frequency and intensity of heavy precipitation events is
also likely to increase in the near term but will appear more sig-
nificantly towards second half of the 21st century. Towards the
end of the century, the monsoon period is projected to increase
with an early onset and a delayed retreat. The volume of the
rainfall in the monsoon period is projected to show an increasing
trend (IPCC, 2013a). Kumar et al. (2006) also forecast a decrease in
the number of rainy days expected over much of India, along with
increased frequency of heavy rainfall during the monsoon period

Fig. 2. Energy infrastructure spread for coal (2013).

Source: MoSPI (2013a).
230 A. Garg et al. / Energy Policy 81 (2015) 226–238
Fig. 3. Energy infrastructure spread for crude oil (2013).
Source: MoSPI (2013a).
by the 2030s. Chaturvedi, et al. (2012) project an increase in the
frequency of extreme rainfall for the decades after the 2060s in
RCP 6.0 and RCP 8.5. These are likely to increase the physical and
supply chain risks for the energy infrastructure.
Similarly, cyclones are creating risks for coastal energy infra-
structure, most of which has come up in the recent years. For
example, in the end-century cyclones in the Asia-Pacific region are
projected to witness more intense storms (Murakami et al., 2013;
Emanuel, 2013). In the North Indian Ocean, the Bay of Bengal and
the Arabian Sea show contrasting projections. According to Knut-
son et al. (2010), the fractional changes in tropical cyclone fre-
quency in the region vary from  52 to 79 percent with their
present-day control simulations. Murakami et al. (2013) project a
46 percent increase in tropical cyclone frequency in the Arabian
Sea for the last quarter of the century (2075–2099). Even if we
know the overall projections for the Indian Ocean, they may have
different implications for a port in the western and eastern regions
of India. The projections for climate variables globally as well as for
India indicate that there is a wider than current range of climatic
variability that the energy infrastructure assets have to be pre-
pared for. These conditions are creating newer risks and enhancing
the ones that already exist. Therefore, there is a strong need to
study important energy infrastructure assets and adopt appro-
priate risk management plans for them proactively. Climate
proofing energy infrastructure assets would require the upcoming
assets to keep in mind the possible extremes and increases in
averages while planning for them. Along the same line, the ex-
isting energy infrastructure must also plan for upgrades in order to
live up to the planned life span to be more resilient. We study two
such infrastructures now.
3. Case of Kandla port
3.1. Overview and climate concerns for Kandla port
Gujarat has India's longest and has 40 minor ports and 1
(Kandla) major port that are well connected by rail-road to the
Delhi–Mumbai corridor (Aiyar, 2008; Bhatt, 2009). An adminis-
trative definition classifies major ports as those that are under the
central government administration and minor ports as those un-
der the state government administration (De, 2009). Kandla port is
one of largest ports of India and tranships about 30 percent of the
total petroleum and oil lubricant (POL) products handled by the 12
major ports of India. The Vadinar off shore facilities of Kandla port
handled 54 million tonnes of POL in 2013 (Fig. 4). The terminal has
Essar and Reliance refineries in the vicinity. A pipeline network
connects Kandla to the refineries in the northern states of Punjab
and Haryana. All these factors makes Kandla port an important
hub for energy infrastructure in India. Damage to port, either in
 A. Garg et al. / Energy Policy 81 (2015) 226–238
Fig. 4. Commodity traffic at Kandla port.
Source: Indian Ports Association (2013).
the form of physical damage infrastructure assets or in the form of
operating time loss due to hindered operations at the Kandla port
can create disruptions in supply of primary energy to many parts
of India.
The changing frequency and variability of climate variables is
likely to create concerns for the port. Adverse weather phenomena
disrupt the operations at the port and in severe cases, can cause
damage to the physical infrastructure. Kandla port is vulnerable to
high wind speeds and cyclonic disturbances. In the event of wind
speeds exceeding 35 kmph, the taller structures have to be closed,
to avoid any damage to the port infrastructure or to human life.
The Regional Specialized Meteorological Centre (RSMC), New Delhi
has a set guidelines for port operations the event of cyclonic dis-
turbances (RSMC-IMD, n.d.-a). Kandla port maintains an ob-
servatory for monitoring elementary weather phenomenon but
cyclone warnings are issued from the regional meteorological
centers (RSMC-IMD, n.d.-b).
Cyclonic disturbances pose damage risks for stored goods
specially food grains and fertilizers. In 2012–2013, Kandla port
handled a cargo of 93.6 million tonnes and therefore a large vo-
lume of goods is exposed. Supply disruption problem is severe in
the case of petroleum, oil, and lubricants (POL) because of the
large volume transhipped at the port, as it has implications for
Indian energy security. This was experienced in the June 1998
Porbandar cyclone, that washed away approach roads to the port;
damaged storage tanks, pipelines, and other critical infrastructure
(Sibal and Jadeja, 2011). The government had to resort to emer-
gency management of crude oil by diverting oil to refineries in the
north from other ports to meet the shortage (Business Standard,
1998).
3.2. Vulnerability of Kandla port
3.2.1. Conceptualization and methodology
Vulnerability to climate change is the degree to which a system
is susceptible to i.e. propensity or predisposition to be adversely
affected, adverse effects of climate change including climate
variability (IPCC, 2013b; MoEF, 2010). Vulnerability also has a
component of consequence and is a function of hazard and ex-
posure (Shukla et al., 2003). The hazard in this case is the damage
potential of cyclonic disturbances while the exposure levels de-
pend on the infrastructure and the concentration of economic
activities at the port. As Kandla port because is a “large point
origin” for POL supplies to the country, therefore it is an important
energy infrastructure node in the overall energy network of India.
231
Extended interruptions can have serious implications on the en-
ergy security of India.
