European Commission
Directorate-General Environment
Unit D.1. Water
ENV.F.2 (BU-5 00/122)
Background Document on Vulnerability Indicators
for the project
Climate Adaptation –
modelling water scenarios and sectoral impacts
Contract N° DG ENV.D.2/SER/2009/0034
8. September 2010
Alterra
In co-operation with
CESR – Center for Environmental Systems Research,
Wissenschaftliches Zentrum für Umweltsystemforschung
Kurt-Wolters-Str. 3, D – 34109 Kassel, Tel.: +49 561- 804-3266, Fax: +49 561- 804-3176
Introduction
Vulnerability to climate change impacts in the water sector mostly relates to extreme events of
a wide variety of types and duration. This summary paper first describes how climate change
can exacerbate the vulnerability to current climate variability and the associated different types
of hydrological events. It then summarizes various indicators that can be used to monitor and
project changes in water problems due to climate change. In the appendices maps are included
representing those vulnerability-related indicators that can currently be quantified at a
European level.
Definition of water-related problems
Water resources managers are familiar with the variability of water quantity and quality
parameters as planning and management of water resources systems requires knowledge about
the statistical properties of hydrological events. Figure 1 depicts for example long term time
series of precipitation. On its basis not only long term annual average, but also rare extreme
events such as frequency of occurrence of one in 50 years flooding and drought can be
estimated.
Figure 1. Long term precipitation time series.
The estimations, made on basis of statistical analysis can reveal, for example, the development
of water scarcity conditions in the region of consideration (Fig.2).
1
Figure 2. Longterm water
availability series,
indicating water
scarcity.
In order to make clear distinction between water scarcity and drought, we will introduce the
following definitions:
Droughts are temporary decreases of the average water availability. They are natural
phenomena, originating from prolonged deficiency in precipitation over an extended region.
Droughts are periods during which relevant hydrological variables such as precipitation, river
discharge, and aquifer level or soil moisture are continuously below a certain threshold. Based
on the specific part of the hydrological cycle which is affected, different types of droughts can
be identified:
- meteorological (mainly due to changes in precipitation in combination with other climatic
factors such as high temperature, wind and low humidity),
- agricultural (when soil moisture and plant growth is affected),
- hydrological (leading to low water levels in streams, reservoirs, lakes and reduced
groundwater levels).
When making this classification, it should be borne in mind, however, that a drought in one
stage of the cycle often leads to a drought also in other stages.
Although droughts are natural phenomena, they can be enhanced significantly by
anthropogenic factors such as population and economical growth and land use changes.
(Tallaksen and van Lanen 2004)
Water scarcity1 is defined as a situation where insufficient water resources are available to
satisfy long-term average requirements. It refers to long-term water imbalances, combining low
water availability with a level of water demand exceeding the natural recharge.
1
This definition is taken from the second Interim report “Water Scarcity and Droughts”, prepared by DG
Environment – European Commission, June 2007
2
There are various ways in which hydrological events can be characterised statistically, but for
the extreme, hazardous events such as flooding and droughts, the magnitude of the event (its
severity or intensity) duration and timing (time of the year when they appear) (see Fig.3) are of
great importance. The importance of these characteristics differs for the different events and
sectors: for example the magnitude determines a flooding categorization, while for droughts it is
the combination of magnitude and duration, which represents the cumulative water deficit. The
timing of an event such as delay in the start of the rainy season could have significant impact on
agriculture, depending on the seasonal phenology2 of the crops, but could be irrelevant for
industry.
Figure 3. Main
characteristics of
hydrological events.
Another key characteristic of the extreme events is the return period, defined as frequency or
recurrence interval of the event with given intensity and duration over an extended period of
time (red circles on Fig 1). It is particularly of interest for risk analysis, for example for planning
of structures able to withstand an event with a given intensity such as dykes and building (see
Fig 4) or for building water reservoirs.
Climate change is projected to have significant impacts on the hydrological cycle. Changes in
temperature, evaporation and precipitation will impact the quantity and distribution of river
flows, soil moisture and groundwater recharge. Current global climate change models indicate
that the magnitude and frequency of extreme events could increase due to climate change.
