Arab J Geosci (2016) 9: 13 DOI 10.1007/s12517-015-2045-7
ORIGINAL PAPER
Simulation of temperature and precipitation climatology for the CORDEX-MENA/Arab domain using RegCM4 Mansour Almazroui 1 & M. Nazrul Islam 1 & A. K. Alkhalaf 1 & Fahad Saeed 1,2 & Ramzah Dambul 1 & M. Ashfaqur Rahman 1
Received: 13 February 2015 / Accepted: 10 September 2015 / Published online: 8 December 2015 # Saudi Society for Geosciences 2015
Abstract The performance of a regional climate model RegCM4.3.4 (RegCM4) in simulating the climate characteristics of the Middle East and North Africa (MENA) region has been evaluated. The simulations carried out in this study contribute to the joint effort by the international regional downscaling community called Coordinated Regional climate Downscaling Experiment (CORDEX). The model has been forced with the boundary conditions obtained from the global reanalysis dataset ERA-Interim for the period 1979–2010. An east–west cold bias is found in the northern part of the MENA domain in RegCM4 that is absent in the ERA-Interim driving forcings, whereas a large warm bias is found over the southern Arabian Peninsula (Yemen/Oman) for both RegCM4 and ERA-Interim. The possible causes leading to the warm bias over Yemen/Oman in the RegCM4 are discussed. The model performed well in capturing the salient features of precipitation which includes ITCZ, Mediterranean cyclones as well as precipitation minima over the deserts. Moreover, the annual cycles of precipitation and mean temperature over the prominent river basins of the region have been ably captured by the model. Temperature-precipitation relationship revealed that the ERA-Interim driving forcings stay closer to the observations; however, RegCM4 remains competent for most of the Koppen-Geiger climate classification types. Performance of the model in capturing the near surface winds and specific humidity is also presented. Based on the results of this study, it is concluded that RegCM4 is well suited to conduct
* Mansour Almazroui
[email protected] 1
Center of Excellence for Climate Change Research/Department of Meteorology, King Abdulaziz University, P. O. Box 80234, Jeddah, Saudi Arabia
2
Sustainable Development Policy Institute, Islamabad, Pakistan
long-term high-resolution climate change projection for the future period over the CORDEX-MENA/Arab domain. Keywords Temperature . Rainfall . Simulation . Regional climate model . CORDEX-MENA/Arab domain
Introduction The Arab region, also called as Middle East and North Africa (MENA), consists of 22 Arab League countries, has remained under pressure to adapt to water scarcity and heat waves throughout the course of the history (WWAP 2012). The ancient societies of MENA region, which is also called as Bcradle of civilization,^ have developed various technical solutions and institutional mechanisms over the period of time in order to deal with such constraints. However, the scale at which the havoc of climate change is looming over the MENA region is alarming, and its impacts are likely to go beyond the coping range of many communities and countries (Sowers et al. 2011). For a region that is already vulnerable to many non-climate stresses, the multi-faceted impacts of climate change on physical and socio-economic systems are likely to exacerbate this vulnerability, leading to large-scale instability. According to the International Panel on Climate Change (IPCC) Fifth Assessment Report (AR5), most of the MENA region is expected to become hotter and drier (IPCC 2013). Besides many indirect affects, the region is specifically susceptible to weather extremes such as dust storms, heat waves, heavy rainfall, freak flash floods, rising sea levels, and multi-annual droughts, which are on the rising trends under climate change (IPCC 2013). Recognizing the vulnerability of the MENA region to climate change, on the request of the Arab Ministerial Water Council and the Council of Arab Ministers Responsible for the Environment, Arab League and United Nations joined hands for the creation of RICCAR
13 Page 2 of 14
(Regional Initiative for the assessment of Climate Change impacts on water resources and socio-economic vulnerability in the Arab Region). This regional initiative aims at assessing the impact of climate change on water resources and its associated implications for socio-economic vulnerability in the MENA region which can be found in Almazroui (2015). The first challenge in RICCAR initiative was to construct the information of changes in the future climate for the Arab region (Almazroui 2015). Today, coupled Atmosphere Ocean General Circulation Models (AOGCMs), representing physical processes in the atmosphere, ocean, cryosphere, and land surface, are considered to be the most advanced numerical tools for simulating the global climate system for both the present and future periods. The typical horizontal resolution of