Abstract

Based on the simulations of 22 CMIP5 models in combination with socio-economic data and terrain elevation data, the spatial distribution of risk levels of flood disaster and the vulnerability to flood hazards in China are projected under the RCP8.5 for the near term period (2016–2035), medium term period (2046–2065) and long term period (2080–2099), respectively. The results show that regions with high flood hazard levels are mainly located in Southeast China, while the vulnerability to flood hazards is high in eastern China. Under the RCP8.5 greenhouse gas emissions scenario, future high flood risk levels will mainly appear in the eastern part of Sichuan, in major part of East China, and in the provinces of Hebei, Beijing, and Tianjin. The major cities in Northeast China, some areas in Shaanxi and Shanxi, as well as the coastal areas in southeastern China will also encounter high flood risks. Compared with the baseline period, the regional flood risk levels will increase towards the end of the 21st century, although the occurrences of floods change little. Due to the coarse resolution of the climate models and the indistinct methodology for determining the weight coefficients, large uncertainty still exists in the projection of flood risks.

Keywords

RCP8.5 scenario ; flood risk ; projection

1. Introduction

Flood risk assessment is currently a high-ranked research topic. Under the background of global warming, the variation in future flood risks is the main focus. As China is a country prone to a variety of natural disasters, flooding is the major meteorological disasters affecting many regions. Floods in China cause huge losses to the national economy, e.g., in 2013 flooding caused annual direct economic losses of CN¥188.38 billion. Much work on flood risk assessment research has been done. For example, Ma et al. (2011) made assessment and warning research on China’s meteorological-observation-data-based flood hazard risks. They developed warning products according to the assessment results, which were combined with the distribution data of population and transport facilities in dangerous zones and other basic data. Li et al. (2013) made a rainstorm and flood risks division of the Zhanghe River Basin according to the principle of disaster risk (Zhang, 2010 ). Huang et al. (2001) and Ma et al. (2005) separately made flood risk assessment based on river network density, while Shi (2003) and Benito et al. (2004) made flood risk assessment using disaster statistics. Zhou et al. (2000) , Zhang (2006) , and Liu et al. (2011) made flood risk assessment according to indexation of rainfall, terrain, and vegetation of the river network and land-use type. In general, these researches provide a lot of information on historical floods, but little on future flood risks.

Numerical models provide a good methodology for the study of future climatic changes and the corresponding extreme climatic events. Especially with the improvement of high resolution climate models, it is now possible to do research on extreme climatic events and to make further assessments and estimations on disaster risks due to such events. For example, Wu et al. (2012a) and Yao et al. (2012) used regional and global climate simulations to analyze the 20-year return period of extreme temperature and precipitation variations. Additionally, Wu et al. (Wu et al., 2012b  and Wu et al., 2011 ) assessed the flood risks for different future periods. In 2012, IPCC published the Special Report: Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation (SREX) (IPCC, 2012 ), in which comprehensive analysis on the frequency of occurrence, the intensity, the vulnerability, and the degree of exposure to extreme events and corresponding disaster risks are presented.

More than 50 climate model results participated in the Coupled Model Intercomparison Project Phase 5 (CMIP5) were applied in more than 20 global climate model groups of the IPCC AR5, which provides the basis for the study of future extreme climatic events and associated disaster risks. Estimations of future climatic changes at decadal to 100-year scale can be calculated under different greenhouse gases emission scenarios and on the basis of historical climate change simulations of climate models (Xu and Xu, 2012a , Xu and Xu, 2012b , Yao et al., 2013  and Xu, 2010 ), similar calculations can also be done for extreme climatic events and future disaster risks.

Therefore, this research study uses simulation results of 22 global climate models provided by CMIP5, calculated extreme precipitation index and combined with socio-economic and geographic information data of China, to conduct a preliminary analysis for future flood disaster risks in China. This will further lead to the provision of references for disaster prevention and reduction policies and measures, as well as for disaster risk management.

