• Carson, M., A. Köhl, D. Stammer, A. B. A. Slangen, C. A. Katsman, R. S. W. van de Wal, J. Church, and N. White, 2016: Coastal sea level changes, observed and projected during the 20th and 21st century. Climatic Change, 134, 269281, doi:10.1007/s10584-015-1520-1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chen, H.-P., J.-Q. Sun, and C. Xiao-Li, 2013a: Future changes of drought and flood events in China under a global warming scenario. Atmos. Ocean. Sci. Lett., 6, 813.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chen, Q., C. Wu, G. Huang, and S. Wu, 2013b: Study on characteristics of short duration rainstorm in Guangzhou City (in Chinese). Water Res. Power, 31, 46.

    • Search Google Scholar
    • Export Citation
  • Cuo, L., T. K. Beyene, N. Voisin, F. Su, D. P. Lettenmaier, M. Alberti, and J. E. Richey, 2011: Effects of mid-twenty-first century climate and land cover change on the hydrology of the Puget Sound basin, Washington. Hydrol. Processes, 25, 17291753, doi:10.1002/hyp.7932.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Elshamy, M. E., I. A. Seierstad, and A. Sorteberg, 2009: Impacts of climate change on Blue Nile flows using bias-corrected GCM scenarios. Hydrol. Earth Syst. Sci., 13, 551565, doi:10.5194/hess-13-551-2009.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Favre, A.-C., S. El Adlouni, L. Perreault, N. Thiémonge, and B. Bobée, 2004: Multivariate hydrological frequency analysis using copulas. Water Resour. Res., 40, W01101, doi:10.1029/2003WR002456.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Fischer, T., M. Gemmer, L. Liu, and B. Su, 2010: Trends in monthly temperature and precipitation extremes in the Zhujiang River basin, south China (1961–2007). Adv. Clim. Change Res., 1, 6370, doi:10.3724/SP.J.1248.2010.00063.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Fischer, T., B. Su, Y. Luo, and T. Scholten, 2012: Probability distribution of precipitation extremes for weather index–based insurance in the Zhujiang River basin, south China. J. Hydrometeor., 13, 10231037, doi:10.1175/JHM-D-11-041.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gemmer, M., T. Fischer, T. Jiang, B. Su, and L. L. Liu, 2011: Trends in precipitation extremes in the Zhujiang River basin, south China. J. Climate, 24, 750761, doi:10.1175/2010JCLI3717.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Harding, B. L., A. W. Wood, and J. R. Prairie, 2012: The implications of climate change scenario selection for future streamflow projection in the upper Colorado River basin. Hydrol. Earth Syst. Sci., 16, 39894007, doi:10.5194/hess-16-3989-2012.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hay, L. E., R. L. Wilby, and G. H. Leavesley, 2000: A comparison of delta change and downscaled GCM scenarios for three mountainous basins in the United States. J. Amer. Water Resour. Assoc., 36, 387397, doi:10.1111/j.1752-1688.2000.tb04276.x.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hosking, J. R. M., and J. R. Wallis, 2005: Regional Frequency Analysis: An Approach Based on L-Moments. Cambridge University Press, 244 pp.

  • IPCC, 2007: Summary for policymakers. Climate Change 2007: The Physical Science Basis, S. Solomon et al., Eds., Cambridge University Press, 1–18.

    • Search Google Scholar
    • Export Citation
  • IPCC, 2013: Summary for policymakers. Climate Change 2013: The Physical Science Basis, T. F. Stocker et al., Eds., Cambridge University Press, 3–29.

  • Kay, A. L., and D. A. Jones, 2012: Transient changes in flood frequency and timing in Britain under potential projections of climate change. Int. J. Climatol., 32, 489502, doi:10.1002/joc.2288.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kendall, M. G., 1975. Rank Correlation Methods. 4th ed. Charles Griffin, 272 pp.

