d8.3 report on needs in hydropower sector - imprex · v.09 31/07/2016 a. castelletti integration of...

D8.3 Report on needs in

hydropower sector

IMPREX has received funding from the European Union Horizon 2020 Research and Innovation

Programme under Grant agreement N° 641811 2

Deliverable D8.3 Report on needs in hydropower sector

Related Work Package: WP8

Deliverable lead: POLIMI

Author(s): Andrea Castelletti, Yu Li, Matteo Giuliani, Maria Helena

Ramos, Hector Macián-Sorribes, Manuel Pulido Velasquez,

David Gustafsson, Rodolfo Soncini Sessa

Contact for queries [email protected]

Grant Agreement Number: n° 641811

Instrument: HORIZON 2020

Start date of the project: 01.10.2015

Duration of the project: 48 months

Website: www.IMPREX.eu

Abstract This report aims at reviewing the existing knowledge and

needs for weather and climate services in the hydropower

sector. It presents the results of a survey designed to

objectively assess the current practices and future needs of

the hydropower industry. The survey results show that all

11 respondents have already been using forecasts

products of various forms. Public free weather and climate

services tend to be the main data source to retrieve

forecasts information, and the majority of the respondents

hold a positive evaluation of the quality of current

products. In addition, respondents expect future

improvements to be focused on enhancing the forecast of

extreme events and extending the forecast lead-time.

http://www.powerstep.eu/

3

Deliverable n° D8.3

Dissemination level of this document

X PU Public

PP Restricted to other programme participants (including the Commission

Services)

RE Restricted to a group specified by the consortium (including the

European Commission Services)

CO Confidential, only for members of the consortium (including the European

Commission Services)

Versioning and Contribution History

Version Date Modified by Modification reasons

v.01 01/07/2016 A. Castelletti, M.

Giuliani

Definition of deliverable structure,

contents, and partner’s contributions

v.02 24/07/2016 Y. Li, R. Soncini-

Sessa, A. Castelletti,

H. Macián-Sorribes,

M. Pulido-Velazquez

Description of Italian (YL, AC) and Spanish

stakeholders (HM, MP)

v.03 25/07/2016 Y. Li, M. Giuliani, A.

Castelletti

Description of survey and analysis of

replies.

v.04 26/07/2016 M.H. Ramos Description of French stakeholder and

contribution to Sections 1 and 2


Giuliani

Review of sections 1-3 and conclusions.



v.06 27/07/2016 M.H. Ramos, H.

Macián-Sorribes, M.

Pulido-Velazquez

Review of section 4 (MHR) and of the

Spanish stakeholder (HM, MP).


Giuliani

Final review of full report.

v.08 31/07/2016 M.H. Ramos Integration of review comments received

from J. Hunink (Future Water) on Section 2

and overall review of the report.

v.09 31/07/2016 A. Castelletti Integration of review comments received

from J. Hunink (Future Water) on Section 2

and overall review of the report and

appendix 2.

v.10 1/08/2016 M. Giuliani Review of report format and Appendix 1.

v.11 1/08/2016 D. Gustafsson Addition to the Swedish case study

v.12 9/08/2016 M. Giuliani, A.

Castelletti, M.H.

Ramos

Review of report implementing comments

by Bart van den Hurk

v.13 08/09/2016 H. Macián-Sorribes,

M. Pulido-Velazquez,

D. Gustafsson

Detailed info on stakeholders added

v.14 19/09/2016 M. Giuliani, A.

Castelletti

Final review

v.15 03/07/2017 M.H. Ramos Abstract added. Final version after remarks

from EU Project Officer and reviewers.

5


Table of Contents

List of figures ................................................................................................................................................................... 7

List of tables .................................................................................................................................................................... 7

Executive Summary ....................................................................................................................................................... 8

1 Introduction ............................................................................................................................................................ 9

2 Climate services for hydropower ................................................................................................................. 11

2.1 Overall aspects related to the provision of climate services ................................................ 11

2.2 Some recent results related to users in the energy sector ................................................... 13

2.3 Particular features of the hydropower sector .............................................................................. 14

2.4 Challenges and opportunities for the hydropower sector ..................................................... 16

3 Review of stakeholder knowledge and needs in the hydropower sector ................................ 19

3.1 Methodology ............................................................................................................................................... 19

3.2 Description of stakeholders .................................................................................................................. 21

3.2.1 A2A (Italy) ............................................................................................................................................ 22

3.2.2 EDF (France) ........................................................................................................................................ 26

3.2.3 Vattenfall (Sweden) ......................................................................................................................... 29

3.2.4 Iberdrola (Spain) ............................................................................................................................... 32

4 Analysis of stakeholders’ responses .......................................................................................................... 36

4.1 Profiles of the respondents .................................................................................................................. 36

4.2 Current use of W&C services .............................................................................................................. 39

4.3 Application of forecasts to decision-making ................................................................................ 42

4.4 Expectation from W&C services ......................................................................................................... 43

5 Conclusions ........................................................................................................................................................... 47

6 References ............................................................................................................................................................. 48

7 APPENDIX 1 .......................................................................................................................................................... 52

7.1 Respondents’ background .................................................................................................................... 52



7.2 Use of W&C services ............................................................................................................................... 54

7.3 HP company profile ................................................................................................................................. 66

7.4 Additional questions ................................................................................................................................ 68

8 APPENDIX 2 .......................................................................................................................................................... 72

8.1 Current use of W&C services .............................................................................................................. 72

8.2 Application of forecasts to decision-making ................................................................................ 77

8.3 Expectation from W&C services ......................................................................................................... 78

7


List of figures

FIGURE 1: GROWTH OF GROSS MAXIMUM CAPACITY (TOP PANEL; GSE, 2016) AND NUMBER OF HYDROPOWER PLANTS (BOTTOM PANEL;

GSE, 2015). ......................................................................................................................................................... 22 FIGURE 2: HYDROELECTRIC ENERGY PRODUCTION IN ITALY FOR 2014 (GSE, 2015). .................................................................... 23 FIGURE 3: ORGANIZATION STRUCTURE OF A2A GROUP (SOURCE: A2A GROUP WEBSITE). ............................................................. 23 FIGURE 4: GEOGRAPHICAL AREA OF A2A ACTIVITIES (SOURCE: A2A GROUP WEBSITE). ................................................................. 24 FIGURE 5: EDF GROUP INSTALLED CAPACITY AND ELECTRICITY GENERATION IN 2015 (SOURCE: EDF ANNUAL REPORT IN EDF WEBSITE). 28 FIGURE 6: DISTRIBUTION OF EDF HYDROELECTRIC ENERGY INSTALLED CAPACITY IN FRANCE (LEFT) AND LOCATION OF THE EDF HYDRO

PLANTS IN FRANCE (RIGHT) (SOURCE: EDF WEBSITE). ..................................................................................................... 29 FIGURE 7: NET ELECTRICITY PRODUCTION IN SWEDEN FROM 1971 TO 2013 (SOURCE: SWEDISH ENERGY AGENCY AND STATISTICS

SWEDEN. SEA, 2015. NOTE: THE HYDROPOWER ITEM INCLUDES WIND POWER UP TO AND INCLUDING 1996). ........................ 30 FIGURE 8: VATTENFALL HYDROPOWER GENERATION FROM 2011 TO 2015 (SOURCE: VATTENFALL WEBSITE). ................................... 31 FIGURE 9: IBERDROLA INSTALLED CAPACITY BY COUNTRY (LEFT) AND BY TYPE (RIGHT) (SOURCE: IBERDROLA WEBSITE). ........................ 32 FIGURE 10: IBERDROLA INSTALLED CAPACITY IN SPAIN (LEFT) AND IBERDROLA PRODUCTION IN SPAIN (RIGHT) BY ENERGY SOURCE

(SOURCE: IBERDROLA WEBSITE). ................................................................................................................................. 33 FIGURE 11: MAP OF IBERDROLA NUCLEAR AND HYDROPOWER PLANTS (LEFT) AND VIEW OF LA MUELA DE CORTES FACILITY (RIGHT)

(SOURCE: SELF-MADE USING INFORMATION FROM THE JUCAR RIVER BASIN AUTHORITY AND IBERDROLA, AND THE IBERDROLA BLOG

FOR THE PHOTO). .................................................................................................................................................... 34 FIGURE 12: SUMMARY OF THE RESULTS FROM QUESTIONS: ‘WHAT IS THE FINEST TEMPORAL RESOLUTION OF THE FORECASTS THAT YOU

USE?’ (PANEL A), ‘WHAT IS THE FINEST SPATIAL RESOLUTION OF THE FORECASTS THAT YOU USE?’ (PANEL B), AND ‘WHAT IS THE

MAXIMUM FORECAST HORIZON (LEAD-TIME) OF THE FORECASTS THAT YOU USE?’ (PANEL C). EACH COLOUR CORRESPONDS TO

DIFFERENT ANSWERS FROM THE RESPONDENTS. ............................................................................................................. 41 FIGURE 13: SUMMARY OF THE RESULTS FROM THE QUESTIONS ABOUT THE INTEREST IN A NUMBER OF OPTIONS OF FORECAST

INFORMATION. ....................................................................................................................................................... 44 FIGURE 14: RESULTS FROM QUESTION “PLEASE RANK YOUR INTEREST FOR THE FOLLOWING OPTIONS OF IMPROVED FORECAST

INFORMATION USING A SCORE FROM 1 (LOW INTEREST) TO 9 (HIGH INTEREST)?". THE RESPONDENTS’ PROFILES ARE

INDICATED IN X AXIS, WITH DIFFERENT COLOURS SHOWING THE REPORTED RANKS.......................................................... 46

List of tables

TABLE 1: LIST OF THE MAIN STAKEHOLDERS INVOLVED IN WP8. ................................................................................................ 21 TABLE 2: LIST OF MAJOR HYDROELECTRIC PLANTS OPERATED BY A2A GROUP IN LOMBARDY. .......................................................... 25 TABLE 3: LIST OF MAJOR DAMS OPERATED BY A2A GROUP IN LOMBARDY (SOURCE: A2A GROUP WEBSITE)............................... 25 TABLE 4: LIST OF HYDROPOWER PLANTS IN THE JUCAR RIVER BASIN, IBERDROLA (SOURCE: JUCAR RBMP). ........................................ 34 TABLE 5: LIST OF HP STAKEHOLDERS AND THEIR RESPECTIVE CHARACTERISTICS, WITH GREY BACKGROUND USED TO IDENTIFY IMPREX

