On Future Changes in Mediterranean Winter Temperatures, Precipitation and Relative Humidity

Bachelor Thesis, 2017

43 Pages, Grade: 3,0




Index of abbreviations

1. Introduction

2. The representative Concentration Pathways

3. Origin of the data

4. Future periods 2021 – 2050 and 2071 - 2100
4.1 Winter mean temperatures
4.1.1 Warm and cold days
4.2 Extreme values of the maximum and minimum temperatures
4.3 Annual mean relative humidity
4.4 Annual mean precipitation

5. Reference period 1971 – 2000
5.1 Winter mean temperatures
5.1.1 Warm and cold days
5.2 Extreme values of the maximum and minimum temperatures
5.3 Annual mean relative humidity
5.4 Annual mean precipitation

6. Differences between reference and future periods
6.1 Differences in winter mean temperatures
6.1.1 Differences in warm and cold days
6.2 Differences of the extreme values
6.3 Differences in relative humidity
6.4 Differences in precipitation

7. Conclusion

8. References

9. List of figures


Index of abbreviations

Abbildung in dieser Leseprobe nicht enthalten


This thesis is going to identify the differences for the Representative Concentration Pathway (RCP) 2.6 W/m² scenario of the 2 meter temperature, daily precipitation and relative humidity in the Mediterranean area (30°N - 50°N and 10°W - 45°E) compared to the reference period 1971 – 2000; and also the development and main characteristics of the RCPs especially for the chosen RCP 2.6 scenario. All four RCPs deal with the time lapse from the year 1850 to 2100 and expect a continuous radiative forcing, except RCP 2.6, which is also called RCP 3-P.D. that means peak and decline. The chosen data is from the CORDEX project being part of the CMIP5, has a EUR-11 resolution (0.11°) with EC-Earth as driving GCM and RCA4 as RCM and also r12i1p1 ensemble in RCP 2.6 scenario with the periods 1971 – 2000 (reference period) and 2021 – 2050; 2071 – 2100 (experiment period). This thesis considers the Paris Agreement, which targets a maximum global warming of 1.5°C. Under the RCP 2.6 conditions the first target of the Paris Agreement, maximum 1.5°C global warming, will be achieved in the Mediterranean area compared to the chosen reference period.

1. Introduction

The following thesis focuses on the Mediterranean, in particular the region between the coordinates 30°N to 50°N and 10°W to 45°E. This work approaches the question “how the selected Representative Concentration Pathway (RCP) 2.6 W/m² values will be changing in the near (2021 - 2050) and far future (2071 - 2100) compared to the reference period (1971 – 2000)”. The Mediterranean area is characterized by mild and wet winters but also by hot and dry summers. Moreover, in the Mediterranean north the climate is arid and temperate whilst in its south it is rainy (Giorgi et al. 2008). Geographically, the Mediterranean area counts 21 countries (Sundseth, 2009) and lies between the subtropical and temperate climate zone (Bildungsserver, 2015). This supports the evaporation along with the decreasing of soil moisture and the flow of rivers. Such a development may produce a greater risk of future droughts and heatwaves (Bildungsserver, 2017). Very strong summers in the Mediterranean area are associated with Asian and African monsoons and a strong geopotential blocking (Giorgi et al. 2008). In the future, annual mean temperatures and annual precipitation over the entire European region will increase stronger than globally. It is projected that in Europe the largest warmings and the most increasing precipitation will occur in the Mediterranean area. In addition, the number of rainy days per year will mainly increase in the north. Moreover, it is expected that there will be higher wind speed for the European region (IPCC, 2007).