Seasonal patterns and return periods are projected to be modified, too (Kundzewicz et al. 2008).
As a result, the main characteristics of the hydrological events, depicted in Fig. 3 may alter as
shown in Fig. 5
2
Phenology studies periodic plant and animal life cycle events and how these are influenced by seasonal
and interannual variations in climate
3
Figure 4. Example of level of protection against floods as a function of the land use, Source: EEA
2001, adapted from Saelthun and Tollan, 1996.
Figure 5. Altered
characteristics of
hydrological events
due to climate
change.
Figure 5 suggests that although other main characteristics might be altered, this does not
necessarily alter mean changes in the average. One of the mechanisms that may lead to such a
result is a simultaneous change in the magnitude and duration of one event, yielding shorter,
but more intensive events, as in Fig 5. Fig. 6 depicts another mechanism that may produce the
4
same result on a longer time scale – intensification and increase of the number of both droughts
and flooding that do not change the annual average.
Figure 6. Example of Long-term
precipitation time series, depicted
in Fig. 1, are altered by climate
change, but the annual average
stays the same.
Fig. 6 shows that significant changes in the return periods of flooding and droughts can occur,
too: 1-in-100 years flooding might become 1-in-50 years flooding and 1-in-50 years drought
might become 1-in-10 years drought. In addition to these two mechanisms, the models may
project a decrease in the availability of freshwater resources in some places and periods, and
their increase in other places and periods, which implies that the long-term annual average will
be altered in some places. Main consequences of change in annual average are the possibility of
augmentation of water scarcity and water logging conditions.
Vulnerability
Currently large regions in Europe suffer from water-related hazards and water scarcity. The
expected changes due to global warming might further aggravate them, increasing vulnerability
of socio-ecological systems. Therefore these changes in the water cycle and their impact have to
be studied, monitored and assessed in order to decrease vulnerability and adapt to them
successfully. Such assessments can be done only on basis of indicators, allowing current
vulnerability to be compared with future vulnerability due to climate change.
There are many different definitions of vulnerability, dependent on the framing of the problem
that is considered. For the purpose of this project, we adopt the definition that is often used in
the context of adaptation to climate change, e.g. by the IPCC and EEA. Vulnerability in this
context is often referred to as having three components (see also Fig. 7):
- exposure being the “nature and degree to which a system is exposed to significant climatic
variations” (exposure to climate factors);
- sensitivity being the “degree to which a system is affected, either adversely or beneficially,
by climate-related stimuli” (sensitivity to change); and
- adaptive capacity being the “ability of a 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.”
5
UNECE (2009) considers the exposure as an external and the sensitivity and adaptive capacity as
an internal dimension of vulnerability. They also distinguish between the current and future
vulnerability including the coping capacities. Hence, for future adaptation, efforts should aim to
reduce vulnerability by reducing exposure and sensitivity (potential impacts) and increasing
adaptive capacity. Indicators can be useful to assess the effectiveness of measures targeting
these different components. Further in this document we will use this general conceptual
structure, but we will make it more operational in order to derive associated vulnerability
indicators. A more detailed description of each selected indicator is presented in Appendix 1. In
the following pages, we will summarize definitions of the components of vulnerability and
provide some examples of indicators. Because of lack of data, quantification of these indicators
is only possible for a very limited number of the indicators at the European level. For pragmatic
reasons, we will focus on those indicators, acknowledging that they not give a complete picture
of different aspects of vulnerability, while it is also to be kept in mind that the Europe-wide
maps should be interpreted with caution, since actual vulnerability to specific climate-related
water problems is determined by local characteristics.
Figure 7. Conceptual
model for climate
change impacts,
vulnerability and
adaptation. Based on
Isoard, Grothmann and
Zebisch, 2008)
6
Indicators for water scarcity
Exposure to water scarcity is the degree to which a system is exposed to available long-term
average water quantity. The exposure will affect long-term water availability and thus long-term
water supply.
Examples of possible indicators: Average precipitation, average river discharge, average soil
moisture, groundwater level. This kind of indicator is available from climate modelling work;
those in italics are included in Appendix 1.
