ACRP - Austrian Climate Research Program

Power through Resilience of Energy Systems: Energy Crises, Trends and Climate Change (PRESENCE)

Extreme Events and Resilience Concept Working Paper for the First Review-Workshop 3rd September 2012

Matzenberger J., Kranzl, L., Totschnig, G., Redl, C., Schicker I., Formayer H., Gorgas T.

Austrian research cooperation Funded by Climate and Energy Fund

PRESENCE, First Review Workshop Seite 1 von 53 Contents 0 Introduction ...... 4 0.1 Starting point and motivation of the project ...... 4 0.2 Review Workshop Objective ...... 4 1 Methodological Framework for the concept of resilience (D3.1) ...... 7 1.1 Emergence of the resilience – concept ...... 7 1.2 Resilience, adaptive capacity, vulnerability? ...... 11 1.3 Definition of resilience in various disciplines ...... 14 1.3.1 Definition of resilience in ecosystem sciences ...... 14 1.3.2 Definition of resilience in engineering ...... 14 1.3.3 Definition of resilience in social sciences (communities) ...... 15 1.4 From concept to applicable resilience indicators ...... 19 1.5 Making the definition of resilience workable in the project PRESENCE ...... 22 2 Discussion paper on exogenous reference-scenarios and shocks of energy price and supply ...... 23 2.1 Introduction...... 23 2.2 Storylines and exogenous parameters in the Scenarios ...... 23 2.3 Primary energy prices and price/supply shocks ...... 25 2.4 Calculation of the household prices ...... 28 3 Extreme Events in Relation to the Energy System and Climate Change – First considerations and Open questions ...... 29 3.1 What can be considered an extreme event in terms of the Energy System?...... 30 3.2 Methods for investigating Extreme Events in Atmospheric Sciences ...... 31 3.2.1 Definition of extreme events: ...... 31 3.2.2 Changes of extreme events ...... 33 3.2.3 Description of extreme events by climate models ...... 33 3.2.4 On quantification and classification of extreme events ...... 33 3.3 Existing extreme events in the past (intensity, duration,…) – what can we learn?...... 35 3.4 How often will there be extreme events in the future? Or: how do the conditions change that create critical situations in the future energy system? ...... 35 3.5 The PRESENCE – approach to assess extreme events in the energy system ...... 36 3.6 Open questions: ...... 37 4 Method to identify extreme events in the power system with the simulation model HiREPS ...... 38 4.1 Identifying critical situations (Black outs) ...... 38 4.2 HiREPS Model description ...... 40 4.3 Typical model application cases ...... 41 4.3.1 Variability of RES-electricity generation ...... 41 4.3.2 Simulation of thermal power plants ...... 41 4.3.3 Simulation of future power prices ...... 42

PRESENCE, First Review Workshop Seite 2 von 53

4.3.4 Detailed simulation of hydropower ...... 43 Simulation of the market value of wind and solar power ...... 44 4.3.5 Simulation of the transmission grid limitations ...... 44 Literature: ...... 46 ANNEX 1 ...... 50 Resilience indicators in the energy sector ...... 50

PRESENCE, First Review Workshop Seite 3 von 53

0 Introduction

0.1 Starting point and motivation of the project

The current energy system is a major driver of climate change. At the same time, the energy system itself is affected by climate change. Some elements of energy supply will change the characteristics of its availability (e.g. hydro power) and a modification in energy demand will occur (e.g. heating and cooling). Therefore, simultaneous mitigation and adaptation has to take place. The energy system in the next decades will face fundamental restructuring. Climate mitigation scenarios show the requirements of shifting towards zero- and low-carbon energy solutions. The availability of fossil resources (first of all oil) as well as global conflicts might cause energy shortages leading to energy crises. Demographic and social changes as well as technology developments could lead to additional challenges and opportunities. These trends, possibly occurring energy crises and climate change partly are potential sources of heavy vulnerability of the energy system. The question arises how mitigation efforts, adaptation measures and responses to changing side conditions might be integrated.

The core objective of this project is to provide measures and pathways how to increase the resilience of energy systems in the view of climate change, possible trends and energy crises as well as the transformation of our energy system into a low- and zero carbon future for the Austrian case.

The core and final result of this project will be recommendations on pathways how to increase the resilience of the Austrian energy system in the light of climate change, demographic trends, technological change and eventual energy crises and shocks. These pathways will include adaptation measures for adapting to climate change while simultaneously contributing to climate change mitigation and taking into account essential exogenous trends and developments.

0.2 Review Workshop Objective

This document includes four project deliverables, compiled for the first Review Workshop: 1. Methodological Framework for the concept of resilience (section 1) 2. Discussion Paper on exogenous Scenarios (section 2) 3. Extreme Events (section 3) 4. Energy System Modeling Approach (section 4)

Our objective is to get a in depth-feedback and review of our methodological approach and concepts. Being aware that this document got some length, we want to put some emphasize on the most important parts: - Introduction to all 4 sections - Section 1.2 on the definition of resilience to be we applied in our project and PRESENCE, First Review Workshop Seite 4 von 53

- section 1.5 on the overall approach how to make this definition workable. - Section 2.3 gives and overview on the energy price and supply shocks that we want to take into account as part of the “stress test” of the energy system. - Section 3.5 summarizes our basic approach how to investigate extreme events related to the energy system. - Section 4.1 finally describes the methodology how to identify critical situations (extreme events) in the power sector and the conditions of their occurrence (both energetic and climatic).

Agenda of the Review Workshop:

9:30-10:30 Welcome, round of introduction, project overview – Lukas Kranzl, Julian Matzenberger TUW/EEG 1. Short introduction of each participant 2. Objective and schedule of the project 3. Discussion on the overall approach of the project 10:30-10:50 Resilience Concept in PRESENCE - Julian Matzenberger TUW/EEG 4. Presentation of the resilience concept in the literature and 5. the resilience in the context of energy systems (Project deliverables D3.1, D3.2 10:50-11:05 Short discussion, review comments - Geoff O’Brien, Harald Katzmaier, all Coffee break 11:15-11:35 Extreme Events – Herbert Formayer, Irene Schicker BOKU/MET 6. Literature review on existing methodologies for dealing with extreme events and 7. Methodological concept of the threshold approach applied in the project PRESENCE (Project deliverables D2.1. D2.2.) 11:35-11:50 Short discussion, review comments - Geoff O’Brien, Harald Katzmaier, all Lunch break 13:00-13:20 Electricity System Modeling in PRESENCE with HiREPS, resilience assessment and identification of critical conditions Gerhard Totschnig TUW/EEG 13:20-13:35 Short discussion, review comments - Geoff O’Brien, Harald Katzmaier, all 13:35-14:00 Detailed review comments - Geoff O’Brien, Northumbria University 8. Comments, ideas, suggestions on the concept, notion and definition of resilience in PRESENCE 9. Comments, ideas, suggestions on the assessment of extreme events in PRESENCE 10. Comments, ideas, suggestions on the electricity system modeling 11. General comments, ideas, suggestions to the review-document and the further steps in the project PRESENCE 14:00-14:25 Detailed review comments - Harald Katzmaier, fas research 12. Comments, ideas, suggestions on the concept, notion and definition of resilience in PRESENCE 13. Comments, ideas, suggestions on the assessment of extreme events in PRESENCE

PRESENCE, First Review Workshop Seite 5 von 53

14. Comments, ideas, suggestions on the electricity system modeling 15. General comments, ideas, suggestions to the review-document and the further steps in the project PRESENCE

14:25-15:30 Discussion and conclusions for the methodology of resilience and extreme events assessment in the energy sector and the modeling approach – Lukas Kranzl, TUW/EEG

15:30-16:00 Summary, outlook and next steps – Lukas Kranzl, TUW/EEG

PRESENCE, First Review Workshop Seite 6 von 53

1 Methodological Framework for the concept of resilience (D3.1)

The scope of this document is to outline different notions of the term resilience used in the scientific literature and explore how the concept of resilience can be applied to energy systems. Characteristics of the concept in various contexts, social and techno-economic systems, will be outlined and applicability for energy systems analysis will be discussed. This shall enable to make an informed decision to choose upon a methodology and underlying exemplary indicator set. Thus the two major questions to be addressed are:  Which definitions and underlying concepts of resilience are used in the scientific literature?  How can resilience be defined with respect to energy systems and which underlying principles can be identified? (Bearing in mind that an applicable methodology for assessment is needed)  and how to deal with the resilience concept in PRESENCE

Finally, along the way a selection of indicators to specifically assess the resilience of the energy system will be gathered.

1.1 Emergence of the resilience – concept

The concept of resilience has emerged relatively recently in the scientific debate. The number of publications dealing with resilience is strongly increasing over the last years. Taking into account a general increase in publications per year (about doubled since 1995), scientific articles containing the keyword resilience grew more than ten-fold since 1995, corresponding to a larger application of the resilience concept and a wider diffusion to other scientific areas. Figure 1 shows the number of publications dealing with resilience in all scientific disciplines. Searching for the keyword “resilience” in (only) scientific articles on the scientific database web of knowledge (www.webofknowledge.com) yields 9,272 results (Sept. 2011)

PRESENCE, First Review Workshop Seite 7 von 53

Fig. 1 Publications dealing with resilience in all scientific disciplines

Historically one of the first definitions of the term resilience dates back to 1824. Encyclopedia Britannica, since the seventh edition, defines resilience as “1: the capability of a strained body to recover its size and shape after deformation caused especially by compressive stress” or “2: an ability to recover from or adjust easily to misfortune or change”. [Encyclopedia Britannica 1824, derived online 2011]. (Surprisingly, given all the past efforts to define and reshape resilience, these two definitions still seem to hold true). Originally resilience has been used first in medical and material sciences, related to the ability to recover. Another definition though, is the ratio of energy given up in recovery from deformation to the energy required to produce the deformation.

More recently a wider concept of resilience has emerged. First, the obviously very figurative and inspiring term, has been taken up by Holling [1973] to describe ecosystems and has since then been used in other contexts, lately increasingly in social sciences (psychology) to describe community or individual resilience. Table 1 shows the broad variety of resilience definitions. Some are even led to argue that, after thirty years of academic analysis and debate, the definition of resilience has become so broad as to render it almost meaningless. [Klein et al. 2003] Clearly resilience is becoming increasingly important for approaches fostering sustainable development. In general it may be seen as a framework for understanding how to strengthen adaptive capacity in a complex environment.