Abbildung in dieser Leseprobe nicht enthalten

Figure 1: Global average temperature 1850-2016 (UK Met Office, 2017)

The motivation for this topic is that the Mediterranean area will be the most prominent hotspot for the climate change in the upcoming periods. An increase of aerosols in Central Europe, Africa and Asia will significantly influence the future climate of the Mediterranean area (Giorgi et al., 2008). Due to strongly decreasing precipitation in the central Mediterranean area in the recent years the soil moisture is massively affected. This caused an increasing danger of heatwaves and droughts. Compared to previous periods the average temperature of the Mediterranean area increased by 1°C (Bildungsserver, 2015). Figure 1 shows that the global mean temperature increases at the same time by 0.8°C, which is 0.2°C less than in the Mediterranean area.

2. The representative Concentration Pathways

The Representative Concentration Pathways (RCPs) are various climate models with different conditions for long-term and near-term modeling (Van Vuuren et al. 2011 a). All RCPs assume full participation of all countries (Van Vuuren et al. 2011 b). There are four different RCPs, which cover the period 1850-2100; RCP 2.6, RCP 4.5, RCP 6 and RCP 8.5. All RCPs are named by their total radiative forcing, measured in W/m², until the end of the period. In all RCP scenarios the radiative forcing will increase except in RCP 2.6. This is why the RCP 2.6 scenario is also called RCP 3-P.D. scenario, which stands for peak and decline (IPCC, 2013). In the middle of the 21st century the RCP 2.6 reaches around 3.1 W/m² warming and returns to 2.6 W/m² until the end of the century. RCP 2.6 is inspired by the Paris Agreement which has the first aim that the global warming will be kept under 1.5°C compared to the preindustrial level (Rogelj et al. 2016). The RCPs show a smooth transition from historical to future because they have been harmonized with the historical data (Van Vuuren et al. 2011 a). All RCPs show that the arctic regions are warming more rapidly compared to the global mean warming (figure 2). Other assumptions in the RCPs are that the volume of the sea ice will decrease, which can be seen mainly in RCP 8.5. Only in RCP 2.6 the sea ice will be constant until the year 2050. Furthermore, the PH-value will sink in every RCP. The development of precipitation differs throughout the scenarios; in RCP 2.6 there will be less, and in RCP 8.5 there will be more precipitation. Moreover the oceans will warm during the 21st century: For the first 100 meters the warming will be about 0.6°C (RCP 2.6) – 2°C (RCP 8.5), while in the first 1000 meters it will be about 0.3°C (RCP 2.6) – 0.6 (RCP 8.5). In addition, the sea level will increase about 0.4 meters in RCP 2.6 and 0.7 meters in RCP 8.5 until the year 2100 (IPCC, 2013).

Abbildung in dieser Leseprobe nicht enthalten

Figure 2: Maps from RCP 2.6 and 8.5 of CMIP5 project (IPCC, 2013)

To develop scenarios like the RCPs four criteria has to be followed. Firstly, the scenario should be based on an already existing climate model. Secondly, the scenario should give information about the atmospheric chemistry like greenhouse gases (GHG) or air pollutants (cf. Rodhe, 1990). The needed data for the model has to be provided, too. Thirdly, the transition between analyses of future periods and historical periods has to be smoothed. At last, the scenario should cover a period until the end of a century, like the year 2100 to compare it with other scenarios (Van Vuuren et al. 2011 a). The required changes especially for the RCP 2.6 scenario were explored by the IMAGE model (cf. Bouwman et al. 2006). The subcomponents of this were the global energy model TIMER (cf. de Vries et al. 2001), a land use model, the carbon cycle model MAGICC6 (cf. Magicc Wiki, 2015) and the climate policy model FAIR-SiMCaP (Van Vuuren et al. 2011 b).