AOGCMs is about 200–300 km, and many regional-scale climatic processes go beyond the scope of AOGCMs (Murphy and Mitchell 1995). The horizontal resolution of the AOGCMs used in the recent AR5 is in the order of 100–200 km (IPCC 2013). However, AOGCMs in conjunction with nested Regional Climate Models (RCMs) have the potential to provide geographically and physically consistent estimates of regional climate information which are required in impact analysis (Giorgi and Hewitson 2001; Jones et al. 2004). In the last decade, a rapid growth in the development and application of RCMs has been observed. The higher resolution simulations of RCMs have become less expensive than AOGCMs due to a limited model domain. RCMs now run on a grid size between 50 to 10 km and on time scales up to 150 years (Jacob et al. 2007; Saeed et al. 2012). Today, this technique is becoming popular among the developing countries as well (Pal et al. 2007). A brief review of regional climate modeling, from its ensuing stages in the last 1980s to the recent past, can be found in Giorgi (2006). Fig. 1 Orography (in m) of the RegCM4 domain with CORDEX-MENA/Arab boundaries
Arab J Geosci (2016) 9: 13
The World Climate Research Program (WCRP) has initiated a coordinated effort with the international regional downscaling community in order to provide an ensemble of high-resolution regional climate projections to be used for studying the impacts of climate change in different parts of the world. This effort, referred to as CORDEX (COordinated Regional climate Downscaling EXperiment), currently involves more than 20 regional climate modeling groups around the globe. Considering the acute vulnerability of the MENA region to climate change, this region is considered to be a region which has been neglected in most of the earlier climate studies in general and regional climate modeling studies in particular. Therefore, the first step in RICCAR initiative was to develop high-resolution climate change scenarios for this region using RCM for further impact studies. Initially, this initiative was started before CORDEX being involved. However, it was realized by the partners involved in the RICCAR activity that in CORDEX alongside with other regional domains, public and private researchers can pursue regional climate modeling based on a uniform set of limiting boundary conditions and a common set of assumptions that allows for inter-comparison and exchange on climate change projections conducted at the Arab regional level. Therefore, the initiative was officially recognized by CORDEX and given the domain name as CORDEX-MENA/Arab (Almazroui 2015). The Abdus Salam’s International Center for Theoretical Physics (ICTP) Regional Climate Model (RegCM) is considered to be the first limited area model developed for long-term regional climate simulation (Giorgi et al. 2012). This model has been implemented by a large community for a wide range of regional climate studies, from process studies (Qian 2008) to paleo-climate and future projections (Giorgi and Mearns 1999). Although RegCM has
Arab J Geosci (2016) 9: 13 Fig. 2 Difference of modeled and observed (CRU) summer (JJA) mean air temperature (at 2 m height) in °C; a RegCM4 simulation and b ERA-Interim driving field
Page 3 of 14 13
13 Page 4 of 14
been implemented for different studies over the parts of the MENA domain (e.g., Almazroui 2012; Islam and Almazroui 2012). however, it has not been applied on Fig. 3 Difference of modeled and observed (CRU) annual mean air temperature (at 2 m height) in °C; a RegCM4 simulation and b ERA-Interim driving field
Arab J Geosci (2016) 9: 13
the bigger CORDEX-MENA/Arab domain as a whole. In this study, we are using the fourth version of the model, RegCM4. The details about this version and the
Arab J Geosci (2016) 9: 13
improvements from the previous versions in RegCM4.1 can be seen in Giorgi et al. (2012). Center of Excellence on Climate Change Research (CECCR) is the only institute inside the Arab region which, along with a couple of European institutes, is involved in the activity of downscaling IPCC AR5 based Inter-comparison Project Phase 5 (CMIP5) multi-model database using RegCM4, in order to provide the high-resolution climate change information for further impact and adaptation studies. However, as a first step, the evaluation of the ability of RegCM4 to capture the climate features of the CORDEX-MENA/Arab domain has to be done. Therefore, in the present study, RegCM4 has been integrated using reanalysis data as boundary forcing to evaluate its performance against observations in the present climate. This paper has been organized as follows; the BExperimental setup^ section describes the
Page 5 of 14 13
experimental setup. Evaluation of results is presented in the BEvaluation of model results^ section and summary and conclusions are given in the BSummary and conclusions^ section.