Sensitivity to water scarcity is determined by the influence of the stress factor (e.g. climate
change or changes in hydrological regime) on the functioning of the system; For example in
some basins a reduction of 10% rainfall results in a 30% reduction in water availability while in
other basin reduction in rainfall only have a marginal impact of on stream flow. Also systems
where water demand is high are more sensitive to water scarcity than systems where water
demand is low. Indicators for sensitivity related to both supply and demand of water in a
particular for the exposure we propose to focus on demand and supply related indicators.
Examples of possible indicators: Change in water demand in the future, compared to some base
period. This demand could be further disaggregated according to different users and sectors:
domestic, agriculture, industry, energy production, tourism. This kind of indicator is available
from modelling work (e.g., WaterGAP); those in italics are included in Appendix 2 and given in
the background document on scenarios).
Impacts of water scarcity. If water availability exceeds water demand, the difference between
long-term water supply and water demand will result in a water “gap”, which will negatively
influence socio-ecological systems. These influences we will label (potential) “impacts”.
Examples of possible indicators: Loss of industrial and agricultural production, of jobs, income
and livelihoods; desertification and land degradation. These indicators have not yet been
quantified in any systematic way.
Adaptive capacity to water scarcity is determined by the ability/possibility of regions or sectors
to close the gap between water demand and supply. It could be achieved by enhancing the
societal ability to increase water supply, decrease water demand or some combination of both.
Adaptive capacity is a very challenging concept and difficult to make operational as indicator.
Therefore we will adapt a more practical approach, looking at the capacity to implement
measures, necessary for the reduction of the identified vulnerability, and at cultural,
technological, financial and institutional barriers which hamper this implementation.
Vulnerability to water scarcity is determined by the gap between water supply and demand and
is expressed as a % change of current and expected future water (in-)sufficiency in comparison
with a baseline.
Examples of possible indicators: Water Stress3; Water Exploitation Index (WEI)4 ; Falkenmark
index5. This kind of indicator is available from modelling work (e.g., WaterGAP); those in italics
are included in Appendix 3.
3
Measured as the ratio of total water availability to total water withdrawals
4
Water Exploitation Index (WEI) for a country is calculated as the mean annual total demand for
freshwater divided by the long-term average freshwater resources
5
Measures renewable water recourses per capita
7
Indicators for droughts
Exposure to droughts is the degree to which a system is exposed to climatic variations of the
quantity, magnitude, frequency and seasonality of available water. The exposure will affect
temporal water availability and thus temporal water supply.
Examples of possible indicators: Severity, duration, return periods and timing of drought events
due to temporal decrease of precipitation, river discharge, soil moisture, groundwater and water
stored in lakes and dams below some threshold level. This kind of indicator is available from
climate and hydrological modelling work; those in italics are included in Appendix 1.
Sensitivity to droughts in this project is mostly determined by the temporal changes in the
water demand of the sectors (necessary quantity of water with desired quality).
Examples of possible indicators: Water demand during drought events. This demand could be
further disaggregated according to different users and sectors: domestic, agriculture, industry,
energy production, tourism. This kind of indicator could be made available from modelling work
for specific situations. See appendix 1 for generic information on demand.
Impacts of droughts: If water demand exceeds water availability (supply), the difference
between water supply and water demand will result in a water “gap”, which will influence
negatively socio-ecological systems. These influences we will label “impacts”.
Indicators: Temporal loss of industrial and agricultural production, jobs and income. These
temporal losses may become permanent if socio-ecological systems can not recover from
recurring droughts and a sequence of droughts leads to long term water scarcity. These
indicators have not yet been quantified in any systematic way.
Adaptive capacity to droughts and water scarcity is determined by the ability/possibility of the
sectors to close the gap between water demand and supply over time. It could be achieved by
enhancing the societal ability to increase water supply, decrease water demand or some
combination of both. Adaptive capacity is a very challenging concept and difficult to make
operational as indicator. Therefore we will adapt a more practical approach, looking at the
capacity to implement measures, necessary for the reduction of the identified vulnerability, and
at cultural, technological, financial and institutional barriers which hamper this implementation.