STAKEHOLDERS. NOTE: “-” MEANS “I DON’T KNOW” ANSWER. ........................................................................................ 38 TABLE 6: SUMMARY OF RESULTS FROM THE QUESTIONS: “HOW ARE THE WEATHER FORECASTS USED?” AND “ON AVERAGE, HOW OFTEN

DO YOU USE THE WEATHER FORECASTS?” THE NUMBERS IN BRACKETS INDICATE THE NUMBER OF RESPONDENTS HAVING CHOSEN THE

OPTION INDICATED. THE OPTIONS ARE PRESENTED FROM THE MOST FREQUENT ANSWER TO THE LEAST FREQUENT ANSWER. ......... 42



Executive Summary

In a society moving towards a low-carbon economy, there is a growing interest in the role

of hydropower systems to produce clean energy per se and supporting, by load balancing,

the production from other renewable energy sources. Hydropower systems operations are

challenged by the increasing variability of hydro-meteorological processes and occurrence

of extreme events. In the IMPREX project, Work-package 8 aims at investigating the value of

improved predictions of hydro-meteorological extremes at short-, medium-, and long-range

in the hydropower sector. This sector needs accurate and reliable forecasts over different

spatial and temporal scales. Short-term forecasts can improve flood control, which may

induce spill of water with losses of production and, consequently, of economic revenue;

medium-range forecasts can support the optimal management of the production; long-term

predictions can help in anticipating the effects of seasonal changes in water availability and

implement drought management plans. Despite the anticipated benefits from using weather

and climate services, as in other sectors, the actual adoption of such services by

practitioners is still limited. In this report, we focus on reviewing the existing knowledge and

needs for weather and climate services in the hydropower sector. To cover the spectrum of

potential collaborations between weather and climate services providers, water resources

researchers, and European energy production companies, we interviewed, by means of an

online survey, hydropower companies in France, Italy, Spain, and Sweden. Additional

companies, not directly involved in IMPREX, have also responded the same survey and

contributed to the analysis presented in this report. Despite the limited size of the sample

(11 responses), the analysis offers valuable insights regarding the current state of the use of

weather and climate services in the hydropower sector. The survey results show that all

respondents have already been using forecasts products of various forms. Public free

services tend to be the main data source to retrieve forecast information, and the majority

of the respondents hold a positive evaluation of the quality of current products. This is

particularly true for those who indicated they buy the information from meteorological

institutions. In addition, respondents expect future improvements to be focused on

enhancing the forecast of extreme events and extending the forecast lead-time.

9


1 Introduction

Predictions in hydroelectricity systems aim to help managers to optimize energy production

and the economic value of water resources, as well as stakeholders to guarantee people

safety and dam security (in the case of reservoir-based systems) against extreme events. In a

society moving towards a low-carbon economy, hydropower (HP) has the advantage of

being a renewable source of energy that can be stored and reallocated in space and time. It

can thus better handle the natural variability of hydro-meteorological hazards and the

occurrence of extreme events and/or peak demands. HP water reservoirs may also be

operated for multiple and, possibly, conflicting purposes: not only for energy production,

but also for domestic and agriculture water supply, environment protection, tourism, flood

protection, etc.

The IMPREX project proposes to investigate the value of improved predictions of hydro-

meteorological extremes at short-, medium- and long-range in a number of water sectors,

including hydropower (Hurk et al., 2016). In the project, work-package 8 (WP8) is fully

devoted to analyse how the hydropower sector can benefit from better hydro-

meteorological predictions and improved reservoir management strategies. Four HP systems

in France, Italy, Spain, and Sweden are investigated in order to cover the broad spectrum of

potential collaborations between weather and climate services providers, water resources

researchers, and European energy production companies. These case studies concern

different geographical areas (south-east France, northern Italy, central-eastern part of the

Iberian-Peninsula and northern Sweden), but share common goals: i) the evaluation of

improved predictability of inflows and extreme events on hydropower decision models and,

consequently, ii) the assessment of the operational value of forecasts, at short to medium

up to seasonal time scales, and of the impacts of climate predictions on the adaptability of

reservoir operation rules in a multi-sector perspective. They are supported by close

collaborations between scientists and operational modelling and forecasting centres of

European energy production companies.



This report aims at introducing the topic and research work of the IMPREX partners involved

in WP8, particularly focusing on reviewing the existing knowledge and needs for weather

and climate services in the HP sector. The four HP companies of the four case studies have

been interviewed using face-to-face interviews and an online survey. Additional HP users,

not directly involved in IMPREX, have also responded the same survey and contributed to

the analysis presented here. We note that the focus in the IMPREX project is on hydropower

only (we are thus not including other renewable energies or non-renewable energy sources).

We are dealing with hydropower companies operating a variety of types of power plants:

run-of-the-river, impoundment type (from small to large reservoir-based power plants) and

pumping-storage type. These types are all included in the first group of survey respondents

reported here.

This report is organized as follows: in Section 2, we propose a general overview on the use

of climate services for hydropower. Section 3 introduces the four hydropower companies

involved as stakeholders in the IMPREX project, highlighting their current status and needs

for weather and climate services. We focus on how these services support their operations

and decision-making. Section 4 focuses on the presentation of a survey designed by the

WP8 IMPREX partners to objectively assess the current practices and future needs of the

hydropower industry. Finally, Section 5 presents the main conclusions from our study and

points out to the challenges and open opportunities that may further strengthen the useful-

ness of climate and weather inputs for the hydropower sector.

11


2 Climate services for hydropower

There are numerous enterprises in the water sector that are exposed to weather and climate

variability and extremes, including the energy sector. It is, therefore, essential to promote

actions that support and improve the uptake by weather-sensitive hydrological services of

weather and climate (W&C) services. By weather and climate services, we refer to the

generation and provision of a wide range of information on past, present and future

weather and climate, with the aims of using this information to support decision-making at

all levels in a socio-economic sector. The Global Framework for Climate Services (GFCS), a

United Nations initiative led by the World Meteorological Organization (WMO), was

launched in 2009 “to guide the development and application of science-based climate

information and services in support of decision-making in climate sensitive sectors”. Its

implementation plan notes that there are “considerable benefits to be obtained through

climate services in relation to the water sector on all time scales.” (GFCS, 2014). The use of

W&C services presents a number of challenges, with some of them particularly related to

the hydropower sector.

2.1 Overall aspects related to the provision of climate services

The EU research and innovation Roadmap for Climate Services (EC, 2015) recognizes that

“Climate services have the potential of becoming a supportive and flourishing market, where

public and private operators provide a range of services and products that can better inform

decision makers at all levels, from public administrations to business operators, when taking

decisions for which the implications of a changing climate are an issue”. Although the

perceived importance of weather (several days to weeks) and climate (several months to

decades) services is usually high (WMO, 2015), a number of challenges remain related to the

provision of W&C services and the effective use by (and feedback from) the economic

sectors. Some are listed below, just to name a few:



The increasing complexity and amount of information produced by W&C services and

requested by a diversity of stakeholders from diverse geographic regions should not act

as a disincentive for innovation in research and operations.

At least two translating issues need special attention: translating users’ needs into

services and translating services into added socio-economic value. The production of

increasingly skilful predictions in weather, climate and hydrology should translate into

social benefits or economic added value to society and businesses, with a better

understanding of the links between quality and usefulness of predictions.

Progress requires transdisciplinary scientific approaches and inter-sectoral impact

modelling, supported by more creative strategies to efficiently engage stakeholders in

supporting and providing feedback to research and innovation.

Improvements on the scientific understanding of natural processes and the prediction of

high-impact events should go together with improvements on impact modelling and

economic assessment, ensuring that one can continuously benefit and integrate

knowledge from the other.

Tailoring W&C information to the level of scale and detail needed by water systems is

crucial to move from the stage of having predictions issued by a model available to the

stage of having predictions integrated in the decision-making processes. For some users,

tailoring comprises also the integration of W&C services from external providers to in-

house products.

Increasing professional capacity from both communities of providers and users of W&C

services to communicate, access, understand and use services appropriately is essential.

It goes hand in hand with building confidence and developing credibility in W&C

services.

An example of stakeholder consultation for a specific climate service is provided by Soares

and Dessai (2016), within the works of the EU FP7 EUPORIAS project. Based on data

collected from interviews with organisations across Europe and different economic sectors,

their study indicated that the main barrier to the use of seasonal climate forecasts in Europe

was the perceived lack of reliability (or high level of uncertainty) of these forecasts.

13


In some sectors, the needs for using these data were not clear to users and the services

were perceived as not relevant. It was also noted by the authors that the “lack of

awareness”, i.e., absence of knowledge of the type of product available, could also be a

barrier for the use of a climate service by some companies. Enablers to enhance the use of

seasonal forecasts were also listed by the authors, among which: the relationships with the

producers/providers, the level of resources to commit and expertise needed, and the

accessibility to the service.

2.2 Some recent results related to users in the energy sector

The FP7 EUPORIAS project reported on users’ practices and needs from an extensive

analysis of over 80 interviews and 489 survey responses, obtained from organizations

representing different economic sectors (Dessai and Soares, 2015). The assessment focused

on the potential for using seasonal to decadal (S2D) predictions. For the energy sector, a

total of 14 interviews were analysed, mainly coming from big (more than 1000 employees)

and private companies, acting at the national, European and international levels. The energy

sector represented 14% of the respondents of the EUPORIAS survey (about 70 responses).