Experimental setup After a few preliminary meetings under RICCAR platform, the partners agreed upon following two criteria for the selection of the domain; 1. All the 22 Arab League countries should be covered in the MENA domain. Only exception for this criterion was Comoros, being too far in the South, and hence left out.
Fig. 4 Precipitation (in mm) accumulated over the summer months (JJA) and averaged over the period 1980–2010; a RegCM4 simulation, b ERAInterim driving field, and c CRU observation
13 Page 6 of 14
2. For the selection of domain, special emphasis was given to the regions constituting the fresh water flowing into the MENA region. Therefore, Tigris-Euphrates and Nile have actually defined the northern and southern extent of the CORDEX-MENA domain, respectively. Similarly, the east and west extent are defined following the south Asian summer monsoon and Atlantic disturbances, respectively. The final CORDEX-MENA/Arab domain (7° S–45° N and 26° W–76° E) based on Almazroui (2015) is shown in Fig. 1. RegCM4 has been driven at a resolution of 50 km over this domain. In the setup, wet option in Grell scheme over land and the EMAN scheme over oceans (GLEO) has been used as suggested by Almazroui et al. (2015b). Details of the domain selection and convection scheme selection can be found in
Arab J Geosci (2016) 9: 13
Almazroui (2015) and Almazroui et al. (2015b). respectively. Moreover, boundary layer scheme according to Holtslag et al. (1990) and SUBEX (Pal et al. 2000) moist physics scheme has been applied. A continuous simulation is made for the years 1979–2010; however, the results are presented for 31 years from 1980 to 2010 (1979 is spin up year). The lateral boundary conditions have been taken from the global reanalysis data of ERA-Interim (Simmons et al. 2006) at a horizontal resolution of approx. 0.7°×0.7° and interpolated to match the model grid. To assess the quality of the RegCM4 simulations over the CORDEX-MENA/Arab domain, the monthly mean temperature and the monthly mean precipitation from the Climatic Research Unit (CRU) TS3.21 (referred to CRU, hereafter) observational dataset are used (New et al. 2000; Mitchell
Fig. 5 Annual accumulated precipitation (in mm) averaged over the period 1980–2010 a RegCM4 simulation, b ERA-Interim driving field, and c CRU observation
Arab J Geosci (2016) 9: 13
and Jones 2005). For the analysis, all the three datasets i.e., RegCM4, CRU, and ERA-Interim has been interpolated onto a 0.5°×0.5° global grid.
