C limate L ocal I nformation in the M editerranean - R esponding to U ser N eeds

1. Climate Local Information in the Mediterranean - Responding to User Needs Melanie Davis, Climate Forecasting Unit (CFU)

2. Presentation Contents • 1. Energy status (European Union) • 2. Introduction CLIM-RUN • 3. Climate predictions • 4. Climate variables for renewable energy • 5. Examples of research results • 6. Questions to ask

4. 2009 100 €c/litre €40 €55 2011 130 €c/litre

5. In Europe: For every $10 rise in the barrel price = one tenth less GDP Instituto para la Diversificación y el Ahorro Energético (IDAE): • No other country receives so much oil from Libya as Spain • No country is so dependent on importation of fossil fuels (80% importations) • No country uses so much energy per unit of GDP (energy intensity)

6. EU Energy Generation EU Renewable Energy (RE) Target Energy Consumption of EU27 20% by 2020 10.3% in 2008

7. Renewable Challenge Energy Demand ''The amount of usable solar and wind energy far exceeds the world's total energy demand, with current technology feasibility considered'' 2009 American Institute of Physics 00.00 10.00 16.00 22.00 Time: One Day

8. Presentation Contents • 1. Energy status (European Union) • 2. Introduction CLIM-RUN • 3. Climate predictions • 4. Climate variables for renewable energy • 5. Examples of research results • 6. Questions to ask

9. CLIM-RUN Research Project Improve the provision of adequate climate information, that is relevant to and usable by different sectors of society

10. CLIM-RUN: Work Package 7 Illustrate how climate information can play an important role in future changes and developments in the energy sector

11. A Renewable Europe Power Grid System Power Stations Wind Farms Solar Farms

12. A Renewable Spain Export to France Power Grid System Power Stations Wind Farms Solar Farms Export to Africa

13. Climate Data and RE 1. Site selection 2. Predicted annual energy yield 3. Long-term energy yield performance 4. Frequency when energy yield below a defined threshold

14. Presentation Contents • 1. Energy status (European Union) • 2. Introduction CLIM-RUN • 3. Climate predictions • 4. Climate variables for renewable energy • 5. Examples of research results • 6. Questions to ask

15. Climate Predictions – Current Status Timeline (years) 0 1 2 3 4 5 10 20 30 40 mins - hours days - months weatherpredictions CLIMATE PREDICTIONS – CLIM-RUN PROJECT seasonvariationwithmultipleyears season and annual variation with decades seasons - year Implications: Results…??? Many… Assumed consistency in RE climatic resources Considerable multiplication of RE costs

16. Climate Prediction Sensitivities ´´Components of uncertainty are commonly based on subjective estimations rather than on statistical sound data analysis´´ Heinz-Theo Mengelkamp et al. 2010, Risk analysis for a mixed wind farm and solar power plant portfolio. Investment influence using inter-annual climate resource variability Example: planning of a solar power plant in Spain •Typical size: 50 MW, cost €300 million •Guaranteed price per unit of electricity generated: 0.20 €/kWh •This provides a annual yield of €31 million • Assumptions: small solar irradiance variation • Uncertainty of 1% leads to: - Annual increase or decrease of total revenue = €310000 • Across the investment return period = €8 million or ~ 15% investment

17. Climate Prediction Sensitivities ´´The fact that a trend has existed in the recent past is no certain guarantee of its continuation in to the future e.g. rainfall may readily reverse or disappear over a period of a few decades´´ Climate Impact on Energy Systems, World Bank Study, 2011 CLIM-RUN activities 1. Characterising the climate using statistical analyses 2. Improving the reliability of databases and techniques 3. Collaboration with energy stakeholders

18. Climate Prediction – Current Status Aims to provide climate predictions from days to decades into the future. Climate predictions are produced with numerical models of the climate system. Used alongside observed climate patterns in order to project to future timescales. Improves understanding of how the climate works and helps predict how it will act and react in the future.

