SIMULATION OF EXTREME EVENTS FOR THE CENTRAL ASIA
CORDEX DOMAIN BY USING THE REGCM 4.0
Tuğba ÖZTÜRK1, Hamza ALTINSOY1, Murat TÜRKEġ2 and M. Levent
KURNAZ1
1.
2.
Boğaziçi Üniversitesi, Fizik Bölümü, Ġstanbul, Türkiye, [email protected],
[email protected]
Çanakkale Onsekiz Mart Üniversitesi, Coğrafya Bölümü, Fiziki Coğrafya Anabilim Dalı,
Çanakkale, Türkiye.
Abstract
The Coordinated Regional Climate Downscaling Experiment is a framework devoted to
coordinate international efforts on regional climate simulations. Region 8 of the CORDEX
domains basically covers the Central Asia with corners of the domain at (11.05°E;
54.76°N), (139.13°E; 56.48°N), (42.41°E; 18.34°N) and (108.44°E; 19.39°N) with a
horizontal resolution of 50 km. In the study, the results of an experiment with the
RegCM4.0 model that is ran for analysis of extreme precipitation is presented. The
experiment consists of one simulation from 1970 to 2000 by using the ERA40 reanalysis
data as boundary condition, and another simulation for the period 2070-2100 using the
ECHAM5 A1B scenario data as forcing. Between these two experiments we have
determined the probable changes in the frequency of the extreme events for the Region 8
of CORDEX.
Key Words: Climate change; extreme events, Central Asia, model projections; regional
climate downscaling.
1. INTRODUCTION
The supercontinent of Eurasia is covering about 10.6% of the Earth's surface with a
52,990,000 km2 area located primarily in the eastern and northern hemispheres.
Geographically, it is a single continent, comprising the traditional continents of the Europe
and the Asia (Fig. 1). The arid and semi-arid Central Asia is a core region of the Asian
continent from the Caspian Sea in the west, China in the east, Afghanistan in the south,
and Russia in the north. It is also geographically known as the Inner Asia. Varied
geography of the Central Asia region includes high mountains (e.g. Tian Shan,
Karakurum, Himalayas, etc.), large deserts (e.g. Kara Kum, Kyzyl Kum, Taklamakan,
Gobi, etc.) and treeless, grassy large semi-arid steppes.
475
Fig. 1 General physical/relief map of the Eurasia continent and its surroundings.
Water is the most important resource in the arid and semi-arid Central Asia region,
because the region is a kind of large continental rain shadow basin surrounded by the high
mountains described above (TürkeĢ 2010a) and the large quantities of water are stored in
the mountain glaciers. The Amu Darya, the Syr Darya and the Hari River are main rivers
of the region, and the Aral Sea and Caspian Sea are major water bodies. Complex
precipitation occurrences and temperature and precipitation regimes over the region due to
the various pressure and wind systems during the year and various physical geographical
factors and conditions dominated over the region cause many difficulties in understanding
and modeling climate in such an arid and semi-arid region (TürkeĢ 2010a, 2010b).
Besides simulating changes in average temperature and precipitation, estimating frequency
and intensity of extreme events is also very important. Climate change also has an effect
on severity of extreme events such as heat waves, cold waves, storms, floods and
droughts. Predicting changes in extreme weather events and their impacts on human health
and environment is difficult. As summarized in the Intergovernmental Panel on Climate
Change Fourth Assessment Report (IPCC, 2007), since 1950, the occurrence of heat
waves has increased in number and space. Due to decrease in precipitation while
evaporation has increased, some regions are affected by droughts. On the other hand,
because of increase in the numbers of heavy daily precipitation events, flooding have also
increased in specific regions. Again, according to the IPCC 4th Assessment Report, the
Central Asia‘s environment, ecological and socio-economic systems are under serious
threat of climate change and future increase in number of severe extreme weather events,
particularly because of the semi-arid nature of the region. Agricultural production of the
region has already decreased quantitatively, and water resources are at risk of climatic
change (Perelet 2007). The region consists of mostly arid and semi-arid lands, grasslands,
rangelands, deserts and some woodland (TürkeĢ 2010a, 2010b). With respect to the
climate change in the Central Asia, grasslands, livestock and water resources are one of
the most vulnerable areas in the region due to the location mostly in marginal
physiographic areas. On the other hand, several regions of Central Asia are dealing with
extreme weather conditions such as widespread flooding in Pakistan and also in large parts
of Asia, while other regions are affected by drought and heat wave in Russia. There will
be more frequent and intense extreme weather events due to climate change according to
the IPCC projections. The Central Asia region is vulnerable to future extreme climate
476
conditions. However, there are few works in studying extreme climate conditions of
Central Asia domain (Meehl 2000, Manton 2001), whereas climate impact studies also
exist for the corresponding region (Lioubimtseva 2005).