Evaluation of model results Temperature and precipitation spatial characteristics The difference between simulated and observed summer (June through August) mean air temperature (at 2 m height) over the CORDEX-MENA/Arab domain is shown in Fig. 2. A clear cold bias of more than 4 °C can be seen throughout the zonal belt in the north of the domain in RegCM4 simulation as compared to CRU observations (Fig. 2a). However, this cold bias is absent in the ERA-Interim driving field (Fig. 2b). Another very prominent warm bias exceeding 8 °C can be seen over the southern Arabian Peninsula in particular over Yemen and Oman which is higher in the ERA-Interim driving forcings than the RegCM4 simulation. Almazroui (2012) reported similar large bias in driving fields (ERA-40 and ECHAM5) compared to RegCM3 for annual temperature simulations over this region, particularly over the southwestern Arabian Peninsula. They concluded that RegCM3 reduced the warm bias, means usually RegCM simulate lower temperature compared to the driving forcings. It should be noted that although the cold bias seems to be an artifact of RegCM, the presence of warm bias in both RegCM4 and driving Era-Interim led us to look into this bias in more detail. Fig. 6 Masks of four major river basins in the CORDEX-MENA/ Arab domain used in this study after Hagemann and Dümenil (1998)
Page 7 of 14 13
In a recent study, Wang and Xubin (2013) found a common bias in different independent reanalysis datasets over the region of Yemen/Oman. This is the (only) region of acutest bias common in all the four reanalysis datasets (ERA-Interim, ERA-40, Merra, and NRA) they have analyzed. Although reanalysis are also model generated datasets, however, they are forced/nudged with observations at regular time intervals in order to make them as close to the observation as possible. Still, they show the same bias over Yemen and Oman region. This behavior raises another question that whether this warm bias is due to some error or low dense network data coverage in the observations. The difference between simulated and observed annual mean air temperature over the CORDEX-MENA/Arab domain is shown in Fig. 3. The reduction of the warm bias over Yemen and Oman in annual plots can be clearly seen. This implies that the bias is only during summer months. If that would have been the problem of error or bad data coverage in the observations, we should have a similar bias in other seasons as well. Therefore, the role of observations cannot be blamed here. Saeed et al. (2009) found a similar bias of more than 6 °C over the intensely irrigated region of Indus and Ganges using REMO model. They got rid of the bias with an adequate representation of irrigation in their model and hence concluded that it was the lack of representation of a characteristic process of the region (i.e., irrigation) in their model which was causing the bias. However, the climate of Yemen and Oman is very dry, and therefore, irrigation cannot be the reason of such a large bias over this region, but there is a
13 Page 8 of 14
Arab J Geosci (2016) 9: 13
Fig. 7 Annual cycles of 2 m temperature (in °C) of selected river basins over CORDEXMENA/Arab domain. Red, blue, and black curves represent RegCM4, ERA-Interim and CRU results, respectively
possibility that a similar characteristic process may not be represented in the climate models leading to this unusual bias over the region. Similarly, there can be other reasons for this bias as we know that all the climate models use global datasets at the lower boundary which include vegetation parameters as well as soil properties such as type, texture, albedo, porosity, density, conductivity, etc. There is a possibility that the data of soil type depicted in these global datasets may contain incorrect values. These values may be rechecked with the field Fig. 8 Annual cycles of precipitation [in mm/day] of selected river basins over CORDEX-MENA/Arab domain. Red, blue, and black curves represent RegCM4, ERA-Interim and CRU results, respectively
surveys maps which can be obtained from local authorities which are out of the scope of this paper. If such maps are not available, then perhaps field surveys may be conducted in order to have better data coverage of vegetation and soil properties in near future. In contrast to 2 m temperature, RegCM4 simulated precipitation has shown much better results over the CORDEX-MENA/Arab domain. It can be seen from Fig. 4 that during summer months, although the amounts are little underestimated, but still RegCM4 has very ably
Arab J Geosci (2016) 9: 13
Page 9 of 14 13
Fig. 9 The derived KoppenGeiger climate classification taken from ORNL-DAAC (2009)
Fig. 10 Major climate types of CORDEX-MENA/Arab domain on annual time scale. Each dot represents the monthly mean value of precipitation (in mm/day) and temperature (in °C) for each month of the year. The mean for variables, temperature (in °C) and precipitation (in
mm/day), is represented by a circle. The bars represent the standard deviation. The name of the climate type and its percentage fraction is given in the title for each plot
13 Page 10 of 14
Arab J Geosci (2016) 9: 13
Fig. 11 Same as Fig. 10 but for climate type arid/desert/hot on seasonal time scale
simulated the location as well as extent of precipitation associated with Inter-Tropical Convergence Zone (ITCZ) over Africa. Moreover, the dry parts of MENA region are also captured very nicely by RegCM4 in conjunction with CRU and ERA-Interim. Similarly, RegCM4 simulated annual accumulated precipitation shows the appearance of precipitation band over the southern part of the domain associated with the seasonal march of ITCZ (Fig. 5). Moreover, the appearance of precipitation over the northern part of the domain can also be seen which can be attributed to the cyclones originating from the Mediterranean Sea in the winter season causing precipitation in the neighboring land areas and travel eastward towards the Middle East and central Asia.