2. Data and method

2.1. Data

According to the capability of CMIP5 models in providing daily data, for this study we select the simulation results of 22 global climate models, and calculate the maximum consecutive 5-day precipitation (Rx5day ) and number of days with at least 20 mm of precipitation (R20mm ) (see further model information in Table 1 ). Due to the different resolutions of the used global climate models, different resolution data were interpolated to a grid of 0.5° × 0.5° applying the bilinear interpolation method. Since the multi-model average is more consistent with observations than the single model result ( Xu et al., 2010 , Xu, 2010  and Zhou and Yu, 2006 ), in this study the flood risks assessment are done with an equally weighted average of the multi-model results. The data consists of three emission scenarios (RCP2.6, RCP4.5 and RCP8.5) estimated for the period of 1986–2100. Three future periods, 2016–2035, 2046–2065, and 2080–2099, are considered as the early, middle, and late 21st century, respectively. Focusing on the highest risks in the future, the most extreme high emission scenario RCP8.5 is selected for analyzing the future flood risks.

Table 1. Characteristics of 22 models of CMIP5
Model name Institutions and country Resolution (lon×lat)
CCESS 1.0 CSIRO-BOM, Australia 192×145
BCC-CSM1.1 BCC, China 128×64
BNU-ESM GCESS, China 128×64
CanCM4 CCCMA, Canada 128×64
CanESM2 CCCMA, Canada 128×64
CCSM4 NCAR, U.S. 288×192
CESM1-BGC NSF-DOE-NCAR, U.S. 288×192
CMCC-CM CMCC, Italy 480×240
CNRM-CM5 CNRM-CERFACS, France 256×128
CSIRO-Mk3-6-0 CSIRO-QCCCE, Australia 192×96
NorESM1-M NCC, Norway 144×96
GFDL-ESM2G NOAA GFDL, U.S. 144×90
GFDL-ESM2M NOAA GFDL, U.S. 144×90
INMCM4 INM, Russia 180×120
IPSL-CM5A-LR IPSL, France 96×96
MIROC5 MIROC, Japan 256×128
MIROC-ESM MIROC, Japan 128×64
MIROC-ESM-CHEM MIROC, Japan 128×64
MIROC4h MIROC, Japan 640×320
MPI-ESM-LR MPI-M, Germany 192×96
MPI-ESM-MR MPI-M, Germany 192×96
MRI-CGCM3 MRI, Japan 320×160

Source: http://cmip-pcmdi.llnl.gov/cmip5/

The socio-economic data used in this paper includes population density (POP), gross domestic product (GDP), percentage of arable land, and terrain elevation. The estimated population and GDP data are taken from the global 50 km × 50 km grid data (1990–2100) of the GGI A2 scenario (equivalent to RCP8.5) of the International Institute for Applied Systems Analysis (IIASA) with a temporal resolution of 10 years. Data were elected from the years 1990, 2030, 2060, and 2090 to represent the population and GDP situation in the 4 terms of 1986–2005, 2016–2035, 2046–2065, and 2080–2099, respectively. The percentage of historical arable land is available from 1700 to 2007 (http://www.sage.wisc.edu/download/crop1700/hist_croplands.html ); the data of 1990 is taken for the baseline period. As no future data of land use is available, we assume that no variation will happen in the areal extent of cultivated land area after 2007, and hence will use the latest information available, i.e. from 2007. The resolution of the terrain elevation data is 0.5° × 0.5° (http://eros.usgs.gov/#/Find_Data/Products_and_Data_Available/gtopo30_info ).