  • Kendon, E. J., D. P. Rowell, R. G. Jones, and E. Buonomo, 2008: Robustness of future changes in local precipitation extremes. J. Climate, 21, 42804297, doi:10.1175/2008JCLI2082.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Li, J., Y. D. Chen, L. Zhang, Q. Zhang, and F. H. S. Chiew, 2016: Future changes in floods and water availability across China: Linkage with changing climate and uncertainties. J. Hydrometeor., 17, 12951314, doi:10.1175/JHM-D-15-0074.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Li, X., J. L. Zhang, and Z. G. Gao, 2011: Discussion on semiempirical prediction method for sea level change of China. Mar. Sci. Bull., 30, 540543.

    • Search Google Scholar
    • Export Citation
  • Liu, L.-L., T. Jiang, J.-G. Xu, J. Q. Zhai, and Y. Luo, 2012: Responses of hydrological processes to climate change in the Zhujiang River basin in the 21st century. Adv. Climate Change Res., 3, 8491, doi:10.3724/SP.J.1248.2012.00084.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lutz, A. F., W. W. Immerzeel, A. Gobiet, F. Pellicciotti, and M. F. P. Bierkens, 2013: Comparison of climate change signals in CMIP3 and CMIP5 multi-model ensembles and implications for central Asian glaciers. Hydrol. Earth Syst. Sci., 17, 36613677, doi:10.5194/hess-17-3661-2013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ma, M., S. Song, L. Ren, S. Jiang, and J. Song, 2013: Multivariate drought characteristics using trivariate Gaussian and Student t copulas. Hydrol. Processes, 27, 11751190, doi:10.1002/hyp.8432.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mann, H. B., 1945: Non-parametric tests against trend. Econometrica, 13, 245259, doi:10.2307/1907187.

  • McIntosh, P. C., J. A. Church, E. R. Miles, K. Ridgway, and C. M. Spillman, 2015: Seasonal coastal sea level prediction using a dynamical model. Geophys. Res. Lett., 42, 67476753, doi:10.1002/2015GL065091.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Meehl, G. A., C. Covey, K. E. Taylor, T. Delworth, R. J. Stouffer, M. Latif, B. McAvaney, and J. B. Mitchell, 2007: The WCRP CMIP3 multimodel dataset: A new era in climate change research. Bull. Amer. Meteor. Soc., 88, 13831394, doi:10.1175/BAMS-88-9-1383.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Min, S.-K., X. Zhang, F. Zwiers, H. Shiogama, Y.-S. Tung, and M. Wehner, 2013: Multimodel detection and attribution of extreme temperature changes. J. Climate, 26, 74307451, doi:10.1175/JCLI-D-12-00551.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mladjic, B., L. Sushama, M. N. Khaliq, R. Laprise, D. Caya, and R. Roy, 2011: Canadian RCM projected changes to extreme precipitation characteristics over Canada. J. Climate, 24, 25652584, doi:10.1175/2010JCLI3937.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Nelsen, R. B., 2006: An Introduction to Copulas. 2nd ed. Springer, 272 pp.

  • O’Gorman, P. A., and T. Schneider, 2009: Scaling of precipitation extremes over a wide range of climates simulated with an idealized GCM. J. Climate, 22, 56765685, doi:10.1175/2009JCLI2701.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Panday, P. K., J. Thibeault, and K. E. Frey, 2015: Changing temperature and precipitation extremes in the Hindu Kush–Himalayan region: An analysis of CMIP3 and CMIP5 simulations and projections. Int. J. Climatol., 35, 30583077, doi:10.1002/joc.4192.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Petrow, T., and B. Merz, 2009: Trends in flood magnitude, frequency and seasonality in Germany in the period 1951–2002. J. Hydrol., 371, 129141, doi:10.1016/j.jhydrol.2009.03.024.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Raff, D. A., T. Pruitt, and L. D. Brekke, 2009: A framework for assessing flood frequency based on climate projection information. Hydrol. Earth Syst. Sci., 13, 21192136, doi:10.5194/hess-13-2119-2009.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rahmstorf, S., 2007: A semi-empirical approach to projecting future sea-level rise. Science, 315, 368370, doi:10.1126/science.1135456.