The vulnerability of Kandla port can be defined as a function of
three categories of variables, namely, system condition variables
(SCV), sustainable development variables (SDV) and climate
change variables (CCV) (Garg et al., 2007). System conditions are
intrinsic factors that determine the resilience and exposure of the
system. Likewise, sustainable development variables are external
factors that contribute to system resilience. Depending on their
nature, some of the variables from these two categories are flex-
ible and can be altered to some extent, while others like the lo-
cation can be inflexible. Climate change variables are external to
the system and encompass the hazard.
Extending this methodology proposed by Garg et al. (2007),
with each of these variable categories, the vulnerability of the port
to climate change can be conceptualized. The exposure to the port
is due to cyclonic disturbances, which is the critical climate change
variable for the port. These cyclonic disturbances can lead to a
range of losses, from losses in operating hours to losses due to
physical damage. The important sustainable development variable
that emerges from literature and interaction with experts in this
case is the mangrove cover surrounding the coast. Mangroves are
known to protect flooding from cyclones and reduce their impact
by breaking the strength of the winds and waves, stabilizing se-
diments and reducing shoreline erosion, and by hindering floods
(Hirway and Goswami, 2007; Parthasarathy and Gupta, 2014).
There are many system condition variables that determine the
resilience of the port. Important ones include: proper infra-
structure planning, road connectivity to the interiors, strength and
mass of tall structures likes cranes to withstand the expected
strong parameters, warehousing, communication, insurance, etc.
We take insurance as a system condition variable, which is a
palliative risk management mechanism as the premiums are de-
termined by market mechanisms, which tend to take into account
the factors determining system's resilience. With one variable in
each class, we can define the vulnerability of the port in the form
of an index. This vulnerability index quantifies and represents the
vulnerability of the port in one number. The following metho-
dology has been used to formulate the index:
n
Vulnerability index (VI) = fn(CCV, SCV, SDV) = ∑ wi × Yi
i (1)
where
(Xi − Min Xi )
Yi =
(Max Xi – Min Xi ) (2)
1
wi = m × 2 Var(Yi) m = n
∑ 2 Var(Yi)
i (3)
X1: operating hours lost due to a cyclonic disturbance (CCV).
X2: percentage of area under mangrove cover (SDV).
X3: percentage of assets covered under insurance (SCV).
The base case, probability distributions are based on historical
data. The data on mangrove cover data is from State of Forest
Reports published by the Forest Survey of India (1991–2011). The
distribution of operating hours lost is based on the weather
warnings issued during the last 20 years by the Regional Me-
teorological Centre, Ahmedabad to the Kandla port. The percen-
tage of assets under insurance is based on annual reports. The
future probability distribution for mangrove cover is based on
expert interaction. The future probabilities for cyclonic dis-
turbances in the Arabian Sea have been taken from published
literature. The proportion of assets under insurance cover has been
232 A. Garg et al. / Energy Policy 81 (2015) 226–238
Fig. 5. Vulnerability index projections.
Source: Author's estimates.
assumed to remain the same. Probability distributions for each of
these variables were created for the base case and for the future. A
2000 trial Monte Carlo simulation was done to simulate various
possible values of these variables for future scenarios of RCP 4.5,
RCP 6.0, and RCP 8.5 for mid-century (2021–2050) and end-cen-
tury (2071–2100).
3.2.2. Results
The following graph shows the spread of the vulnerability in-
dex (VI) for various scenarios (Fig. 5). The mean value of the index
is shifting to the right indicating a shift towards higher values of
the vulnerability index in the end-century. The maximum pro-
portion of the vulnerability index values lie in the range of 0.4–0.6
in the base case, which shifts to the 0.3–07 range in the end-
century scenarios. Similarly, the mean value has also increases
from 0.467 in the base case to 0.563 (RCP 6.0) and 0.588 (RCP 8.5)
in the end-century. For the extreme scenarios, a larger proportion
of the vulnerability index lies in the higher values evident from the
fat right tails. The fat right tail of RCP 8.5 end-century dominates
the tails of the other scenarios. The right shift of peaks and med-
ians indicates an increasing vulnerability in the extreme scenarios.
Small intervals show that the spread of the Vulnerability Index is
greater in the extreme scenario showing that the range of possi-
bilities of the Index values is higher in the extreme scenarios to-
wards the end-century.
The pattern of this index shows that there will be interruptions
in the port operations. Being a critical port, even short-term in-
terruptions can create supply chain risks. In the event of a cyclone
in the western coast, the scope of channelizing crude oil from
other neighboring ports to ease the supply crunch is also re-
stricted. Reserves can meet the demand only for some days and
supplies must be restored soon to avoid any energy security issues.
This methodology of index building through simulation takes
into account the present and future probability distributions of the
selected variables. It therefore gives a general trend of the vul-
nerability based on which directions for policy pathways can be
chosen. This methodology is also appropriate as representative
variables have been selected from each class of variables high-
lighted by Garg et al. (2007) and holds good when there are a
limited set of representative variables. It can be replicated for
studying socio-economic vulnerability. The use of distributions
also makes it possible to develop various scenarios. The limitation
of this index building methodology is that it gives general trends
and not specific numbers unlike other statistical tools such as re-
gression or economic methods like cost benefit analysis.
4. Case of Konkan Railways
4.1. Overview and climate concerns for Konkan Railways
Indian Railways network is an import network provider for
transportation of primary energy. Coal transportation contributed
41 percent to the total freight revenue and 300 million tonnes of
coal out of the 455 million tonnes transported by Indian railways
was for use by thermal power plants (Indian Railways, 2011). The
changing climate poses threats of various types of extreme events
that can hinder the operations of the railway network. These
events are location specific and in this section we detail the cli-
mate vulnerability of Konkan Railways.