They answered general questions on the organisations’ general characteristics, their

decision-making processes, the use of weather and climate information, the use of S2D

climate predictions, and dealing with uncertainty. There were no questions targeting the

specificities of the energy sector in the EUPORIAS survey (as there were for the sectors of

forestry, agriculture and tourism). In Dessai and Soares (2015), it was noted that the

organizations that were interviewed from the energy sector were particularly focused on

seasonal forecasts and long-term planning (5 to 30 years). The importance of the short-term

activities (day-to-day operations) was however highlighted in a citation coming from one

organization. The report also noted that “(…) in the energy sector the main concern lies on

variations in temperature, wind, solar radiation, and precipitation as these affect both energy

production and consumers’ demand.” The results on the use of weather and climate

information indicated that 71% of the organizations of the energy sector use historical



data/past observations, 36% use weather forecasts (up to 1 month), 43% use seasonal

forecasts, and 14% use climate change projections/scenarios. Another specificity from the

energy sector referred to its preference for forecasts to be provided “monthly rather than as

3 month outlooks”, contributing to highlight the importance of the time step of the

information when integrating W&C services into in-house modeling tools. Also, it was noted

that “a couple of organisations in the energy sector compare seasonal forecasts from

different sources as a way of reducing/sampling uncertainty”. Finally, from the results of the

survey, the energy sector stood out as a sector that does not seem extremely sensitive

(either positively or negatively) to most of the weather events and impacts listed in the

survey (which were: floods, droughts, landslides and storm surge, ice and forest fires,

lightning, snow, low/high temperatures, high/low wind, high/low rainfall).

The results obtained by the assessment of users’ needs in the EUPORIAS project need to be

understood in the light of the profile of the organizations interviewed and/or responding to

the survey. This is crucial in the energy industry since it comprises a variety of activities with

different focuses (e.g. energy planning, production, transmission, distribution and trading)

and dealing with energy resources coming not only from climate-related sources. These

features translate into very different needs and practices in regards to weather and climate

services. The majority of energy companies interviewed in the EUPORIAS project showed, for

instance, a current use of weather and climate information for the estimation of electricity

demand or the forecast of peak loads. Hydropower production companies were not

specifically targeted in the EUPORIAS project, as it is the case for the IMPREX project, which

makes the results of these two projects complementary with regard to stakeholders’ needs

and practices from the energy sector.

2.3 Particular features of the hydropower sector

Energy systems search to optimize their production and improve their resilience to extreme

weather events and climate change. The needs of hydropower users for accurate and

reliable weather forecasts cover a wide range of space and time scales depending on the

type and dimension of the HP system: short-term forecasts (from few hours up to 2-3 days

ahead), for example, for flood protection of the population living downstream the facilities

15


and for the security of installations; medium-range forecasts (up to 7-15 days ahead) for the

optimisation of hydropower production; and long-term (months ahead) streamflow

forecasts, for instance, for water resources management and environment protection

measures during drought periods.

Additionally, extreme hydro-meteorological events affect hydropower business activities not

only in terms of water availability (power production), but also of water demand for power

(load). Finally, the hydropower industry is also concerned by hydrological predictions based

on future climate conditions and projected trends, as the effects of expected changes in

precipitation and temperature may lead to changes in runoff volume, extremes and

seasonality, directly affecting the potential for hydropower generation (Kumar et al., 2011).

Schaefli (2015) highlights the strong link between investing in real-time forecasting systems

and new management strategies under future climates. Identifying future forecast needs

today can play a key role in the capacity of hydropower systems to cope with climate

change impacts tomorrow.

Links between W&C and hydropower also exist in the integration of hydropower with other

variable renewable energy (VRE) sources (François et al., 2013). Hydropower production has

the advantage of being a VRE that can be stored in space and time and thus better handle

the natural variability of hydro-meteorological hazards and the occurrence of extreme

events and/or peak demands. VRE integration draws attention to optimizing reservoir

management strategies and anticipating W&C conditions at various scales, which is needed

for planning power supply and demand, operating the distribution and transmission

systems, and assessing price trends in the energy market. With the expansion of intermittent

solar and wind energy production, the optimized management of the water storage capacity

of reservoirs for hydropower production is expected to play a major role in the future

energy mix.

Furthermore, hydropower water reservoirs are often storage facilities that operate in a

sharing environment. In this case, water resources are used not only for energy production,



but also for domestic and agriculture water supply, environment protection, tourism, flood

protection, etc. Conflicts of use may arise when resources are not abundant (e.g., Anghileri

et al., 2013). In this case, optimization tools and adaptive strategies may be required

(Giuliani et al., 2014a, b; Tilmant et al., 2008). These modelling tools use weather and climate

input data, hydrologic inflows, parameterized (variable) reservoir capacities and constraints,

together with information on energy prices to optimize management rules and evaluate the

potential economic gains of the entire modelling system (Giuliani et al., 2015). New

constraints may also be taken into account, which, in additional to the climate constraints of

natural variability of water resources, rely on other users, legal/policy requirements and

socio-economic aspects.

2.4 Challenges and opportunities for the hydropower sector

Troccoli et al. (2014) compiled several works that explicitly illustrate the links between

weather (and climate) and energy, and the effects of meteorological (or climatic) scales on

the energy business, spanning from facility construction and maintenance, production

planning, day-to-day operations, up to crisis management when dealing with extreme

events. Several crucial issues are raised, including: the characteristic two-way interaction

between climate and energy systems (i.e., energy impacts climate and climate impacts

energy); the two-level dependence of the energy balance on climate (i.e., energy systems

depend on climate in both terms of the supply/demand balance); the influences at local

levels (at a specific facility or power plant) but also at the scale of the overall system (across

national boundaries and energy sources); the needs of considering weather forecasting

capabilities and climate change adaptation solutions in the context of current operational

rules; the potential of space-based remote sensing to assist decision-making in the energy

sector; the modelling of decisions and businesses responses to weather and climate; and the

importance of strengthening partnerships between the energy sector and the W&C service

community.

Following the aims of the IMPREX project, in order to understand how the hydropower

sector copes with hydro-meteorological events today to increase its capacity to adapt and

mitigate climate change effects, some general challenges and opportunities for the use of

17


W&C services in the hydropower sector are identified. Traditionally, climate services have

played a predominant role in monthly to seasonal dam management and long-term (years

ahead) energy planning. In the last decade, opportunities have grown towards exploring the

benefits of weather information also to short-term planning, including run-of-the-river

generation (i.e., run-of-the-river hydroelectricity is a type of hydroelectric generation plant

whereby little or no water storage is provided and the generation is therefore subject to

natural river flows) and energy demand, which are both highly dependent on weather

variability and extremes. Today, the range of applications of climate information services is

more diversified, including production, exploration, transport and operations of power

plants. Getting the right W&C products at the right spatio-temporal scales for the energy

sector is both a challenge and an opportunity for W&C providers.

With the opening of the electricity market to competition and the rapid development and

integration of decentralized variable renewable energy (VRE) production in the electricity

grids (e.g., solar and wind power plants), a large number of energy projects has emerged. In

order to guarantee short-term energy security (i.e., make sure that an energy system is able

to react promptly to sudden changes within the supply-demand balance), the role of

hydropower storage in regulating the temporal variability of intermittent sources of energy

production is enhanced. The increased need for energy storage for better integration of

renewables translates into an increased need for water (as “energy storage”). Beside

improved W&C services, the HP sector needs a parallel improvement of hydrological

modelling for estimating the energy storage. The influence of both weather and hydrology

in quantifying and managing hydropower storage opens opportunities to improve the skill

of weather and hydrologic forecasts from several days to months ahead, depending on the

characteristic response time of the watershed and the capacity of the storage (water

reservoir) system.

With the changing to a diverse mix of energy sources, a variety of energy producers has

also emerged. This means that stakeholders in the energy sector are more diverse and,

https://en.wikipedia.org/wiki/Hydroelectricity



consequently, their needs and dependencies on W&C information for their decision-making

are also diverse. Users of W&C services in the energy sector vary greatly in their degree of

sophistication. Sophisticated users will collect their own data, run their own hydro-

meteorological forecasting chains and express high expectations towards providers of W&C

services. Less sophisticated users will, in general, focus their activities on energy distribution

and market, with a higher dependence on finalized products coming from W&C services

providers. This variety of profiles and (current and potential) innovation level increases the

challenge of assessing stakeholders’ current practices, their level of knowledge on W&C

services, as well as their needs of and potential for in-house integration of new W&C

products. This is a critical issue for scientist and modellers in the W&C services domain,

notably if we consider that the energy industry is facing a rapidly changing context, with

high-impact changes in energy policies and regulations, which affect both public and private

energy companies. Under this context today, the process of engaging stakeholders to

efficiently elicit their views and operational needs becomes clearly a more complex exercise.

19


3 Review of stakeholder knowledge and needs in the hydropower sector

3.1 Methodology

In this report, a review of existing knowledge and needs on the HP sector is carried out in

order to better understand the hydro-meteorological impacts on both energy production

(operation systems) and consumption. In order to achieve this goal, various face-to-face

interviews were carried out with our stakeholders and the information obtained was

complemented with an online Google Survey 1 , which was implemented in different

languages (English, Spanish, German, Italian) to directly reach our stakeholders. The survey

was set up to collect the responses to a large number of questions and to offer to WP8 the

possibility to apply the survey to other stakeholders from other energy companies (not

necessarily directly involved in the IMPREX project). The hydropower companies involved in

the IMPREX WP8 were selected since since they had already been working (and building

confidence) with the Imprex partners for several years (the case in Sweden and France); they

were interested in expanding their usage of weather and climate services (case in Italy); or

they were starting the development of multi-use water management strategies at the

catchment scale (notably the case in Spain). Although the WP8 stakeholders represent the

major leading HP companies in Europe, the interviewees and survey respondents are

essentially just a sample of the employees working for those companies and concerned by

the use of climate and weather services. They do not constitute a statistically significant

sample of the population and, therefore, the analysis provided in this section is to be taken

cautiously. When designing the survey, we referred to Burgess (2001) as a general guideline.

In order to clarify the way the online survey was structured and the rationale behind each

question, a brief description of its implementation is provided hereafter.

1 The survey is available at the following LINK.

https://goo.gl/forms/YBrHfqd4dEFJUdFf2



The types of questions used throughout the survey vary from the ‘rated response type’ to

the ‘scaling (ranking) type’. The principle is to cover the major areas of interest and ensure

the continuity of responses, while also providing enough flexibility to allow respondents to

skip any question that does not concern them directly. This is a crucial point in the survey

since there is a large variety of types of work and job positions in a hydropower company,

spanning from engineers to operational hydrological forecasters, reservoir managers, energy

traders and optimization specialists. Most questions in the survey are tabulated and allow a

single response. Open text questions are used to record any additional comments, while

scaling-type questions are used whenever quantitative responses are needed, such as when

participants are asked to rate the quality of their forecasts. In addition, we implemented the

technique of conditional questions, which automatically makes the participant move to a

different set of questions according to their previous response. The aim was to tailor the

survey according to some of the respondent’s answers, notable those concerning, for

instance, the existence (or not) of in-house climate services or their interest (or not) in

having them available in the future.