Abbildung in dieser Leseprobe nicht enthalten

Figure 3: Trends in GHG concentrations a) for CO2 b) for CH4 and c) for N2O - Grey area indicates the 98th and 90th percentiles (Van Vuuren et al, 2011 a)

Regarding the main GHG concentrations in the RCPs figure 3 shows that the CO2 (Carbon dioxide) concentration in RCP 2.6 will increase until the year 2050 and decrease to approximately 400 ppm. RCP 8.5 shows the highest concentration of CO2 at the end of the century with 950 ppm (cf. Mactavish et al. 2013). Between both scenarios there are the RCP 4.5 and 6, which are around 500-600 ppm CO2 at the end of the century. The CH4 (methane) concentration is decreasing in every RCP, only in RCP 8.5 the CH4 concentration is still increasing until the year 2100. The peak of the CH4 concentration in RCP 2.6 is earlier than the peak of the CO2 concentration. This is due to the shorter lifetime of CH4. N2O (nitrous oxide), the last of the three main GHG, stagnates only in RCP 2.6. All other scenarios assume an increasing N2O concentration. Figure 4 shows the associated GHG emissions. The CO2 values are decreasing for every RCP except RCP 8.5. In RCP 2.6 there is a peak in the CO2 emissions at the year 2020, thenceforward it is decreasing. In RCP 4.5 and 6 the peaks occur later in the years 2040 and 2060 (a). Looking at the courses, the CH4 emissions are similar to the CO2 emissions; it has a peak and declines in RCP 2.6, 4.5 and 6, only in RCP 8.5 it is increasing. The CO2 and CH4 emissions have a large gap between RCP 2.6, 4.5, 6 and RCP 8.5 (b). The N2O emissions values are more stable than the other. RCP 2.6 has decreasing emission values, RCP 4.5 has constant values. Only in RCP 6 and 8.5 the values are increasing, but in RCP 6 it stagnates in the year 2070 (c). Under every RCP it is assumed that the air pollution control becomes more stringent and that there are more controls. The most stringent scenario is the RCP 2.6 with the lowest air pollutant emissions; it is the opposite of RCP 8.5. Figure 5 shows that all intermediate energy scenarios RCP 2.6, 4.5 and 6 assume a population that stagnates in the year 2070. RCP 4.5 expects the lowest population at the end of the century, circa 8.5 billion. RCP 2.6 assumes circa 9 billion and RCP 6 circa 9.5 billion people. Only in RCP 8.5 the population is still increasing at the end of the century, with 12 billion people in the year 2100. In figure 6 three RCPs imply an almost constant oil use, while in RCP 2.6 the oil use for energy is decreasing. The most increasing oil use is assumed in RCP 4.5 and 6. In every RCP there is an increasing amount of coal use for energy. The highest coal use lists the RCP 8.5 scenario. There is also an increasing use of non-fossil fuels in all RCPs like renewable energy and bio-energy. The use of bio-energy is an important factor for the RCP 2.6. In every RCP scenario the energy use will increase compared to the year 2000. Land use, shown in figure 7, is also an important and big part of the scenarios. In the last years there was an increase of croplands and anthropogenic land-use (Jones et al. 2013). This is caused by the fact that the population was rapidly increasing in the last years. The trends show that the croplands are stagnating for the RCPs 6 and 8.5, only for RCP 2.6 it is increasing. In RCP 4.5 the cropland is decreasing. Only in RCP 8.5 the use of grassland is rising until the end of the century, in all other RCPs it is falling or stagnating like for RCP 2.6. The vegetation increases only in the intermediate scenarios RCP 4.5 and 6. In RCP 2.6 and 8.5 the vegetation is decreasing (Van Vuuren et al. 2011 a)1.

Abbildung in dieser Leseprobe nicht enthalten

Figure 4: Trends in GHG emissions a) for CO2 b) for CH4 c) for N2O - Grey area indicates the 98th and 90th percentiles (cf. Clarke et al. 2010) - dotted lines indicate four of the SRES marker scenarios. (Van Vuuren et al, 2011 a)

Abbildung in dieser Leseprobe nicht enthalten

Figure 5: Trends in population. The dotted lines indicate four of the SRES scenarios. Grey area indicates 98th and 90th percentiles (Van Vuuren et al, 2011 a)

Abbildung in dieser Leseprobe nicht enthalten

Figure 6: Development of primary energy use (Van Vuuren et al, 2011 a)