Konkan Railways is a prominent railway line on the western
coast of the Indian Railways and transports over a million ton each
of POL and coal freight annually (KRCL, 2014; Ministry of Finance-
GoI, 2009). The 762 km railway line passes through the hilly ter-
rain of the Western Ghats, which is one of the heaviest rainfall
prone regions in India (KRCL, 2014). The annual average rainfall
from the period 1998–2012 was close to 4000 mm (KRCL-RTI,
2011a). Extreme rainfall events in the past have led to frequent
interruptions in train operations due to boulder falling and soil
slippage occurrences (BFSS) particularly in the monsoon season
when 90 percent of the season's rain occurs (KRCL, 2013). The
railway line caters to the requirements of many power plants
along the coastal belt of the three states of Maharashtra, Goa, and
Karnataka (Ranade, 2009). It also provides connectivity from ports
to power plants like the Udipi Power Corporation Limited that run
on imported coal. Konkan Railways does not have a double line on
the entire stretch therefore, in case of any extreme event, the
supplies are interrupted and there is no other parallel network to
fill the supply gaps (Mid Day, 2014). The year 2014 also witnessed
a series of disruptions in transport of goods after a train derail-
ment due to falling boulders in the monsoon season (Mid Day,
2014). Fig. 6 shows the spread of BFSS incidents from 1998–2011
and more than two-thirds of them occur in the monsoon
months of June to September. These incidences hinder train op-
erations prevent smooth movement of energy freight to proces-
sing sources.
 A. Garg et al. / Energy Policy 81 (2015) 226–238
Fig. 6. Monthly spread of BFSS (1998–2011).
Source: KRCL-RTI (2011b).
Fig. 7. Probability of extreme rainfall days.
Source: Authors' estimates.
These BFSS incidents do not have a uniform spread and have a
high occurrence rate in the northern districts of Ratnagiri and
Sindhudurg, which witnessed more than 50 percent of the total
incidents. Each incident is a potential case of major accident and
traffic interruptions. There were 7 major freight train accidents
and 25 passenger train accidents during this period, which implies
that 3.37 percent of BFSS incidents have the potential to turn into
fatal accidents (KRCL-RTI, 2011c). The railway track has been
carved out of the mountainous Western Ghats region and the
mountain cuttings are prone to boulder fallings during the mon-
soon (KRCL, 2013). The red laterite soil of the mountains is porous
and transfers water to the clay layer beneath it, which then exerts
pressure on the loose boulders to fall (Ranade, 2009). There are
564 such mountain cuttings of height 1–50 m along the railway
track.
The stability of slopes and strength of soil is rainfall dependent,
which is the critical climate variable in this case. Nagrajan et al.
(2000) have identified that the threshold for rainfall before it can
cause damage is to the tune of 200 mm in 24 in the region. We use
this threshold to see the probability of days when this threshold
will be surpassed for future scenarios. Fig. 7 shows the future
probability of rainfall when it is greater than a specified threshold
for a representative grid on the railway track for mid-century
(2021–2050) and end-century (2071–2100). This data is statisti-
cally downscaled data from the Indian Institute of Technology,
Gandhinagar and is a seven model ensemble.
233
This figure shows that rainfall thresholds will be crossed many
times in future. The probability spread for days that cross rainfall
thresholds in the range of 150–350 mm per day shows a great
difference among the scenarios. This implies future scenarios will
have more rainy days and more of them fall in the rainfall range of
150–350 mm in 24 h. The extreme scenarios will witness much
more rainfall in this range. Similarly, the tails are trailing indicat-
ing that there is a possibility of very extreme rainfall cases as well.
The present system is designed in a way that it has a threshold of
200 mm of rainfall in a day. In future if there is a shift in the
average rainfall in the range of 150–350 mm, then the system's
threshold will have to be pushed to surpass this range. Konkan
Railways will also have to consider if it should enhance the sys-
tem's resilience for a higher threshold of 150–350 mm where a
higher proportion of rainy days will fall or go an extra mile and
account for really extreme cases of 800 mm of rainfall per day. The
system's resilience in future will have economic implications for
electricity supply, refineries, and ports dealing in energy products
located in the region.
4.2. Vulnerability of Konkan Railways
4.2.1. Conceptualization and methodology
Konkan Railways is a very dynamic track and many variables
affect the vulnerability of track to climate change. The variables
discussed above can be classified as system condition variables
234 A. Garg et al. / Energy Policy 81 (2015) 226–238
(SCV), sustainable development variables (SDV) and climate
change variables (CCV) (Garg et al., 2007). The state of these
variables establishes the vulnerability of the railway track. As
discussed above, climate-induced risk is a function of hazard and
vulnerability (Shukla et al., 2003). Risks represent a potential of
loss and therefore, a function representing potential damage de-
pends upon the probability of occurrence of a hazard and how
much loss the hazard can bring about. This can be split into two
parts. In the first part, the probability of the hazard is identified
and in the second part, the loss per event is identified. The prob-
ability of hazard is the probability of the occurrence of BFSS.
Climate induced risk = fn(hazard, vulnerability)
P(hazard) = fn (SCV,i SDV,j CCVk)
Expected loss = P (hazard) × loss per hazard
In this specific case, the hazard is the rainfall induced BFSS
incident. The number of times this threshold is breached is more
important as compared to the precise magnitude of the rainfall as
rainfall beyond 200 mm in a day has a high damage potential
(binary variable). Therefore, the climate change variable is taken as
days with more than 200 mm of rainfall. An important sustainable
development variable is the percentage of area under forest cover
in the region. Forests hold the soil and prevent incidents of BFSS.
Two important system condition variables are location in the
north or south (binary variable) and the amount of investment in
geo-safety works at constant prices made in the previous year.
