International cooperation on detailed climate scenario dissemination by Meteorological Research Institute for climate change impact assessment and development of adaptation measures

1. Introduction

People are interested in how the climate in their area changes as a result of global warming. However, it is impossible to project such detailed local climate change with global climate models such as general circulation models (GCMs) or earth system models (ESMs) even as computer technology has advanced. Therefore, a method called dynamical downscaling is used to project local future climate conditions with regional climate models (RCMs). RCMs are applied over a limited-area domain with boundary conditions from GCM/ESM output to calculate fine climate scenarios. High-resolution climate change scenarios are a basic requirement for assessing climate change impacts, vulnerability, and risks and adapting policy making at local and regional scales.

The Meteorological Research Institute (MRI) has developed the Non-Hydrostatic Regional Climate Model (NHRCM; Sasaki et al., 2008). The MRI (Figure 1) has been conducting research collaboration on and disseminating dynamical downscaling techniques of the NHRCM with researchers from collaborating countries in the Coordinated Regional Downscaling Experiment (CORDEX), including India, the Philippines, Vietnam, Indonesia, Malaysia, and Thailand (Figure 2).

CORDEX is an initiative by the World Climate Research Programme to provide RCM downscaling technology. Such a collaborative effort on future climate projection with the NHRCM is in the best interest of collaborating countries and organizations within the Southeast Asia division of CORDEX, because dynamical regional downscaling is time-consuming, resource-intense, and the NHRCM is a state-of-the-art RCM in the world. A simplified explanation of the research collaboration is described below.

Figure 1: Meteorological Research Institute in Tsukuba, Japan.
Figure 1: Meteorological Research Institute in Tsukuba, Japan.
Figure 2: Members of the Coordinated Regional Downscaling Experiment in Southeast Asia (CORDEX-SEA)
Figure 2: Members of the Coordinated Regional Downscaling Experiment in Southeast Asia (CORDEX-SEA)

2. Research collaboration

The MRI has invited researchers from the collaborating countries to the MRI in Tsukuba, Japan, for two months to implement custom downscaling technology with a super computer for each region and to adjust the NHRCM program (source code) to improve reproductivity of local climate conditions. Thus far, meteorology/climatology or environment researchers with knowledge of LINUX, Fortran, and meteorological/climatological dynamics have participated in the joint study. This research collaboration has been supported by the Ministry of Land Infrastructure Transport and Tourism (MILT), Japan Society for the Promotion Science (JSPS), Program for Risk Information on Climate Change, and Integrated Research Program for Advancing Climate Models operated by the Ministry of Education, Culture, Sports, Science and Technology, Japan (MEXT). A schedule of joint study is indicated in Figure 3, and relevant organizations are shown in Figure 4.

Figure 3: Visiting schedule and cost payers of the joint research between collaborating Southeast Asian countries.
  • SOUSEI
  • TOUGOU
Figure 3: Visiting schedule and cost payers of the joint research between collaborating Southeast Asian countries.
  • Ministry of Land, Infrastructure, Transport and Tourism
  • Japan Meteorological Agency
  • Meteorological Research Institute, Japan
  • MEXT
Figure 4: Relevant organizations

3. Joint study models and downscaling framework

(1) NHRCM

The NHRCM, which was developed by the MRI, is based on an operational non-hydrostatic model (NHM). Details of this NHM are given by Saito et al. (2006). The NHM was modified for climate projection as the NHRCM (Sasaki et al., 2008). In the NHRCM, the MRI/JMA Simple Biosphere Model (Hirai and Ohizumi, 2004) was used to describe biosphere processes, surface temperature, and snow depth. Kain and Fritsch’s (1993) scheme was used to parameterize cumulus convection.

(2) Boundary conditions

The MRI developed a 60 km horizontal resolution GCM named MRI-AGCM3.2H (MRI-AGCM60), which was modified from an operational numerical weather-prediction model and provided information on potential climate change induced by global warming, including future changes in tropical cyclones, East Asian monsoons, extreme events, and blockings (Mizuta et al., 2012). MRI-AGCM60 future projections use four representative concentration pathway scenarios for initial and boundary conditions of the NHRCM (Figure 5). In Phase I, NHRCM with a grid interval of 25 km was used in a slightly wider area, and the research collaboration team conducted a high-resolution experiment with a grid interval of 5 km for Phase II (Figure 6) for future projections.