  • Sansom, J., and J. A. Renwick, 2007: Climate change scenarios for New Zealand rainfall. J. Appl. Meteor. Climatol., 46, 573590, doi:10.1175/JAM2491.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Shiau, J.-T., S. Feng, and S. Nadarajah, 2007: Assessment of hydrological droughts for the Yellow River, China, using copulas. Hydrol. Processes, 21, 21572163, doi:10.1002/hyp.6400.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sklar, A., 1959: Fonctions de répartition à n dimensions et leurs marges. Publ. Inst. Stat. Univ. Paris, 8, 229231.

  • Slangen, A. B. A., J. A. Church, X. Zhang, and D. P. Monselesan, 2015: The sea level response to external forcings in historical simulations of CMIP5 climate models. J. Climate, 28, 85218539, doi:10.1175/JCLI-D-15-0376.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Smith, A., P. Bates, J. Freer, and F. Wetterhall, 2014: Investigating the application of climate models in flood projection across the UK. Hydrol. Processes, 28, 28102823, doi:10.1002/hyp.9815.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Smith, I., J. Syktus, C. McAlpine, and K. Wong, 2013: Squeezing information from regional climate change projections—Results from a synthesis of CMIP5 results for south-east Queensland, Australia. Climatic Change, 121, 609619, doi:10.1007/s10584-013-0956-4.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Taylor, K. E., R. J. Stouffer, and G. A. Meehl, 2012: An overview of CMIP5 and the experiment design. Bull. Amer. Meteor. Soc., 93, 485498, doi:10.1175/BAMS-D-11-00094.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Vermeer, M., and S. Rahmstorf, 2009: Global sea level linked to global temperature. Proc. Natl. Acad. Sci. USA, 106, 21 52721 532, doi:10.1073/pnas.0907765106.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wu, C., G. Huang, and S. Wu, 2014a: Risk analysis of combinations of short duration rainstorm and tidal level in Guangzhou based on Copula function (in Chinese). J. Hydraul. Eng., 33, 3340.

    • Search Google Scholar
    • Export Citation
  • Wu, C., G. Huang, H. Yu, Z. Chen, and J. Ma, 2014b: Impact of climate change on reservoir flood control in the upstream area of the Beijiang River basin, south China. J. Hydrometeor., 15, 22032218, doi:10.1175/JHM-D-13-0181.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wu, C., G. Huang, and H. Yu, 2015: Prediction of extreme floods based on CMIP5 climate models: A case study in the Beijiang River basin, south China. Hydrol. Earth Syst. Sci., 19, 13851399, doi:10.5194/hess-19-1385-2015.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wu, C., Z. Xian, and G. Huang, 2016: Meteorological drought in the Beijiang River basin, south China: Current observations and future projections. Stochastic Environ. Res. Risk Assess., 30, 18211834, doi:10.1007/s00477-015-1157-7.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wuebbles, D., and et al. , 2014: CMIP5 climate model analyses: Climate extremes in the United States. Bull. Amer. Meteor. Soc., 95, 571583, doi:10.1175/BAMS-D-12-00172.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yang, T., Q. Shao, Z.-C. Hao, X. Chen, Z. Zhang, C.-Y. Xu, and L. Sun, 2010: Regional frequency analysis and spatio-temporal pattern characterization of rainfall extremes in the Pearl River basin, China. J. Hydrol., 380, 386405, doi:10.1016/j.jhydrol.2009.11.013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yao, Y., Y. Luo, J. Huang, and Z. Zhao, 2013: Comparison of monthly temperature extremes simulated by CMIP3 and CMIP5 models. J. Climate, 26, 76927707, doi:10.1175/JCLI-D-12-00560.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, D.-L., Y. Lin, P. Zhao, X. Yu, S. Wang, H. Kang, and Y. Ding, 2013: The Beijing extreme rainfall of 21 July 2012: “Right results” but for wrong reasons. Geophys. Res. Lett., 40, 14261431, doi:10.1002/grl.50304.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, H., K. Fraedrich, R. Blender, and X. Zhu, 2013: Precipitation extremes in CMIP5 simulations on different time scales. J. Hydrometeor., 14, 923928, doi:10.1175/JHM-D-12-0181.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhao, Y., X. Zou, L. Cao, and X. Xu, 2014: Changes in precipitation extremes over the Pearl River basin, southern China, during 1960–2012. Quat. Int., 333, 2639, doi:10.1016/j.quaint.2014.03.060.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • View in gallery

    Location of Guangzhou city in southern China.