PRESENCE, First Review Workshop Seite 8 von 53

Table 1 Definitions of resilience, advanced from [Norris et al. 2008] First Author, Level of Definition Year Analysis Gordon, Physical The ability to store strain energy and deflect elastically under a load 1978 without breaking or being deformed Bodin, 2004 Physical The speed with which a system returns to equilibrium after displacement, irrespective of how many oscillations are required Holling, 1973 Ecological The persistence of relationships within a system; a measure of the ability system of systems to absorb changes of state variables, driving variables, and parameters, and still persist Waller, 2001 Ecological Positive adaptation in response to adversity; it is not the absence of system vulnerability, not an inherent characteristic, and not static Klein, 2003 Ecological The ability of a system that has undergone stress to recover and return to system its original state; more precisely (i) the amount of disturbance a system can absorb and still remain within the same state or domain of attraction and (ii) the degree to which the system is capable of self-organization (see also Carpenter et al. 2001) Longstaff, Ecological The ability by an individual, group, or organization to continue its 2005 system existence (or remain more or less stable) in the face of some sort of surprise….Resilience is found in systems that are highly adaptable (not locked into specific strategies) and have diverse resources Resilience Ecological The capacity of a system to absorb disturbance and reorganize while Alliance, system undergoing change so as to still retain essentially the same function, 2006 structure and feedbacks—and therefore the same identity. (Retrieved 10/16/2006 from http://www.resalliance.org/564.php) Adger, 2000 Social The ability of communities to withstand external shocks to their social infrastructure Bruneau, Social The ability of social units to mitigate hazards, contain the effects of 2003 disasters when they occur, and carry out recovery activities in ways that minimize social disruption and mitigate the effects of future earthquakes Godschalk, City A sustainable network of physical systems and human communities, 2003 capable of managing extreme events; during disaster, both must be able to survive and function under extreme stress Brown, 1996 Community The ability to recover from or adjust easily to misfortune or sustained life stress Sonn, 1998 Community The process through which mediating structures (schools, peer groups, family) and activity settings moderate the impact of oppressive systems Paton, 2000 Community The capability to bounce back and to use physical and economic resources effectively to aid recovery following exposure to hazards Ganor, 2003 Community The ability of individuals and communities to deal with a state of continuous, long term stress; the ability to find unknown inner strengths and resources in order to cope effectively; the measure of adaptation and flexibility Ahmed, Community The development of material, physical, socio-political, socio-cultural, and 2004 psychological resources that promote safety of residents and buffer adversity Kimhi, 2004 Community Individuals’ sense of the ability of their own community to deal successfully with the ongoing political violence Coles, 2004 Community A community’s capacities, skills, and knowledge that allow it to

PRESENCE, First Review Workshop Seite 9 von 53

participate fully in recovery from disasters Pfefferbaum, Community The ability of community members to take meaningful, deliberate, 2005 collective action to remedy the impact of a problem, including the ability to interpret the environment, intervene, and move on Masten, Individual The process of, capacity for, or outcome of successful adaptation despite 1990 challenging or threatening circumstances Egeland, Individual The capacity for successful adaptation, positive functioning, or 1993 competence…despite high-risk status, chronic stress, or following prolonged or severe trauma Butler, 2007 Individual Good adaptation under extenuating circumstances; a recovery trajectory that returns to baseline functioning following a challenge Norris, 2008 Community A process linking a set of adaptive capacities to a positive trajectory of functioning and adaptation after a disturbance Renn, 2002 Community Countermeasure to uncertainties by avoiding irreversibilities and vulnerabilities Rose, 2007 System the ability of an entity or system to maintain function (e.g. continue producing) when shocked Cutter, 2008 Community ability of a social system to respond and recover from disasters O’Brien, System ability to withstand and adjust to disruptions whilst still retaining 2010 function Walker, System capacity of a system to absorb disturbance and reorganize while 2004 undergoing change so as to still retain essentially the same function, structure, identity, and feedbacks Rose, 2009 Economic process by which a community develops and efficiently implements its capacity to absorb an initial shock through mitigation and to respond and adapt afterward so as to maintain function and hasten recovery, as well as to be in a better position to reduce losses from future disasters

Given the broad diversity of concepts is not easy to find common characteristics, still most definitions emphasize a capacity for successful adaptation in the face of disturbance, stress, or adversity. It can be concluded, that a general consensus exists on two important properties of the resilience definitions in literature [Norris et al. 2008]:  Resilience is better conceptualized as an ability or process than as an outcome  Resilience is better conceptualized as adaptability than stability

With respect to energy systems it appears that definitions focusing on the system’s function are more applicable than other ones. A definition of resilience which can be used for energy systems (used by eg. Coaffe 2008, Rose 2009] is “the ability of an entity or system to maintain function (e.g. continue producing) when shocked“ [Rose 2007]

The above definition has one important shortcoming we would like to take into account. I would see (1) a changing environment as given, that (2) system are too complex to understand or map all interdependencies and (3) not only one equilibrium state does exist - change is the normal state. Resilience therefore is a learning process and recognizes that no steady-state exists

PRESENCE, First Review Workshop Seite 10 von 53

I would like to take not only into account the maintenance of function to external stress or disaster but also the possibility (probability) to adapt to opportunities or innovation. As resistance may not only act as an opposing force, but may also act as a catalyst and lead to a tipping point of changing behavior, a more resilient system is also enabled to adapt to newly emerging patterns. From an engineering point of view this might be interpreted as the ability to adapt to technological change or innovation. (Where as the traditional notion of engineering resilience would make the system return to pre-defined steady state after disturbance). The definition most suited I found in literature thus is: [Walker et al. 2004] “ The capacity of a system to absorb disturbance and reorganize while undergoing change so as to still retain essentially the same function, structure, identity, and feedbacks” In general or as [O’Brien and Hope 2010] frame it for energy (systems):

‘‘A resilient energy exhibits adaptive capacity to cope with and respond to disruptions by minimising vulnerabilities and exploiting beneficial opportunities through socio-technical co- evolution. It is characterized by the knowledge, skills and learning capacity of stakeholders to use indigenous resources for energy service delivery.’’

1.2 Resilience, adaptive capacity, vulnerability?

It becomes obvious from the above definitions that resilience, vulnerability and adaptability are very much interlinked and it seems to be not always clear where the line between the different terms is. Vulnerability and resilience are somewhat generic concepts, and the underlying or driving factors often overlap to make distinctions between the two unclear. [O’Brien et al. 2004] Does vulnerability influence adaptive capacity, or does adaptive capacity determine vulnerability? Does decreasing sensitivity enhance adaptive capacity? Does reduced vulnerability always lead to increased resilience? To articulate the relationship between vulnerability, resilience, and adaptive capacity, Cutter [et al 2008] grouped the different notions of the terms in global environmental change and (environmental) hazards research. The big difference is in moving from single stressors (hazards) to multiple stressors (global change). Resilience is either perceived is an integral part of adaptive capacity (3a), as a main component of vulnerability (3b) or as nested concepts within an overall vulnerability structure (3c). The Third Assessment Report (TAR) of the IPCC defines vulnerability as: “The degree to which a system is susceptible to, and unable to cope with, adverse effects of climate change, including climate variability and extremes. Vulnerability is a function of the character, magnitude, and rate of climate change and variation to which a system is exposed, its sensitivity, and its adaptive capacity” [Hinkel 2011]. Hence, IPPC defines adaptive capacity as a part of vulnerability. (3b)

In hazards research, the definition of resilience is refined to mean the ability to survive and cope with a disaster with minimum impact and damage. It incorporates the capacity to reduce or avoid losses, contain the effects of disasters, and recover with minimal (social) disruptions. Resilience

PRESENCE, First Review Workshop Seite 11 von 53 within hazards research is generally focused on engineered and social systems, and includes pre- event measures to prevent hazard-related damage and losses (preparedness) and post-event strategies to help cope with and minimize disaster impacts [Cutter et al. 2008]

Fig. 2 Conceptual linkages between vulnerability, resilience, and adaptive capacity [Cutter et al. 2008]

Another example is the definition of “Vulnerability” as used e.g. in the Project Regions2020) ◦ V=f(E,S,AC) V=Vulnerability, E=Exposure, S=Sensitivity, AC=Adaptive Capacity ◦ Enhanced to a causal chain with temporal dependency: V(t)=f(IM(t), AC(t)) IM=Impacts, SC=Social Capital, INV=Investments IM(t)=g(E(t-1), S(t-1)) AC(t)=h(SC(t-1), INV(t-1), INV(t-2),...)

Fig. 3. Time dependence in the vulnerability concept (Project: Regions 2020)

The ESPON CLIMATE Inception Report (2009) follows another notion of the term vulnerability: • Exposure: The nature and degree to which a system is exposed to significant climatic variations. • Sensitivity: The degree to which a system is affected, either adversely or beneficially, by

PRESENCE, First Review Workshop Seite 12 von 53

climate related stimuli. The effect may be direct or indirect. (cause-effect relationship) • (Climate) Impacts: Consequences of climate change on natural and human systems. Depending on the consideration of adaptation, one can distinguish between potential and residual impacts. • Adaptive capacity (or adaptability): The ability of a natural or human system to adjust to climate change (including climate variability and extremes) to moderate potential damages, to take advantage of opportunities, or to cope with the consequences. • Vulnerability: The degree to which a system is susceptible to, or unable to cope with, adverse effects of climate change, including climate variability and extremes. Vulnerability is a function of the character, magnitude, and rate of climate variation to which a system is exposed, its sensitivity, and its adaptive capacity.

I‘d like to argue, that conflicting “optimization criteria” could exist in framework that includes resilience in vulnerability. Increasing diversity (for e.g. fuel supply) and therefore resilience, inherently would make the system more complex, which in return negatively affects system vulnerability. I would propose to use a framework where the understanding of vulnerability refers a risk hazard approach widely referred to in engineering and economic literature. The risk-hazard approach is useful for assessing the risks to certain valued elements (“exposure units”) that arise from their exposure to hazards of a particular type and magnitude. (For an extensive discussion of the different notations of vulnerability see eg. Füssel [2007] or Hinkel [2011].) Vulnerability is the pre- event, inherent characteristics or qualities of the system that create the potential for harm. [Cutter et al. 2008]. Seen like this, energy systems are not vulnerable to climate change, but to its impacts (increasing risk hazard). On the contrary adaptation measures to climate change are very well possible. Being resilient to extreme events, disaster, catastrophe does not necessarily mean to be resilient against climate change as it unfolds on another timescale. A definition like Mechler’s could suffice very well: [Mechler et al. 2010]: Adaptation can be described as all activities aimed at preparing for or dealing with the impacts of climate change, be it at the level of individual households, communities and firms, or of entire economic sectors, governments and countries. Adaptation serves to reduce the damage resulting from the unavoidable impacts of climate change, as well as to protect lives and livelihoods. Vulnerability is often seen as the “antonym of resilience” [Füssel 2007, O’brien et al. 2004] which would explicitly read resilience = 1/Vulnerability. I would argue that vulnerability and adaptability are parts of the resilience concept. Especially the capability of a system to adapt should be part of its resilience, but not part of a vulnerability approach.