These variables emerge from interaction with experts on railways
and Konkan region and from existing literature on Konkan Rail-
ways. Vulnerability is location specific and in this case a general
classification into north and south has been made as the number
of BFSS incidents is higher in the north. Similarly about 79 percent
of the total investment in geo-safety works goes into flattening
and stabilizing of slopes and cuttings and holding soil, which, di-
rectly arrests BFSS. The risk function assesses the incidence of
boulder falling and soil slippage for a location, given that the
rainfall has crossed threshold. Therefore, the BFSS risk can be
formulated as
logit (BFSS) = β0 + β1CCV + β2SDV + β3SCV1 + β4 SCV2
4.2.2. Results
The logistic regression output based on actual data since the
inception of the railway network is as follows:
logit (BFSS) = − 1. + 1.187 × rainfall (1, 0)
+ 0.716 × location (1, 0) − 0.162 × geosa fety
− 0.029 × forest
The odds ratios give a sense of how the presence of one vari-
able affects the presence or absence of another variable. The odds
ratio for rainfall is 3.28, that is, the rainfall crossing threshold is
3.28 times more likely to cause an incident of BFSS. The odds ratio
Table 4
Preventive costs.
Average costs per grid for 1998–2011 (2004–2005 North South
prices)
Geo-safety worksa 14,858,220 15,491,249
Bridges & tunnels maintenance costsb 7,635,217 3,758,720
Permanent wayb 21,515,231 23,428,886
a
KRCL-RTI (2011a).
b
KRCL-RTI (2011b).
Table 5
Palliative costs.
Average costs per grid for 1998– North South
2011 (2004–2005 prices)
Passenger Goods Passenger Goods
Accidentsa 19 1 6 6
Damages and compensationa 8,904,489 360,000 986,333 2,413,533
Train operations lossa 348,950 47,973 345,270 44,602
Salariesb 75,230 59,400 68,528 91,064
Total 9,328,670 467,373 1,400,131 2,549,199
a
KRCL-RTI (2011a).
b
(4) KRCL-RTI (2011b).
(5) for the location variable is 2.05 or being in the northern region of
railway line is 2.05 times more likely to cause BFSS. Similarly,
(6) higher investment in geo-safety is 1.12 times more likely to cause a
non-event.
The second part of determining the expected loss is the cal-
culation of losses per accident, which can be classified preventive
and palliative costs. Preventive costs are those that are incurred to
avoid adverse impacts, while palliative costs are post event costs
that borne to restore normal operations. Investment in preventive
geo-safety works has historically been shown to significantly ar-
rest incidents of BFSS (KRCL-RTI, 2011b). Preventive maintenance
costs on permanent way and on bridges and tunnels have to be
incurred regularly and they influence safety in the overall opera-
tions of the railway line.
Tables 4 and 5 show the pattern of investment in preventive
and palliative measures taken up by Konkan Railways. Specific
preventive maintenance costs in geo-safety works were to the
tune of Rs. 14.5–15.5 million (2004–2005 prices) per grid where
each grid represents the section of railway line that passes through
0.5° latitude and 0.5° longitude. The above function shows that
investing in geo-safety works and increasing the green cover can
reduce the vulnerability of the region. Fig. 8 shows various com-
binations of investment in geo-safety works and area under forest
cover that will maintain the existing levels of BFSS probability.
This gives possible combinations of geo-safety works and forest
cover to maintain the existing risk levels. It is assumed that the
forest cover will not drastically change and therefore from the 20
(7)
year average, an operating range for the forest cover has been
derived. The normal expenditure made in the base case for geo-
safety works will be far less than the incremental climate-induced
expenditure. For the RCP 4.5 mid-century the range is Rs. 19.83–
23.63 million per grid per year (2004–2005 prices) in the north for
46–47 percent forest cover.
The costs of building a green belt around the railway track and
undertaking intensive plantation in order to maintain the safety
(8) levels are Rs. 1.22 million per grid (2004–2005 prices) in the north
for RCP 4.5 mid-century and Rs. 27.18–32.23 million per grid in
RCP 8.5 mid-century. For the Rs. 4 million incremental invest-
ments in geo-safety work investment per grid, the trade-off is
1 percent of forest cover. These greening costs have been calcu-
lated on the basis of the investment done by Konkan Railways in
vetiver plantations. However, factors like time period for benefits
are not the same in these strategies. Geo-safety work is expensive
but provides immediate protection while forest cover provides
sustained protection but takes time to show results. Nevertheless,
the positive spill over of an increase in forest cover has many
sustainable development co-benefits that cannot be ignored.
This methodology involving logistic regression and expected
loss calculations separates the losses from occurrences of BFSS.
Therefore, given a specific set of conditions the regression equa-
tion gives the risk potential or the probability of occurrence. The
consequential losses are separated from this equation. The trade-
 A. Garg et al. / Energy Policy 81 (2015) 226–238
Fig. 8. Trade-offs between investing in geo-safety works and forest cover.
Source: Authors' estimates.
off curves provide an estimate of various combinations of sus-
tainable development variable and system condition variable to
maintain a specific risk level. This methodology can as well be
extended to study vulnerabilities and trade-offs among adaptive
strategies for infrastructure assets. The limitation of this metho-
dology is that it is extending the present conditions into future.
Therefore, this analysis has relevance only to the extent that it
helps prioritize current adaptive actions. Due to limitations in
availability of data, the location has been taken as a binary vari-
able; however, if specific locational details are available, multi-
nomial analysis is possible.
5. Discussion
The two cases presented above show the vulnerability of im-
portant energy infrastructure assets in India related to the trans-
port sector. They are important nodes in the energy network and
must function seamlessly. The vulnerability of the Kandla port to
cyclonic disturbances increases by 21 percent in RCP 6.0 end-
century scenario and 25 percent in RCP 8.5 scenario w.r.t. to the
base case. Similarly, for Konkan Railways, just to maintain the
existing levels of the BFSS probability, an annual expenditure on
geo-safety works will be close to Rs. 500 million in RCP 4.5 end-
century scenario and Rs. 1300 million in RCP 8.5 end-century
scenario as compared to the present expenditure of Rs. 180 million
(2004–2005 prices). This higher amount under more severe cli-
mate change (RCP8.5) is around 10 percent of initial capital ex-
penditure to build this railway line, a huge financial burden to
climate proof an existing asset. Our analysis shows a trade-off
wherein green cover contributes to holding loose boulders and
preventing soil slippage, and is a cheaper alternative to investment
in geo-safety works in the long-run. The immediate vegetation
next to the track is very effective; however, vetiver grass planta-
tion was carried out only once since the inception of the railway
line.