2.2. Methods

In order to facilitate the comparison of the results with similar domestic studies, this paper, calculated the flood disaster risk based on the method of Wu et al. (2012b ). This method has a high applicability, as the flood risk is understood as an extreme climate index, and the vulnerability of the flood hazard is based on an exposure and vulnerability level, with the specifics as follows:

(1) Hazard factors assessment: A normalization is conducted for the Rx5 day , R20mm and terrain elevation (E ). In order to avoid big value differences of the standardized E , the 100 m elevation is chosen as the numerator and the standardization formula is: 100/(E +145) ( Wu et al., 2012b ). The above 3 normalized indicators are weighted on the proportions of 0.5, 0.4 and 0.1, respectively. Then the normalized results are added up to form the flood hazard risk index. Based on the value, the flood hazard degree is divided into five levels, namely 0–0.3 (I), 0.3–0.4 (II), 0.4–0.5 (III), 0.5–0.6 (IV), 0.6–1 (V).

(2) Vulnerability assessment: First, the normalized POP, GDP, and the percentage of arable land are weighted based on experience (Wu et al., 2012b ). A projection model of the vulnerability to the hazard is established using the formula VF  = 0.34 × DPOP  + 0.38 × DGDP  + 0.27 × Pf , where VF refers to the vulnerability index from the projected area, DPOP , DgdP  and Pf refer to the normalized POP, GDP and the percentage of arable land of the projected area, respectively. Second, the calculated vulnerability index is standardized, where 1 is regarded as the maximum value of VF in all grids, so the standardized ratio of VF of each grid is the vulnerability level of each grid (Wu et al., 2012b ).

(3) Disaster risk assessment: According to the formula Risk = hazard risk × vulnerability (vulnerability and exposure), the flood disaster risk index (0–1) over China can be calculated. According to the value of the risk level, the flood disaster risk will be divided into 5 levels, namely, 0–0.02 (I), 0.02–0.05 (II), 0.05–0.1 (III), 0.1–0.2 (IV), and 0.2–1.0 (V).

3. Projection of flood disaster risk changes

Before analyzing the changes in flood disaster risks in the future, the simulation capability of the global climate models used for the projection of flood risks needs to be validated. Such validations of several CMIP3 and CMIP5 global climate models of extreme precipitation-related indices in China have been conducted by Xu (2010) , Jiang et al. (2009) , Xu et al. (2009) , and Sillmann et al. (2013) . Their results indicate that although the resolution of the global climate models is coarse, they can well simulate the spatial distribution of extreme precipitation climate fields and their linear trends in China. Furthermore, they assess that the multi-model set simulation ability is better than most of the single model simulations.

Based on the simulations of Rx5day and R20mm over China from 22 CMIP5 models from 1986 to 2005, these studies show that the correlation coefficients between single model simulations and observations were 0.60–0.80, while the correlation coefficients for the multi-model ensemble results were 0.85–0.88. The spatial distribution of the multi-model ensemble results is consistent with the observed values, which can well describe the distribution characteristics of extreme precipitation. Hence, the projection of flood risks based on the RCP8.5 and the integration of 22 CMIP5 models is conducted below.

3.1. Change in rainfall intensity and number of rainy days

The analysis of temporal and spatial changes in annual means of R x5day of different periods over China show that for the baseline period (Figure omitted) the national mean is at 89 mm, and the maximum is 273 mm (in the southern region of the Qinghai-Tibet Plateau). The areas with Rx5day ranging from 150 mm to 175 mm are mainly distributed in the southern coastal provinces such as Fujian, Guangdong, Guangxi and Hainan and Hunan. In the near term (Fig. 1 a), the national average is projected at 93 mm and the maximum at 280 mm. The areas with Rx5day above 150 mm extend northward to the provinces of Jiangxi and Zhejiang. In the late 21st century (Fig. 1 b), with increased concentrations of greenhouse gases, the national mean Rx5day will reach up to 109 mm, with a maximum of 352 mm, and the area with above 150 mm will continue to expand northward to Henan, while Fujian and Hainan will experience above 175 mm.


Distribution of annual meanRx5day(a, b) andR20mm(c, d) over China under RCP8.5 ...


Figure 1.