Annual cycle of temperature and precipitation One of the main objectives of the RICCAR initiative is to produce high-resolution regional climate change information as input to physical models for the climate impact research and adaptation work related to climate change. Many impact models need such information on river basin scale. Therefore, in this study, the annual cycles of precipitation and
temperature over the four major river basins (Fig. 6) of the region are analyzed. As it can be seen from Figs. 7 and 8 that RegCM4 has simulated the annual cycles of both the variables very well for these four basins. Considering the fact that these basins are situated in different climatic zones, the model’s performance appears even more satisfactory. It is evident from Fig. 8 that for Tigris-Euphrates, the wet (dry) season is in winter (summer) as opposed to the rest of African basins. However, this behavior is captured very well by RegCM4, giving us confidence in its precipitation simulation. Moreover, the model has ably reproduced the precipitation cycle for Nile basin; however, for Niger and Lake-Chad, the precipitation is a little underestimated in the wet season which can also be seen from the spatial plots of Fig. 5. For the case of temperature, however, RegCM4 shows a little cold bias for Niger basins during boreal summer as shown in Fig. 7. Nevertheless, the shape of the curve for annual cycles for both temperature and precipitation are captured really well by RegCM4. Details Fig. 12 RegCM4 simulated lower level (1000 hPa) wind velocity superimposed with specific humidity difference (RegCM4 minus ERA-Int) averaged over the period 1980–2010 for a annual, b summer, and c winter
Arab J Geosci (2016) 9: 13
Page 11 of 14 13
13 Page 12 of 14
of the precipitation and temperature biases in 11 sub-domains within the CORDEX-MENA/Arab domain can be found in Almazroui (2015) and Almazroui et al. (2015b) which are in line with the present study. Temperature-precipitation relationship Although the CORDEX-MENA/Arab domain mainly has an arid climate due to the presence of large desert areas, however, the southern and northern parts of the domain also have a tropical and temperate climate, respectively. Therefore, we use the climate type classification after Koeppen-Geiger, which is used by Jacob et al. (2012) for the validation of their RCM over different domains. Here, we have used the Koeppen-Geiger climate classification map (Fig. 9) obtained from ORNL DAAC (2009). For each of the climate types, a mask is generated and is subsequently applied to RegCM4, ERA-Interim, and CRU data in order to group the data into similar climate types. Moreover, all regions attributed to a climate type that is below an areal fraction of 6 % of all land points of the respective domain are excluded from the analysis. This threshold is introduced to only consider climate types that are representative of the domain. Figure 9 shows the climate types having an areal fraction of more than 6 %. Understandably, the arid/desert/hot climate constitutes the largest fraction followed by Tropical/Savannah close to the equator. Figure 10 shows the results of monthly values of precipitation-temperature regimes for the four dominant climate types. On the yearly time scale, RegCM4 has shown higher standard deviation than the other two datasets. Understandably, the results of ERA-Interim stay closer to the CRU datasets; however, there have been instances such as tropical/savannah in Fig. 10 where RegCM4 remains closer to the observations for the case of 2 m temperature. Figure 11 shows the seasonal plots for the biggest climate type in the region i.e., arid/desert/ hot which constitutes most of the Arab region. In DJF, RegCM4 results stay very close to that of CRU and ERA-interim; however, the standard deviation remains high. In MAM, the cold bias becomes more prominent with ERA-Interim staying much closer to the observations. However, for JJA, RegCM4 results stay closer to the CRU data. For SON, RegCM4 simulated temperature, and precipitation remains high as compared to CRU and ERA-Interim. It may be noted that RegCM4 has simulated higher values of precipitation for few months which appears to be like outlier from the plots in Figs. 10 and 11, especially for precipitation axis. This is probably the reason why RegCM4 has shown much higher standard deviations for both temperature and precipitation for both temperature and precipitation.