Abbildung in dieser Leseprobe nicht enthalten

Figure 7: Land use a) for Cropland, b) for Grassland c) for Vegetation (Van Vuuren et al. 2011 a) - Grey area indicates the 90th percentile (cf. Smith et al. 2010)

The RCPs are supplemented with the Extended Concentration Pathways (ECPs). Normally the RCPs only run until the year 2100, but the ECPs allow climate models to run up to the year 2300 (Van Vuuren et al. 2011 a). The ECPs expect inverse CO2 emissions. It will be more CO2 eliminated by a natural way then is ejected (Meinshausen et al. 2011). The RCPs are also the successors of the Special Report of Emission Scenarios (SRES) which were developed in the years from 1990 to 1992 and evaluated in 1995; at that time SRES covered a wider range of energy structures than the IS92 scenarios, which are the predecessors of the SRES. There are 40 SRES scenarios divided in four families, A1, A2, B1 and B2.The A1 family is also splitted in A1FI, A1T and A1B which is the most popular (figure 8).

Abbildung in dieser Leseprobe nicht enthalten

Figure 8: CO2 emissions in a) the A1- b) the A2- c) the B1- and d) the B2 scenario (IPCC, 2010)

The A1 family assumes a rapidly economic growth and a global population with a peak in the middle of the century. The three smaller families, A1FI, A1T and A1B, describe a different direction. The A1FI stands for an intensive fossil use, the A1T suggests that most energy sources are non-fossil and the A1B balances these two sources. The A2 family describes a heterogeneous world with primary regional economic development. The B1 family is about a convergent world with the same population assumptions like in A1. At last, the B2 family is about a world with the emphasis on local solutions to economic, and also a continually increasing population with intermediate levels of economic development. The current scenarios (RCPs) were developed because the assumptions of the SRES became outdated (IPCC, 2010). SRES did not include climate policy measures, but RCPs do (Van Vuuren et al. 2011 a). It is hardly possible to reach the low-emission scenario RCP 2.6. But the IMAGE calculations show that it is technically feasible to achieve the conditions until the end of the century. By reaching the conditions the global warming will be between 1.5°C and 2°C. (Van Vuuren et al. 2011 b).