Figure 5: Boundary conditions of the joint study; ERA-Interim is a global atmospheric reanalysis (assimilated observation) from 1979, continuously updated in real time. MRI-AGCM: Meteorological Research Institute Atmospheric General Circulation Model; NHRCM: Non-Hydrostatic Regional Climate Model; RCP: Representative Concentration Pathway.
Figure 5: Boundary conditions of the joint study; ERA-Interim is a global atmospheric reanalysis (assimilated observation) from 1979, which is continuously updated in real time(https://www.ecmwf.int/en/forecasts/datasets/archive-datasets/reanalysis-datasets/era-interim). MRI-AGCM: Meteorological Research Institute Atmospheric General Circulation Model; NHRCM: Non-Hydrostatic Regional Climate Model; RCP: Representative Concentration Pathway.
Figure 6: Downscaling framework in the joint research. AGCM: Atmospheric General Circulation Model; AOGCM: Atmosphere-Ocean General Circulation Model; CMIP: Coupled Model Intercomparison Project; RCM: regional climate model; SST: Sea Surface Temperature.
Figure 6: Downscaling framework in the research collaboration. AGCM: Atmospheric General Circulation Model; AOGCM: Atmosphere–Ocean General Circulation Model; CMIP: Coupled Model Intercomparison Project ; RCM: regional climate model; SST: Sea surface temperature.

4. NHRCM performance check (observation vs NHRCM)

The performance of the NHRCM in the various collaborating countries was evaluated, and the results are described below.

(1) Philippines

Cruz and Sasaki (2017) evaluated the performance of the NHRCM in simulating the present climate over Southeast Asia to determine its applicability to downscaling climate projections in the region. Simulations from 1989 to 2008 were conducted over the region at a 25km resolution using boundary conditions from the European Centre for Medium-Range Weather Forecasts ERA-Interim reanalysis (Dee et al., 2011).

The NHRCM captured the seasonal spatial patterns in rainfall and winds, especially along coastlines and over high topography, but overestimated rainfall on the windward side of mountains (Figure 7).

In order to assess the model performance over Southeast Asia, the model output was compared to daily temperature and rainfall data over land from the 0.25° resolution Asian Precipitation–Highly Resolved Observational Data Integration Towards Evaluation of Water Resources (APHRODITE) project(Hamada et al. 2011; Yasutomi et al. 2011; Yatagai et al. 2012) in boxes of Figure 8.

Furthermore, the NHRCM was able to reduce overestimated rainfall in the ERA-Interim, particularly over the eastern Philippines and on the Maritime Continent, with improvements in spatial patterns (Figure 9). Both seasonality and daily distribution of rainfall were represented in most regions. On the other hand, there was a tendency to underestimate the number of wet days, especially during the respective wet seasons, and to overestimate daily rainfall intensity. The NHRCM had an overall cold model bias, which reduced the warm bias in the ERA-Interim, except for some parts of Indochina during boreal winter and spring (Figure 10). These results indicate improved representation of the present climate in Southeast Asia using the NHRCM and its potential applicability in downscaling climate projections to increase projected climate scenarios for the region.

Figure 7: Seasonal mean rainfall (mm/day) and winds (m/s) at 850 hPa at 0.25° resolution from the Asian Precipitation–Highly Resolved Observational Data Integration Towards Evaluation of Water Resources (APHRODITE) project, ERA-Interim (observation), and Non-Hydrostatic Regional Climate Model (NHRCM) for June to August (JJA) and December to February (DJF). Note that the spacing between wind vectors has been adjusted for display (Cruz and Sasaki, 2017).
Figure 7: Seasonal mean rainfall (mm/day) and winds (m/s) at 850 hPa at 0.25° resolution from the Asian Precipitation–Highly Resolved Observational Data Integration Towards Evaluation of Water Resources (APHRODITE) project, ERA-Interim (observation), and Non-Hydrostatic Regional Climate Model (NHRCM) for June to August (JJA) and December to February (DJF). Note that the spacing between wind vectors has been adjusted for display (Cruz and Sasaki, 2017).
Figure 8: Terrain elevation (m) used in the Non-Hydrostatic Regional Climate Model at 25 km resolution (shaded) over the simulation domain, namely regions R1–R20 (Cruz and Sasaki, 2017).
Figure 8: Terrain elevation (m) used in the Non-Hydrostatic Regional Climate Model at 25 km resolution (shaded) over the simulation domain, namely regions R1–R20 (Cruz and Sasaki, 2017).
Figure 9: Evaluation of the Non-Hydrostatic Regional Climate Model (NHRCM) performance compared with the ERA-Interim. Seasonal mean rainfall (a) percentage difference (%) and (b) spatial correlation of the NHRCM and ERA-Interim relative to the Asian Precipitation–Highly Resolved Observational Data Integration Towards Evaluation of Water Resources (APHRODITE) project for regions 1–20 (land only) defined in Figure 7. Note that rainfall biases exceeding ±100% are colorless (Cruz and Sasaki, 2017).
Figure 9: Evaluation of the Non-Hydrostatic Regional Climate Model (NHRCM) performance compared with the ERA-Interim. Seasonal mean rainfall (a) percentage difference (%) and (b) spatial correlation of the NHRCM and ERA-Interim relative to the Asian Precipitation–Highly Resolved Observational Data Integration Towards Evaluation of Water Resources (APHRODITE) project for regions 1–20 (land only) defined in Figure 8. Note that rainfall biases exceeding ±100% are colorless (Cruz and Sasaki, 2017).
Figure 10: (a) Seasonal mean temperature bias (℃) and (b) spatial correlation of the Non-Hydrostatic Regional Climate Model (NHRCM) and ERA-Interim relative to the Asian Precipitation–Highly Resolved Observational Data Integration Towards Evaluation of Water Resources (APHRODITE) project for regions 1–20 (land only) defined in Figure 7 (Cruz and Sasaki, 2017).
Figure 10: (a) Seasonal mean temperature bias (℃) and (b) spatial correlation of the Non-Hydrostatic Regional Climate Model (NHRCM) and ERA-Interim relative to the Asian Precipitation–Highly Resolved Observational Data Integration Towards Evaluation of Water Resources (APHRODITE) project for regions 1–20 (land only) defined in Figure 8 (Cruz and Sasaki, 2017).