Distribution of annual mean Rx5day  (a, b) and R20mm  (c, d) over China under RCP8.5 for different periods

The analysis of spatial distribution of annual mean R20mm shows that for the baseline period (Figure omitted) the national average is 7 d, the maximum is 39 d, and the areas with above 15 d are mainly concentrated in the provinces of Hunan, Jiangxi, Zhejiang, Anhui and Fujian, as well as in the southern parts of Anhui and Hubei, in the northern parts of Guangdong and Guangxi, and in central Sichuan. In the near future (Fig. 1 c), the national average is projected to be 8 d, the maximum will be 40 d, and areas of above 15 d will be similar to the baseline period. In the medium and late 21st century, with increasing concentrations of greenhouse gas emissions, the annual average R20mm will reach 9 d and 10 d, respectively, and the maximum will reach up to 41 d and 43 d, respectively. In the late 21st century, areas with more than 20 d will be larger than for the baseline period and the early 21st century, and will extend to the west of Henan, to the southern part of Shaanxi, and to the western parts of Guangxi and Guizhou. The provinces of Jiangxi, Fujian, and Zhejiang will experience more than 20 d, and central Sichuan will experience more than 25 d (Fig. 1 d).

3.2. Temporal and spatial changes in flood hazards

Based on the results from Rx5day , R20mm and terrain elevation data, the levels of the flood hazards can be assessed by overlaying the indices (Fig. 2 ). The results show that the national mean flood hazard level of the baseline period is 0.26, while for the near, medium and late 21st century the level is at 0.27 with little variation. Both in the baseline and future periods, areas with the highest flood hazard levels are concentrated in southeastern China. The regions with level III continue to expand in the future. In the late 21st century, the area with level IV, which was originally located in Zhejiang, Jiangxi, Fujian, Guangdong and the eastern part of Guangxi, will get smaller than during the baseline period and the beginning of the 21st century. The hazard areas with level III will extend from central Sichuan to the east, and northward to the south of Shanxi. In Table 2 the percentage of areas with different flood hazard levels accounting for the total land area are shown. It can be seen that the major areal changes in flood hazards will occur in the middle and long term periods, with little differences for the near term compared with the baseline period. In the middle period, the areas of flood hazards of level I and II account for 73.0%, while level III to V areas cover 26.9% of the country. The area with level III will expand from 14.2% during the baseline period to 18.3% in the late 21st century. The area with level V will slightly increase in the future (2.1%) compared to the baseline period (1.9%).


Distribution of flood hazard levels over China under RCP8.5 for different ...


Figure 2.

Distribution of flood hazard levels over China under RCP8.5 for different periods

Table 2. Proportion of areal flood hazard levels under RCP8.5 for different periods (unit: %)
Level 1986–2005 2016–2035 2046–2065 2080–2099
I 61.2 60.0 59.5 59.2
II 14.5 15.0 13.5 12.5
III 14.2 13.9 15.8 18.3
IV 8.2 8.3 8.9 7.8
V 1.9 2.1 2.2 2.1

3.3. Spatial and temporal changes in the vulnerability to flood hazards

According to the aforementioned calculation of the vulnerability to flood hazards, the percentage of population, GDP and arable land in 1990, 2030, 2060 and 2090 has been chosen to estimate the vulnerability distribution over China for the past and future periods (Wu et al., 2012b ) (Fig. 3 ).


Distribution of flood disaster vulnerability over China under RCP8.5


Figure 3.

Distribution of flood disaster vulnerability over China under RCP8.5

The national mean vulnerability to flood hazards during the baseline period is 0.065, while the highest value is about 0.47. For the near future period the national average is projected at 0.064, and the maximum value is at 0.57. For the middle future period the national average vulnerability value will be 0.068, and the highest value will be up to 0.68. In the late 21st century, the national mean vulnerability will be 0.07, with the highest value of 0.70. Although there are little changes in the mean value of the country’s vulnerability to flood hazards in future periods of the 21st century, compared to the baseline period the maximum values will gradually increase by up to 49% in the late 21st century.