This analysis aims to highlight the preventive risk management
actions by having a balance in the sustainable development vari-
ables and the flexible system condition variables can play a sig-
nificant role in climate proofing infrastructure. This generates a
paradoxical situation wherein these energy infrastructures be-
come more susceptible to climate change concerns due to their
very creation and existence. For instance, Kandla port construction
235
has resulted in part destruction of mangroves in the region that
could provide one layer of protection against cyclones, and Konkan
Railway construction has increased human settlements around the
railway stations and railway corridor wherein deforestation hap-
pens and could promote more BFSS cases. The assets, therefore,
have to be aligned with the broader area development to manage
the climate change risks better, and we have shown how sus-
tainable development and flexible system condition variables can
contribute to reducing risks.
Although, sustainable development variables emerge as strong
adaptive risk management, they have a longer time frame and
cannot solve all problems pertaining to climate-induced risks.
Risks can only be managed and cannot be completely eliminated.
The palliative financial burden, as shown in Table 5, is high and its
economic implications can only be evaluated till the first or the
second order. Therefore, the total direct impact may be lower than
of the actual losses that many sectors may face. The choice of the
right adaptation practice may not always be easy to determine as
the costs are unambiguous and but not the potential losses pre-
vented. Consequently, it becomes important to plan for potential
climate-induced risks keeping in view other factors like the time
frame for results in case of a particular adaptive practice or costs
for inducing adaptive measures, risk coverage etc.
Another policy decision is on how much risk to cover and how
much to leave uncovered, i.e. determining the right balance be-
tween preventive and palliative adaptation. We could term this as
adaptation dilemma. In the cases above we have focussed only on
preventive adaptation, and climate change adaptation measures
heavily depend on the risk perception and the carrying capacity of
the infrastructure. Managing risks through adaptation is an ex-
pensive proposition. According to the Draft TFYP of the Govern-
ment of India (2013), adaptation costs for new infrastructure could
be in the range of 3–10 percent of the total investment, although
for certain sectors and locations this may be higher (Planning
Commission-GoI, 2013). This translates into adaptation costs for
new assets alone to the tune of 6–20 billion USD each year for the
period 2012–2017. This number for existing infrastructure is likely
to be as high as 25 percent of the present costs and could run into
several billion dollars considering the annual gross fixed capital
formation in India was Rupees 20020 billion in 2012–2013 (at
2004–2005 prices) (MOSPI, 2013b) or 1424.7 billion USD in 2011 in
terms of purchasing power parity (World Bank, 2014).
While owners of energy infrastructure assets have to carefully
236 A. Garg et al. / Energy Policy 81 (2015) 226–238
plan climate proofing the assets, the level and extent of this
adaptation is an important decision variable. Mal-adaptation could
lead to excess expenditure (due to more than desired preventive
actions) or carrying excess risk (not taking enough preventive
actions). Excessive adaptation and over estimation of risk leads to
type 1 or α error. Under investment in risk mitigation and adap-
tation strategies leads to type 2 or β error. The policy dilemma
therefore is how much to invest in adaptation. A higher risk per-
ception and estimation by insurance agencies would lead to an
increase in premium amounts, while a similar perception by asset
owners could additionally lead to increased investment in making
the asset more resilient. Climate proofing infrastructure does not
mean that all possible risks are eliminated; it just implies that the
infrastructure asset is more resilient towards climate-induced
risks. For Kandla port, it translates into improper judgement by the
insurance companies, which would reflect in the premium prices.
Similarly, the premium prices and risk estimation at the port's end
would determine the coverage. For Konkan Railways excessive
preventive adaptation costs would mean more than required in-
vestment in geo-safety works in trying to completely proof the
system, which would result in sunk costs. Under adaptation means
that risks are left uncovered. Thus the adaptation dilemma re-
volves around choosing an acceptable level of risk from a wide
spectrum.
6. Conclusion and policy recommendations
The paper has highlighted the climate change risks faced by
transport nodes in Indian energy infrastructure through two ex-
amples. Energy infrastructure assets have two key features; first,
they are long life assets; and second, they are exposed to climate.
Whereas each energy infrastructure has its own purpose and
function, the energy services are delivered by infrastructures
forming a network. Climate risks from each energy infrastructure
may therefore spill-over the entire network and beyond. The
owners of a specific infrastructure may take measures to mitigate
risks at their end. However, appropriate policy measures are
needed to mitigate the risk spill-overs that add to systemic risk,
which affect vital national sustainability indicators like energy
access and energy security.
Risk profiles of infrastructure assets change with time and
under various climate change scenarios. The burden of preventive
adaptation costs is much higher in extreme scenarios and has
trade-offs based on the choices. In Konkan Railways, our analysis
shows that there is a difference of about 108 million rupees in geo-
safety expenditure between the RCP 4.5 and RCP 8.5 mid-century
scenarios to maintain the existing risk levels for the entire track i.e.
a difference of 8 million rupees in the north (and 10 million rupees
in the south). Similarly, there are trade-offs between sustainable
development variables and system condition variables. This im-
plies that the project level risk and vulnerability assessments have
to be carried out and two have to be carried out on a more regular
basis. The risk profiles of future depend not just on the future
climate scenario but also on the investment decisions made today.