Arab J Geosci (2016) 9: 13
Specific humidity and low level circulation To attain further confidence in our simulations, we have also analyzed the performance of the model in simulating specific humidity and low level wind circulation on annual as well as seasonal basis as shown in Fig. 12. The difference in specific humidity between RegCM4 simulation and ERA-Int shows that in general, the model has underestimated (overestimated) the specific humidity over the land (ocean). The main areas of underestimation include the regions of equatorial belt, Yemen/Oman, Iran/Pakistan, etc., which is consistent with the results of Fig. 4 for precipitation where an underestimation of RegCM4 is found over the same areas as compared to ERA-Int. Moreover, the underestimation of specific humidity over land areas is generally found over the coastal/near-coastal regions or the regions near to the boundary of the domain, specifically where the movement of wind is from ocean to the land. The presence of overestimation/ underestimation of humidity over ocean/land, in a way points towards lack of advection of moisture from ocean towards land. Moreover, the presence of overestimation of specific humidity over Arabian Sea near to the coast of Yemen and underestimation over the land areas of Yemen/Oman may partially explain the link between the role of moisture advection through winds and the warm bias over these countries, discussed in earlier sections. However, as mentioned earlier, the issue of warm bias over Yemen/Oman will be explored in detail in a separate document. Although the direct comparison of the magnitude of the wind circulation against ERA-Int data is not presented, however, the model has captured the general circulation of winds quite nicely. The seasonal reversal of the winds between summer and winter over the Arabian Sea is captured quite well by the model. Moreover, the West African monsoon system driven by the positioning of ITCZ between summers and winter is also captured well by the model. The West African monsoon, which is an alternation of the southwesterly wind and the harmattan at the surface, and driven by the positioning of ITCZ between summers and winter is also captured well by the model (Sylla et al. 2010). An obvious occurrence of harmattan (northeasterlies) and strong southwesterlies during winters and summers, respectively, over the West African monsoonal region can be seen from the Fig. 12. Moreover the strong westerlies occurring over the Mediterranean Sea, responsible for bringing rain and snow over Saudi Arabia, Middle East, Iran, Afghanistan, and Pakistan, during winter season are also captured quite well by the model (Almazroui et al. 2015a).
Summary and conclusions In this study, the motivation and background behind the RICCAR initiative in conducting the simulation over the
Arab J Geosci (2016) 9: 13
Arabian region is presented. We have discussed the ability of RegCM4 in capturing the climatic feature over the CORDEXMENA/Arab domain and detailed for four large river basins in the domain. A general cold bias is found over the zonal belt in the northern part of the analysis domain in RegCM4 simulation, whereas a warm bias is found over Yemen/Oman region for both RegCM4 and ERA-Interim driving forcings. The possible causes leading to this warm bias are discussed. Other climate models as well as different reanalysis datasets have shown the similar bias over this particular region, and therefore, it is concluded that some characteristic process is missing from all these models. The other reason could be the deficiency in the data fed into the lower boundary of these models which can be rectified either by consulting the local authorities or carrying the field surveys afresh. Further, RegCM4 has done a very satisfactory job in capturing the salient features of precipitation which includes ITCZ, Mediterranean cyclones as well as minima over the deserted Arab region. Moreover, the model has very ably captured the annual cycles of precipitation