The survey is organized in four sections, with the first half including compulsory questions,

while the last two sections are optional. The first section aims to identify the respondents’

decision-making context, such as their responsibility in the company and years of work

experience. Documenting this information is important since we expect that the background

of users may influence how they perceive the value of W&C services and the way they use

the services. The second section of the survey has questions regarding the adoption of

W&C services, namely the current status of usage and perception of forecast information,

as well as the expectation for future improvements. Specifically, for the stakeholders who

currently do not use any W&C services, the survey aims to understand the main barriers and

the factors that could motivate their future adoption. For the ones who are already using

W&C services, the survey addresses the types of forecast information used and the

perception users have on the quality of this forecast information. In addition, we included

questions on how the forecast information fits into their decision context, on their main

interests concerning additional forecast information and on the direction for improvement

they see in the near-future.

21


The third and fourth sections of the survey are optional. They were designed to profile the

HP companies’ assets with respect to their power plants and hydroelectric generation

capacity, and to collect any other additional information that may be of interest. In

particular, in the last part of the survey the respondents are asked to express their interest

on various issues related to their operational practice, ranging from their interest in

improving data collection of different hydro-meteorological variables to improving

communication and training programs. Therefore, this section aims to provide clues to

potential areas that may be valuable for the long-term development of W&C services, even

though some of them might well be beyond the scope of the IMPREX project.

3.2 Description of stakeholders

WP8 stakeholder group includes four major HP companies in France, Italy, Spain and

Sweden. These are Electricité de France, A2A, Iberdrola and Vattenfall, respectively (Table 1).

Table 1: List of the main stakeholders involved in WP8.

WP8 stakeholder Country

Total

installed

capacity

[GW]

HP installed

capacity

[GW]

Total

production

[TWh]

HP

production

[TWh]

Number of

employees

A2A, A2A Trading and

Edipower (A2A subsidiary) Italy

10.4 1.9 12.9

(in 2015)

4.5 TWh

(2015)

12,083

(in 2015)

EDF DTG Grenoble France

134.2 21.5 619.3

(in 2015)

43.4

(in 2015)

158,161

(in 2014)

Iberdrola Spain

26.2

(in Spain)

9.7

(in Spain)

55.5

(in Spain, in

2015)

12.4

(in Spain, in

2015)

28,836

(worldwide in

2015)

Vattenfall AB and

Vattenreglerigsföretagen AB

(data provider) Sweden

16.8 8.2 GW 82 TWh

(in 2014)

31 TWh

(2014)

500



In this section, we provide a short description of these stakeholders and the context where

they operate.

3.2.1 A2A (Italy)

Italy is the world’s 14th largest producer of hydroelectric power, with a total of 50,545 GWh

produced in 2014 (GSE, 2015). Electric energy from hydropower production accounts for

about 18% of the national electricity production. While the development of large

hydropower schemes is no longer a national priority, the number of active plants has

increased of nearly 80% from 2001 to 2014 (Figure 1, bottom), mostly through small hydro

plants (e.g. run-of-the-river facilities), reaching 3,432 active plants in 2014. Amongst those

plants only 302 had more than 10 MW of power capacity in 2014, but nevertheless they

constituted almost 83% of the total installed hydropower capacity at the national level. The

gross maximum capacity achieved around 18,531 MW in 2015 (Figure 1, top).

Figure 1: Growth of gross maximum capacity (top panel; GSE, 2016) and number of hydropower

plants (bottom panel; GSE, 2015).

Hydropower production is mostly concentrated in the northern part of the country

(Figure 2), where abundant snow accumulation and steep slopes created the perfect

requisites for hydropower development in the past century across most of the Alps. For

instance, the hydropower plants located in north Lombardy, Piedmont and Trentino-Alto

Adige contribute for almost 60% of the total hydropower capacity in Italy.

23


Figure 2: Hydroelectric energy production in Italy for 2014 (GSE, 2015).

Most of the hydropower plants with large installed capacity (>10 MW) have been operated

by three companies, namely Enel, Edison, and A2A Group (A2A and Edipower). A2A

(including Edipower) is involved in IMPREX as stakeholder of the Italian case study (the Lake

Como system). A2A is currently the largest Italian multi-utility company and a leader in the

Italian domestic energy, environment, heat, and networks sectors, and its portfolio includes

a large share of renewable sources, from which it obtains 53% of the energy generated. It is

also the second largest operator in the distribution networks for electricity and one of the

largest in gas and water cycle networks. The activities of A2A are organized in 6 business

units (Figure 3) and are geographically spread over Europe (Figure 4).

Figure 3: Organization structure of A2A Group (source: A2A Group website).

A2A Group

Generation and Trading Business

Unit

Generation

Sector

Trading

Sector

Sale Business

Unit

Environmental Business Unit

Heat and

Services Business Unit

Networks Business Unit

EPCG

Business Unit

Other Services and Corporate

http://www.a2a.eu/en/company/group.html



Figure 4: Geographical area of A2A activities (source: A2A Group website).

The hydroelectric plants of the A2A Group are both run-of-the-river and storage based

(Table 2 and Table 3). Most of them were constructed between the 1920 and the 1960. A

substantial number of these plants is concentrated in the mountainous areas of northern

Italy (Lombardy and Friuli Venezia Giulia regions), with a few plants located in the southern

Calabria region (Figure 4).

Although A2A is the company involved as one of the WP8 stakeholders, during the survey

we also interviewed other Italian leading energy companies, namely the Enel and Edison Spa.

Together with A2A, these three energy companies represent well the Italian domestic

hydropower producers. Enel is the multinational manufacturer and distributor of electricity

and gas, and the largest energy producer in Italy, who shared around 25% of the national

production in 2012. It is also the company who owns the most large-scale (>500 MW)

hydroelectric plants in Italy. Edison Spa is the oldest power company in Italy, which covers

around 6% of the Italian electricity market. We believe that the inclusion of these two

companies in the assessment of the stakeholders’ needs and practices contributes with

useful complementary information to the IMPREX project.


25


Table 2: List of major hydroelectric plants operated by A2A Group in Lombardy.

Name of plant Installed capacity [MW] Catchment area [km2]

Isolato Spluga 43 25

Isolato Madesimo 16 25

Prestone 24 120

San Bernardo 34 14

Mese 170 190

Chiavenna 60 207

Prata 3,3 207

Gordona 14 410

Gravedona 13 78

Braulio 19 108

San Giacomo 9 270

Premadio 226 361

Grosio 428 712

Stazzona 30 990

Lovero 49 919

Grosotto 10 124

Boscaccia 3.3 208.4

Table 3: List of major dams operated by A2A Group in Lombardy (source: A2A Group website).

Name of dam Capacity [ x106 m3] Type of dam

Cancano dam 123 Gravity dam

San Giacomo dam 64 Gravity dam

Val Grosina dam 1.2 Gravity dam

Spluga dams 32 Gravity dams

Lago Truzzo dam 20 Gravity dam

Isolato dam 1.4 Arch dam




3.2.2 EDF (France)

Following the massive development of nuclear energy during the 1970s, the electricity

sector in France is dominated by nuclear power production (over 70%). The country has,

however, worked on a strategy for controlling energy consumption and promoting the

development of renewable energies within its territory since the major national consultation,

the "Grenelle Environment Forum", in 2007. In 2013, renewable energy sources generated

19% of electricity production, with 74% coming from hydropower.

The French electricity generation and power market is highly concentrated and dominated

by Electricité de France (EDF), who owns also the French transmission system operator (RTE)

and the distribution network operator (ERDF). EDF is a French private energy company,

largely owned by the French state. It was founded in 1946 by the State and became S.A.,

“société anonyme”, in 2004. EDF is the first provider of electricity in France. In Europe, it is

mainly established in France, United Kingdom, Italy and Belgium. In 2015, the EDF Group

generated, globally, 619.3 TWh of electricity, with 7% of electricity generation and 16% of

EDF installed capacity attributed to hydropower (Figure 5).

In France, EDF has 640 dams and 439 hydroelectric plants, ranging from about 10 kW to

1,800 MW (source: EDF website). The total installed capacity is of 20 GW, spread across

France, with higher amounts in the French Alps in the upper parts of the Rhone River basin

(Figure 6). The total volume of water stored in EDF dams is about 7 billion m3. The

hydropower plants are controlled by four hydroelectric control centres located in Lyon,

Toulouse, Sainte-Tulle and Kembs.

Hydro-meteorological forecasting is an important activity at EDF. The aim is to improve the

evaluation of water resources and the management of water-related risks spatially and at

several time scales. EDF operates a hydro-meteorological forecasting chain in two

operational centres, located in Grenoble and Toulouse. The operational centre in Grenoble is

the one involved in IMPREX as the stakeholder of the French case study. Desaint et al.

(2009) list the following targets of the hydro-meteorological forecasting activity at EDF:

monitoring hydrological risk (90 flow thresholds are monitored over more than 50 points

in 31 watercourses),

short-term (1 to 7 days) flow forecasting on 115 points spread over 50 watercourses,

http://energie.edf.com/fichiers/fckeditor/Commun/En_Direct_Centrales/collection_nos_energies/edf_hydraulique_bd_va.pdf

27


monitoring the risk of extreme events, storms on the French Southwest, strong winds

and sticky snow throughout France,

low flow forecasting (leadtimes of weeks to months) on a few large French river basins,

the prediction of long-term (forecast horizon of a few months) inflows to reservoirs

located at upstream valley areas,

short term water temperature forecasting in six major rivers, and, more recently,

sediment transport prediction.

According to Desaint et al. (2009), forecasts and products are distributed to more than 300

users. Two distinct EDF internal users are listed in Ramos et al. (2010): 1) the unit

responsible for the optimisation/trading of resources (energy purchase, production and sale)

and the guarantee of energy delivery to clients, and 2) the local hydraulic centres,

responsible for dam security and reservoir management. Desaint et al. (2009) note a key-

difference between these two users: while the former is more used to probabilities, the latter

is much less. Garçon et al. (2009) note the importance of understanding the two different

roles of the forecaster and the decision-maker in the EDF forecasting system: the former

provides decision-support to the decision-maker, translated into probabilities of possible

future scenarios. Finally, in addition to the two above mentioned users of the forecast

products, Garçon et al. (2009) note also the environmental protection use, which comprises

the respect of the requirements related to the preservation of the environment (in particular,

the protection of aquatic environments).