  • View in gallery

    Comparison of empirical and theoretical accumulative probabilities for the (a) Clayton and (b) AMH copula.

  • View in gallery

    Uncertainty range of trend magnitudes in (a) monthly temperature, (b) monthly precipitation, and (c) AM1R and AM3R for the period 2020–50 under all the emissions scenarios of the eight climate models. The red dotted lines indicate the average values of trend magnitudes in monthly temperature and precipitation, and the blue dashed lines indicate 95% confidence interval for the average values of trend magnitudes.

  • View in gallery

    Predicted changes in monthly temperature under the (a) A1B, (b) RCP2.6, (c) RCP4.5, and (d) RCP8.5 scenario for the future period 2020–50 relative to the historical period 1970–2000. The projected changes are computed as the long-term mean monthly values for the period 2020–50 minus those for the period 1970–2000. (e) The average values of the projected changes for all the emissions scenarios of eight climate models (red dotted line) and the corresponding 95% confidence interval (blue dashed lines).

  • View in gallery

    Predicted changes (%) in monthly precipitation under the (a) A1B, (b) RCP2.6, (c) RCP4.5, and (d) RCP8.5 scenario for the future period 2020–50 relative to the historical period 1970–2000. The projected percentage changes represent the percentage ratio of the projected differences to the long-term mean monthly values for the period 1970–2000, where the projected differences are computed as the long-term mean monthly values for the period 2020–50 minus those for the period 1970–2000. (e) The average values of the projected percentage changes for all the emissions scenarios of eight climate models (red dotted line) and the corresponding 95% confidence interval (blue dashed lines).

  • View in gallery

    Contours showing the projected changes (%) in simultaneous probability for the future period 2020–50 relative to the historical period 1970–2000, for four different emissions scenarios.

  • View in gallery

    Contours showing the projected changes (%) in waterlogging probability for the future period 2020–50 relative to the historical period 1970–2000 for four different emissions scenarios of eight climate models.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 261 261 36
PDF Downloads 193 193 28

Assessing the Impact of Climate Change on the Waterlogging Risk in Coastal Cities: A Case Study of Guangzhou, South China

View More View Less
  • 1 School of Civil Engineering and Transportation, South China University of Technology, Guangzhou, China
  • | 2 School of Civil Engineering and Transportation, and State Key Laboratory of Subtropical Building Science, South China University of Technology, Guangzhou, China
© Get Permissions
Full access

Abstract

Climate warming is expected to occur with an increased magnitude of extreme precipitation and sea level rise, which leads to an increased probability of waterlogging in coastal cities. In this paper, a combined probability model is developed to evaluate the impact of climate change on waterlogging in Guangzhou by using eight climate models with four emissions scenarios [Special Report on Emissions Scenarios (SRES) scenario A1B and representative concentration pathway (RCP) scenarios RCP2.6, RCP4.5, and RCP8.5]. The copula method was applied to derive the bivariate distributions of extreme rainfall and tidal level. The uncertainty in the projected future temperature, extreme rainfall, sea level, and the combined extreme rainfall and tidal level probability were discussed. The results show that although there is a large uncertainty driven by both climate models and emissions scenarios in the projection of climate change, most modeling results predict an increase in temperature and extreme precipitation in Guangzhou during the future period of 2020–50, relative to the historical period of 1970–2000. Moreover, greater increases are projected for higher emissions scenarios. The sea level is projected to increase in the range of 11.40–23.37 cm during the period 2020–50, consistent with climate warming. Both simultaneous probability and waterlogging probability are projected to show an upward trend in the future period 2020–50, with the largest and smallest increases in the RCP4.5 and RCP2.6 scenarios, respectively. The results of this paper provide a new scientific reference for waterlogging control in Guangzhou under climate change conditions.