The paper, through the second case study of Kandla port, has
demonstrated that vulnerability and risk assessment vary de-
pending on the type of asset. The vulnerability index and the in-
struments for risk assessment hence need to be developed and
customized for specific projects and monitored regularly. Reverse
impact matrix creation, identification of critical climate change,
system condition, and sustainable development variables is a key
task in customizing vulnerability index. Project assessments have
to evolve into multi-criterion assessments as even generic classes
of energy infrastructure assets will face different types of risks
based on their location, connectivity with other energy
infrastructure assets in the network, volume of economic activity
etc. The complex nature of reverse impact on projects cannot be
confined to one criterion assessment. As demonstrated in the case
of Kandla port, the insurance cover emerges as a suitable instru-
ment that captures the vulnerability of the project dynamically so
as to optimally adapt to climate change related risks. The index
also captures other variables that influence the vulnerability of the
port. It would be advisable for the government to make such as-
sessment mandatory for all large energy (and also other) infra-
structure projects since energy network interruptions are ex-
pensive and impede energy security and access.
For high investment assets such as energy infrastructure, risk
management has to be preferably driven by preventive adaptation
with lesser reliance on palliative measures due to very high damage
costs. Instruments such as Catastrophe Bonds could also be de-
ployed for risk management. The monetary compensation in the
event of a major damage to the node may at best add a little mo-
mentum to the restoration process. A broader energy security
strategy has to be planned at a sub-regional level for such energy
nodes since a climate change induced disruption at a major energy
importing port such as Kandla could create major challenges. In the
case of Konkan Railways, we can see that the palliative costs in the
base case are high. More than the monetary losses, it is the spill
over of losses due to disrupted operations and loss of life that
concerns the authorities. Preventive adaptation expenditure has
pivotal role for safe operations in future given the uncertainty in
future climate and the broad range of weather phenomenon that
energy infrastructure assets are likely to face. Konkan Railways has
brought down the accident rates by investing in geo-safety works.
Even to maintain the existing risk levels the preventive expenditure
can be as high as Rs. 500 million for mid-century RCP 4.5 scenario.
At a national level, this also implies that appropriate policy driven
incentives for insurance cover must be provided because risk
pooling is in the larger interest. The government should institute
policies that mandate hedging of systemic climate induced risks
and fair apportioning of hedging costs among the players. Besides,
the government may consider setting up climate change adaptation
fund which may tap into the UNFCCC's adaptation fund.
The Konkan Railways case shows that sustainable development
variables, like forest cover, have positive spill over effects. Existing
risk levels can be maintained through an incremental investment
of Rs. 4 million in geo-safety works or by increasing the forest
cover by 1 percent. Increasing the green cover is cheaper, but both
the interventions have different time lines as geo-safety works
gives immediate protection but green cover will give long-term
and sustained protection against heavy rainfall. Gaining positive
spill-overs from sustainable development variables would require
policies to engage local agencies and agents who will also gain
local co-benefits. Cooperation will need to be institutionalized for
continuous monitoring and correction. Besides mandates such as
compensatory plantations targets, the policies should incentivize
sustainable development measures that promote resilience of
critical energy infrastructure assets.
To sum-up, the following are the key policy recommendations:
i) mandatory vulnerability assessment of future climate change
risks for all existing and future energy infrastructures; ii) Using
conventional or developing appropriate tools and methodologies
for multi-criterion risk and vulnerability assessment such as de-
veloping a vulnerability index that includes project level risks and
systemic risks; iii) monitoring climatic parameters over the life
of the asset and corresponding implementation of preventive
adaptation measures through appropriate tools such as adaptation
funds or insurance for unmitigated climate risks; and iv) pro-
moting sustainable development actions along energy infra-
structure assets that enhance climate related resilience and
simultaneously deliver co-benefits to local agents.
 A. Garg et al. / Energy Policy 81 (2015) 226–238
In India, the government policies on environmental impacts
assessment of projects have remained one-sided in the sense that
they focus solely on assessing the impacts of the project on the
environment and not the reverse impacts of changing environ-
ment, for example, future climate change, on the project's per-
formance. The analysis in this paper shows that these policies
require an urgent relook with a view to recognize the impacts of
future climate change on long-life assets like energy infrastructure
assets and to institute proactive adaptation measures.
Acknowledgements
The research for this paper was supported from UNEP project
“Promoting Low Carbon Transport in India” which was funded by
the German Government (BMU) under the International Climate
Initiative” (Project No. 10_I_129_M_Low Carbon Transport). The
authors appreciate the valuable comments received from the
anonymous reviewers which helped in improving the paper.
References
Aiyar, S.S., 2008. The Benefits of Port Liberalization: A Case Study from India. 7.
Center Global Liberty and Prosperity: Development Policy Analysis, pp. 1–19 ,
http://object.cato.org/sites/cato.org/files/pubs/pdf/DPA7.pdf.
Asif, M., Muneer, T., 2007. Energy supply, its demand and security issues for de-
veloped and emerging economies. Renew. Sustain. Energy Rev. 11, 1388–1413.
Bhatt, B.K., 2009. The Gujarat Economy, 2nd ed. Vrinda Publications Ltd., New
Delhi.
Briceño-Garmendia, C. Estache, A., Shafik, N., 2004. Infrastructure services in de-
veloping countries: access, quality, costs and policy reform. World Bank Policy
Research Working Paper, 3468, December 2004, pp. 1–33.
Brown, C.E., 2002. World primary energy overview. In: World Energy Resources.
Springer-Verlag, Berlin, Heidelberg, New York, pp. 61–81.
Business Standard, 1998. Plans drawn up to face oil crisis. 〈http://www.business-
standard.com/article/specials/plans-drawn-up-to-face-oil-crisis-
198061201031_1.html〉, June 12, 1988 (retrieved 20.04.14.).
Central Electricity Authority-GoI, 2011. Growth of Electricity Sector in India from
1947–2011. Ministry of Power, Central Electricity Authority, New Delhi.