(2) Vietnam

Kieu-Thi et al. (2016) evaluated the performance of the NHRCM in Vietnam. The simulated precipitation showed that heavy rainfall centers were well captured in the seasonal change. In the near and far future, the projected rainfall by the NHRCM using outputs of MRI-AGCM3.2 (horizontal resolution is 60km) with the RCP8.5 scenario was analyzed. Monthly rainfall from the rain gauge data and NHRCM in the three sub-regions and entire Vietnam during the baseline and future periods are plotted in Figure 11. The model adequately simulated precipitation during December–April in northern Vietnam over the baseline period, but it underestimated rainfall during May–November, especially May–August, in the first half of the tropical cyclone season (Figure 11a).

Interestingly, the NHRCM captured the increasing tendency of rainfall, albeit slightly overestimated, in central Vietnam during May–June when early summer floods frequently occur. The simulated maximum rainfall occurred in September instead of October as with the observed rainfall (Figure 11b), leading to an overestimation during August–September and an underestimation during October–November. For southern Vietnam, the simulated rainfall agreed with the observed rainfall from December to May. However, the NHRCM overestimated rainfall from June to September when the summer monsoon prevails in the region (Figure 11c). This is probably due to NHRCM physics that are not consistent with monsoonal processes in low latitudes.

For the entire Vietnam, the NHRCM reproduced somewhat lower or higher rainfall for an intensive amount of 200–350 mm per month during the rest of the year (Figure 11d).

Figure 11: Monthly rainfall of the rain gauge data and Non-Hydrostatic Regional Climate Model (NHRCM) in (a) north, (b) central, (c) south, and (d) all of Vietnam during 1979–2003, 2015–2039, and 2075–2099 (Kieu-Thi et al., 2016).
Figure 11: Monthly rainfall of the rain gauge data and Non-Hydrostatic Regional Climate Model (NHRCM) in (a) north, (b) central, (c) south, and (d) all of Vietnam during 1979–2003, 2015–2039, and 2075–2099 (Kieu-Thi et al., 2016).