According to Figure 3 , the areas with high vulnerability values are located in China’s eastern region. In the baseline period, areas with high vulnerability are mainly located in Henan, Anhui, and Shandong, as well as the eastern region of Sichuan. Shanghai shows the highest level of vulnerability. In the near future, the high vulnerable area will stretch further north to the Beijing-Tianjin region, and the vulnerability of parts of the Sichuan basin will be higher than 0.5. In the late 21st century, the area with high vulnerability will expand eastward and the intensity of vulnerability will also gradually increase. Especially in the late 21st century, the vulnerability of East China and parts of the Pearl River Delta will be very high. In addition, high values in vulnerability will become more in major cities of Northeast China, as well as Wuhan, Changsha and Nanchang, which is mainly due to their high POP and GDP. Areas with medium and low vulnerability are mainly located in southeastern China, Yunnan, Xinjiang, Inner Mongolia and Tibet.

In Table 3 , the percentages of the different flood disaster vulnerability levels are shown on the basis of the country’s total area. It can be seen that with the population growth and economic development in the future, the areas above the 0.4 flood vulnerability level will be constantly expand from 0.1% in the baseline period to 1.4% in the late 21st century.

Table 3. Proportion of areal flood disaster vulnerability levels under RCP8.5 for different periods (unit: %)
Level 1986–2005 2016–2035 2046–2065 2080–2099
0–0.02 48.3 49.3 49.8 50.0
0.02–0.04 6.6 7.5 7.5 7.6
0.04–0.06 6.5 7.6 7.5 7.3
0.06–0.1 12.8 11.8 10.5 9.9
0.1–0.2 16.5 14.7 14.3 14.1
0.2–0.3 7.5 6.5 6.3 6.3
0.3–0.4 1.6 2.5 3.4 3.2
0.4–0.5 0.1 0.1 0.7 1.3
0.5–1.0 0.0 0.1 0.1 0.1

3.4. Projection of flood disaster risks

Through integrating and standardizing the above bearing body, the flood risk distribution over China in the baseline term and the future each term can be found.

The results of the calculation of the flood hazard and vulnerability show that in various future periods areas with high flood hazard risk levels are mainly concentrated in the central and eastern regions of China. The areas with the highest flood risks will be located in eastern Sichuan, Chongqing, and the middle and lower reaches of the Yangtze River, and stretching as far north as the Beijing-Tianjin region. In addition, high-risk flood areas will also be found in the major cities in Northeast China, Shaanxi, Shanxi, as well as some coastal areas in southeastern China (Fig. 4 , showing the results only for the periods of 2016–2035 and 2080–2099).


Spatial distribution of flood risk levels over China under RCP8.5 for different ...


Figure 4.

Spatial distribution of flood risk levels over China under RCP8.5 for different periods

By comparing each period with the baseline period, little variation in flood hazard risks can be found. Due to a significant decrease in arable land area but a significant increase in POP and GDP in the future, the vulnerability to flood hazards does not reduce, and therefore the high-risk flood hazard areas (level IV and V) will stay similar to the baseline period. As can be seen from Table 4 , in the late 21st century the area with the risk level IV is slightly less than in the baseline period, while the area with the risk level V is slightly increasing. It can be seen that in the future, with increasing greenhouse gas emissions, the flood prone areas vary slightly, but the regions with high risk levels will increase.

Table 4. Proportion of areal flood risk levels under RCP8.5 for different periods (unit: %)
Level 1986–2005 2016–2035 2046–2065 2080–2099
I 54.7 54.8 54.8 54.9
II 9.0 9.1 9.3 9.5
III 13.6 13.5 13.6 13.6
IV 11.6 11.4 10.8 10.4
V 11.0 11.2 11.5 11.6

4. Conclusions and discussion

Based on the simulations from 22 CMIP5 models and in combination with data on POP, GDP, arable land, and terrain elevation, the spatial distributions of the flood risk levels are calculated and analyzed under RCP8.5 for the baseline period (1986–2005), the near term period (2016–2035), the medium term period (2046–2065), and the long term period (2080–2099).