In this study, Regional climate model of the Abdus Salam International Center for
Theoretical Physics (ICTP), RegCM version 4.0 is used to simulate climate extremes for
the Central Asia domain. The originality of work comes from the fact that domain of
regional climate model covers all Central Asia region. It is the first time that RegCM4.0 is
applied to the large region of Central Asia. We used ERA-40 global dataset to simulate
present climate. Investigation of seasonal extreme climate conditions of the Central Asia
region was carried out with temperature and precipitation. Then we applied the RegCM4.0
to investigate climate extremes of the Central Asia domain by downscaling the ECHAM5
global dataset for future period. We first defined extreme events in temperature and
precipitation for the period 1970-2000 by taking 95th percentile of climate variables. And
we examined future change of extreme event frequency of domain for the period 20702100 with A1B scenario of ECHAM5 global dataset.
2. MODEL DESCRIPTION
In the framework of the present study, RegCM4.0 code, which was first developed by
Giorgi (1993a, b) was applied as a regional climate model. The dynamical structure of
RegCM includes the hydrostatic version of the National Center for Atmospheric Research
(NCAR) of the Pennsylvania State University mesoscale model version 5 MM5 (Grell
1994). For surface process representation, RegCM includes the Biosphere-Atmosphere
Transfer Scheme (BATS, Dickinson 1993) as well as the Community Land Model (CLM )
version 3.5 as an option in its dynamical core for land surface processes. For radiative
transfer, RegCM is modeled by using the radiation package of NCAR Community Climate
Model, version CCM3 (Kiehl 1996). Solar radiative transfer of the model follows the δEddington approximation of Kiehl (1996). Cloud radiation part contains three parameters
including the cloud fractional cover, the cloud liquid water content and the cloud effective
droplet radius. RegCM is using the planetary boundary layer scheme developed by
Holtslag (1990) based on a nonlocal diffusion concept. Convective precipitation schemes
of a model are computed using one of three schemes which are modified-Kuo scheme
(Athens 1977), Grell scheme (Grell 1993) and MIT-Emanuel scheme (Emanuel 1991,
Emanuel 1999). This Regional Climate Modeling system has been effectively used for
climatic change studies applications during the last decade (Giorgi 1999).
Forcing Data and Experiment Design
ERA40, which is 2.5° x 2.5° resolution forty year global re-analysis dataset, was used as
lateral boundary conditions for present day simulation of years 1970-2000. ERA40 global
dataset is developed by the European Centre for Medium-Range Weather Forecasts
(ECMWF). ECHAM5 (Roeckner 2003), which is the 5th generation of the ECHAM
general circulation model developed by the Max Planck Institute for Meteorology, was
used for future run simulations for the period 2070-2100 with A1B emission scenario.
ECHAM5 is the most recent version of the ECHAM models originally developed from the
spectral weather prediction model of the ECMWF. There are many advances in ECHAM5
compared to the previous version, ECHAM4, in both numerical and physics of the model.
According to Roeckner (2003), these changes include a flux-form semi-Lagrangian
transport scheme for water components and chemical tracers, a new long-wave radiation
code, separate treatment of cloud liquid water and cloud ice, a new cloud microphysical
and cloud cover parameterization formulations, sub-grid scale orographical effects. The
477
model domain covers the Central Asia with a spatial resolution of 50 km. The regional
model simulation spans in the time interval between the 1970-2000 with the ERA-40
dataset for present-day simulation and time period of 2070-2100 for future simulation. We
used 95th percentile for defining extreme conditions in temperature and precipitation
climatology. Change in frequency of extreme conditions was investigated for future period
according to present day climate extremes.