© 2017 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author: Guoru Huang, huanggr@scut.edu.cn

Abstract

Climate warming is expected to occur with an increased magnitude of extreme precipitation and sea level rise, which leads to an increased probability of waterlogging in coastal cities. In this paper, a combined probability model is developed to evaluate the impact of climate change on waterlogging in Guangzhou by using eight climate models with four emissions scenarios [Special Report on Emissions Scenarios (SRES) scenario A1B and representative concentration pathway (RCP) scenarios RCP2.6, RCP4.5, and RCP8.5]. The copula method was applied to derive the bivariate distributions of extreme rainfall and tidal level. The uncertainty in the projected future temperature, extreme rainfall, sea level, and the combined extreme rainfall and tidal level probability were discussed. The results show that although there is a large uncertainty driven by both climate models and emissions scenarios in the projection of climate change, most modeling results predict an increase in temperature and extreme precipitation in Guangzhou during the future period of 2020–50, relative to the historical period of 1970–2000. Moreover, greater increases are projected for higher emissions scenarios. The sea level is projected to increase in the range of 11.40–23.37 cm during the period 2020–50, consistent with climate warming. Both simultaneous probability and waterlogging probability are projected to show an upward trend in the future period 2020–50, with the largest and smallest increases in the RCP4.5 and RCP2.6 scenarios, respectively. The results of this paper provide a new scientific reference for waterlogging control in Guangzhou under climate change conditions.

© 2017 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author: Guoru Huang, huanggr@scut.edu.cn

1. Introduction

It is widely recognized that atmospheric concentrations of greenhouse gases have been increasing since industrialization due to human activities, leading to significant climate change. According to the Intergovernmental Panel on Climate Change (IPCC)’s Fourth Assessment Report (AR4), coastal areas, in particular delta areas in Asia, are projected to have an increased flooding risk in the future (IPCC 2007). Climate change causes sea level rise and intensifies the global hydrological cycle, most notably making the spatiotemporal distribution of rainfall more uneven, resulting in an increased frequency and severity of water disasters (flood and drought) over many global regions (IPCC 2013).

Climate change has received wide attention, and numerous scientists have focused on the assessment of climate change impacts in the past two decades (Sansom and Renwick 2007; Petrow and Merz 2009; Cuo et al. 2011; Kay and Jones 2012; Li et al. 2016). In hydrologic modeling studies for climate change impact assessment, historical and future climate simulations by global climate models (GCMs) are generally used as inputs to hydrologic models (Elshamy et al. 2009; O’Gorman and Schneider 2009; Raff et al. 2009; Wu et al. 2014b, 2015). However, GCM outputs are coarse and cannot be directly used in hydrologic modeling for climate impact studies at the regional scale. Downscaling methods are therefore utilized to link coarse-resolution GCM outputs with catchment-scale climatic variables. Based on the data provided by GCMs, numerous studies have investigated the effects of climate change on extreme rainfall (Kendon et al. 2008; Yang et al. 2010; Min et al. 2013; D.-L. Zhang et al. 2013; H. Zhang et al. 2013), drought (Wu et al. 2016), floods (Smith et al. 2014), and sea level (Carson et al. 2016; McIntosh et al. 2015; Slangen et al. 2015).

China has already suffered from the environmental impacts of climate change through increased surface and ocean temperatures and sea level. Because of the effects of climate warming, the uneven distribution of water resources in China has been exacerbated in recent decades, leading to an increased occurrence of climate-related disasters such as floods in southern China. Several studies have indicated that increased extreme climate events and floods are very likely to appear in southern China throughout the twenty-first century (Gemmer et al. 2011; Fischer et al. 2012; Chen et al. 2013a; Wu et al. 2014b, 2015). Guangzhou, located in the estuary of the Pearl River, is the economic center of southern China. It has the largest number of heavy rain days in China. Some previous studies have observed that Guangzhou is facing a high waterlogging risk due to intense rainstorms and sea level rise (Chen et al. 2013b; Wu et al. 2014a). Zhao et al. (2014) analyzed the spatial and temporal changes in precipitation extremes over the Pearl River (Zhujiang) basin and found that the daily intensity shows a significant positive trend over the study area, including Guangzhou city, due to the impact of the Pacific decadal oscillation and Southern Oscillation index.