and temperature over the four prominent river basins of the region having different characteristics. In spite of cold bias in 2 m temperature discussed earlier, the performance of the model in capturing the annual cycles gives us confidence in the ability of RegCM4 for simulating the climate of the region. In temperatureprecipitation relationship plots, understandably the results of ERA-Interim driving field stay closer to the observations, still RegCM4 has done a satisfactory job for most of the KoppenGeiger climate classification types in capturing the mean, spread and standard deviation of temperature and precipitation. RegCM4 has done a satisfactory job in capturing the main characteristics of near surface wind circulation including West African Monsoon, Indian Monsoon and winter time westerlies over Mediterranean Sea. Moreover, RegCM4 has shown reasonable performance in simulating specific humidity. However, underestimation over the land areas, especially where the flow is from ocean towards land is observed. Basing on the performance of RegCM4 in capturing the characteristic features of CORDEX-MENA/Arab domain, we conclude that RegCM4 is well suited to conduct longterm high-resolution climate change projection in downscaling CMIP5 database. In future, the same analysis will be repeated for other regional climate models applied over CORDEX-MENA/Arab domain to do the inter-comparison of their performances. Acknowledgments The authors would like to acknowledge the grant by the NSTIP strategic technologies program in the Kingdom of Saudi Arabia– Project No. 12-ENV3197-03 to complete this work– and the Science and Technology Unit, King Abdulaziz University for technical support. ICTP, Trieste, Italy, is acknowledged for providing the model and the CRU data were acquired from their website. The simulations in this work were performed using Aziz Supercomputer at King Abdulaziz University’s High Performance Computing Center. The League of Arab States (LAS) and the United Nations Economic and Social Commission
Page 13 of 14 13 for Western Asia (UN-ESCWA) are also acknowledged for leading and facilitating the BRegional Initiative for the Assessment of Climate Change Impacts on Water Resources and Socio-economic Vulnerability in the Arab Region (RICCAR)^.
References Almazroui M (2015) RegCM4 in climate simulation over CORDEXMENA/Arab domain: selection of suitable domain, convection and land surface schemes. Int J Climatol. doi:10.1002/joc.4340 Almazroui M, Awad AM, Islam MN, Al-Khalaf AK (2015a) A climatological study: wet season cyclone tracks in the East Mediterranean region. Theor Appl Climatol 120: 351–365. doi:10.1007/s00704014-1178-z Almazroui M, Islam MN, Al-Khalaf AK, Saeed F (2015b) Best convective parameterization scheme within RegCM to downscale CMIP5 multi-model data for the CORDEX-MENA/Arab domain. Theor Appl Climatol. doi:10.1007/s00704-015-1463-5 Almazroui M (2012) Dynamical downscaling of rainfall and temperature over the Arabian Peninsula using RegCM. Clim Res 52:49–62 Giorgi F, Mearns LO (1999) Introduction to special section: regional climate modelling revisited. J Geophys Res 104:6335–6352 Giorgi F, Hewitson B (2001) Regional climate information-evaluation and projections. In Climate Change: The Scientific Basis, Contribution of Working Group I to the Third Assessment Report, Houghton JT, Ding Y, Griggs DJ, Noguer M, van der Linden PJ, Xioaosu Giorgi F (2006) Regional climate modelling: Status and perspectives. Journal de Physique 4 139: 101–118. D (eds). Cambridge University Press, Cambridge, United Kongdom and New York, US, 583–638 pp., ISBN: 0521 01495 6 Giorgi F, Coppola E, Solmon F, Mariotti L, Sylla MB, Bi X, Elguindi N, Diro GT, Nair V, Giuliani G, Turuncoglu UU, Cozzini S, Guttler I, OBrien TA, Tawfik AB, Shalaby A, Zakey AS, Steiner AL, Stordal F, Sloan LC, Brankovic C (2012) RegCM: model description and preliminary tests over multiple CORDEX domains. Clim Res 52:7– 29 Hagemann S, Dümenil L (1998) A parameterization of the lateral waterflow for the global scale. Clim Dyn 14(1):17–31 Holtslag A, de Bruijn E, Pan HL (1990) A high resolution air mass transformation model for short-range weather forecasting. Mon Weather Rev 118:1561–1575 IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 1535 pp, doi:10.1017/CBO9781107415324 Islam MN, Almazroui M (2012) Direct effects and feedback of desert dust on the climate of the Arabian Peninsula during the wet season: a regional climate model study. Clim Dyn 39:2239–2250. doi:10. 