19. Climate Prediction with RE Better understanding of : - Confidence in energy yield forecasts - Assist decision making - Understand mechanism to accelerate RE generation efficiently

20. Climate Prediction with RE Guidance for: - Site selection and system planning - Offsetting yield variability - Infrastructure adjustments - Future energy demand/requirement

21. Climate Prediction with RE Protect against: - Excess costs for renewable energy operation and maintenance - Vulnerability of industry and society

22. Climate Prediction with RE Contribute to: - Climate change adaptation policy - Energy security policy - Building codes and other regulations - Investment opportunities

23. CLIM-RUN Questions Worst case scenarios: Worst case scenarios: How representative is current climate data for estimating the performance of a RE plant over its lifetime (e.g. 30 years)? ? ? Can we characterise the frequency, amplitude and duration of high energy demand (extreme heat/cold periods) and low RE yield climatic resources? How confident can we be about the energy yield forecasts? ? What are the likely lowest level of energy yield from a RE project in a season/year? (known as ´´climate droughts´´) ? How can solar and wind climatic resources co-vary to supply a more consistent stream of energy? ?

24. Presentation Contents • 1. Energy status (European Union) • 2. Introduction CLIM-RUN • 3. Climate predictions • 4. Climate variables for renewable energy • 5. Examples of research results • 6. Questions to ask

25. Climate Variables Both wind & solar: Air temperature (oC) : stability Air density (ρ) : environment Solar radiation (W/m2) with wind speed (m/s): stability

26. Climate Variables - Region • Wind only: • Wind speed (m/s) • Direction (degrees) • Consistency/Direction frequency (degrees, %) • Power density (W/m2) • Vertical wind shear (m/s) • Turbulence profile/Turbulence Intensity (TI)

27. Challenges: Wind ! Wind resource highly variable (spatially) compared to solar and is complicated by complex land orography Wind speed and direction must be taken into account but can complicate the statistical procedures Large-scale land use change has an unknown impact on regional climate ! ! !

28. Climate Variables - Region • Solar only: • Surface Direct Natural Irradiance, DNI (W/m2) • Surface Global Horizontal Irradiance, GHI (W/m2) Both affected by: - Cloud cover and type • Concentration of aerosols (anthropogenic and natural) Absorb and/or scatter solar radiation

29. Challenges: Solar ! Solar surface irradiance varies dramatically with cloud cover and aerosols GHI is most often the only available solar radiation data so conversion models are used to derive estimates of DNI (Perez et al, 1987) ! !

30. Climate Variables - Continent • Climate Phenomena • Seasonal: • - Tropical Pacific: El Niño Southern Oscillation (ENSO) • North Atlantic Oscillation (NAO) Inter-annual: • - Pacific Decadal Oscillation (PDO) • - Atlantic Multi-decadal Oscillation (AMO)

31. Climate Variables - Others Anthropogenic : land use, industry etc.. Natural Events: volcanoes etc.. Anthropogenic? Natural? Anthropogenic? Natural?

32. Presentation Contents • 1. Energy status (European Union) • 2. Introduction CLIM-RUN • 3. Climate predictions • 4. Climate variables for renewable energy • 5. Examples of research results • 6. Questions to ask

33. Climate Prediction - Results Renné et al, 2008, Solar Resource Assessment, NREL 1998-2005 > 1961-1990 Up to 10% higher 1998-2005 < 1961-1990 Up to 10% lower Map background: average global radiation data from 1998-2005 (kWh/m2/day) Points: difference annual average between 1961-1990 and 1998-2005 (kWh/m2/day)

34. Climate Predictions - Results Difference in annual mean value of global irradiance between 2003 and 1998-2005 (Watt-hours/m2/day)

35. Climate Predictions - Results By using more years of data for the analysis, there is less chance that anomalous climate events or patterns could influence the results. By using more years of data for the analysis, there is less chance that anomalous climate events or patterns could influence the results. Awareness of the differences between short-term (monthly/annual) datasets and longer-term means. Winter Spring Difference in seasonal mean value of global irradiance between 2003 and 1998-2005 (Watt-hours/m2/day) Summer Autumn

36. Presentation Contents • 1. Energy status (European Union) • 2. Introduction CLIM-RUN • 3. Climate predictions • 4. Climate variables for renewable energy • 5. Examples of research results • 6. Questions to ask

37. Questions: Wind ? Are there dominant climate patterns associated with high winds? Is there an interplay between i) large scale & local scale, ii) multi-annual & decadal, climate patterns? What is the frequency and intensity of such predictions over annual or decadal timescales? ? ? ?

38. Questions: Solar ? How well can we estimate inter-annual & intra-annual variations of surface solar irradiance using observed datasets? What is the long-term climate effect of changes in atmospheric aerosols on solar radiation? ? ?

39. Conclusion For the RE sector as a whole, simple and reliable climate predictions are needed. Higher-quality RE climate resource assessment can accelerate technology deployment by making a positive impact on decision making and reducing uncertainty of financial investments.