Chaturvedi, R.K., Joshi, J., Jayaraman, M., Bala, G., Ravindranath, N.H., 2012. Multi-
model climate change projections for India under representative concentration
pathways. Curr. Sci. 103 (7), 791–802.
De, P., 2009. Globalisation and the Changing Face of Port Infrastructure: The Indian
Perspective. Peter Lang AG, Berlin.
Emanuel, K.A., 2013. Downscaling CMIP5 climate models shows increased tropical
cyclone activity over the 21st century. Proc. Natl. Acad. Sci. USA, 1–6.
Forest Survey of India, 1991&;2011. India State of Forest Reports. Ministry of En-
vironment and Forests, Government of India, Dehradun.
Garg, A., Shukla, P.R., Kapshe, M., 2007. From climate change impacts to adaptation:
a development perspective for India. Nat. Resour. Forum 31, 132–141.
Hallegatte, S., 2009. Strategies to adapt to an uncertain climate change. Glob. En-
viron. Change 19, 240–247.
Hirway, I., Goswami, S., 2007. Valuation of coastal resources: the case of mangroves
in Gujarat. Academic Foundation and Centre for Development Alternatives,
Ahmedabad.
IEA, 2014. Special Report: World Energy Investment Outlook. International Energy
Agency, Paris.
Indian Ports Association, 2013. Major Ports of India: A Profile. Indian Ports Asso-
ciation, New Delhi.
Indian Railways, 2011. Indian Railways Annual Report 2010–11. Ministry of Rail-
ways (Railway Board), New Delhi.
IPCC, 2013a. Climate Change 2013: The Physical Science Basis. Contribution of
Working Group I to the Fifth Assessment Report of the Intergovernmental Panel
on Climate Change. Cambridge University Press, Cambridge (UK).
IPCC, 2013b. Climate Change 2014: Impacts. Adaptation, and Vulnerability; Con-
tribution of Working Group II to the Fifth Assessment Report of the Inter-
governmental Panel on Climate Change. Cambridge University Press, Cam-
bridge (UK).
ISPRL, 2014. About Us: ISPRL, Indian Strategic Petroleum Reserves Ltd. 〈http://www.
isprlindia.com/aboutus.htm〉, 11 March, 2014 (retrieved 30.09.14.).
Kapshe, M., Shukla, P.R., Garg, A., 2003. Climate change impacts on infrastructure
and energy systems. In: Shukla, P.R., Sharma, S.K., Ravindranath, N.H., Garg, A.,
Bhattacharya, S. (Eds.), Climate Change and India: Vulnerability Assessment
and Adaptation, 1st ed. Universities Press, Hyderabad, pp. 291–325.
Knutson, T.R., McBride, J.L., Chan, J., Emanuel, K., Holland, G., Landsea, C., Held, I.,
Kossin, J.P., Srivastava, A.K., Sugi, M., 2010. Tropical cyclones and climate
change. Nat. Geosci. 3, 157–163.
237
Kousky, C., Cooke, R.M., 2009. Climate change and risk management: challenges
for insurance, adaptation, and loss estimation. Resources for the Future,
pp. 1–29.
KRCL, 2014. About Us. Konkan Railways Corporation Limited. 〈http://www.konkan
railway.com/content/operating-statistics〉 (retrieved 20.04.14.).
KRCL, 2013. Operating Statistics. Konkan Railways Corporation Limited. 〈http://
konkanrailway.com/english/about-us/〉 (retrieved 20.10.14.).
KRCL-RTI, 2011a. Right to Information Application: KR/CO/PR/IA/11/16/47/393.
Mumbai, India.
KRCL-RTI, 2011b. Right to Information Application: KR/CO/PR/IA/11/16/52/398.
Mumbai, India.
KRCL-RTI, 2011c. Right to Information Application: KR/CO/PR/IA/11/16/57/403.
Mumbai, India.
Kumar, R.K., Sahai, A.K., Kumar, K.K., Patwardhan, S.K., Mishra, P.K., Revadekar, J.V.,
Kamala, K., Pant, G.B., 2006. High-resolution climate change scenarios for India
for 21st century. Curr. Sci. 90 (3), 334–345.
Kumar, U., Jain, V.K., 2010. Time series models (Grey-Markov, Grey Model with
rolling mechanism and singular spectrum analysis) to forecast energy con-
sumption in India. Energy 35, 1709–1716.
Kurian, A.L., Vinodan, C., 2013. Energy security: a multivariate analysis of emerging
trends and implications for South Aisa. India Q.: J. Int. Aff. 69 (4), 383–400.
Mid Day, 2014, August, 24. National News: Mid Day. 〈http://www.mid-day.com/ar
ticles/over-500-tonnes-of-food-grain-lie-on-tracks-as-train-derails/15553178〉
(retrieved 2.10.14.).
Ministry of Coal-GoI, 2011. Report of the Working Group on Coal and Lignite for
Formulation of the Twelfth Five Year Plan 2012–17. Ministry of Coal, New Delhi
(Retrieved from: http://planningcommission.gov.in/aboutus/committee/index.
php?about ¼ 12strindx.htm).
Ministry of Coal-GoI, 2012. Annual Report 2012–13. Ministry of Coal, New Delhi.
Ministry of Coal-GoI, 2013. Annual Report 2012–13. New Delhi: Ministry of Coal.
Retrieved from: 〈http://coal.nic.in〉.
Ministry of Finance-GoI, 2009. Economic Survey: Chapter 9. Energy, Infrastructure,
and Communications. Ministry of Finance, New Delhi (Retrieved from: 〈http://
indiabudget.nic.in/es2008-09/esmain.htm〉).
Ministry of Petroleum and Natural Gas-GoI, 2014. Indian Petroleum and Natural Gas
Statisitcs. Economics and Statistics Division, Ministry of Petroleum and Natural
Gas, New Delhi.