In view of potentially severe damages caused by urban waterlogging, the likelihood of such events occurring more frequently due to future climate change, along with an escalating magnitude and greater incurred losses, is thus an important research question to be addressed. On the basis of eight climate models with four emissions scenarios, the objectives of this paper are 1) to derive the bivariate distributions of extreme rainfall and tidal level and develop a conditional probability model in Guangzhou using the copula method; 2) to evaluate projected changes in temperature, extreme rainfall, sea level, and waterlogging probability (i.e., simultaneous probability and waterlogging probability) in Guangzhou under future climate change scenarios; and 3) to quantify the uncertainties driven by GCMs and emissions scenarios in predictions of waterlogging probability. The results will improve our understanding of the changing waterlogging risk under changing environmental conditions and will also help us cope with future natural hazards in Guangzhou.

2. Study area and data

a. Study area

Guangzhou (spanning from 22°26′ to 23°56′N and 112°57′ to 114°03′E) is the capital of Guangdong Province and is the third largest city in China with a total area of 7434.4 km2 (Fig. 1). The Pearl River, the third largest river in China, runs through Guangzhou and is navigable to the South China Sea. The climate of this region is characterized by a subtropical climate, with an annual average temperature of 21.9°C. The relative humidity is approximately 77%, whereas annual rainfall in the metropolitan area is over 1600 mm. Guangzhou has a lengthy monsoon season, from April to September. In recent years, with expansion of city size and urban population, waterlogging has become more frequent in Guangzhou, in which extreme rainfall and tidal lockup are the key factors of waterlogging (Wu et al. 2014a).

Fig. 1.
Fig. 1.

Location of Guangzhou city in southern China.

Citation: Journal of Hydrometeorology 18, 6; 10.1175/JHM-D-16-0157.1

b. Data

The data used consist primarily of temperature and precipitation data and climate model data. Long-term (1969–2011) observed daily temperature and precipitation data are available at Wu-shan Station located in the central urban area of Guangzhou. The data are provided by the China Meteorological Data Service Center, China Meteorological Administration (http://data.cma.cn). The tidal level data (1970–2010) are obtained from the Zhongda tide gauge station of the Pearl River, which is located in the Yuexiu district of Guangzhou. The data are provided by the Hydrology Bureau of Guangdong Province, China.

The GCMs provided by phases 3 and 5 of the Coupled Model Intercomparison Project (CMIP3 and CMIP5) have been widely used in the assessment of climate change (Harding et al. 2012; Smith et al. 2013). CMIP3 provides a basis for hundreds of peer-reviewed papers and plays a prominent role in the IPCC’s AR4 of climate change (Meehl et al. 2007), while CMIP5 provides a framework for coordinated climate change experiments that use new insights on the climate system and the processes responsible for climate change and variability (Taylor et al. 2012). In this study, the downscaling results (daily temperature and precipitation data) driven by 11 CMIP3 GCMs and 13 CMIP5 GCMs [Table 1 in Wu et al. (2014b)] were used. The simulation data include 1) a historical simulation for both CMIP3 and CMIP5 for the period 1970–2000, 2) the CMIP3 Special Report on Emissions Scenarios (SRES) scenario A1B for the future period 2020–50, and 3) the CMIP5 representative concentration pathway (RCP) scenarios RCP2.6, RCP4.5, and RCP8.5 for the future period 2020–50. An introduction to the downscaling method is described in detail in Wu et al. (2014b).

To enhance the credibility of model predictions, one regional climate model (RegCM4) and five individual CMIP5 GCMs (BCC_CSM1.1, CanESM2, CSIRO Mk3.6.0, GISS-E2-R, and MPI-ESM-LR) were also chosen for analysis. RegCM4 was driven by a historical run of BCC_CSM1.1, and experiments were conducted for both the RCP4.5 and RCP8.5 scenarios. RegCM4 data include one historical simulation for the historical period 1970–2000 and two emissions scenarios, RCP4.5 and RCP8.5, for the future period 2020–50. Five individual GCM outputs were downscaled to the study region using a statistical downscaling method described in Wu et al. (2015). The data are from one historical simulation (1970–2000) and three emissions scenarios, RCP2.6, RCP4.5, and RCP8.5, for 2020–50. To reduce the simulation system deviation, a delta method (Hay et al. 2000) was further used in the bias correction of these eight climate model simulations. Finally, all the bias-corrected data were interpolated to a high-resolution grid (i.e., 0.25° × 0.25°) of the study area using the bilinear interpolation method.