3. Origin of the data

The chosen data was downloaded from the Earth System Grid Federation (ESGF) which works together with the Linköping University (ESGF-LiU). ESGF-LiU maintains a global system of federated data centers (ESFG-LiU, 2017). The treated data includes the variables temperature (tas), precipitation (pr) and relative humidity (hurs). Tas stands for the Near-Surface Air Temperature; pr is about the precipitation flux and hurs stands for the Near-Surface Relative Humidity. In the years 1971 – 2000, the reference period took place. The future period from 2021 – 2100 gets split into two smaller periods of 30 years each to have a better comparison with the reference period. The first experiment period takes place from 2021 - 2050, the second one from 2071 – 2100. The data is part of the fifth phase Coupled Model Intercomparison Project (CMIP5) which is built of processes and successes of earlier CMIP phases. CMIP5 is performed by 20 modelling groups using more than 50 models and includes long-term and near-term integration. Long-term integration is on a century timescale; near-term is only 10 to 30 years to the future. The long-term experiments, which include the RCPs, are built on CMIP3 and include additional runs to have a better understanding of climate change. The near-term prediction is a new addition to CMIP5 which also uses models with higher resolution than CMIP3. Many smaller projects are based on CMIP5 to get high resolution downscaled climate data (up to 10 km) and a more accurate representation of localized extreme events, like the Coordinated Regional climate Downscaling Experiment (CORDEX) does (Taylor, 2012 b). The chosen data is also part of the CORDEX project which results have a higher resolution than the older ENSEMBLES project; both have been compared to results of SRES. ENSEMBLES was carried out under SRES A1B which follows the A1 family. The CORDEX project uses the RCPs and is divided into 14 domains (Christensen et al. 2014). A domain is a region for which the regional downscaling is taking place. The data is part of the EURO-CORDEX domain, which covers all countries in the European Union (Jacob, 2014). The driving Global Climate Model (GCM) of the data is EC-Earth, which is an earth system model to provide information of the future climate (EC-Earth, 2017 b). EC-Earth is based on the forecast system model of the European center of Medium-Range Weather Forecast (ECMWF).It can simulate large scale physical characteristics in the atmosphere, ocean and sea-ice. The EC-Earth model is made of different components: An atmosphere model, a land surface component, an ocean model, a sea-ice model and a coupler. The atmosphere model is based on an Integrated Forecasting System (IFS) which can be run in different resolutions. The IFS has a time step about 1 hour what makes it very efficient. An indirect aerosol forcing is also implemented. The land surface component has four soil layers with 2.89 meters depth and is modeled by TESSEL which stands for Tiled ECMWF Scheme for Surface Exchanges over Land. TESSEL does not consider capillary rise (groundwater) or horizontal exchange of soil water. It is part of the IFS and used for describing the evolution of soil, vegetation, and snow over the continents in different spatial resolutions (cf. Balsamo et al., 2008). The ocean model is made of NEMO Version 2, which consists of the Ocean Parallelise Version 9 (OPA9) and the ice model consists of the second version of LIM2, which has plastic dynamics and a thermodynamic model with three layers (cf. Madec, 2016). IFS, TESSEL, NEMO and LIM2 are coupled with the Ocean Atmosphere Sea Ice Soil Version 3 (OASIS3). From IFS via OASIS3 coupler, NEMO receives surface heat flux, snowfall, freshwater flux and wind. The solar and non-solar part in heat-flux is separately calculated over water and sea-ice. The coupling frequency of OASIS3 between ocean and atmosphere is about three hours (cf. Valcke, 2013). EC-Earth is used for transient climate integration in CMIP5 (Hazeleger, 2012). The EC-Earth consortium is made of 22 research institutes which develop newer versions of EC-Earth (EC-Earth, 2017 a). The Regional Climate Model (RCM) is the Rossby Center regional atmospheric model (RCA4). RCA4 takes the boundary conditions of the GCM; large parts of the RCA4 simulated climate are attributed to the driving GCM. Data from the GCM is used as input for the RCM, as forcing data. The RCM makes its own climate inside and adds details to the results to get a higher resolution. RCA4 is based on the numerical weather prediction model HIRLAM and can be coupled with OASIS3 to NEMO or LIM3. RCA4 is part of the SMHI and CMIP5. The simulations of RCA4 can cover the period from the year 1961 to 2100. With RCMs the resolution gets higher (Strandberg, 2011). The ensemble of the data is r12i1p1. Ensemble members include realization (r), initialization (i) and physics (p). These three points can have influence to the results (European Network for Earth System modelling [enes], 2011 b). Realization, initialization and physics have also positive values and should start with one; only time independent variables are marked with zero. The realization is the measurement of one model at a different time. Each value of the initialization gets associated with another initialization method. For example: r12i1p1 is the twelfth realization of the first perturbed physics model. It is recommended that the historical run has the same realization like the experiment run; to merge them easily together (Taylor, 2012 a). The data has the EUR-11 domain with a resolution of 0.11°. EUR stands for Europe and the number behind this for the Grid resolution in native CORDEX simulations. Native grid resolutions are about 0.44° or 0.11° (EUR-44; EUR-11) and have rotated poles. Interpolated data with back rotated poles and slightly modified data are marked with an “i” at the end of the domain e. g. EUR-11i. Because of the interpolation the grid resolution changed in EUR-44i to 0.5° and in EUR-11i to 0.125°. Interpolation gets done mostly for monthly and semi-annual data (European Network for Earth System modelling [enes], 2011 a). The experiment of the data is the RCP 2.6 W/m², as discussed in chapter 2. Summarized, the data is from the CORDEX project, which is part of the CMIP5, has a EUR-11 resolution with EC-Earth as driving GCM and RCA4 as RCM and also r12i1p1 ensemble in RCP 2.6 scenario. The variables are 2 meter temperature, precipitation and relative humidity, in the periods 1971 – 2000 (reference period) and 2021 – 2050; 2071 – 2100 (experiment).