1007/s00382-012-1293-4 Jacob D, Baerring L, Christensen OB, Christensen JH, de Castro M, Deque M, Deque M, Giorgi F, Hagemann S, Hirschi M, Jones R, Kjellstrom E, Lenderink G, Rockel B, Sanchez Sanchez E, Schar CH, Senevirantne S, Somot S, van Ulden A, van den Hurk B (2007) An inter-comparison of regional climate models for Europe: design of the experiments and model performance. Clim Chang 81:31–52 Jacob D, Elizalde A, Haensler A, Hagemann S, Kumar P, Podzun R, Rechid D, Remedio AR, Saeed F, Sieck K, Teichmann C, Wilhelm C (2012) Assessing the transferability of the regional climate model REMO to different COordinated Regional Climate
13 Page 14 of 14 Downscaling EXperiment (CORDEX) Regions. Atmosphere 3(1): 181–199 Jones RG, Noguer M, Hassell DC (2004) Generating high-resolution climate change scenarios using PRECIS. Met Office Hadley Centre, Exeter, UK Mitchell TD, Jones PD (2005) An improved method of constructing a database of monthly climate observations and associated highresolution grids. Int J Climatol 25:693–712. doi:10.1002/joc. 1181 Murphy JM, Mitchell JFB (1995) Transient response of the Hadley Centre coupled ocean–atmosphere model to increasing carbon dioxide. Part II. Spatial and temporal structure of response. J Clim 8:57– 80 New M, Hulme M, Jones P (2000) Representing twentieth–century space–time climate variability. Part II: development of 1901–1996 monthly grids of terrestrial surface climate. J Clim 13(13):2217– 2238 Oak Ridge National Laboratory Distributed Active Archive Center (ORNL DAAC) (2009) FLUXNET Network Map. Available online [http://webmap.ornl.gov/wcsdown/wcsdown.jsp?dg_id=10012_1] from ORNL DAAC, Oak Ridge, Tennessee, U.S.A Pal JS, Small E, Eltahir E (2000) Simulation of regional-scale water and energy budgets: representation of subgrid cloud and precipitation processes within RegCM. J Geophys Res 105:29579–29594 Pal JS, Giorgi F, Bi X, Elguindi N, Solmn F, Gao X, Rauscher SA, Francisco R, Zakey A, Winter A, Ashfaq M, Saeed SF, Bell JL, Diffenbaugh NS, Karmacharya J, Konare A, Martinez D, da
Arab J Geosci (2016) 9: 13 Rocha RP, Sloan LC, Steiner A (2007) Regional climate modeling for the developing world: the ICTP RegCM3 and RegCNET. Bull Am Meteorol Soc 88(9):1395–1409 Sylla MB, Coppola E, Mariotti L, Giorgi F, Ruti PM, Dell’Aquila A, Bi X (2010) Multiyear Simulation of the African climate using a regional climate model (RegCM3) with the high resolution ERA-interim reanalysis. Clim Dyn 35:231–247 Qian JH (2008) Why precipitation is mostly concentrated over islands in the maritime continent. J Atmos Sci 65:1428–1441 Saeed F, Hagemann S, Jacob D (2012) A framework for the evaluation of the South Asian summer monsoon in a regional climate model applied to REMO. Int J Climatol 32:430–440. doi:10.1002/joc.2285 Saeed F, Hagemann S, Jacob D (2009) Impact of irrigation on the South Asian summer monsoon. Geophys Res Lett 36:L20711. doi:10. 1029/2009GL040625 Simmons A, Uppala S, Dee D, Kobayashi S (2006) ERA-Interim: new ECMWF reanalysis products from 1989 onwards. ECMWF Newslett 110:25–35 Sowers J, Vengosh A, Weinthal E (2011) Climate change, water resources, and the politics of adaptation in the Middle East and North Africa. Clim Chang 104(3–4):599–627 Wang A, Xubin Z (2013) Development of global hourly 0.5° land surface air temperature datasets. J Clim 26:7676–7691 WWAP (World Water Assessment Programme) (2012) The United Nations World Water Development Report 4: managing water under uncertainty and risk. UNESCO, Paris