Vulnerability to droughts and water scarcity is determined by the gap between water supply
and demand and is expressed as a % change of current and expected future water insufficiency
in comparison with a baseline.
Examples of possible indicators: Palmer Drought Severity Index (PDSI)6, Crop moisture index7.
This kind of indicator could be derived from modelling work but is not yet available to the
project.
6
Measures meteorological droughts. PDSI is based on the cumulative difference between normal
precipitation and precipitation needed for evapotranspiration (Palmer, 1965). Alley (1985) adjusted it to
measure the hydrological drought as well.
7
Measured as the difference between the actual and expected weekly evapotranspiration (Palmer, 1968)
8
Indicators for flooding
Exposure to flooding is the degree to which a system is exposed to climatic variations of the
quantity, magnitude, frequency and seasonality of high waters and to sea level rise.
Examples of possible indicators: Severity, duration, return periods and timing of flooding events
due to increase of precipitation and river discharge above some threshold level and sea level
rise. This kind of indicator is available from climate and hydrological modelling work; those in
italics are included in Appendix 1.
Sensitivity to flooding is mainly determined by the share of socio-ecological systems, located in
the flood-prone areas. Please note that sensitivity would be decreased by the presence of
wetlands or protective infrastructure, while the sensitivity of a natural river bed can be lower
than for a canalized river. Because of lack of data about these factors the sensitivity indicators
below reflect the potential sensitivity rather than necessarily the full sensitivity.
Examples of possible indicators: Number of people, infrastructure, crops, livestock, forests, and
industrial production capacities located in flood-prone area. Some of these indicators are
available for Europe (e.g., at JRC) and will be taken into account in the project at a later stage.
Impacts of flooding are the impacts, emerging from the changes in the flood exposure
parameters and the part of socio-ecological system, located in the flood-prone zone.
Examples of possible indicators: (temporal) loss of shelter, jobs and livelihoods due to
flooding. These indicators have not yet been quantified in any systematic way.
Adaptive capacity to flooding is determined by the ability/possibility to protect the system
against flooding. Adaptive capacity is a very challenging concept and difficult to make
operational as indicator. Therefore we will adapt a more practical approach, looking at the
capacity to implement measures, necessary for the reduction of the identified vulnerability, and
at cultural, technological, financial and institutional barriers which hamper this implementation.
Vulnerability to flooding is determined by the extent to which the human-ecological system can
(or can not) be protected against flooding (expressed as a % change of current and expected
future casualties in comparison with a baseline).
Examples of possible indicators: Relative vulnerability for flooding8, People flooded, Total flood
damages, Flood damages in different sectors. Some of these indicators can be made available
for Europe and will be taken into account in the project at a later stage when new JRC LISFLOOD
results become available.
8
Measures the ratio between number of people killed by the number of people exposed (UNDP)
9
Discussion and disclaimer
In our approach, the indicators for exposure (summarized in Fig. 8) do not depend on socioeconomic factors (who is exposed?, such as water users, or people and economic activities in
flood-prone areas) and hence are basically sector-independent. However, the other components
of vulnerability add the socio-economic dimension, differing for each sector and user group.
Table 1 presents the exposure indicators for each sector, but now combined with the indicators
for sensitivity of the sectors and the climate-related water impacts on these sectors. It should be
noted that availability of data limits the valuation of the indicators presented, particularly at the
more detailed level.
Water
scarcity
Changes in long
term Water
availability
Mean
precipitation
Mean evapotranspiration
Mean river
discharge
Long term trend
in soil moisture
Long term
changes in
groundwater level
Long term
changes in water
reservoirs
Drought
Changes in
temporal Water
availability
Flooding
Changes in
temporal Water
quantity
Change in occurrence (Return
period), intensity, duration and
timing of 1-in-X y drought
Change in occurrence (Return
period), intensity, duration and
timing of 1-in-X y flooding
Change in occurrence (Return
period), intensity, duration and
timing of 1-in-X y low
precipitation
Change in occurrence (Return
period), intensity, duration and
timing of 1-in-X y high
precipitation
Change in occurrence (Return
period), intensity, duration and
timing of 1-in-X y low flow
Change in occurrence (Return
period), intensity, duration and
timing of 1-in-X y high flow
Temporal Evapo-transpiration
above given local threshold
Temporal soil moisture above
given local threshold
Temporal soil moisture below
given local threshold
Temporal changes in water
reservoirs above given local
threshold
Temporal changes in
groundwater below given local
threshold
Temporal changes in water
reservoirs below given local
threshold
Figure 8. Climate-related indicators for exposure.