4. Future periods 2021 – 2050 and 2071 - 2100

Firstly, the Mediterranean area was selected of the normal daily data with the coordinates as described before (10°W-45°E and 30°-50°N). Secondly, the data was converted into the needed units; the data of the temperature was in Kelvin and got converted into degrees Celsius and also the precipitation needed to be in mm/day. The relative humidity got calculated into percent. The next step was taking the months to work with, in this case December, January and February. A further step was to convert the calendar to the same standard setting and select the years for the periods, 2021 - 2050 and 2071 - 2100. The last step was calculating the different and final parameters. The final parameters are a time mean, an annual mean in form of a field mean and a zonal mean for each variable; temperature, precipitation and relative humidity. A time mean indicates averaged values over the whole period for each grid point. The field mean indicates daily or averaged values over the whole field for every time step. A zonal mean indicates averaged values over the whole period for every degree of latitude. The data of the maximum and minimum daily temperature was processed similarly to the normal daily data; the differences are only about calculating the final parameter. To check the percentage of days that are warmer than 16.2°C in their daily maximum temperature or colder than -8.9°C in their daily minimum temperature; the 95th and 5th percentile of the averaged daily maximum or minimum winter temperatures of the reference period were calculated to achieve the defined temperatures for a cold or warm day

4.1 Winter mean temperatures

The differences between both periods are very small. Figure 15 a) and d) show the mean of the respective period (time mean). A larger difference is above the Black Sea, many values are changing from the 5-10 degrees Celsius span to the 10-15 degrees Celsius span. Considering the entire Mediterranean area there are many little changes in their values, so most likely it will get warmer in the latest future period. Looking at c) and f), there are again not many and major differences to see. The minimum and maximum of the first RCP 2.6 period are 0.6°C and 11.8°C; for the second and last period they are 0.8°C and 12.2°C. The maximum is at 36° north and the minimum at 49° north. So in the zonal mean the values are also increasing in the second period from 2071 – 2100 compared to the first 2021 – 2050. B) and e) show the anual mean in form of a field mean with an overlaid 10 year run mean. The 10 year run mean builds the mean over the specified time steps to better recognize the trend; it shows that the temperatures are increasing in the first period from 6.5°C to 7.3°C until the year 2040. In the second period the 10 year run mean indicates that the temperatures are increasing at the beginning and decreasing slower at the end. The difference of the lowest and highest value of the 10 year run mean is about 0.4°C.

Figure 15 : 2 meter Temperature in the period 2021-2050 a) over the whole time, b) the winter mean with 10 year run mean c) in the zonal mean; and in the period 2071-2100 d) over the whole time, e) in the winter mean with 10 year run mean, f) in the zonal mean


1 Detailed Information about the RCP 2.6 conditions in Attachment 1

Excerpt out of 43 pages


On Future Changes in Mediterranean Winter Temperatures, Precipitation and Relative Humidity
Justus-Liebig-University Giessen
Catalog Number
ISBN (eBook)
ISBN (Book)
Klimawandel, Klimaszenarien, Climate change, Mittelmeer, Future
Quote paper
Tim Sperzel (Author), 2017, On Future Changes in Mediterranean Winter Temperatures, Precipitation and Relative Humidity, Munich, GRIN Verlag, https://www.grin.com/document/456198


  • No comments yet.
Read the ebook
Title: On Future Changes in Mediterranean Winter Temperatures, Precipitation and Relative Humidity

Upload papers

Your term paper / thesis:

- Publication as eBook and book
- High royalties for the sales
- Completely free - with ISBN
- It only takes five minutes
- Every paper finds readers

Publish now - it's free