EDF real-time data network feeds its streamflow forecasting system with precipitation,

temperature and discharge data. It is a shared network and contributes to the national

hydrological and climatological databases. It comprises measurements of precipitation,

temperature, discharge, water level and snow package, as well as water temperature and

sediment flow. The forecasting system takes into account rainfall forecast uncertainties and

hydrological model forecast uncertainties by post-processing model outputs (Zalachori et al.,

2013). The main steps of the chain include: (1) Pre-processing of meteorological ensembles

(temperature and rainfall bias and reliability correction), (2) streamflow forecasting using a



rainfall-runoff model and streamflow data assimilation and (3) post-processing of streamflow

ensembles based on error distributions obtained from model calibration. Verification is

carried out against observations for discharge, precipitation and temperature. The

probabilistic (ensemble-based) forecasts have an “appraised value”, but there is a clear

demand towards quantifying the economic value of the streamflow forecasts. Cost-benefit

analyses are today not strictly performed. There has been some quantification of benefits to

show the gains in increased preparedness (increased lead times) comparatively to

deterministic forecasts, but not of the benefits of the probabilistic feature of the forecasts

(i.e., of the uncertainty quantification approach). Some first studies to quantify the economic

value of 7-day ensemble streamflow forecasts were carried out in 2013 in a EDF/Irstea

collaboration and have been pursued within the IMPREX project.

Figure 5: EDF Group installed capacity and electricity generation in 2015 (source: EDF Annual

report in EDF website).

https://www.edf.fr/sites/default/files/contrib/finance/Annual%20Report%20VA/2015/edf_performance_2015_va.pdf

https://www.edf.fr/sites/default/files/contrib/finance/Annual%20Report%20VA/2015/edf_performance_2015_va.pdf

29


Figure 6: Distribution of EDF hydroelectric energy installed capacity in France (left) and location

of the EDF hydro plants in France (right) (source: EDF website).

3.2.3 Vattenfall (Sweden)

The Swedish electricity production is largely based on hydropower and nuclear power, with

an increasingly expansion of wind power and biofuels. According to the Swedish Energy

Agency, hydropower contributed with 64 TWh to the Sweden’s energy system in 2014

(Figure 7) and accounted for 41% of the total installed electricity production capacity.

According to OECD/IEA (2012), Sweden is the 10th country in the world with regard to

hydropower generation. Hydropower is considered a stable source of power in the energy

system, with a relatively constant level of production in the past decades, although

variations from one year to another can be expected due to climatic variations. There are

about 2000 hydropower plants in Sweden, with about 200 large plants (capacity greater

than 10 MW). A large part of hydropower production is located in the mountainous areas in

the Northern part of Sweden.

http://energie.edf.com/fichiers/fckeditor/Commun/En_Direct_Centrales/collection_nos_energies/edf_hydraulique_bd_va.pdf



Figure 7: Net electricity production in Sweden from 1971 to 2013 (source: Swedish Energy

Agency and Statistics Sweden. SEA, 2015. Note: The hydropower item includes wind power up to

and including 1996).

Vattenfall AB is involved in IMPREX as the stakeholder of the Sweden case study. The case

study covers the upper part of the River Umeälven, a typical north European catchment,

highly influenced by snowmelt runoff and volumes for planning the hydropower production

for the current and next winter seasons. The study area includes four major reservoirs and

hydropower stations operated by Vattenfall AB.

Vattenfall is a Swedish state-owned company. In 2015, the company’s electricity generation

amounted to approximately 173.0 TWh, of which 39.5% came from hydropower (source:

Vattenfall website). Figure 8 presents an overview of Vattenfall hydropower generation from

2011 to 2015 in the various countries in which Vattenfall is active.

https://www.energimyndigheten.se/globalassets/statistik/overgripande-rapporter/energy-in-sweden-till-webben.pdf

https://www.energimyndigheten.se/globalassets/statistik/overgripande-rapporter/energy-in-sweden-till-webben.pdf

https://corporate.vattenfall.com/globalassets/corporate/investors/vattenfall-in-brief_2015.pdf

31


Figure 8: Vattenfall hydropower generation from 2011 to 2015 (source: Vattenfall website).

For Vattenfall hydropower reservoir management, seasonal forecasts of snowmelt runoff

volumes are key inputs to their decision models. Forecasts for the accumulated runoff

volumes during April-July are issued once a month from January until the start of the

snowmelt season in April. For the survey we present here, answers were collected from

Vattenfall AB and Vattenregleringsföretagen AB. These two companies represent on one

hand the hydropower producer operating the power plants (Vattenfall AB) and on the other

hand the common provider of hydro-meteorological information and reservoir operation

(Vattenreglerigsföretagen AB) for all hydropower producers operating in the same river.

This situation is rather common in Sweden, where several hydropower producers have

power plants in the same river and even share the same upstream seasonal reservoirs. This

is also the case in the Swedish case study, even though Vattenfall AB is the only producer in

most of the upper part of Umeälven. Vattenregleringsföretagen AB provides to all

producers, daily updates of hydrological forecasting and water resources availability, while

the producer companies are using their own decision models for the production planning.

https://corporate.vattenfall.com/about-vattenfall/vattenfall-in-brief/



The production planning is shared secretly from each producer to

Vattenregleringsföretagen AB, which without revealing this information to the other

competing companies, distributes the production and reservoir management on the

available resources for the most efficient use of water in the system and to stay within the

maximum and minimum flows and levels stated by the regulating permissions. Thus, even

though Vattenfall AB is the main stakeholder in the case study; Vattenregleringsföretagen

AB is clearly also a stakeholder in their role providing the hydro-meteorological information

to the hydropower producers. It should also be noted that the hydrological forecasting

systems used by the Swedish hydropower companies have been almost exclusively

developed by the Swedish Meteorological and Hydrological Institute based on the HBV

model.

3.2.4 Iberdrola (Spain)

Starting more than 100 years ago, Iberdrola has a total installed capacity of 44,393 MW,

distributed between Europe, North America and South America (Figure 9). The company’s

largest share of installed capacity for energy generation corresponds to renewable energy

sources (hydro, solar and wind), for which climate services are needed as their production is

heavily linked to weather and climate patterns.

Figure 9: Iberdrola installed capacity by country (left) and by type (right) (source: Iberdrola

website).

In Spain (Figure 10), Iberdrola’s installed capacity is equal to 26,187 MW, with hydropower

as the energy source with the highest installed capacity (9,712 MW). However, nuclear

power plants are the most important source of energy production of Iberdrola, which is

33


explained by the fact that they are not subject to climate uncertainty and they are able to

operate at full capacity the whole time. The unbalance between installed capacity and

production in hydropower implies that there is room for improvement in the use of climate

services for the optimization of energy generation.

Figure 10: Iberdrola installed capacity in Spain (LEFT) and Iberdrola production in Spain (RIGHT)

by energy source (source: Iberdrola website).

Iberdrola is involved in IMPREX as the stakeholder of one the Spanish case studies, the Jucar

river basin. In the Jucar river basin, Iberdrola owns 8 hydropower plants with more than 10

MW of installed capacity (Table 4), as well as 1 nuclear power plant with an installed

capacity of 1,092 MW (see Figure 11). Between these infrastructures, the most remarkable is

the La Muela de Cortes pumped-storage facility, which connects the Cortes II reservoir in

the Jucar River with an artificial off-stream reservoir of 20 Mm3 capacity built at the top of a

mountain. Divided into two separate units (La Muela I and La Muela II), this facility has an

installed capacity of 1,516.53 MW, one and a half times the installed capacity of the nearby

Cofrentes nuclear power plant. This facility generates electricity during demand peaks, water

being pumped from the Jucar River when energy generation exceeds demand. The

Cofrentes nuclear power plant takes and releases water for cooling from the Jucar River in a

partially-closed cycle, with reduced net water consumption.



Figure 11: Map of Iberdrola nuclear and hydropower plants (left) and view of La Muela de

Cortes facility (right) (source: self-made using information from the Jucar River Basin Authority

and Iberdrola, and the Iberdrola blog for the photo).

Table 4: List of hydropower plants in the Jucar river basin, Iberdrola (source: Jucar RBMP).

NAME RIVER TYPE CAPACITY INTENSITY

MW m3/kWh

Contreras II Cabriel Storage power plant 51.62 6

Lucas de Urquijo Guadazaon Storage power plant 39.6 4

Alarcon Jucar Storage power plant 16.43 13

El Picazo Jucar Storage power plant 17.58 9

Cofrentes Jucar Storage power plant 121.69 4

La Muela de

Cortes

Off-line Storage power plant 1,516.53 0.84

Cortes II Jucar Storage power plant 290.5 5

Millares II Jucar Storage power plant 70.6 4

35


The Jucar river basin is a typical south Mediterranean basin, with an important share of

water for irrigated agriculture (80%) and tensions on water allocation in a multi-reservoir

system. Given that the Spanish law concedes priority to urban and agricultural uses over

power generation, and considering that the upstream reservoirs are owned by the farmers

and public administrations (although Iberdrola is a co-owner of the Alarcon reservoir),

Iberdrola is largely constrained in the management of its facilities. It is committed to release

from the Millares II reservoir (the Naranjero reservoir) the same amount that is released

from the upstream reservoirs, which is determined by the Jucar River Basin Authority (CHJ).

Once satisfied this constraint, Iberdrola is able to manage its reservoirs following its own

rules.

The operation of the Jucar river basin power plants is made in the Center of Operation and

Control (COC) of the Mediterranean Spain of Iberdrola, placed in Cortes de Pallàs (Valencia,

Spain), beside the La Muela de Cortes, Cortes II, Millares II and Cofrentes hydropower plants.

The operation schedule is defined in Madrid, based on the hourly electricity price, and

transmitted to all the Iberdrola COCs in order to jointly operate the Iberdrola facilities. If the

electricity production exceeds the demand, either by a reduced demand (night periods) or

by an excess of supply (hours in which wind power operates at full capacity), then water is

pumped from the Cortes II reservoir to the La Muela de Cortes off-stream storage. Water

turbines are turned on in periods of peak energy demand.