We thus propose that resilience be:

PRESENCE, First Review Workshop Seite 13 von 53

As resilience (aside from the traditional usage in material sciences) is merely a metaphoric concept a single measure or indicator of resilience would seem to be very abstract. I therefore would suggest to use (actually to cluster the indicators into) two different indicator sets for vulnerability and adaption capacity, which would make the underlying assumptions more comprehensive.

1.3 Definition of resilience in various disciplines

1.3.1 Definition of resilience in ecosystem sciences To make the overall concept more clear a short look into resilience in ecosystems: A concept often cited when referring to ecosystem resilience is the adaptive renewal cycle, first framed by Holling [2004]. The Adaptive renewal cycle is a heuristic model, generated from observations of ecosystem dynamics, of four phases of development driven by discontinuous events and processes. [Folke 2006]

 periods of exponential change (the exploitation or r phase),  periods of growing stasis and rigidity (the conservation or K phase),  periods of readjustments and collapse (the release or Ω phase) and  periods of re-organization and renewal (the α phase).

“Resilience is not only about being persistent or robust to disturbance. It is also about the opportunities that disturbance opens up in terms of recombination of evolved structures and processes, renewal of the system and emergence of new trajectories. In this sense, resilience provides adaptive capacity that allow for continuous development, like a dynamic adaptive interplay between sustaining and developing with change. Too much of either will ultimately lead to collapse. It does not imply that resilience is always a good thing.” [Folke 2006]

1.3.2 Definition of resilience in Fig. 4 Adaptive renewal cycles – ecosystem resilience [Folke 2006] engineering

In engineering resilience can also be understood as the time a system needs to “bounce back” to its previous state after an event (disaster, surprise).

PRESENCE, First Review Workshop Seite 14 von 53

a) b)

Fig. 5ab Time resilience [Tierney and Bruneau 2007] / [Rose 2009]

Figure 5 shows the notion of time resilience. The time an infrastructure needs to acquire its pre- event state (in case (a) an earthquake in case (b) a power outage). Rose takes into account not only the speed of recovery but also the pattern of recovery. As discussed above, in a larger context, the possibilities to adapt or learn from an event and switch to another state are outside the boundaries of this framework.

1.3.3 Definition of resilience in social sciences (communities) Norris developed a model of stress resistance and resilience over time (in c :

Fig. 6 Model of stress resistance and resilience over time [Norris et al. 2008]

Resistance occurs when resources are sufficiently robust, redundant, or rapid to buffer or counteract the immediate effects of the stressor such that no dysfunction occurs. Total resistance is hypothesized to be rare in the case of severe, enduring, or highly surprising events, making transient situational dysfunction the more likely and normative result in the immediate aftermath of PRESENCE, First Review Workshop Seite 15 von 53 disasters. Resilience occurs when resources are sufficiently robust, redundant, or rapid to buffer or counteract the effects of the stressor such that a return to functioning, adapted to the altered environment, occurs. For human individuals and communities, this adaptation is manifest in wellness. Vulnerability occurs when resources were not sufficiently robust, redundant, or rapid to create resistance or resilience, resulting in persistent dysfunction. The more severe, enduring, and surprising the stressor, the stronger the resources must be to create resistance or resilience. [Norris et al. 2008]

Table 2. Properties of resililient (energy) systems 4R Framework, MCEER nach Tierney et Bruneau, 2007 O'Brien and Hope, 2010 Walker et al, 2004 Robustness Appropriateness Latitude Redundancy Indigenous RES Resistance capacity enhancing, Resoucefulness adaptable and upgradeable Precariousness Rapidity easy to repair and maintain Panarchy

PRESENCE, First Review Workshop Seite 16 von 53

The “4R” Concept In examining the attributes and determinants of resilience, Multidisciplinary Center for Earthquake Engineering Research (MCEER) investigators developed the R4 framework of resilience [Tierney and Bruneau 2007]:  Robustness—the ability of systems, system elements, and other units of analysis to withstand disaster forces without significant degradation or loss of performance;  Redundancy—the extent to which systems, system elements, or other units are substitutable, that is, capable of satisfying functional requirements, if significant degradation or loss of functionality occurs;  Resourcefulness—the ability to diagnose and prioritize problems and to initiate solutions by identifying and mobilizing material, monetary, informational, technological, and human resources; and  Rapidity—the capacity to restore functionality in a timely way, containing losses and avoiding disruptions.

Norris [et al 2008] revises the “4R” properties and defines similar attributes, leaving out resourcefulness (“critical thinking”) due to its more static approach. MCEER investigators identified four dimensions or domains of resilience: the technical, organizational, social, and economic (TOSE) [Tierney and Bruneau 2007]:  The technical domain refers primarily to the physical properties of systems, including the ability to resist damage and loss of function and to fail gracefully. The technical domain also includes the physical components that add redundancy  Organizational resilience relates to the organizations and institutions that manage the physical components of the systems. This domain encompasses measures of organizational capacity, planning, training, leadership, experience, and information management that improve disaster-related organizational performance and problem solving. The resilience of an emergency management system, therefore, is based on both the physical components of the system—such as emergency operations centers, communications technology, and emergency vehicles—and on the properties of the emergency management organization itself—such as the quality of the disaster plans, the ability to incorporate lessons learned from past disasters, and the training and experience of emergency management personnel.  The social dimension encompasses population and community characteristics that render social groups either more vulnerable or more adaptable to hazards and disasters. Social vulnerability indicators include poverty, low levels of education, linguistic isolation, and a lack of access to resources for protective action, such as evacuation.  Local and regional economies and business firms exhibit different levels of resilience. Economic resilience has been analyzed both in terms of the inherent properties of local economies—such as the ability of firms to make adjustments and adaptations during nondisaster times—and in terms of their capacity for postdisaster improvisation, innovation, and resource substitution (3). In general, social and economic resilience relate to the ability to identify and access a range of options for coping with a disaster - the more limited the options of individuals and social groups, the lower their resiliency.

PRESENCE, First Review Workshop Seite 17 von 53

O’Brien and Hope [2010] identify four underlying principles of a resilient energy system in a socio technical co-evolution of energy systems for at transition to a low carbon economy:  Appropriateness  based on indigenous renewable sources  capacity enhancing, adaptable and upgradeable  easy to repair and maintain

Walker [et al. 2004] has a more conceptual framework, he identified four aspects are critical for these definitions:  Latitude: the maximum amount the system can be changed before losing its ability to recover; basically the width of the basin of attraction. Wide basins mean a greater number of system states can be experienced without crossing a threshold (L, Fig. 1).  Resistance: the ease or difficulty of changing the system; related to the topology of the basin—deep basins of attraction (R, or more accurately, higher ratios of R:L) indicate that greater forces or perturbations are required to change the current state of the system away from the attractor.  Precariousness: the current trajectory of the system, and how close it currently is to a limit or “threshold” which, if breached, makes recovery difficult or impossible (Pr).  Panarchy: how the above three attributes are influenced by the states and dynamics of the (sub)systems at scales above and below the scale of interest (Pa).

Fig. 7 Conceptual “resilience landscape”[Walker et al 2004]

A completely different, but none the less interesting concept is presented by Coaffe, defining resilience as function of sustainability and security. Which, with respect to the energy system, can be understood as a sustainable, more autarkic, energy supply would also make the system more secure. Considering environmental shocks as well as terrorist attacks on the energy system. Coaffe argues that synergies between the isolated policies of increasing security and sustainability can be merged in an integrated resilience policy. [Coaffe 2008]

In short, the above three(four) concepts might be useful to identify new indicators and cluster them treelike: PRESENCE, First Review Workshop Seite 18 von 53

Resilience vulnerability II adaptability robustness II redundancy II resourcefulness II rapidity Another, from my point of view underrepresented, aspect of resilience is fiscal (economic?) resilience. The ability to invest in mitigation measures, insurance or reconstruction. Although the occurrence of floods follows a random pattern, one can address vulnerability and design emergency management actions well in advance. The major challenge is to determine the limit to which one is willing to invest in resilience.[Renn et al. 2011]

It might be perceived that society is not resilient against cheap fossil energy, in a way that an addict is not resilient to certain “substances”. Even if the perspective of a favorable development path is within sight, we can observe a “lock in” effect. The deployment of innovation does not occur as structures do not adapt. Can we incorporate effects like that? In a (very comprehensive) risk management approach, Renn and Klinke embed resilience as one out of three (complexity, uncertainty and ambiguity) system properties. Resilience acts as a countermeasure to uncertainties by avoiding irreversibilities and vulnerabilities. Resilience therefore only acts as countermeasure to one of the risk management objectives. In fact, we have to bear in mind, that in this context the objective of resilience, through increasing diversity, may also increase complexity and therefore negatively affect the overall systemic risk.

Table 3. Risk Management Challenges and corresponding strategies [Klinke and Renn 2002]

1.4 From concept to applicable resilience indicators

Helio has developed an indicator set for assessing the energy system. The assessment has been carried out mostly for developing countries, but can prospectively be applied to developed countries as well. Two separate indicator sets for vulnerability and adaptability have been used. Although the methodological concept does not extend to a final indicator set of resilience the combination of both indicator sets for vulnerability and adaptability are used to describe resilience

PRESENCE, First Review Workshop Seite 19 von 53 of the energy system. (It is stated that “The level of resilience is based on a system’s adaptive capacity”.) A complete list of indicators can be found in Annex 1 In addition, a number of recommendations to increase the resilience of energy systems were made [Williamson et al. 2009]: 1. Current and future energy systems must be “climate-proof”. This is achieved through systematic assessment and monitoring of energy systems; including improved data collection on climate change and adaptation and mitigation strategies; mapping of potential renewable energy areas susceptible to the impacts of climate change; and using the indicators to develop monitoring protocols. 2. Climate and poverty factors should be included when assessing new energy systems. 3. Medium and long-term strategies must be developed to secure a diversified, decentralized, accessible, affordable and modern energy system, more resilient to climate change. 4. Improve energy efficiency as an essential adaptation measure. This will reduce the need for new energy sources and lower the burden on existing ones. Using renewable, low energy sources will reduce the vulnerability of the energy supply sector to the adverse effects of climate change. 5. Local expertise within countries needs to be developed about the effects of climate change on energy supply, production, distribution and use. 6. Ecosystem services that support existing and future energy production must be protected and developed. These include water and biomass resources, which support power plant production and provide millions of households with their main energy source. 7. Transparent technology transfer and financing procedures should be established. 8. Consultation and involvement of end-users in decision-making processes for energy systems must be ensured in order to develop appropriate energy systems that meet both energy and community needs.