3. RESULTS
Here we present the maximum number of consecutive dry days of thirty year period of
1970-2000 in Central Asia domain in Fig. 2. We defined dry day as a day which gets
precipitation less than 1 mm. However, the number of consecutive dry days is meaningful
regarding of extreme events. The maximum number of consecutive dry days shows the
strength of drought in the region. Southwest part of the domain was affected by drought
around 2000 days at maximum. The second dry region consists of Iran, Iraq and Arabia
which have around 300 days of dry condition. Other parts of domain have relatively less
dry days consecutively. Future projection of the maximum number of consecutive dry
days is presented in Fig. 3. Southwest part of the region is the most vulnerable part which
has around 2000 consecutive dry days which central region of domain has 250 dry days in
average. Northern part of domain has again relatively less dry days.
Fig 2. Maximum number of consecutive days that have precipitation amounts less than
1mm for the period1970-2000.
478
Fig. 3 Maximum number of consecutive days, which have precipitation amounts less than
1mm for 2070-2100 period.
In another analysis, we find maximum precipitation amount of each year of 1970-2000
period and 95th percentile of these amounts to get number of years which has greater
precipitation amount than 95th percentile. The numbers of years which get precipitation
amount greater than 95th percentile of max precipitation for period of 1970-2000 are
presented in Fig 4 for Central Asia domain. Same analysis for future period of 2070-2100
is done and shown in Fig 5. However in this analysis we used 95th percentile of max
precipitation of future period namely itself. In Fig 6 we did same analysis by using 95th
percentile of max precipitation of past period of 1970-2000. It shows that the number of
years greater than 95th percentile of maximum precipitation of past period is slightly
increasing in the future period all around the region. The total number of days which get
greater amount than 95th percentile for all around the map is changing from 132976 for
past period to 221113 for future period. There is slight increase in the number of
maximum days.
Fig. 4 The number of years that get precipitation amounts greater than 95th percentile of
maximum precipitation for the period of 1970-2000.
479
Fig. 5 The number of years that get precipitation amounts greater than 95th percentile of
maximum precipitation for the period of 2070-2100.
Fig. 6 The number of years that gets precipitation amounts greater than 95th percentile of
maximum precipitation of past years of 1970-2000 for the period of 2070-2100.
In Fig 7 maximum precipitation amount in a year for thirty year period of 1970-1999 is
presented while same graphic for future period of 2070-2099 is shown in Fig 8. And
finally, both past and future maximum precipitation amounts are shown in the same plot
(Fig. 9). In Fig 9 red line refer to future period where as blue line refer to past period. We
can see from plots for future period maximum precipitation amounts are shifted to the
right means that amount of maximum precipitation will be increasing in the future period.
There is also one peak around zero value; we get these values due to the region‘s
complexity. Region consists of Siberia and the second biggest desert at the same time. The
study in extreme frequency of temperature is not presented here due to limitation of paper
size.
480
Fig 7 Frequency distribution of the maximum precipitation amounts in a year for the
period of 1970-1999.
Fig 8 Frequency distribution of the maximum precipitation amount in a year for the period
of 2070-2099.
481
Fig 9 Density distribution of the maximum precipitation amount in a year for the period of
1970-1999 (blue line) and for the period of 2070-2099 (red line).
CONCLUSION
In this study, we investigated the annual time-scale performance of the RegCM4.0 in
simulating extreme climate for the Central Asia region. We evaluated annual change of
extreme precipitation of the region by running the model for two thirty year period of
1970-2000 and 2070-2100. We used the ERA40 reanalysis datasets as a forcing data to the
regional climate model for present period, while we used the EH5OM global dataset for
the future period with the A1B scenario. In this work, we studied dry spells and extreme
precipitation of the Central Asia domain for both past and future periods. We come up
with the idea that southern part of the region is the most vulnerable part of the domain to
the droughts in the future period. On the other hand we get also results that show increase
in the extreme precipitation for future period in almost all parts of the region. These results
show that the Central Asia domain will be vulnerable to climate change by means of not
only the changes in precipitation amounts but also the changes in the extreme precipitation
events.
Acknowledgement
We thank to all RegCM team of ICTP for helping in providing forcing data, in installation
and running of the model. We want to give special thanks to Ivan Guttler (Meteorological
and Hydrological Service of Croatia) for helping in parameterization of the regional
climate model. Finally, we thank to Muhammed Zeynel Öztürk (Research Assistant in
Department of Geography, Physical Geography Division, Çanakkale Onsekiz Mart
University, Çanakkale, Turkey) for his help on plotting map.