Iberdrola’s current use of climate services, according to the information provided to the

IMPREX partner UPV research team by Iberdrola’s experts, consists in forecasting the inflows

to their reservoirs. For that, Iberdrola relies on measurements from its own precipitation and

streamflow gauging stations, sharing information with the CHJ’s Automatic Hydrological

Information System (SAIH). Using these variables and some hydrological and hydraulic

techniques, they are able to forecast future inflows with approximately 6 to 8 hours in

advance.



4 Analysis of stakeholders’ responses

4.1 Profiles of the respondents

As explained in Section 3, the survey was designed having in mind different typical profiles

and roles in a hydropower company. Specifically, we considered potential respondents from

the energy trading department, the infrastructure and plant operations, and the operational

and R&D hydrological offices. For some IMPREX stakeholders, we were able to get respons-

es from more than one profile interested in or already using W&C services. All in all, also

including non IMPREX stakeholders, we were able to sample all the three profiles in this first

application round of the survey (Table 5). Below, we provide a brief description of these

three typical profiles, their characteristics and responsibilities. In the next sections, we report

and discuss the responses to the survey. Given the limited size of the sample (11 respond-

ents in total), the analysis of these first results is not based on statistics and has to be con-

sidered simply as a reporting of the detailed point of view of some preselected ex-

perts/operators from the hydropower sector.

Energy trading operators are mainly concerned with selling or buying shares of energy at a

given price to make a profit. This is a complex task, especially when dealing with renewable

energies since their production is closely related to the natural variability of the hydro-

climatic conditions where the power facilities are installed. The energy trading operator has

to anticipate the expected energy production and demand, as well as the elasticity of the

energy price in the market. To this end, they often deal with decisions for the short term

(i.e. next 24 hours) to the medium term (i.e. weekly to monthly time horizons). Reliable W&C

services are valuable pieces of information to assist them in maximizing the profit.

The second representative profile of respondents consists of persons involved in the

infrastructure and plant operations (generically referred as the ‘reservoir operators’

hereafter). Their primary duties are to perform a variety of technical tasks related to the

regulation of the flow of water from the dam and through the turbines to meet the power

generation requested by the energy trading office. Despite the existence of general

operating rules, the reservoir operator is given a certain degree of freedom and is expected

to make optimal decisions. These are usually done on a daily or sub-daily basis and may

37


result from a trade-off among multiple conflicting objectives. Reservoir operation decisions

are generally supported by a comprehensive set of information across different time scales,

particularly on the reservoir inflows and weather conditions. Accurate W&C information may

therefore better inform the operators in their decisions and improve their performance.

Finally, the third profile of respondents includes employees in the hydrological office

(hereafter, referred as ‘hydrologists’). Their tasks can involve the collection and quality

control of real-time data and weather forecasts, the assessment of the main variables of the

water cycle and of the real-time flow conditions, the production of in-house streamflow or

dam inflow forecasts using hydrological and hydraulic models, the communication of the

forecasts to its users in the company and the provision of technical support to decision-

makers in the operation and, potentially, also in the trading departments. Engineers in the

hydrological officers generally have to take decisions at the level of the forecasts to

communicate to in-house users. These decisions mainly concern the level of confidence in

the forecasts or the severity of a critical extreme situation (e.g., flood or drought). Due to

the nature of their scientific and natural system modelling background, hydrologists from

hydropower companies are stakeholders that appreciate the current and potential quality of

W&C services. They are usually keen to having forecast information across different lead-

times and, depending on their decision contexts, they use different sources of W&C

information as input to their modelling framework. They usually also have in-depth

understanding about the limits of the current forecast products and the uncertainties

involved in the modelling process.



Table 5: List of HP stakeholders and their respective characteristics, with grey background used to

identify IMPREX stakeholders. Note: “-” means “I don’t know” answer.

Country HP company Working area Working

experience

Main duty

Italy A2A trading energy market,

trading or

strategic

planning

from 11 to 15

years

research and

development

Italy Edipower (A2A) operation from 3 to 5

years

operation

Italy A2A trading energy market,

trading or

strategic

planning

less than 2

years

research and

development

French EDF hydrology more than 20

years

research and

development

Sweden Vattenfall hydrology more than 20

years

operations

Sweden Vattenregleringsf

öretagen

hydrology more than 20

years

operations

Spain Iberdrola dam or

hydraulic

operation

from 11 to 15

years

operation

Italy Enel operation more than 20

years

operations

Italy Enel hydrology from 3 to 5

years

-

Italy Enel hydrology more than 20

years

technical

support

Italy Edison energy market,

trading or

strategic

planning

from 3 to 5

years

research and

development

39


4.2 Current use of W&C services

With no exception, all respondents said they are currently using weather forecasts. All but

one respondent answered that they use public free data sources, and 4 out of 11

respondents use weather forecast data coming from specific meteorological services.

Interestingly, those respondents indicating the use of data from meteorological services

pertain to the group of reservoir operators and hydrologists, which may be an indication of

their specific needs of more advanced, and perhaps tailored, forecasts in their work.

To further reveal the current status of the use of W&C services, a number of questions were

proposed to identify the main facets of the forecast information in use. The main results

show that:

The majority (8 respondents) uses precipitation forecasts at the minimum hourly

temporal resolution. All profile groups are represented by these respondents. For the

spatial resolution, the answers diverge: 1 respondent chose the finer resolution (< 1km),

2 answered to use precipitation forecasts at the resolution range 2-10 km, 4 answered to

use them at 11-25 km, and 1 indicated the > 50km resolution option.

In terms of temperature forecasts, only 5 respondents (out of the 8 respondents

mentioned above) said they use the forecasts at the hourly time resolution. The other 3

respondents answered they use temperature forecasts at the daily time resolution (thus,

at a lower resolution than their precipitation forecasts). In terms of spatial resolution of

temperature forecasts, the results interestingly resemble those indicated for the

precipitation forecasts.

6 respondents indicated they use climate projections, and the daily temporal resolution

was the most used for this type of information. For the spatial resolution, no option

showed a majority of answers and respondents indicated a variety of scales, from < 1km

to > 50 km. The choice of high resolutions may indicate that downscaling methods may

also be used.

3 respondents said they do not use wind forecasts and this number increases to 5 when

it concerns solar radiation forecasts. Wind forecasts are more often used at the hourly



time resolution. For solar radiation, both resolutions, hourly and daily, were equally

indicated. For the spatial resolution, no option showed a majority of answers, as in the

case of climate projections.

Only 2 respondents indicated the use of all suggested forecasts (precipitation,

temperature, wind and solar radiation), one of them belonging to the hydrologist profile

and the other to the energy trading operator profile.

Practically all respondents (9 out of 11) said they use precipitation and temperature

forecasts for a maximum forecast horizon of ‘a few days’. This option is also more often

selected for wind and solar radiation forecasts, although fewer respondents said to be

using these forecasts. As expected, climate projections are more used for horizons of ‘a

few months’ to ‘a few seasons’ ahead.

Figure 12 provides a summary of all responses concerning resolution and lead time. It

clearly shows that precipitation and temperature forecasts are the most used variables,

while forecasts for solar radiation tend to be the least used ones. The predominant option

chosen for the finest temporal resolution for precipitation and temperature forecasts is the

‘hourly’, while the spatial resolution that is more often chosen is smaller than 25 km. It

seems that precipitation forecasts are usually required in a coarser resolution than

temperature forecasts for the temporal resolution, but not necessarily for the spatial

resolution. The use of climate projections and wind forecasts differs significantly across

respondents, with the temporal resolution varying from hourly to monthly scales, and the

spatial resolution varying from less than 1 km to more than 50 km. It is worth mentioning

that some respondents also pointed out other forecast information they used, such as snow

melt, upstream releases and streamflow.

When asked to rate the quality of the information they use (from 1 to 10), all the

respondents that answered the question (10 out of 11) gave a mark higher than 5

(average mark: 6.5), indicating a positive evaluation of the quality of the weather forecast

products they currently use. Moreover, it is also interesting to note that respondents who

indicated they buy information from W&C service providers also reported very high scores

(average mark: 7.2), while those who reported a medium mark of 5 were mainly those that

indicated the use of public free W&C services as information source.

41


Figure 12: Summary of the results from questions: ‘What is the finest temporal resolution of the

forecasts that you use?’ (panel a), ‘What is the finest spatial resolution of the forecasts that you

use?’ (panel b), and ‘What is the maximum forecast horizon (lead-time) of the forecasts that you

use?’ (panel c). Each colour corresponds to different answers from the respondents.



4.3 Application of forecasts to decision-making

Given the fact that most respondents are using weather forecasts in their operations, two

additional questions were asked to assess some details regarding the use of the forecast

information in their specific decision-making contexts: “How are the weather forecasts

used?” and “On average, how often do you use the weather forecasts?” Results are

summarized in Table 6. They highlight the way respondents use their forecasts for the

various frequencies investigated. They reflect the following answers: 7 out of 11 respondents

said they use the forecasts once per day; 2 respondents use them more often, several times

per day, and 2 respondents use them less often, i.e., once per month and once per season.

Table 6: Summary of results from the questions: “How are the weather forecasts used?” and “On

average, how often do you use the weather forecasts?” The numbers in brackets indicate the

number of respondents having chosen the option indicated. The options are presented from the

most frequent answer to the least frequent answer.

Uses of weather forecast Frequency of usage

Quantitatively as input to a hydrological model [2]

Quantitatively as input to a decision support system [2]

To decide the reservoir operation [2]

To generate the inflow forecast [1]


To derive flood alerts [1]

To trigger emergency operations [1]

several times per day


Quantitatively as input to a hydrological model [4]



To derive flood alerts [2]

To trigger emergency operations [2]

Qualitatively as additional knowledge to make decisions [2]


once per day

Quantitatively as input to a hydrological model [1] once per month



Visually to see what the future weather situation might be [1]

Qualitatively as additional knowledge to make decisions [1]

once per season

43


The results show that most of the respondents work with forecasts quantitatively, either

as direct input to a decision support system or as input to a hydrological model for

streamflow predictions. Moreover, 10 out of 11 respondents said that the forecasts are used

for deciding on reservoir operation, which is expected to be a common task among our

respondents. In addition, most hydrologists and reservoir operators also mentioned other

roles in which weather forecast information is involved, such as triggering of emergency

operations, communication of flood alerts and planning the maintenance of infrastructures.