Jansen and Seebregts argue that conventional (economic) approaches in energy security neglect the demand side response in energy security issues. …approaches tend to underexpose essential aspects for resilience performance to address energy services security on the demand side, such as energy efficiency and underlying aspects such as spatial and infrastructure planning and consumption patterns. A demand-side focus is warranted to enhance the resilience level of a defined population in facing vulnerabilities to its access to energy services on long timescales. A credible index to gauge medium/long-term energy services security of a defined population strikes a good balance between measuring demand-side resilience against energy services security vulnerability and the magnitude of the energy services security vulnerability itself. [Jansen and Seebregts 2010]

PRESENCE, First Review Workshop Seite 20 von 53

Consequently they define two indicator sets including demand issues: four diversity based indices and a supply/demand indicator framework . The four diversity-based indices are as following:  Diversification of energy sources in energy supply (I1).  Diversification of imports with respect of imported energy sources (I2).  Long-term political stability in regions of origin(I3).  The fuel resource base in regions of origin, including the home region (I4).

This set of indicators does take into account the vulnerability due to availability of resources but adaption or resilience consideration are not considered. They also note that the resilience of a certain society against shocks in the supply of energy resources driving the provision of societal needs for energy services is not only determined by diversification of – notably external – supply and other non-domestic supply considerations.

Fig. 8 Structure of the Supply/Demand Index [Jansen and Seebregts 2010]

The S/D Indicator is normalized to a 0-100 scale. Austria yields 57 pts whereas Denmark achieves the maximum with 82 pts.

A common index used in energy economics to quantify the diversity of fuel sources (by origin) is the Shannon-Wiener Index. It might be also used to quantify the diversity in the energy mix. H has its maximum at an even distribution of p. (with p the proportion (or share) of p(i) in all p.)

The maybe most experimental idea to discuss, I came up with, is to assess the resilience of a network (grid) with an approach used in ecosystem analysis. (personal communication Ulanowicz)

PRESENCE, First Review Workshop Seite 21 von 53

Ascendency H can be understood as a measure of network articulation: the more probable a path the higher the value. [Fath 2010] (with T , where i represents the origin of the arc and j the node ij in which it terminates.)

An extensive List of indicators is listed in the Annex.

1.5 Making the definition of resilience workable in the project PRESENCE

In section 1.2 we defined resilience as the ratio of adaptability and vulnerability of energy systems:

Moreover, we intensively discussed different concepts and definitions. This is the key background and common notion of our approach. However, the objective of this project is not to derive resilience indicators for different settings of the energy system, different regions etc.

Rather, when it comes to make this notion of “resilience” workable in the project PRESENCE, we suggest the following approach: - We carry out various “stress tests” of the energy system. For this purpose, we define different energy price and supply shock scenarios combined with storylines of the general development of the energy system (see section 2.3). Moreover, weather/climate data in different climate scenarios are taken into account. This refers to the vulnerability of the energy system. Without having in mind to derive quantitative indicators of the vulnerability of different energy system settings, we will be able to derive conclusions regarding the difference of vulnerability of different energy systems in the different ”stress test” situations. - We will identify most critical situations (extreme events) of the energy system (mainly related to black outs in the power sector) and investigate the probability of the occurrence of meteorological conditions leading to such events. - We investigate the impact of different adaptation measures on the stress tests and the extreme events. So, we will be able to assess to which extent various adaptation measures are suitable for increasing the resilience of the system. In the following chapters we want to highlight some of these aspects more detailed.

PRESENCE, First Review Workshop Seite 22 von 53

2 Discussion paper on exogenous reference-scenarios and shocks of energy price and supply

2.1 Introduction

This text should give a short overview on the definition of the techno-economic reference scenarios (especially the qualitative and quantitative basic parameters as well as corresponding exogenous parameters) in the project PRESENCE. In particular, we focus on our assumptions regarding possible price and supply shocks of fossil energy carriers. These will be taken into account in the resilience assessment and identification of extreme events.

2.2 Storylines and exogenous parameters in the Scenarios

In this project three techno- and socio-economic scenarios of the energy sector are analysed in detail:

• Fossil fuel scenario • Renewable-Scenario • Renewable/Efficiency Scenario

To characterize the scenarios for the development of the energy system, the relevant exogenous parameters are identified and possible consistent developments illustrated. Investment decisions in energy technologies and (for the electricity sector) the resulting energy prices are the particular, endogen variables in the model.

Necessary parameters for the scenario development therefore include:1

• Energy consumption patterns (and energy efficiency policies) • Renewable and other environmental policies as well as further energy policies (e.g. security of energy supply)

• CO2 Prices

The scenarios reflect the possible development paths of the energy system and its restrictions. The Austrian electricity (and overall energy) system is part of the (central) European system. As far as necessary, corresponding developments at European level are therefore taken into account in the parameter definition. Each scenario shows a different development illustrating the influence of the political and social decision in the energy sector. For that reason, the chosen scenarios remain in contrast to each other. A vital aspect is to summarize the storylines behind each

1 Technological parameters such as investment costs, market parameters such as interest and discount rates are not discussed in this document. PRESENCE, First Review Workshop Seite 23 von 53 development in order to give a consistent, characteristic picture. So, the storylines2 focus on the long-term, general characteristics.

• Reference scenario: The conflicts of the different stakeholders in the energy sector (politics, energy economy, enterprises and environment organisations) remain unresolved. As there is no clear course, policy target and guideline, only half hearted climate and energy policies are implemented (on a global and European level). The acceptance of subsidies for renewable energies and energy efficiency in the society is small. • Renewable scenario: In this scenario energy policy is focusing on renewable energies. Neither social nor political barriers put strong limitations to the development of renewable energies. Correspondingly, ambitious support policies are implemented. Security of energy supply becomes one of the main objectives of energy policy which is bound to regional renewably energy sources. Development of additional renewable generation capacity on European level will take place. To allow for such a development, the modernisation of the European electricity grid is carried out. Reduction of the conversion losses and reduced final energy consumption are the results of an increased efficiency. The efficiency improvements are mainly influenced by energy prices. E-mobility is widespread used. • Renewable/Efficiency Scenario: In this scenario the energy policy focuses on renewable energy sources and energy efficiency. The main objective of the energy policy is strong reduction of greenhouse gas emissions. Direct subsidies of renewable are lower and is dynamically reduced in comparison to the renewable scenario. Significant efficiency yields are mainly policy driven. On the European (and global) level ambitious efficiency policies are implemented. Efficient use of resources together with sufficient available renewable

technologies leads to lower CO2-prices compared to the renewable scenario. E-mobility reduces partly the achieved reduction of consumption in the electricity sector.

Table 4 summarizes the scenario parameters (qualitatively).

2 There is a significant number of additional sources describing storylines of the energy system. E.g. in ESPON-RERISK (2010) four long-term scenarios for Europe with a time horizon of 30 – 60 years were defined using the following hypotheses: energy prices will remain at a high level, different political responses to this challenge. Some of the patterns of these storylines (“Green High-tech”, “Energy-efficient Europe”, “Business as usual?”) are similar to our storylines, some others are dismissed (e.g. the scenario “nuclear energy for big regions”).

PRESENCE, First Review Workshop Seite 24 von 53

Table 4 qualitative development of price scenario parameters in the presence-scenarios. Source: EC (2010), IEA (2010), own assumptions Scenarios Referenz (fossil) RES RES-Eff Demand + o - E-mobility O ++ + RES support O ++ + Energy and CO2 price Coal + + + Gas/Oil + + +

CO2 O ++ + CCS ? ? ?

2.3 Primary energy prices and price/supply shocks

The development of the fuel costs belongs to the most vital parameters for energy investments and operation. Of course, future projections of energy prices are bound to major uncertainties. Therefore within the developed scenarios three different (sub)-development paths are described as secondary conditions to the three main-scenarios. As an important determinant of the future economic relations, assumptions are made about the energy price path and supply of energy sources. The resilience of the energy supply system is analysed towards constant, sudden or continuous changes in the supply situation.

80

70

60

50 Öl 40 Gas '09/MWh] €

[ 30 Kohle 20

10

0 2005 2010 2020 2030 2040 2050

Fig. 9 Variation of the primary energy prices for coal, oil and gas. Source: IEA (2010) as well an extrapolation until 2050

PRESENCE, First Review Workshop Seite 25 von 53

Assumption of price and suppply shock Sub-scenario A: constant supply situation • There are no supply bottlenecks; prices of considered energy fuel sources are developing synchronously with world prices like in the figure above. • Gross domestic energy consumption is not restricted by supply shocks. Sub-scenario B: Continuous rise in prices • Due to the tense supply situation until 2030 energy prices (wholesale) increase to 240% compared to the reference year 2010. • Gross domestic consumption is continuously decreasing Sub-scenario C: short-term rise in prices • Gross domestic consumption decreases as there is a shortage of supply • In the short term, the energy prices (wholesale) are sharpy rising to 240% compared to WEO base price and going back to the previous level after the energy price shock.

The scope of this project is not to investigate in detail the possible reasons behind these shocks. They might be due to geopolitical crises and constellations and/or due to climatic conditions and extreme events (e.g. cyclones hindering the shipping of fossil oil). However, we just assume that there is some probability connected with these events, why it is relevant to test the resilience of our energy system towards these shocks.

For the development of the oil-price in the European countries it is assumed that the rate of exchange of Dollar-Euro is constant. The development of consumer price of crude oil is driven by different tax levels for each product as well as a fee for energy stockage . In the investigated period we assume these values to be constant. We do not assume increasing cost of conversion and distribution (falling EROI). The weight of this cost component for most energy carriers exceeds the rest of the cost. This partly compensates the high increase of the crude oil-price leading to a lower effect on the consumer retail energy price.

PRESENCE, First Review Workshop Seite 26 von 53

Fig. 10 Variation of the fossil fuel energy price in Base Scenario

Fig. 11 Quantity development of fossil fuel energy sources in Base Scenario

The figures illustrate the patterns of the price and supply shocks. The timing and scaling of the shocks will be element of sensitivity analyses in the project.t

On December 5, 2012 the price of WIT (West Texas Intermediate) lies at $101,5 per barrel i.e. in the scenario “continuous shortage“ the oil-price lies at $240 per barrel in 2030 at a constant world- market price. Gross domestic consumption of petroleum in 2010 amounted to around 550 PJ (449.171 TJ). The scenario “continuous shortage“ assumes decline of available amount for 60% of PRESENCE, First Review Workshop Seite 27 von 53

Energy demand. In 2010 petroleum products accounted for 38% of Austrian´s total gross energy consumption [Statistik Austria 2011].

2.4 Calculation of the household prices

Household retail energy price taxes are calculated based on the price-scenarios of primary energy sources of the World Energy Outlook (WEO) The taxation of energy in Austria includes taxes for electricity, natural gas and coal. The regulatory framework for this and for the reimbursement is based on four different laws.

The coal fuel tax (defined in Kohleabgabegesetz) covers the supply of coal, the consumption of coal through the distributes and producer as well as the consumption of the imported coal. The rate of tax is €0,05 per kg. The taxation of gas (defined in Erdgasabgabegesetz) includes mineral oil, liquid gas and electric energy. This tax is in Austria 0,66 €/Nm3.