(1) Areas with higher flood hazard risk levels in the future are concentrated in southeastern China, and the areas with the risk level III continue to expand. The major changes in flood hazard risks will occur in the medium and long term future.

(2) In future, the areas with high vulnerability to flood hazards will be located in China’s eastern region. In the middle and late 21st century, the extent of the high vulnerability area will expand eastward and its intensity will gradually increase. The highest vulnerability values are found in Beijing, Tianjin, Hebei, Henan, Anhui, Shandong, Shanghai, Jiangsu, and parts of the Pearl River Delta. Furthermore, the major cities in Northeast China, as well as Wuhan, Changsha and Nanchang are highly vulnerable.

(3) The regions with high flood risk levels will be located in eastern China, in the middle and lower reaches of the Yangtze River and stretching northward to Beijing and Tianjin. High-risk flood areas are also occurring in major cities in Northeast China, in some parts of Shaanxi and Shanxi, and in some coastal areas in southeastern China.

(4) Compared to the baseline period, the high flood risks will increase on a regional level towards the end of the 21st century, although the areas of flood hazards show little variation.

In this paper, the projected future flood risks for different periods were analyzed under the RCP8.5 emission scenario. By comparing the results with the simulations under the RCP 2.6 and RCP 4.5 scenarios, both scenarios show no differences in the spatial distribution, but in the intensity of flood hazard risks, which are weaker than for the RCP8.5 scenarios.

By using the simulations from climate model ensembles to project future flood risks, uncertainty exists for various factors, such as the coarse resolution of global climate models, different approaches to flood assessments, the selection of the weighting coefficients, as well as the used greenhouse gas emission scheme, and the estimations of future POP, GDP, and arable land. Therefore, further analysis is needed to reduce the uncertainties of future flood risks.

Acknowledgements

This paper is supported by the China Meteorological Administration Special Public Welfare Research Fund (GYHY201306019), the National Natural Science Foundation of China (41275078), the Grant Projects of China Clean Development Mechanism Fund (121312), and the Climate Change Foundation of the China Meteorological Administration (CCSF201339).