4.4 Expectation from W&C services

In this last section of the survey, a number of questions were proposed to identify the

respondents’ focal points about the potential interests for additional information and the

directions of improvement. Basically, we asked respondents to rate their interest in a

number of forecast information, choosing from ‘not interested’, ‘little interested’, ‘very

interested’, and ‘highly interested’. Figure 13 shows a summary of the results obtained from

the 11 respondents. We can see that 7 options are considered by a majority of respondents

(more than 6 out of 11) to be of high interest. They are:

1. Streamflow forecast: 9 respondents

2. Short-range precipitation forecast: 8 respondents

3. Energy prices forecast: 6 respondents

4. Flood forecast: 6 respondents

5. Forecast from different meteorological centres: 6 respondents

6. Sub-seasonal to seasonal streamflow forecasts: 6 respondents

7. Sub-seasonal to seasonal climate forecasts: 6 respondents

Some respondents show ‘no interest’ or ‘little interest’ toward particular types of

information, such as:

Heat wave forecast: 5 respondents

Meteorological drought indexes: 4 respondents

Decadal climate projections: 4 respondents

Energy prices forecast; energy demand forecast: 3 respondents (mainly from the

profile group of hydrologists).



Figure 13: Summary of the results from the questions about the interest in a number of options

of forecast information.

45


Regarding future improvements of the forecast information, Figure 14 reports the

respondents’ interests on the direction of improvement for the forecasts, based on a

number of options given in the survey. In this case, the higher rank should indicate a higher

necessity of further development, especially in the near future. Results show that most of

the respondents have high levels of interest (marks greater than 7 points) in the following

options:

Better forecasts of weather extremes (8 out of 11 respondents),

Better streamflow forecasts (7 out of 11 respondents),

Weather forecasts for longer lead times (6 out of 11 respondents),

Better probabilistic forecasts2 (5 out of 11 respondents).

Moderate importance is given to the improvement towards having more weather forecast

scenarios. Low levels of interest are more often seen on improvements towards: more

scenarios of hydrological forecasts, better river level forecasts and higher spatial resolution

of weather forecasts.

Lastly, the respondents were asked to provide an answer to three questions about the ideal

spatial resolution, temporal resolution and lead time of a weather forecast, given the current

capability of using forecast information to support operations and decision-making in their

company. The ideal spatial resolution of the weather forecasts that was chosen by the

highest number of respondents (5 out of 11) was ‘about 10 by 10 km’. The ideal temporal

resolution was chosen to be the ‘hourly’ resolution by 7 out of 11 respondents, and the

ideal lead time more often chosen (4 out of 11 respondents) was ‘a couple of days’. When

confronting these answers with previous responses about the characteristics of the forecast

information currently in use and the users’ expectations about future products, these

2 Probabilistic forecasts are those that provide the quantiles of the forecasted variable, while forecast scenarios give the

variables directly.



answers seem to suggest that our respondents (and the HP companies they represent) have

adapted well their operations to the forecast products available today. Further refinement of

spatial/temporal resolution and enhancement of forecast lead time might actually require

comparable efforts from the users to expand their capability in order to better utilize the

improved forecasts.

Figure 14: Results from question “Please rank your interest for the following options of

improved forecast information using a score from 1 (low interest) to 9 (high interest)?". The

respondents’ profiles are indicated in x axis, with different colours showing the reported ranks.

47


5 Conclusions

The demand for weather and climate (W&C) services in the water sector, in general, and in

the hydropower sector, in particular, has increased as forecasting and modelling capabilities

have improved, and informed decisions have gained in social relevance and economic value.

The hydropower sector is a user of W&C services with broad objectives along the chain

of energy generation, management and planning. Its interests comprise: multi-use

reservoir management, optimal space-time allocation of water resources for energy

generation, flood and drought risk mitigation, integration with other, mainly intermittent,

climate-related renewable energy sources (e.g., wind and solar power), climate adaptation,

as well as strategic and sustainable energy planning to secure economic growth and

environmental preservation. Such a rich context requires strong transdisciplinary

collaborations and partnerships between science and stakeholders, as well as between

public and private organizations.

Within the works of the WP8 in the IMPREX project, we have searched to enhance our

understanding of the current practices and needs concerning W&C services in the

hydropower sector, in general, and in the energy companies that play a crucial role in the

project as stakeholders associated to specific case studies in Italy, France, Spain and Sweden.

This has been done by directly interviewing different HP companies through face-to-face

meetings and using an online survey, specifically designed for the project. The first results

obtained are reported in this deliverable D8.3.

Despite the limited size of the sample we have collected so far (11 respondents to the

online survey), the analysis of the responses offers valuable insights regarding the current

state of the use of W&C services, as seen from the perspective of the users of these services

in the hydropower sector. The survey results show that all respondents have already been

using forecasts products of various forms. Public free W&C services tend to be the main

data source to retrieve forecasts information, and the majority of the respondents hold a

positive evaluation of the quality of current products. This is particularly true for those

who indicated they buy the information from meteorological institutions. In addition,



respondents expect future improvements to be focused on enhancing the forecast of

extreme events and extending the forecast lead-time.

At the time of this report, we had 11 survey responses available. They covered the

stakeholders involved in the modelling case studies of the IMPREX projet in France, Spain,

Italy and Sweden, and some other additional stakeholders. However, since a larger survey

sample would provide more robustness to our analyses, a follow-up report is planned to

include additional responses from a larger sample of countries, including non-European

stakeholders from USA, Canada, Brazil, and Thailand. We have already a total of 27

responses and their analysis will be integrated to those here reported and made publicly

available within the IMPREX project. This will allow us to better assess the most common

features among the users of W&C services in the sector both in Europe and worldwide.

In the future, we will also continue the interactions that we have started with the

stakeholders participating to the case studies of the IMPREX project. This will allow us to

bring additional knowledge to some questions and issues that could not be treated in

details in the online survey. For instance, we will be able to better detail the way decisions

are taken and how changes in forecasts might impact the way users work today with these

forecasts in their modelling chain. We have observed that users often have their operations

adapted to the current properties of the forecast products they use. The refinement of these

products could require additional in-house efforts to adapt the user’s capability to handle

these new data and information (e.g. efforts could be needed to change the resolution of

hydrological models that use these forecasts as input and to recalibrate its parameters, or to

adapt the optimization models used for energy production). The involvement of operational

forecast users in research projects such as IMPREX can be an asset to optimize these efforts

effectively.

6 References

Anghileri, D., A. Castelletti, F. Pianosi, R. Soncini-Sessa, and E. Weber, 2013: Optimizing

Watershed Management by Coordinated Operation of Storing Facilities. J. Water

Resour. Plann. Manage., 139(5): 492–500.

Burgess, T.F. (2001). Guide to the Design of Questionnaires. Gen. Introd. Des. Quest. Surv.

Res. iss.leeds.ac.uk/downloads/top2.pdf

49


Desaint B., Nogues P., Perret C., Garçon R., 2009: La prévision hydrométéorologique

opérationnelle : l’expérience d’Electricité de France. La Houille Blanche, 5: 39-46.

Dessai, S., and Soares, M. B., 2015: Report summarising users’ needs for S2D predictions.

EUPORIAS (308291) WP12, Deliverable 12.3, 115 p. Available at (last seen on

21/07/2016): http://www.euporias.eu/system/files/D12.3_Final.pdf

EC, 2015: A European research and innovation Roadmap for Climate Services. European

Commission, EU, KI-06-14-177-EN-N, 56p. DOI: 10.2777/702151.

François, B., Borga, M., Anquetin, S., Creutin, J. D., Engeland, K., Favre, A. C., Hingray, B.,

Ramos, M. H., Raynaud, D., Renard, B., Sauquet, E., Sauterleute, J. F., Vidal, J. P. and

Warland, G., 2014. Integrating hydropower and intermittent climate-related renewable

energies: a call for hydrology. Hydrol. Process., 28(21): 5465-5468.

Garçon R., Houdant, B., Garavaglia F., Mathevet T., Paquet E., Gailhard J. (2009) Expertise

humaine des prévisions hydrométéorologiques et communication de leurs

incertitudes dans un contexte décisionnel [Human assessment of

hydrometeorological forecasts and communication of their uncertainties in a decision

making context]. La Houille Blanche, 5: 71-80. [in French]

GFCS, 2014. Implementation Plan of the Global Framework for Climate Services. WMO,

Geneva, Switzerland, 81p.

Giuliani, M., F. Pianosi, and A. Castelletti, 2015: Making the most of data: an information

selection and assessment framework to improve water systems operations. Water

Resources Research, doi:10.1002/2015WR017044

Giuliani, M., A. Castelletti, F. Amigoni, and X. Cai, 2014a: Multiagent Systems and Distributed

Constraint Reasoning for regulatory mechanism design in water management, Journal

of Water Resources Planning and Management, doi:10.1061/(ASCE)WR.1943-

5452.0000463.

Giuliani, M., J.D. Herman, A. Castelletti, and P.M. Reed, 2014b: Many-Objective Reservoir

Policy Identification and Refinement to Reduce Institutional Myopia in Water

Management, Water Resources Research, doi:10.1002/2013WR014700.

http://dx.doi.org/10.1002/2015WR017044



GSE. (2016). Rapporto Statistico: Energia da fonti rinnovabili 2015. Retrieved from Gestore

Servizi Energetici Website.

GSE. (2015). Rapporto Statistico: Energia da fonti rinnovabili 2014. Retrieved from Gestore

Servizi Energetici Website.

Hurk, B. van den, L. Bouwer, C. Buontempo, R. Döscher, E. Ercin, C. Hananel, J. Hunink, E.

Kjellström, B. Klein, M. Manez, F. Pappenberger, L. Pouget, M.-H. Ramos, P. Ward, A.

Weerts, J. Wijngaard, 2016: Improving Predictions and Management of Hydrological

Extremes through Climate Services: www.imprex.eu. Climate Services, Volume 1,

March 2016, Pages 6-11, doi:10.1016/j.cliser.2016.01.001.