The taxation of oil covers the different intended purpose and requirement in Mineralölsteuergesetz. This tax is currently 60 €/t. All price components cover value added tax of 20%.

Fig. 12 Variation of household prices of energy fuel sources in the defined scenarios

PRESENCE, First Review Workshop Seite 28 von 53

3 Extreme Events in Relation to the Energy System and Climate Change – First considerations and Open questions The energy demand and supply is highly sensitive to extremes in temperature, wind, and precipitation which can have not only local effects but can also influence e.g. regions downstream of heavy precipitation events. Intensification of extreme events most likely has impacts on the energy system. Extreme events in relation to the energy system, for example, can be extreme weather events having impacts on the infrastructure of the energy system. However, there are also other events that can be classified as extreme events, e.g. long hot, dry periods with high cooling demand, low generation of hydro and wind power (and low share of PV in the electricity generation system). Energy crises, energy price shocks etc. in addition can contribute to critical situations in the energy system. Not necessarily do extreme weather events affect the energy system. Thus, how to best quantify the impact of extreme events on the energy system? Following questions are crucial:

1. What concept of resilience, vulnerability and adaptation is useful for the energy system? We tried to find a suitable approach for this question in our project in chapter 1. 2. Which events are extreme events related to the energy system? What is the impact on the system (supply and demand)? The main approach of how to deal with this question for the electricity sector will be addressed in chapter 4. 3. What are the causes of such extreme events in the energy system? Are they related to extreme weather events or related to economic, social, technological disruptions? How to consider changes in the energy system not created through climate change? Several aspects of non-climate related changes, trends and possible crises are addressed in chapter 2. 4. What are the methods for investigating extreme events in atmospheric sciences? 5. Existing extreme events in the past (intensity, duration,…) – what can we learn? 6. How often will there be extreme events in the future? Or: how do the climate and other conditions change that create problems in the future energy system?

The last three questions will be the focus of this section. The analysis will focus on the electricity sector. However, also the heating and cooling sector will be integral part of the analysis. Firstly, we established a link between the techno-economic bottom- up tool INVERT/EE-Lab modeling heating and cooling energy demand and the HiREPS model, ensuring the consideration of climate sensitive heating and cooling related electricity demand. Secondly, for the part of the heating and cooling energy demand that is not covered by electricity, we will define two extreme events:

PRESENCE, First Review Workshop Seite 29 von 53

1. Energy expenses for heating and cooling are above a certain threshold related to mean household income (for households) and related to mean value added (for the tertiary sector). 2. Indoor temperature levels are above a certain threshold for a certain period in case that no cooling device is installed.

3.1 What can be considered an extreme event in terms of the Energy System? For a coarse definition two time frames will be considered here: the short term and the long term events. Short term events are incidents which may affect the energy system in its present situation. 1. Energy System fails to maintain energy supply (short term). This may occur if: ◦ Energy supply is not able to meet the demand ◦ Blackout or shortage of energy supply 2. Critical energy demand in case of: ◦ Increased need for Heating and Cooling ◦ Other short term peaks (Industrial, Transport) 3. Decreased energy supply in case of: ◦ Extreme Weather ◦ Accidents ◦ Lack of resources ◦ Strikes ◦ Load balancing failure ◦ Distinction: Failure of the whole energy system or of specific sources of energy? What is the role of different technologies in the energy mix? ▪ Water power: drought, flooding, ice... ▪ Wind power: calms storms ▪ Thermal power: lack of energy carriers, lack of cooling water, cooling water too hot 4. Economical aspects: Energy prices, prices for raw materials reach critical level (short term) 5. Combination of several of these events

Long-term events are incidents which may affect the energy system in the future 1. Changes of location of storm tracks 2. Increasing needs for Cooling and Heating through e.g. increased in number of heat days 3. Longer periods of dry spells, ….

To understand, define and classify a specific (past and future) extreme event more detailed one needs to answer following questions: 1. Is the event weather / climate related, direct and indirect, or are other factors determining PRESENCE, First Review Workshop Seite 30 von 53

the energy sector extreme event? To which extent do these other factors increase the vulnerability of the energy sector towards weather / climate related extreme events? 2. What is the spatial / temporal scale concerning location of the energy system, regions affected,…? 3. And what spatial and temporal scale shall be defined for future events? 4. Which energy source / which technology is affected? 5. What is the intensity of the event? What thresholds best to use? Should increments be used? 6. In case of meteorological / climatic induced extreme event: Which meteorological parameters are the cause? Is it a combination of more than one meteorological parameter? 7. And what weather conditions / combinations of conditions shall be defined as extreme events for the future?

3.2 Methods for investigating Extreme Events in Atmospheric Sciences This section presents an overview of methods for investigating extreme events in atmospheric sciences. We will start with various approaches for defining extreme events, then discuss changes in extreme events and finally give a short insight into extreme events in climate models.

3.2.1 Definition of extreme events:

Studies dealing with extreme events can be devided into parametric studies and non- parametric studies (Haylock and Goodess, 2004) or return period analyses and direct empirical diagnostics (Frei et al., 2006): Parametric methods: − Return period analyses determine which events can be expected to occur within a certain period (1y, 5y, 10y,...). These methods use manifold approaches pertaining to the field of extreme value statistics. Distributions are fitted to the sample of annual or seasonal extremes predicting the return period of events exceeding certain thresholds. − Peaks over threshold methodology (POT): All observations exceeding a sufficiently high threshold are taken into account. The extremes are clustered to ensure serial independence. Maxima of clusters of exceedances are defined as POT events. The method results in a Poisson process model for event arrivals and utilizes the Generalized Pareto distribution for their magnitudes. − Block maxima method: It is assumed that the long-range dependence at extreme levels is weak. A Generalized Extreme Value (GEV) distribution is used for modeling return values. (Coles, 2001; Paeth and Hense, 2005; Frei et al. 2006; Kysely and Beranova, 2009) Non-parametric methods or direct empirical diagnostics: • Indices:

PRESENCE, First Review Workshop Seite 31 von 53

◦ Frequency of occurence of extreme events. ◦ Intensity: threshold exceedance of events. ◦ Combination: Indices relating frequency and intensity of events: ▪ Number of days where precipitation/ temperature exceed certain thresholds. ▪ Cold/Warm day indices: Number of days below 10th percentile/above 90th percentile of temperatures. ▪ Maxium consecutive numbers of dry days. ◦ Quantiles/percentiles (e.g. 10%, 90%) of a distribution of parameters: ▪ Fraction of the total amount of rainfall due to events above the 95th percentile. ◦ Medians of daily maxima/minima ◦ Dichotomous (“yes/no”) description of extreme events: Event En is defined to occur if total precipitation amounts, max/min temperatures, etc. at a specific location exceed n standard deviations above/below the mean. Mostly used: E2, E3. (Palmer and Raeisaenen, 2002) ◦ Extremes defined from rank of, e.g. daily precipitation within a year. The 10 wettest days are assessed (Unkasevic and Tosic, 2011). ◦ Duration of events: Heat waves, cold waves, dry spells, etc. Dependent on the number of consecutive days with temperatures above/below, precipitation amounts below certain thresholds (Cindric et al. 2010). ◦ Most indices are defined for single meteorological parameters, a few for a combination of different parameters and sub-indices, for example the Climate Extremes Index (CEI, Karl et al., 1996). A climate extreme is described as “large areas experiencing unusual climate values over longer periods of time”. The CEI is composed of percent area with extremes in maximum and minimum temperature (for cold and warm periods), the Palmer Drought Severity Index (dry and wet periods), extreme precipitation and the number of days with precipitation. A similar index has been proposed that also includes parameters important to high-latitude climates, such as extreme snow accumulation and wind (Easterling, 2000). Such indices must be adapted to the regional climate regime and to their individual purpose. Other studies investigate the relationship between the occurrence of extremes and the large scale circulation (e.g. Besselaar et al., 2009). High water and low water events have been related to certain circulation types as defined by MSL-pressure (KLIEN, Nachtnebel et al, 2010). Impact-related: Extreme events can also be defined from their impact on society (Halenka et al., 2006), human health, economic cost/loss, etc. The dichotomous description (defined event occurs or not) of extreme events is feasible for economic cost/loss considerations in terms of (Palmer and Raeisaenen, 2002); − Cost of protecting against a certain extreme event − Loss due to the occurrence of a certain extreme event

PRESENCE, First Review Workshop Seite 32 von 53

3.2.2 Changes of extreme events They are caused by: • changes of the mean of parameter: the frequency of extremes changes non- linearly with the change of the mean of a distribution. • the variance: a change of the variance will have mainly an effect on the frequency of extremes than a change of the mean. • of both, mean and variance (Meehl et al, 2000, Fig. 1): for non-Gaussian distributed variables (e.g. precipitation) a change of the variance also causes a change of the mean. The impact can change due to the change of extreme events, but it can also change independently of the climate system, e.g. if the vulnerability of society and systems (ecosystem, energy system) changes.

3.2.3 Description of extreme events by climate models

In climate/weather modelling simulation results for a specific region depend strongly on the used spatial resolution, i.e. correct representation of local ortography. In general, General circulation (Global climate) models (GCMs) use resolutions (~ 100 – 150 km for non-coupled atmosphere only GCMs) which are not desirable when one wants to estimate future climate for a specific area. Thus, regional climate models (RCMs) are applied with higher spatial resolutions (~10 – 25 km) using data of GCMs as input. For the description of extreme events in climate models coarse resolutions climate models provide acceptably realistic means for a larger area, whereas high resolution is needed for e.g. a realistic simulation of extreme rainfall (Huntingford et al. 2002). In this project data of three available RCM simulations (RegCM3, REMO, ALADIN) are used.

3.2.4 On quantification and classification of extreme events Quantifying extreme events can be crucial. Renn and Klinke have written a comprehensive paper on risk management, that might also be of interest for an approach to classify risks (based on [WBGU 2000]) in PRESENCE. Risk governance draws the attention to the fact that not all risks are simple: they cannot all be calculated as a function of probability and effect. Many risks, which require societal choices and decisions, are adequately characterized as complex, uncertain and/or ambiguous. It is a consistent finding, however, that in most cases they are treated, assessed and managed as if they were simple. The many failures to deal adequately with risks such as genetic engineering, nuclear energy, financial crisis, cyber-terrorism as well as environmental risks such as chemical pollution or eutrophication demonstrate an urgent need to develop alternative concepts and approaches to deal with uncertain, complex and/or ambiguous risks.

PRESENCE, First Review Workshop Seite 33 von 53

It might be difficult to quantify the probability of occurrence of extreme events needed to quantify vulnerability. But the events could be mapped in risk classes according to the approach by [Klinke and Renn 2002] (for a detailed explanation of the risk classes consult me or one of the above mentioned sources). The vulnerabilities to extreme events of major energy generation technologies are shown in Table 6, classes of risk are shown in Fig, 15.