References

  1. Benito et al., 2004 G. Benito, M. Lang, M. Barriendos, et al.; Use of systematic, palaeo and historical data for the improvement of flood risk estimation; Natural Hazards, 31 (2004), pp. 623–643
  2. Huang et al., 2001 S.-F. Huang, M. Xu, D.-Q. Chen; GIS-based extraction of drainage network density and its application to flood hazard analysis; Journal of Natural Disasters (in Chinese), 10 (4) (2001), pp. 129–132
  3. IPCC, 2012 IPCC; Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation: Special Report of the IPCC; Cambridge University Press, New York (2012)
  4. Jiang et al., 2009 Z.-H. Jiang, W.-L. Chen, J. Song, et al.; Projection and evaluation of the precipitation extremes indices over China based on seven LPCC AR4 coupled climate models; Chinese Journal of Atmospheric Sciences (in Chinese), 33 (1) (2009), pp. 109–120
  5. Li et al., 2013 L. Li, Y.-H. Zhou, L.-M. Ye, et al.; Basin rainstorm flood risk regionalization method based on gis rainstorm flood inundation model; Meteorological Monthly (in Chinese), 39 (1) (2013), pp. 112–117
  6. Liu et al., 2011 J.-F. Liu, J. Li, Y.-H. Liang, et al.; Storm flood risk assessment in the typical regions of asia; Scientia Geographica Sinica (in Chinese), 31 (10) (2011), pp. 1266–1271
  7. Ma et al., 2011 G.-B. Ma, J. Li, W.-G. Jiang, et al.; Hazard assessment and early warning research on flood disaster in china based on weather forecast data; Journal of Catastrophology (in Chinese), 26 (3) (2011), p. 817
  8. Ma et al., 2005 Z.-W. Ma, Y.-P. Xu, J.-J. Li; River fractal dimension and the relationship between river fractal dimension and river flood: Case study in the middle and lower course of the Yangtze River; Advances in Water Science (in Chinese), 16 (4) (2005), pp. 530–534
  9. Shi, 2003 P.-J. Shi; Atlas of Natural Disaster System of China; Science Press, Beijing (2003)
  10. Sillmann et al., 2013 J. Sillmann, V.-V. Kharin, X.-B. Zhang, et al.; Climate extremes indices in the CMIP5 multimodel ensemble: Part 1. Model evaluation in the present climate; Journal of Geophysical Research: Atmospheres, 118 (2013), pp. 1–18
  11. Wu et al., 2012a J. Wu, X.-J. Gao, Y. Shi; Changes of 20-year return temperature and precipitation extremes over China simulated by RegCM3; Advances in Climate Change Research (in Chinese), 8 (4) (2012), pp. 243–249
  12. Wu et al., 2011 S.-H. Wu, T. Pan, S.-F. He; Primary study on the theories and methods of research on climate change risk; Advances in Climate Change Research (in Chinese), 7 (5) (2011), pp. 363–368
  13. Wu et al., 2012b S.-H. Wu, E.-F. Dai, Q.-S. Ge, et al.; The Comprehensive Risk Prevention: Chinese Comprehensive Climate Change Risks; Science Press, Beijing (2012)
  14. Xu, 2010 C.-H. Xu; Simulation and projection for extremes climate events in China by global climate models; Graduate University of Chinese Academy of Science (2010)
  15. Xu and Xu, 2012a C.-H. Xu, Y. Xu; The projection of temperature and precipitation over China under RCP scenarios using a CMIP5 multi-model ensemble; Atmospheric and Oceanic Science Letters, 5 (6) (2012), pp. 527–533
  16. Xu and Xu, 2012b Y. Xu, C.-H. Xu; Preliminary assessment of simulations of climate changes over China by CMIP5 multi-models; Atmospheric and Oceanic Science Letters, 5 (6) (2012), pp. 489–494
  17. Xu et al., 2010 Y. Xu, X.-J. Gao, F. Giorgi; Upgrades to the reliability ensemble averaging method for producing probabilistic climate-change projections; Climatic Research, 41 (2010), pp. 61–81
  18. Xu et al., 2009 Y. Xu, C.-H. Xu, X.-J. Gao, et al.; Projected changes in temperature and precipitation extremes over the Yangtze River Basin of China in the 21st century; Quaternary International, 208 (2009), pp. 44–52
  19. Yao et al., 2012 Y. Yao, Y. Luo, J.-B. Huang; Evaluation and projection of temperature extreme over China based on 8 modeling data from CMIP5; Advances in Climate Change Research (in Chinese), 8 (4) (2012), pp. 250–256
  20. Yao et al., 2013 Y. Yao, Y. Luo, J.-B. Huang, et al.; Comparison of monthly temperature extremes simulated by CMIP3 and CMIP5 models; Journal of Climate, 26 (2013), pp. 7692–7707
  21. Zhang, 2010 G.-C. Zhang; Meteorological disaster risk assessment and zoning method (in Chinese); China Meteorological Press, Beijing (2010)
  22. Zhang, 2006 N.-Q. Zhang; Evaluation of flood risk in Poyang Lake region basing on GIS; Nanchang University (2006)
  23. Zhou et al., 2000 C.-H. Zhou, Q. Wan, S.-F. Huang, et al.; A GIS-based approach to flood risk zonation; Acta Geographical Siniac (in Chinese), 55 (1) (2000), pp. 15–24
  24. Zhou and Yu, 2006 T.-J. Zhou, R.-C. Yu; Twentieth century surface air temperature over China and the globe simulated by coupled climate models; Journal of Climate, 19 (22) (2006), pp. 5843–5858
Back to Top

Document information

Published on 15/05/17
Submitted on 15/05/17

Licence: Other

Document Score

0

Views 6
Recommendations 0

Share this document

claim authorship

Are you one of the authors of this document?