References
Anthes, R. A., A cumulus parameterization scheme utilizing a one-dimensional cloud
model, Mon. Wea. Rev., 105, 270–286, 1977.
482
Dickinson RE, Henderson-Sellers A, and Kennedy PJ (1993) Biosphere-atmosphere
transfer scheme (bats) version 1e as coupled to the NCAR community climate
model, Tech. rep., National Center for Atmospheric Research.
Emanuel KA (1991) A scheme for representing cumulus convection in large-scale models,
J. Atmos. Sci., 48:2313–2335.
Emanuel KA, Zivkovic-Rothman M, (1999) Development and evaluation of a convection
scheme for use in climate models, J. Atmos. Sci., 56:1766–1782.
Giorgi F, Bates GT, and Nieman SJ (1993a) The multi-year surface climatology of a
regional atmospheric model over the western united states, J. Climate, 6:75–95.
Giorgi F, Marinucci MR, and Bates GT (1993b) Development of a second generation
regional climate model (regcm2) Boundary layer and radiative transfer processes,
Mon. Wea. Rev., 121:2794–2813.
Giorgi F, Shields C (1999) Tests of precipitation parameterizations available in the latest
version of the NCAR regional climate model (RegCM) over the continental United
States Journal of Geophysical Research, 104:6353–6375.
Grell, G., Prognostic evaluation of assumptions used by cumulus parameterizations,
Mon.Wea. Rev., 121, 764–787,1993.
Grell GA, Dudhia J, and Stauffer DR (1994) Description of the fifth generation Penn
State/NCAR Mesoscale Model (MM5), Tech. Rep. TN-398+STR, NCAR,
Boulder, Colorado, pp. 121.
Holtslag, A. A. M., E. I. F. de Bruijn, and H.-L. Pan, A high resolution air mass
transformation model for shortrange weather forecasting, Mon. Wea. Rev., 118,
1561–1575, 1990.
Kiehl JT, Hack JJ, Bonan GB, Boville BA, Breigleb BP, Williamson D, and Rasch P
(1996) Description of the ncar community climate model (ccm3), Tech. Rep.
NCAR/TN-420+STR, NCAR, Boulder, Colorado.
Manton MJ, Della-Marta PM, Haylock MR, Hennessy KJ and 23 others (2001) Trends in
extreme daily rainfall and temperature in southeast Asia and the South Pacific:
1961–1998. Int J Climatol 21:269–284
Meehl GA, Zwiers F, Evans J, Knutson T, Mearns LO, Whetton P. 2000. Trends
in extreme weather and climate events: issues related to modeling extremes in
projections of future climate change. Bull. Am. Meteorol. Soc. 81:427–36
Lioubimtseva E, Cole R, Adams JM, Kapustin G (2005) Impacts of climate and landcover changes in arid lands of Central Asia, Journal of Arid Environments 62:285–
308.
Perelet R (2007) Central Asia: Background Paper on Climate Change. UNDP Human
Development Report Office Occasional Paper.
Roeckner E, Bäuml G, Bonaventur L, Brokopf R, Esch M, Giorgetta M, Hagemann S,
Kirchner I, Kornblueh L, Manzini E, Rhodin A, Schlese U, Schulzweida U, and
Tompkins A (2003) The atmospheric general circulation model ECHAM5. Part I:
Model description. Max Planck Institute for Meteorology Report 349, 127 pp.
TürkeĢ, M. (2010a). Climatology and Meteorology. First Edition, Kriter Publisher Publication No. 63, Physical Geography Series No. 1, ISBN: 978-605-5863-39-6,
650 + XXII pp., Istanbul (in Turkish).
TürkeĢ, M. (2010b). Analysis of the UN Convention to Combat Desertification with
respect to the Climate, Climate Change and Drought, and Applications in Turkey.
Invited Panel Paper. In: ÇölleĢme Ġle Mücadele Sempozyumu Tebliğler Kitabı,
601-616, 17-18 Haziran 2010, Çorum (in Turkish).
IPCC, 2007: Climate Change 2007: Impacts, Adaptation, and Vulnerability . Contribution
of Working Group II to the Fourth Assessment Report of the Intergovernmental
483
Panel on Climate Change [Parry, Martin L., Canziani, Osvaldo F., Palutikof, Jean
P., van der Linden, Paul J., and Hanson, Clair E. (eds.)]. Cambridge University
Press, Cambridge, United Kingdom, 1000 pp
484