Table 5: Most harmful climatic effects per electricity sector in order of importance [Kirchsteiger et al. 2011]

Fig. 13 Classes of risk [WBGU 2000]

PRESENCE, First Review Workshop Seite 34 von 53

3.3 Existing extreme events in the past (intensity, duration,…) – what can we learn? One example of the recent past is the dry spell in Austria in November 2011 where except in East Tyrol 0 mm of precipitation were measured at the Austrian rain gauges (Fig.3 ) with persistent fog and low wind speeds in great parts of central and eastern Europe.

Fig. 14 Precipitation sum November 2011 for the Alpine region (http://medienportal.univie.ac.at/presse/aktuelle-pressemeldungen/detailansicht/artikel/klimakarten- der-universitaet-wien-zeigen-niederschlagsfreien-november/).

In terms of energy production this example showed that if several meteorological conditions are combined several energy sectors produced less than expected. In November 2011 less energy through wind and hydro power was produced. In some areas, depending on the persistent fog, additionally less solar energy was available. In addition, more coal and gas was burned due to the low gauge height.

3.4 How often will there be extreme events in the future? Or: how do the conditions change that create critical situations in the future energy system? Relating future changes of the energy system and climate change and the interaction are not easy to quantify and several questions arise and should be discussed.

1. What kind of scenarios of the future development of the energy system do exist and can be used in this study? (We discussed the general guidelines and storylines in section 2.) 2. Which aspects should all be taken into account? a. Changes in energy demand, energy production, energy prices? b. Used technologies, available natural resources, structure of the energy system (regional / central), politics,… c. Will there be “business as usual” or “energy transition”? d. Which time frame needs to be considered?

PRESENCE, First Review Workshop Seite 35 von 53

Fig. 15 Schematics of ways of dealing with future changes in the energy sector.

3.5 The PRESENCE – approach to assess extreme events in the energy system

In this part, we will outline the approach that we will use for assessing extreme events in the project PRESENCE: 1. As a first step, we identify potential critical situations in the energy sector. These critical situations are defined in terms of energetic conditions (e.g. demand, energy mix etc) and climatic conditions. So, critical conditions always are considered as a combination of energetic and climatic conditions. We focus on the electricity sector, with links to electricity demand for heating and cooling. The approach of identifying critical situations in the electricity sector will be described in chapter 4 of this paper. 2. As a second step, we determine the probability of the occurrence of corresponding climatic conditions in future climate scenarios: o The climatic condition(s) causing the potential critical situation is (are) identified and being searched for in the BIAS corrected climate scenarios for Austria and Europe. RCM scenario results used in this project are of the REMO model driven by ECHAM5 (MPI Hamburg), ALADIN driven by ARPEGE (CNRM – France), and the RegCM3 model driven by ECHAM5 (ICTP – Italy). These scenarios are part of the ENSEMBLES project (http://www.ensembles-eu.org) and are available on European scale with 25 x 25 km resolution. The temporal output of the variable varies, surface parameters come as daily averages for the time period 1951 – 2100. The time period 1981 – 2010 is considered as control period. For the BIAS correction of the scenario data the gridded observation data sets E-OBS temperature climatology (Haylock et al., 2008), covering Europe, and the Frei and Schär (Frei and Schär, 1998) precipitation data set, covering the Alps, have been used. The method applied is based on the

PRESENCE, First Review Workshop Seite 36 von 53

quantile mapping method of Deque (2007).Here, scenario data are compared to observed reference data correction functions for each single quantile (in steps of one percent) are calculated. In doing so, the whole probability density function of data values is adapted to the observed one. o Empirical statistics as e.g. quantiles or the interquantile range of the meteorological parameters and possible their combinations which have caused the extreme event are calculated for different time slices of 30 years: 1951 – 1980, 1981 – 2010, 2011 – 2040, 2036 – 2065, and 2051 – 2080. Events evaluated will not go beyond the five year event margin. o Comparison of the future time slice results to the statistics derived from the climate scenario control runs. In this step the results of the empirical statistics of the future time slices will be compared to their occurrence in the control period. 3. Based on the probabilities of climatic extreme events that cause critical situations in the energy system we will derive conclusions regarding the urgency and need to take adaptation measures.

3.6 Open questions: Some open questions in different sectors still need to be discussed. • Vulnerability and energy system: ◦ What are the key vulnerabilities of the energy system? A suggestion of how to define critical situations in the energy system will be described in chapter 4. ◦ Adaptive capacity: What time period is considered for adaptive actions? STARDEX, Regions2020 used 2020 (~10y) how must AC be defined for, e.g. 30y, 50y,...? • Change: ◦ What kinds of long-term changes and possible crises other than climate change shall be considered as relevant for extreme events? Some of these aspects have been discussed in chapter 2.

PRESENCE, First Review Workshop Seite 37 von 53

4 Method to identify extreme events in the power system with the simulation model HiREPS

In this section we outline our approach to identify extreme events in the power system. The basic tool for this step is the model HiREPS, a dynamical power system simulation and optimization model and its application to the Austrian and German electricity sector (more detailed documentation in section 4.1 and 4.2.).

So, the analysis will focus on the electricity sector. However, also the heating and cooling sector will be integral part of the analysis. Firstly, we established a link between the techno-economic bottom-up tool INVERT/EE-Lab modeling heating and cooling energy demand and the HiREPS model, ensuring the consideration of climate sensitive heating and cooling related electricity demand. Secondly, for the part of the heating and cooling energy demand that is not covered by electricity, we will define two extreme events: 3. Energy expenses for heating and cooling are above a certain threshold related to mean household income (for households) and related to mean value added (for the tertiary sector). 4. Indoor temperature levels are above a certain threshold for a certain period in case that no cooling device is installed.

4.1 Identifying critical situations (Black outs)

The core critical situation in the power system is considered to be a black-out. For the identification of such critical situations, the HiREPS model is run in a mode which includes the optimization of the power plant portfolios and investment decisions. It allows to identify Black-Outs or near-Black- Outpus. A Black Out as such cannot be simulated with the HiREPS model but if in a certain extreme situation a black out would occur the model will invest for this extreme situation all the required back up capacity to make the power system feasible. As a result of the simulation one can look at the hourly total power generation costs. This cost allocates the power system investments to the hours which causes the cost. Therefore the rare and extreme events causing extra investments to avoid a black out can be clearly identified.

Example: Scenario 2030 with • 10% (now 3%) wind power in Austria as share of electricity demand • 14% (now 8%) wind power in Germany as share of electricity demand • 3% (now 0%) solar power in Austria as share of electricity demand • 10% (now 2.1%) solar power in Germany as share of electricity demand • Total 23% (now 9%) solar and wind power as share of electricity demand in Austria and Germany

PRESENCE, First Review Workshop Seite 38 von 53

This scenario has the following maximum loads and generations: • Solar: 59 GW • Wind: 68 GW • Solar + Wind: 84 GW • Load: 122 GW (1.2% growth since 2010)

In the scenario it is allowed to curtail wind and solar power when the power price gets negative.

Fig. 16 Simulation Results: Residual Load

Fig. 17 Hourly total cost of power generation (including investment costs)

In the above figure the cost spikes at 23.11. 6pm and 13.11. 6pm extend far beyond the vertical axis.

Fig. 18 Analysis of the extreme event at 23.11. 6pm and 13.11. 6pm

At 23.11. 6pm and 13.11. 6pm are the two events with the maximum residual load for the whole simulated year. Therefore these two hours force the power system to install extra reserve capacity to cope with this exceptionally low wind and solar generation situation.

PRESENCE, First Review Workshop Seite 39 von 53

In scenarios extreme events could be: • Exceptional high residual load • Strong ramps in the residual load (e.g. wind and solar generation drops, load increases) • Long periods with low amount of renewables  Storages get empty

So, we will make use of the model HiREPS in order to identify situations where such extreme events (criticial situations) in the power system occur. For this purpose we will take into account the energy price and supply shocks and different storylines according to section 2. Moreover, we will identify critical combinations of weather/climate data with different sets of energetic side conditions. The weather/climate conditions will be subject of a probability assessment according to the methodology described in section 3.

HiRPES, covering the electricity supply subject to a certain demand will be coupled with a model cluster investigating the climate sensitivity of heating and cooling and the related electricity demand. So, climate sensitive heating and cooling electricity demand will be delivered to HiREPS on an hourly basis for all investigated scenarios. So, demand and supply conditions will be taken into account in identifying critical situations (extreme events) of the electricity sector.

4.2 HiREPS Model description

The HiREPS model is a dynamical power system simulation and optimization model. The focus of the model is to analyze the integration of fluctuating renewable electricity generation into the power system - by specifically including endogenously the important system constraints.

In general, critical aspects of RES-electricity generation are: • Variability of RES-electricity generation • Limits of the electricity grid • Limits in the flexibility of thermal power plants • Limits of the hydro power storage capacity • The HiREPS model endogenously addresses these aspects by: • using spatially and temporally highly resolved wind, solar and hydro inflow data • and by including a detailed model of: o hydro power and pumped storage o thermal power plants (including startup costs and efficiency losses at part loaded operation) o load flow calculation (including thermal limits of the electricity grid) o hourly temporal resolution of the model o HiREPS can be used to analyze: • What is the future role of pumped hydro storage and other advanced electricity storage concepts? • Which adjustments are needed for the historically grown power plant portfolios? PRESENCE, First Review Workshop Seite 40 von 53

• How to guarantee operationally and economically the system reliability and supply security? • What are the technically and economically feasible ways for balancing power provision? • What is the importance of future electricity grid extensions? • Which options are provided by the interconnected European transmission grid?

4.3 Typical model application cases

4.3.1 Variability of RES-electricity generation

The HiREPS Model uses historical weather data, to calculate for an assumed distribution of wind turbines, solar photovoltaic and solar thermal power plants the local renewable power generation across Europe. This localized renewable power generation is then used in the unit commitments simulation and the load flow simulations of the HiREPS model.

Fig. 19 Windspeed over Europe. Based on the COSMO-EU model

The wind speed data source for HiREPS is the COSMO-EU model of the German Weather Service DWD. The solar irradiation data is taken from the SOLEMI database of the German Aerospace Agency DLR. The figure 16 depicts the windspeeds over Europe 100 meter above ground for the 1st of February 2 am. The Figure is based on data from the COSMO-EU model.

4.3.2 Simulation of thermal power plants HiREPS dynamically simulates the unit commitment of the thermal power plants by including the technical and economical limitations. The technical constraints are for example maximum ramp rates, efficiency reduction at part loaded operation, minimum stable output, minimum on and off times. Economical contrainst are for example start up costs, fuel costs and CO2 costs.