Ministry of Power-GoI, 2014, October. Power Sector Performance. Ministry of
Power. 〈http://powermin.nic.in/JSP_SERVLETS/internal.jsp#〉 (retrieved 1.11.14.).
Ministry of Power-GoI, 2013, October. Indian Electricity Scenario. Ministry of Power.
〈http://powermin.nic.in/JSP_SERVLETS/internal.jsp〉 (retrieved 20.04.14).
MoEF, 2010. Climate Change and India: A 4  4 Assessment – A Sectoral and Re-
gional Analysis for 2030s. Ministry of Environment and Forests, Government of
India, New Delhi.
MoSPI, 2013a. Energy Statistics 2013. National Statistical Organization, Central
Statistics Office. Ministry of Statistics and Programme Implementation-Gov-
ernment of India, New Delhi.
MOSPI, 2013b. CSO National Accounts Statistical Publications, 2014. Ministry of
Statistics and Programme Implementation, Government of India. 〈http://mospi.
nic.in/Mospi_New/site/Publications.aspx?status ¼ 1&menu_id ¼ 22〉 (retrieved
15.10.14.).
Murakami, H., Sugi, M., 2010. Effect of model resolution on tropical cyclone climate
projections. Sci. Online Lett. Atmos. 6, 73–76.
Murakami, H., Sugi, M., Kitoh, A., 2013. Future changes in tropical cyclone activity in
the North Indian Ocean projected by high-resolution MRI-AGCMs. Clim. Dyn.
40 (7–8), 1949–1968.
Nagrajan, R., Roy, A., Vinod Kumar, R., Mukherjee, A., Khire, M.V., 2000. Landslide
hazard mapping based on terrain and climatic factors for tropical monsoon
regions. Bull. Eng. Geol. Environ., 275–287.
Pargal, S., Banerjee, S.G., 2014. More Power to India: The Challenge of Electricity
Distribution. The World Bank, Washington D.C.
Parthasarathy, R., Gupta, M., 2014. Mangrove management and cyclone risk re-
duction in Kachchh, Gujarat. In: Shaw, R. (Ed.), Disaster Recovery: Used or
Misused Development Opportunity. Springer, Tokyo, Japan, pp. 301–316.
Peter, H., Grimm, T., 2008. Rising natural catastrophe loses: what is the role of
climate change. In: Hansjürgens, B., Antes, R. (Eds.), Economics and Manage-
ment of Climate Change: Risks, Mitigation and Adaptation. Springer Sciene &
Business Media, New York, pp. 13–22.
Planning Commission-GoI, 2013. Twelfth Five Year Plan (2012–17): Economic Sec-
tors. Government of India, Planning Commission. SAGE Publications India Pvt.
Ltd., New Delhi (Retrieved from: 〈http://planningcommission.gov.in/plans/
planrel/12thplan/welcome.html〉).
Power-Technology, 2013, October 10. The top 10 biggest thermal power plants in
India. Retrieved from power-technology Web site: 〈http://www.power-tech
nology.com/features/feature-the-top-10-biggest-thermal-power-plants-in-in
dia/〉.
Ranade, P.S., 2009. Infrastructure Development and its Environmental Impact:
Study of Konkan Railways. Concept Publishing Company, New Delhi.
RSMC-IMD, n.d.-a. Damage potential of tropical cyclone. Regional Specialised Me-
teorological Centre (Cyclone Warning Division, India Meteorological Depart-
ment). 〈http://www.rsmcnewdelhi.imd.gov.in/images/pdf/cyclone-awareness/
damage-potential-cyclone/damage.pdf〉 (retrieved 2.10.14.).
RSMC-IMD, n.d.-b. Port warning and signals. Regional Specialised Meteorological
Centre for Tropical Cyclones over North Indian Ocean 〈http://www.rsmcnew
delhi.imd.gov.in/images/pdf/cyclone-awareness/port-warning-signals/port-
warning.pdf〉 (retrieved 2.10.14.).
238 A. Garg et al. / Energy Policy 81 (2015) 226–238
Santhiyavalli, G., Usharani, M., 2011. Energy Infrastructure in India. Asian J. Res. Soc.
Sci. Humanit. 1 (4), 446–454.
Schenker-Wicki, A., Inauen, M., Olivares, M., 2010. Unmastered risks: from crisis to
catastrophe: an economic and management insight. J. Bus. Res. 63, 337–346.
Shukla, P.R., Sharma, S.K., Garg, A., Bhattacharya, S., Ravindranath, N.H., 2003.
Vulnerability and adaptation: challenges Ahead. In: Shukla, P.R., Sharma, S.K.,
Ravindranath, N.H., Garg, A., Bhattacharya, S. (Eds.), Climate Change and India:
Vulnerability Assessment and Adaptation, 1st ed. Universities Press, Hyderabad,
pp. 389–432.
Sibal, H.K., Jadeja, A.K., 2011, December. Operations management at Kandla Port in
the event of adverse weather phenomenon. (Naswa P., Interviewer).
Stern, N., 2007. The Economics of Climate Change: The Stern Review. Cambridge
University Press, Cambridge.
TERI, 2006. National Energy Map for India: Technology Vision 2030. TERI Press,
New Delhi.
Vivoda, V., 2010. Evaluating energy security in the Asia-Pacific region: a novel
methodological approach. Energy Policy 38 (9), 5258–5263.
Weitzman, M.L., 2007. A review of the Stern review on the economics of climate
change. J. Econ. Lit. XLV, 703–724.
Weitzman, M.L., 2011. Fat-tailed uncertainty in the economics of catastrophic cli-
mate change. Rev. Environ. Econ. Policy 5 (2), 275–292.
World Bank, 2014. Purchasing Power Parities and Real Size of World Economies: A
Comprehensive Report of the 2011 International Comparison Program. The
World Bank, Washington D.C.