References

  • Cruz, F. T., and H. Sasaki, 2017: Simulation of present climate over Southeast Asia using the Non-Hydrostatic Regional Climate Model. SOLA, 13, 13−18, doi:10.2151/sola. 2017-003.
  • Cruz, F. T., H. Sasaki, and G. T. Narisma, 2016: Assessing the sensitivity of the Non‐Hydrostatic Regional Climate Model to boundary conditions and convective schemes over the Philippines. J. Meteorol. Soc. Jpn., 94, 165–179, doi: 10.2151/jmsj.2015-059.
  • Dee, D. P., S. M. Uppala, A. J. Simmons, P. Berrisford, P. Poli, S. Kobayashi, U. Andrae, M. A. Balmaseda, G. Balsamo, P. Bauer, P. Bechtold, A. C. M. Beljaars, L. van de Berg, J. Bidlot, N. Bormann, C. Delsol, R. Dragani, M. Fuentes, A. J. Geer, L. Haimberger, S. B. Healy, H. Hersbach, E. V. Hólm, L. Isaksen, P. Kållberg, M. Köhler, M. Matricardi, A. P. McNally, B. M. Monge-Sanz, J.-J. Morcrette, B.-K. Park, C. Peubey, P. de Rosnay, C. Tavolato, J.-N. Thépaut, and F. Vitart, 2011: The ERA-Interim reanalysis: configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553−597, doi: 10.1002/qj.828.
  • Hamada, A., O. Arakawa, and A. Yatagai, 2011: An automated quality control method for daily rain-gauge data. Global Environ. Res., V15N2, 183−192.
  • Hirai, M., and M. Ohizumi, 2004: Development of a new land-surface model for JMA-GSM. Bull. Amer. Meteor. Soc., 85, 2737‒2743.
  • Jamaluddin, A. F., F. Tangang, J. X. Chung, L. Juneng, H. Sasaki, and I. Takayabu, 2017: Investigating the mechanisms of diurnal rainfall variability over Peninsular Malaysia using the non-hydrostatic regional climate model. Meteorol. Atmos. Phys., 1–23, doi: 10.1007/s00703-017-0541-x.
  • Juneng, L., F. Tangang, J. X. Chung, S. T. Ngai, T. W. Tay, G. Narisma, F. Cruz, T. Phan-Van, T. Ngo-Duc, J. Santisirisomboon, P. Singhruck, D. Gunawan, and E. Aldrian, 2016: Sensitivity of Southeast Asia rainfall simulations to cumulus and air–sea flux parameterizations in RegCM4. Clim. Res., 69, 59−77, doi:10.3354/cr01386.
  • Kain, J., and J. Fritsch, 1993: Convective parameterization for mesoscale models: The Kain-Fritsch scheme. In: Emanuel K.A., Raymond D.J. (eds) The Representation of Cumulus Convection in Numerical Models, Meteor. Mongor. Amer. Meteor. Soc., Boston, MA, 165‒170, doi: 10.1007/978-1-935704-13-3_16.
  • Kieu-Thi, X., H. Vu-Thanh, T. Nguyen-Minh, D. Le, L. Nguyen-Manh, I. Takayabu, H. Sasaki, and A. Kitoh, 2016: Rainfall and tropical cyclone activity over Vietnam simulated and projected by the Non-Hydrostatic Regional Climate Model – NHRCM. J. Meteor. Soc. Jpn., 94A, 135–150, doi: 10.2151/jmsj.2015-057.
  • Mizuta, R., H. Yoshimura, H. Murakami, M. Matsueda, H. Endo, T. Ose, K. Kamiguchi, M. Hosaka, M. Sugi, S. Yukimoto, S. Kusunoki, and A. Kitoh, 2012: Climate simulations using MRI-AGCM3.2 with 20-km grid. J. Meteor. Soc. Jpn., 90A, 233–258, doi: 10.2151/jmsj.2012-A12.
  • Saito, K., T. Fujita, Y. Yamada, J. Ishida, Y. Kumagai, K. Aranami, A. Ohmori, R. Nagasawa, S. Kumagai, C. Muroi, T. Kato, H. Eito, and Y. Yamazaki, 2006: The operational JMA Nonhydrostatic Mesoscale Model. Mon. Weather Rev., 134, 1266‒1298, doi: 10.1175/MWR3120.1.
  • Sasaki, H., K. Kurihara, I. Takayabu, and T. Uchiyama, 2008: Preliminary experiments of reproducing the present climate using the non-hydrostatic regional climate model. SOLA, 4, 25–28, doi:10.2151/sola.2008-007.
  • Sasaki, H., A. Muratal, M. Hanafusa, M. Oh’izumi, and K. Kurihara, 2011: Reproducibility of present climate in a non-hydrostatic regional climate model nested within an atmosphere general circulation model. SOLA, 7, 173–176, doi:10.2151/sola.2011-044.
  • Sasaki, H., A. Muratal, M. Hanafusa, M. Oh’izumi, and K. Kurihara, 2012: Projection of future climate change in a non-hydrostatic regional climate model nested within an atmospheric general circulation model. SOLA, 8, 53−56, doi:10.2151/sola.2012-014.
  • Yasutomi, N., A. Hamada, and A. Yatagai, 2011: Development of a long-term daily gridded temperature dataset and its application to rain/snow discrimination of daily precipitation. Global Environ. Res., V15N2, 165−172.
  • Yatagai, A., K. Kamiguchi, O. Arakawa, A. Hamada, N. Yasutomi, and A. Kitoh, 2012: APHRODITE: Constructing a long-term daily gridded precipitation dataset for Asia based on a dense network of rain gauges. Bull. Amer. Meteor. Soc., 93, 1401− 1415.