Evapo-transpiration
Seasonal distribution of
precipitation Duration of
periods with low precipitation
10
We also note that generic, Europe-wide indicators can hide important differences between
regions and sectors. For example, because of different biophysical, socio-economic and
institutional differences a drought event which is similar in purely climatologically terms may
have quite different impacts in different regions (e.g., Finland or Spain). If a potential flood due
to high river levels actually leads to real impacts depends on local flood protection measures
which do not feature in European databases. In order to capture these differences, local
thresholds would have to be taken into account to determine if a potential climate change
impact actually leads to problems. To take these differences into account, more in-depth
analysis is required that is beyond the scope of this project.
As to response measures, the database compiled for the project has a level of detail beyond the
capacity of Europe-wide databases and models. As is discussed in the material on the inventory
and evaluation of measures, only a limited share of the available measures can be evaluated
with the coarse (large scale) models, and then usually in an aggregated fashion. The role of
adaptive capacity will be addressed later in the project. Vulnerability assessment can lead to the
identification of no-regrets measures to enhance adaptive capacity, but usually cannot provide
justification for costly measures (Patt et al., 2005). According to Patt et al. (2005), the
combination of climate change projections, socio-economic scenarios and estimates of adaptive
capacity for a broad evaluation of vulnerability should actually be avoided and a more narrow
focus on risks of particular communities could provide more meaningful results. However, for a
rough picture of different vulnerabilities across Europe as a method to identify such
communities the assessment and integrated framework in this project could be useful.
Questions to stakeholders regarding vulnerability indicators:

How and for which purposes do you think that the vulnerability indicators could be used?

Do you agree with the current selection of indicators, as it is determined mainly by
availability of data and scope of models? If not, how can they be improved and which data
sources would be used?

How do you see the relationship between generic EU-wide indicators and location-specific
characteristics of vulnerability indicators, e.g. thresholds for impacts?
11
References
Alley, W.M., 1985, The Palmer Drought Severity In Measure of Hydrologie Drought, Water
Resources Bulletin, 21 (1), 105-114
Cancelliere, A., Cubillo, F., Wilhite, D.A. , 2009a, Coping with drought risk in agriculture and
water supply systems: Drought management and policy development in the Mediterranean.
Springer, The Netherlands
Iglesias, A., Garrote, L., Quiroga, S., Moneo, M., 2009, Impacts of climate change in agriculture in
Europe. PESETA-Agriculture study
EEA (2008) Impacts of climate change in Europe: An indicator based report.
Palmer, W.C., Meteorological Drought, Research Paper 45, US Weather Bureau, Washington, 19
65
Palmer, W.C., Keeping Track of Crop Moisture Conditions Nationwide: The New Crop Moisture
Index, Weatherwise, 21 (4), 1968, pp 156-61
Patt, A.,R.J.T.Klein and A. de la Vega-Leinert (2005). Taking the uncertainty in climate-change
vulnerability assessment seriously. Comptes Rendus Geoscience 337, pp. 411-424
PRUDENCE (2007) (Prediction of Regional scenarios and Uncertainties for Defining
EuropeaN Climate change risks and Effects) http://prudence.dmi.dk/
Tallaksen, L.M., and van Lanen, H., 2004, Hydrological Drought – Processes and Estimation
Methods for Streamflow and Groundwater, Developments in Water Science, Elsevier,
Amsterdam
UNECE, 2009, Guidance on Water and Adaptation to Climate Change, United Nation Publications
http://www.unece.org/env/water/publications/documents/Guidance_water_climate.pdf
12