Kumar, A., T. Schei, A. Ahenkorah, R. Caceres Rodriguez, J.-M. Devernay, M. Freitas, D. Hall,

Å. Killingtveit, Z. Liu, 2011: Hydropower. In: IPCC Special Report on Renewable Energy

Sources and Climate Change Mitigation [O. Edenhofer, R. Pichs-Madruga, Y. Sokona,

K. Seyboth, P. Matschoss, S. Kadner, T. Zwickel, P. Eickemeier, G. Hansen, S. Schlömer,

C. von Stechow (eds)], Cambridge University Press, Cambridge, United Kingdom and

New York, NY, USA. Available at: http://srren.ipcc-

wg3.de/report/IPCC_SRREN_Ch05.pdf (last visited on 13/01/2016).

Ramos, M.-H., Mathevet, T., Thielen, J. and Pappenberger, F., 2010: Communicating

uncertainty in hydro-meteorological forecasts: mission impossible? Meteorological

Applications, 17(2): 223-235.

Schaefli, B., 2015: Projecting hydropower production under future climates: a guide for

decision‐makers and modelers to interpret and design climate change impact

assessments. WIREs Water, 2: 271-289.

Soares, M. B. and Dessai, S., 2016: Barriers and enablers to the use of seasonal climate

forecasts amongst organisations in Europe. Climatic Change, 137 (1): 89-103. DOI

10.1007/s10584-016-1671-8.

Tilmant, A., Pinte, D., and Goor, Q., 2008: Assessing marginal water values in multipurpose

multireservoir systems via stochastic programming, Water Resources Research, 44,

W12431, doi: 10.1029/2008WR007024.

Troccoli, A., Dubus, L., Haupt, S.E., 2014: Weather matters for energy. Springer, New York,

528p. DOI 10.1007/978-1-4614-9221-4

http://www.gse.it/it/Statistiche/RapportiStatistici/Pagine/default.aspx

http://www.gse.it/it/Statistiche/RapportiStatistici/Pagine/default.aspx

http://dx.doi.org/10.1016/j.cliser.2016.01.001

http://srren.ipcc-wg3.de/report/IPCC_SRREN_Ch05.pdf

http://srren.ipcc-wg3.de/report/IPCC_SRREN_Ch05.pdf

51


WMO, 2015. Valuing Weather and Climate: Economic Assessment of Meteorological and

Hydrological Services. WMO No. 1153, Geneva, Switzerland, 308p.

Zalachori, I., Ramos, M.H., Garçon, R., Mathevet, T., Gailhard, J., 2012: Statistical processing of

forecasts for hydrological ensemble prediction: a comparative study of different bias

correction strategies. Advances in Science & Research, 8: 135 – 141. doi:10.5194/asr-

8-135-2012.



7 APPENDIX 1

Below we report the questions asked in our survey. As explained in Section 3, the survey is

organized in four different sections, with the first half containing compulsory questions and

the second half that is instead optional.

7.1 Respondents’ background

53




7.2 Use of W&C services

55


In the following, we first assume the respondent uses W&C service.

57


59


61


63


If the respondent replies she/he does not use any W&C service, the following questions will

be asked in order to understand the main obstacles that motivate this decision.

65




7.3 HP company profile

67




7.4 Additional questions

69


71




8 APPENDIX 2

Below we report the responses by the survey participants anonymized. The id numbers have

been randomly resorted and the order does not correspond to the one in Table 5.

8.1 Current use of W&C services

Table A2.1. "Q: Are you using any weather forecast in the operations?".

ID. Answer

1 yes

2 yes

3 yes

4 yes

5 yes

6 yes

7 yes

8 yes

9 yes

10 yes

11 yes

Table A2.2. "Q: how do you obtain the weather forecast data?"

ID. Source of forecast

1 meteorological services

2 public data source







9 don't know



73


Table A2.3. "Q: What is the finest temporal resolution of the forecasts that you use?"

ID. Precipitation

forecast

Temperature

forecast

Climate

projection

Wind

forecast

Solar

radiation

forecast

Others

1 hourly hourly not used daily not used not used

2 hourly daily daily not used not used daily

3 monthly - - daily daily not used

4 hourly hourly not used not used not used not used

5 hourly hourly daily - - -

6 hourly hourly hourly hourly hourly not used

7 hourly daily monthly not used daily not used

8 hourly daily not used hourly not used monthly

9 - - - - - -

10 daily daily monthly hourly hourly daily

11 hourly hourly daily hourly not used hourly

Table A2.4. "Q: What is the finest spatial resolution of the forecasts that you use?"

ID. Precipitation

forecast

Temperature

forecast

Climate

projection

Wind

forecast

Solar

radiation

forecast

Others

1 11 – 25km 11 – 25km not used > 50km not used not used

2 2 – 10km 2 – 10km 11 – 25km not used not used 2 – 10km

3 > 50km - - > 50km > 50km -

4 - - not used not used not used not used



5 11 – 25km 11 – 25km 26 – 50km - - -

6 11 – 25km 11 – 25km 11 – 25km 11 – 25km 11 – 25km 11 – 25km

7 < 1km < 1km < 1km not used < 1km not used

8 11 – 25km > 50km not used < 1km not used 26 – 50km

9 - - - - - -

10 - - - - - 2 – 10km

11 2 – 10km 2 – 10km > 50km 2 – 10km not used 2 – 10km

75


Table A2.5. "Q: What is the maximum forecast horizon (lead-time) of the forecasts that you

use?"

ID. Precipitation

forecast

Temperature

forecast

Climate

projection

Wind

forecast

Solar

radiation

forecast

Others

1 a few days a few days not used a few days not used not used

2 a few days a few days a few days not used not used a few days

3 a few seasons - - - - -

4 a few days a few days not used not used not used not used

5 a few days a few days a few

months

- - -

6 a few days a few days a few days a few days a few days not used


seasons

not used a few seasons not used

8 a few days a few days not used a few hours not used a few

months

9 - - - - - -


seasons

a few days a few days a few days


seasons

a few days not used a few days



Table A2.6. “Q: In your opinion, what is the quality of the weather forecast that use?”

ID. Score

1 7

2 7

3 5

4 7

5 5

6 8

7 5

8 7

9 -

10 6

11 8

77


8.2 Application of forecasts to decision-making

Table A2.7. "Q: How are the weather forecasts used?" & “Q: On average, how often do you

use the weather forecasts?”

ID. Type of usage with forecasts Frequency of usage

1 Quantitatively as input to a hydrological model;

To generate the inflow forecast;

To decide the reservoir operation;

To derive flood alerts;

To trigger emergency operations;

Qualitatively as additional knowledge to make decisions

once per day

2 To generate the inflow forecast;




Visually to see what the future weather situation might be;

Qualitatively as additional knowledge to make decisions;

To plan for maintenance;

once per day

3 Quantitatively as input to a decision support system;


Visually to see what the future weather situation might be;

Qualitatively as additional knowledge to make decisions

once per season

4 To generate the inflow forecast;

Qualitatively as additional knowledge to make decisions;

once per day


Quantitatively as input to a hydrological model;


once per day




once per day

7 Quantitatively as input to a hydrological model; once per month








once per day


Quantitatively as input to a decision support system;





To decide the reservoir

once per day


Quantitatively as input to a decision support system;




To trigger emergency operations


8.3 Expectation from W&C services

Table A2.8. "Q: Rate your interest in the following forecast information."

(“+”: not interested, “++”: little interest, “+++”: very interested; “++++”: highly interested)

1 2 3 4 5 6 7 8 9 10 11

Storm

forecast +++

+++

+

++ +++

+

+++

+

+++

+

+++

+

+++

+ +++ +++

+++

+

Flood forecast +++

+++

+

++ +++

+++

+

+++

+

+++

+

+++

+ +++ +++

+++

+

Short-range

(up to 72

hours ahead)

precipitation

forecast

+++

+

+++

+++ +++

+

+++

+++

+

+++

+

+++

+

+++

+

+++

+

+++

+

79


Medium-

range (up to

10 days

ahead)

precipitation

forecast

+++

+

+++

+++ +++

+++

+++

+

+++

+ +++ ++

+++

+

+++

+

Short-range

(up to 72

hours ahead)

temperature

forecast

+++

+

+++

+++ ++

+++

+++

+

+++ +++ +++

+

+++

+ +++

Medium-

range (up to

10 days

ahead)

temperature

forecast

+++

+

++

+++ ++

+++

+

+++

+

+++ +++ +++

+

+++

+ +++

Heat wave

forecast +++

+++

++ ++

+++

+

+++

++ + +++

+ ++

+++

+

Meteorologic

al drought

indexes

+++

+++

+++

+ ++

++

+++

+++

+ +++ +++ ++ ++

Streamflow

forecast +++

+++

+

+++

+

+++

+

+++

+

+++

+

+++

+

+++

+ +++

+++

+

+++

+

Energy prices

forecast +

+++

+

+++

+ ++

+++

+

-

+++

+

+++

+ +++

+++

+ +

Energy

demand

forecast

+

+++

+++

+ ++

+++

+

-

+++

+ +++

+++

+

+++

+ ++



Table A2.9 "Q: Given your current capability of using forecast information, what would be

the ideal spatial resolution of a weather forecast for you to support operations and

decision-making in your company?".

ID. Less

than

1 km

About

5 x 5

km

About

10 x 10

km

About

25 x 25

km

About

40 x 40

km

More than

100 x 100

km

Not

sure

Others

1 x

2 x

3 x

4 site-

specific

5 x

6 x

7 x

8 x

9 x

10 x

11 x

81



the ideal temporal resolution of a weather forecast for you to support operations and

decision-making in your company?".

ID. Less than

30 mins

Hourly Daily Monthly Seasonally Not sure Other

1 x

2 x

3 X

4 x

5 x

6 x

7 x

8 x

9 x

10 x

11 x




the ideal forecast lead time for you to support operations and decision-making in your

company?".

ID. A couple

of hours

A couple

of days

A couple of

weeks

A couple of

months

Not sure Others

SWE2H x

x

x

ESP1O 5 to 7 days

x

x

x

x

depends

x

depends

x

83


IMPREX has received funding from the

European Union Horizon 2020 Research and

Innovation Programme under Grant

Agreement N° 641811

d8.3 report on needs in hydropower sector - imprex · v.09 31/07/2016 a. castelletti integration of...

Documents