PRESENCE, First Review Workshop Seite 41 von 53

Exemplary results of the Hireps simulation of the power generation in Germany and Austria are shown in the figure 17 for the first two weeks of January 2006. 90000 80000 70000 60000 50000 MW Solar 40000 Wind 30000 Hydro 20000 Gas Coal 10000 Nuclear 0 1.-14. Jannuary 2006 Fig. 20 Exemplary results of the Hireps simulation of the power generation in Germany and Austria

4.3.3 Simulation of future power prices

A fundamental feature of HiREPS is the endogenous calculation of hourly spot market prices. In principle, it is able to calculate locational prices (commonly known as “nodal pricing”), although the model is generally used to calculate zonal prices with regard to currently implemented market designs in Europe’s electricity markets. Market prices do play an important role for the actual operation as well as the future development of electricity markets. On the one hand the utilization of storages and the power plant resource scheduling results from the development of day-ahead prices. On the other hand the availability of storage and conventional generation capacities in turn have again a great influence on the prices. In the long term, the altering of the power system strongly depends on existing investment incentives araising from the temporal characteristics of the market prices as well. To adequately map these interactions, HiREPS utilizes comprehensive data bases with detailed information on technology scale. Based on calculated market prices the investment in new generation capacity and decommissioning of existing capacities are modelled endogenously within the model. In the figure 18 the modeled spot market prices and the corresponding residual load are illustrated on the example of the EEX market area. In this scenario 23% of demand is supplied bv solar and wind power. It can be seen that in times of low residual load resulting from high RES-generation prices partially drop to zero, whereas on the other hand characteristic price peaks can be observed in times of low RES-generation and limited availability of low-cost power generation. For this simulation the curtailment of the RES-E was allowed for avoiding negative prices.

PRESENCE, First Review Workshop Seite 42 von 53

EPEX Spot Price in July: Wind 14% and Solar 9 % of annual demand 140 100

120 80

100 60

80 40 GW 60

Euro/MWh 20 40 Mo Mo Mo Mo Mo 0 20

0 (20) Days EPEX Spot Price Residual Load Fig. 21 The modelled spot market prices and the corresponding residual load of the EEX market area

4.3.4 Detailed simulation of hydropower

Storage hydro power is the most important storage and balancing option to integrate variable renewable energy generation from solar and wind energy. Therefore HiREPS includes a detailed modeling of the hydropower sector in Europe. The hydropower system of France, Italy, Austria, Switzerland, Germany, Norway, Sweden and Spain is modeled in detail. These are the countries in the EU with significant amounts of hydropower. For the other European countries a more aggregated model of hydropower is used. In the figure 19 the HiREPS simulation of several Spanish hydro power plants of the Duero river basin are shown for the year 2007. The Almendra reservoir is the upper reservoir of the pumped hydro power plant Villarino. The other reservoirs in the graph are the upper dams of hydro storage power plants with no pumping. 5000 4500 4000 3500 3000 2500

GWh 2000 1500 1000 500 0 Time [2007] Aguilar Almendra Barrios Porma Riano Santa Teresa

Fig. 22 HiREPS simulation of several Spanish hydro power plants of the Duero river basin in 2007

PRESENCE, First Review Workshop Seite 43 von 53

Simulation of the market value of wind and solar power

The increasing shares of solar and wind power strongly influences the power prices. Because of this the market value of power from a local wind farm depends on the correlation of the local wind with the average European wind production. In the figure 20 the HiREPS simulation of marked value of windpower is analysed for the year 2008 assuming the EEX power prices. One can see that the market value varies quite singificantly.

Fig. 23 HiREPS simulation of marked value of windpower in 2008

4.3.5 Simulation of the transmission grid limitations

In order to analyse possible limitations of the transmission grid and to evaluate the costs and effects of these limitations HiREPS includes the option to simulate endogeneously the load flow in the transmission grid and consequences for the unit commitment in the case of overloaded lines. An example of the endogenously combined simulation of the transmission grid and unit commitment is given in the figure 21. The HiREPS simulation of the transmission line capacity utilization is shown as hourly values for one week in October 2006 for Austria.

PRESENCE, First Review Workshop Seite 44 von 53

100% 90% Mo Tu We Holiday Fr Sa Su 80% 70% 60% 50% 40% Utilitzation 30%

Transmission Capacity 20% 10% 0% Time Fig. 24 Endogenously combined simulation of the transmission grid and unit commitmen

PRESENCE, First Review Workshop Seite 45 von 53

Literature:

Resilience (part 1) Bohunovsky, L., Bruckner, M., Omann, I., Madlener, R., Stagl, S. (2006): Integrierte Nachhaltigkeitsbewertung von Energieszenarien. Eine lokale Fallstudie im Projekt ARTEMIS; Presentattion held at the 9th Symposium Energieinnovation, 15.-17.2.2006, Graz

Coaffee, J. 2008. Risk, resilience, and environmentally sustainable cities. Energy Policy 36: 4633- 4638.

Cutter, SL, Barnes, L, Berry, M, Burton, C, Evans, E, Tate, E, Webb, J. 2008. A place-based model for understanding community resilience to natural disasters. Global Environmental Change 18: 598-606.

Dillon, JH. 1947. Resilience of Fibers and Fabrics. Textile Research Journal 17: 207 -213.

Encyclopedia - Britannica Online Encyclopedia. http://www.britannica.com/bps/dictionary?query=resilience [accessed 5.9.2011]

Folke, C, Carpenter, S, Walker, B, Scheffer, M, Elmqvist, T, Gunderson, L, Holling, CS. 2004. Regime shifts, resilience, and biodiversity in ecosystem management. Annual Review of Ecology, Evolution, and Systematics: 557–581.

Folke, C. 2006. Resilience: The emergence of a perspective for social-ecological systems analyses. Global Environmental Change 16: 253-267.

Füssel, H-M. 2007. Vulnerability: A generally applicable conceptual framework for climate change research. Global Environmental Change 17: 155-167.

Hinkel, J. 2011. “ Indicators of vulnerability and adaptive capacity”: Towards a clarification of the science-policy interface. Global Environmental Change 21: 198-208.

Holling, CS. 1973. Resilience and stability of ecological systems. Annual review of ecology and systematics 4: 1–23.

Holling, CS. 2001. Understanding the complexity of economic, ecological, and social systems. Ecosystems 4: 390–405.

Jansen, JC, Seebregts, AJ. 2010. Long-term energy services security: What is it and how can it be measured and valued? Energy Policy 38: 1654-1664.

Kirchsteiger, C, Rademaekers, K, Van der Laan, J, Boeve, S, Lise, W. 2011. Investment needs for future adaptation measures in EUnuclear power plants and other electricity generationtechnologies due to effects of climate change. Brussels: DG Energy.

Klein, R, Nicholls, R, Thomalla, F. 2003. Resilience to natural hazards: How useful is this concept? Global Environmental Change Part B: Environmental Hazards 5: 35-45.

Klinke, A, Renn, O. 2002. A new approach to risk evaluation and management: risk-based, precaution-based, and discourse-based strategies. Risk analysis 22: 1071–1094.

PRESENCE, First Review Workshop Seite 46 von 53

Mechler, R, Hochrainer, S, Aaheim, A, Salen, H, Wreford, A. 2010. Modelling economic impacts and adaptation to extreme events: Insights from European case studies. Mitigation and Adaptation Strategies for Global Change 15: 737-762.

Norris, FH, Stevens, SP, Pfefferbaum, B, Wyche, KF, Pfefferbaum, RL. 2008. Community resilience as a metaphor, theory, set of capacities, and strategy for disaster readiness. American Journal of Community Psychology 41: 127–150.

O’Brien, G, Hope, A. 2010. Localism and energy: Negotiating approaches to embedding resilience in energy systems. Energy Policy 38: 7550-7558.

O’Brien, G. 2009. Vulnerability and Resilience in the European Energy System. Energy & Environment 20: 399-410.

O’Brien, K, Sygna, L, Haugen, JE. 2004. Vulnerable or Resilient? A Multi-Scale Assessment of Climate Impacts and Vulnerability in Norway. Climatic Change 64: 193-225.

Renn, O, Klinke, A, Asselt, M. 2011. Coping with Complexity, Uncertainty and Ambiguity in Risk Governance: A Synthesis. AMBIO 40: 231-246.

Rose, A. 2007. Economic resilience to natural and man-made disasters: Multidisciplinary origins and contextual dimensions. Environmental Hazards 7: 383-398.

Rose, A. 2009. Economic Resilience to Disasters. CARRI Research Report 8.

Tierney, K, Bruneau, M. 2007. Conceptualizing and measuring resilience: a key to disaster loss reduction. TR News.

Walker, B, Holling, CS, Carpenter, SR, Kinzig, A. 2004. Resilience, adaptability and transformability in social–ecological systems. Ecology and society 9.

WBGU. 2000. World in Transition: Strategies for Managing Global Environmental Risks. Springer Press, Berlin.

Wells, SR, Hill, L. 1913. The Influence of the Resilience of the Arterial Wall on Blood-Pressure and on the Pulse Curve. Proceedings of the Royal Society B: Biological Sciences 86: 180-186.

Williamson, LE, Connor, H, Moezzi, M. 2009. Climate-proofing Energy Systems. Report published by HELIO International [ http://www. heliointernational. org/uploads/Global% 20Report. En. pdf].

Extreme events (part 3) van den Besselaar, E. J. M., A. M. G. Klein Tank, G. van der Schrier, 2009: Influence of circulation types on temperature extremes in Europe. Theor. Appl. Climatol.: 99: 431-439.

Cindric, K., Z. Pasaric, M. Gajic-Capka, 2010: Spatial and temporal analysis of dry spells in Croatia. Theor. Appl. Climatol. 120: 171-184.

PRESENCE, First Review Workshop Seite 47 von 53

Coles S., 2001: An introduciton to statistical modeling of extreme values. Springer, Berlin, Heidelberg, New York, p 208.

Déqué, M., 2007: Frequency of precipitation and temperature extremes over France in an anthropogenic scenario: Model results and statistical correction according to observed values. Global and Planetary change, 57, 16-26.

Easterling, D. R., J. L. Evans, P. Ya. Groisman, T. R. Karl, K. E. Kunkel, and P. Ambenje, 2000: Observed Variability in Extreme Climate Events: A Brief Review. Bull. Amer. Meteor. Soc., 81: 417-425.

ESPON Climate Inception Report (2009): The ESPON 2013 Programme. ESPON CLIMATE – Climate Change and Territorial Effects on Regions and Local Economies. Applied Research Project 2013/1/4 http://www.espon-climate.eu/

Frei., C and C. Schär, 1998: A precipitation climatology of the Alps from high-resolution rain-gauge observations. Int. J. Climatol., 18, 873-900.

Frei, C., R. Schoell, S. Fukutome, J. Schmidli, P. L. Vidale, 2006: Future change of precipitation extremes in Europe: Intercomparison of scenarios from regional climate models. J. Geophys. Res. 111: D06105.

Halenka, T., J. Kalvova, Z. Chladova, A. Demeterova, K. Zemankova, and M. Belda, 2006: On the capability of RegCM to capture extremes in long term regional climate simulation – comparison with the observations for Czech Republic. Theor. Appl. Climatol. 86, 125-145.

Haylock, M.R., C.M. Goodess, 2004: Interannual Variability of European Extreme Winter Rainfall and Links with mean large-scale circulation. Int. J. Climatol. 24: 759-776.

Haylock, M.R., Hofstra, N., Klein Tank, A.M.G., Klok, E.J., Jones, P.D. and New, M., 2008: A European daily high-resolution gridded data set of surface temperature and precipitation for 1950-2006. Journal of Geophysical Research, 113, D20119, doi:10.1029/2008JD010201

Karl, T. R., R. W. Knight, D. R. Easterling, and R. G. Quayle, 1996: Indices of climate change for the United States. Bull. Amer. Meteor. Soc., 77, 279–291.

Kysely, J., R. Beranova, 2007: Climate-change effects on extreme precipitation in central Europe: Uncertainties of scenarios based on regional climate models. Theor. Appl. Climat., 95: 361-374.

Meehl, G. A., F. W. Zwiers, J. Evans, T. Knutson, L. Mearns, P. Whetton, 2000: Trends in extreme weather and climate events: Issues related to modeling extremes in projections of future climate change. Bull. Amer. Meteor. Soc. 81, 427-436.

Nachtnebel, P., R. Steinacker, E. Rudel, P.Stanzel, M. Buechinger, I. Aschauer, A. Lexer, T. Krennert, C. Zingerle, 2010: Trendanalyse von hydro-meteorologischen Extremwerten. Endbericht.

Paeth, H. and A. Hense, 2005: Mean versus extreme climate in the Mediterranean region and its sensitivity to future global warming conditions. Meteorologische Zeitschrift, Vol. 14, No. 3, 329- 347. PRESENCE, First Review Workshop Seite 48 von 53

Palmer, T.N. and J. Raeisaenen, 2002: Quantifying the risk of extreme seasonal precipitation events in a changing climate. Nature, 415, 512-514.

Regional Challenges in the Perspective of 2020 – Phase 2: Deepening and Broadening the Analysis. EU-Contract Study. Final Report.

ReRisk Regions at Risk of Energy Poverty, 2010: The ESPON 2013 Programme, Draft Final Report: http://www.espon.eu/export/sites/default/Documents/Projects/AppliedResearch/ReRISK/RERIS K-Draft-Final-Report_31-03-2010.pdf

Unkašević, M. and I. Tošić, 2011: A statistical analysis of the daily precipitation over Serbia: trends and indices. Theor. Appl. Climatol., 106: 69-78

PRESENCE, First Review Workshop Seite 49 von 53

ANNEX 1 Resilience indicators in the energy sector

Criteria Indicator / Description Source A Adaptability Lifetime (minimize), Modularity (maximize) (Data: GEMIS) [Bohunovsky et al 2006] of energy Flexibility of technology is the ability of an energy system to react flexible on changes of frame system conditions and demand. Many energy supply technologies are capital intensive and cause irreversible costs. On the contrary modular technologies enable stepwise capacity deployment, reducing investment risk and allowing a more optimal, flexible capacity planning (shorter planning horizons) R/ Fiscal Fiscal deficit and state of fiscal space (%GDP Data: Eurostat) [Mechler et al. 2010] A resilience The government’s portfolio of ex ante and ex post financial measures is critically important for the recovery of the economy should a disaster occur. For this reason, an assessment of the government’s asset risk and fiscal resilience is an essential part of disaster risk management. V Coal Number of coal mines plants located at less than one meter above sea level and within the area [Williamson et al. 2009] that could be flooded by a flood with a current recurrence period of 100 years V Oil and Gas Share of offshore oil and gas installations likely to be hit by a storm of more than 70 m/s gusts [Williamson et al. 2009] within the next 20 years (%)

V Oil and Gas Share/number of refineries likely to be hit by a storm of more than 70 m/s gusts within the next [Williamson et al. 2009] 20 years (%) V All Fossil Number of thermal (coal, oil and gas) power plants located at less than one meter above sea level and within the [Williamson et al. 2009] Fuels area that would be flooded by a flood with a current recurrence period of 100 years

V Nuclear Number of nuclear power plants located at less than one metre above sea or river level and [Williamson et al. 2009] within the area that would be flooded by a flood with a current recurrence period of 100 years

V Nuclear Number of incidents/accidents since the plant was built [Williamson et al. 2009] and describe the most significant incidents V Hydro Expected precipitation change over next 20-50 years (%) and/or probability of floods in each [Williamson et al. 2009] watershed

PRESENCE, First Review Workshop Seite 50 von 53

V Hydro Number of multiple-use dams in the country today: volume of water (m3) of each dam [Williamson et al. 2009] Additionally: Describe what % of the water is used for: agriculture; power; drinking Expected additional run-off from glacier melting (million m3)

V Transmission Length of in-country, above-ground transmission and distribution lines (km) [Williamson et al. 2009] Systems

V Transmission Distinguish between: high (transmission ); middle + low voltage lines (distribution) [Williamson et al. 2009] Systems Additionally: Describe any transnational lines V Transmission Number and length of power cuts (differentiate between failures due to weather or equipment [Williamson et al. 2009] Systems failure and those cuts due to rationing) V Transmission Average hours of interruption per year [Williamson et al. 2009] Systems V Transmission Percentage of energy supply requiring regional transport over 50 km [Williamson et al. 2009] Systems V Transmission % that is transportation of fossil fuel [Williamson et al. 2009] Systems V Transmission % that is transportation of biomass [Williamson et al. 2009] Systems V Biomass Proportion of biomass used for energy purposes (%) in total biomass production [Williamson et al. 2009] V Biomass If possible distinguish between different sources and different applications – agricultural biomass harvest; [Williamson et al. 2009] electricity; heat V Biomass Forest (as defined by FAO) biomass harvest: electricity; heat [Williamson et al. 2009] V Biomass Expected precipitation change over next 20-50 years (%) [Williamson et al. 2009] Additional information: Probability of temperature increase beyond biological heat tolerance of key biomass crops within the next 20 years (%)

V Wind Number of wind turbines at less than one metre above sea level [Williamson et al. 2009] V Wind Projected change of average windspeed over the next 20 years, based on regional climate models (%) [Williamson et al. 2009] A Solar Capacity of solar installations already in place (m2) [Williamson et al. 2009] A Solar Distinguish between PV (MW) and thermal (m2) [Williamson et al. 2009] A Solar Describe sites (quality of the insulation and of the building on which systems are installed) and type of ownership [Williamson et al. 2009]

PRESENCE, First Review Workshop Seite 51 von 53

(private, government, public/private partnership, etc.) Expected temperature (°C) increase in the next 20 years relevant for PV capacity Additional information: Projected change in rainfall and cloud cover over next 20 years (%) A Fiscal Domestic capital formation (million USD per year) – Proxy: Domestic savings (million USD per year) [Williamson et al. 2009] A Fiscal Domestic investment in renewable energy (million USD per year) [Williamson et al. 2009] A Knowledge Number of technical engineers graduating annually as a percentage of the total population [Williamson et al. 2009] A Knowledge Availability of hazard maps for floods/droughts [Williamson et al. 2009] A Knowledge Existence and enforcement of power plants siting and construction guidelines taking climate change into [Williamson et al. 2009] consideration (If there is no information available, discuss qualitatively how climate change could effect siting and construction guidelines) A Security Existence of emergency plans to react to extreme meteorological events and availability of local emergency repair [Williamson et al. 2009] teams Comment if possible on the level of implementation A Fiscal A Participation Existence of citizens’ users groups in the energy governance structure (enforcement of participatory decision- [Williamson et al. 2009] making) A Power System Existence and use of a siting map for mines and power plants taking into account projected storms, floods and [Williamson et al. 2009] drought areas A Power System Implementation of national regulations for thermal power plant siting at sites with sufficient cooling water availability [Williamson et al. 2009] over the next 50 years A Hydro Existence of a national plan for optimised operation of hydropower plants under projected flow regimes for systems [Williamson et al. 2009] A Hydro Is such a plan currently in place? [Williamson et al. 2009] A Hydro If not, has the government decided to have one at a future date? [Williamson et al. 2009] A Hydro Number of dams equipped with desilting gates and/or number of up-stream land use management and water [Williamson et al. 2009] catchment plans for each hydro installation A Biomass Research,development and dissemination budget for heat and drought resistant crops, biofuels, agricultural waste [Williamson et al. 2009] for energy and vulnerability of forest (million USD/year) does not include municipal waste as this is usually considered in mitigation plans A Biomass If possible, comment on consistency of funding [Williamson et al. 2009] A Biomass In-country utilisation of biomass fuels not traditionally used by private enterprises and cooperatives (percentage of [Williamson et al. 2009] total fuels) A Biomass Percentage of households using improved woodstoves out of total number of households using woodstoves [Williamson et al. 2009] A Wind Existence and enforcement of national regulations requiring storm proofing of wind power plants to withstand [Williamson et al. 2009] highest anticipated windspeed A Wind Existence of siting maps that detail projected changes in: windspeed; floodplains; and, areas impacted by sea level [Williamson et al. 2009] rise

PRESENCE, First Review Workshop Seite 52 von 53

A Solar Existence of siting maps that detail projected changes in cloud cover [Williamson et al. 2009] A Solar Existence and enforcement of national regulation requiring storm proof concentrating solar power plants (CSP) to [Williamson et al. 2009] withstand the highest anticipated windspeed R Governance Good adaptation under extenuating circumstances; a recovery trajectory that returns to baseline functioning [Williamson et al. 2009] following a challenge V Infrastruct. Land area in 100-year flood plain (%) [Cutter et al. 2008] V Biomass Soil Erosion (%) [Cutter et al. 2008] A Institutional Recent hazard mitigation plan [Cutter et al. 2008] A Fiscal Sylves 2007 nach [Cutter et Municipal expenditures (fire, police, emergency services) (%) al. 2008] R Energy Grid Resilience of Energy Grid (Network resilience vgl. Ulanowicz 2011 in this text) mE R Energy supply Diversity of energy supply (Shannon-Wiener diversity index) mE V Fiscal Micro-economic costs of energy services do not exceed certain thresholds for different parts of the population (e.g. PRESENCE a certain share of household income). V Energy supply supply of energy services does not decrease stronger than a certain threshold within a certain period PRESENCE V Energy supply The electricity system does not exceed a limited number of (near) black outs PRESENCE V Fiscal Macro-economic costs of energy services do not exceed certain thresholds PRESENCE

PRESENCE, First Review Workshop Seite 53 von 53