3. Methodology

a. Copula-based probability model

Copulas, developed by Sklar, are functions for describing the correlation of variables without limiting their marginal distributions (Nelsen 2006; Shiau et al. 2007; Sklar 1959). Using copulas, a number k of marginal distributions (in any form) can be combined to create a multivariate joint probability distribution model. Currently, there are three main families of copulas: elliptical, Archimedean, and quadratic. The Archimedean copulas are the most widely applied in hydrology (Favre et al. 2004; Ma et al. 2013).

Wu et al. (2014a) conducted a copula-based multivariate frequency analysis of extreme rainfall and tide level in the city of Guangzhou and found that Clayton and Ali–Mikhail–Haq (AMH) copulas show the best-fitting precision for the joint probability distributions of rainfall and tide level. On the basis of Wu et al. (2014a), both Clayton and AMH copulas are introduced to build rainfall and tide joint distributions in this study (Table 1). First, the Pearson type-III (P-III) distribution is used for fitting the marginal distribution of rainfall and tide level, respectively. The parameter of the P-III frequency distribution is estimated by using the linear moment (L-moment) method (Hosking and Wallis 2005). Then a nonparametric method (as shown in Table 1) is used for the estimation of the parameter θ of Clayton and AMH copulas. The Akaike information criterion (AIC) and the ordinary least squares (OLS) are used to evaluate the goodness of fit for different copulas to choose the best-fitting rainfall–tide joint distribution function (Wu et al. 2014a).

Table 1.

Two-dimensional copulas used in this study: θ is the copula parameter; τ is the Kendall rank correlation coefficient; and u1 and u2 represent marginal distribution functions of P and Z, respectively.

Table 1.

Figure 2 shows the comparison of empirical and theoretical accumulative probabilities of Clayton and AMH copulas in measuring the joint distribution of daily maximum rainfall and tidal level during the period of 1984–2010. As shown, generally both copulas showed a good agreement between the points of the empirical and theoretical frequencies. When comparing the goodness-of-fit tests of the copula parameter θ, the values of AIC and OLS for the AMH copula are smaller than those for the Clayton copula (Table 2), indicating that the AMH copula showed a better fit than the Clayton copula. Therefore, the AMH copula is used for analyzing the combination of rainfall and tide probability changes in the future period under different climate scenarios.

Fig. 2.
Fig. 2.

Comparison of empirical and theoretical accumulative probabilities for the (a) Clayton and (b) AMH copula.

Citation: Journal of Hydrometeorology 18, 6; 10.1175/JHM-D-16-0157.1

Table 2.

Results of the goodness-of-fit test for θ.

Table 2.

In this study, two main types of conditional probability are considered: 1) simultaneous probability, which means the probability that rainfall and tide level are simultaneously greater than a specific value of a certain return period, is shown in Eq. (1), and 2) waterlogging probability, which means the probability that rainfall or tide level is greater than a specific value of a certain return period, is shown in Eq. (2):
e1
e2
where P and Z represent precipitation and tide level, respectively, and and are the marginal distribution functions of P and Z, respectively.

b. Prediction of sea level rise

The IPCC provides a more reliable prediction for global temperature than for sea level rise, since our knowledge on the dynamics of ice sheets and glaciers, and to a lesser extent that of oceanic heat uptake, is still limited (Vermeer and Rahmstorf 2009). Rahmstorf (2007) developed a semiempirical model to connect global sea level rise to global mean surface temperature and proposed that the rate of sea level rise is generally proportional to the magnitude of temperature increases for time scales relevant to anthropogenic warming. Recently, Li et al. (2011) applied the Rahmstorf model to establish a sea level change prediction method for China under climate warming conditions and found that sea level rise is strongly correlated with global mean surface temperature, with a correlation coefficient of 0.94. The sea level rise in response to warming in China can be approximated by
e3
where is the rate of sea level rise and is the temperature change.

In this study, Eq. (3) was used for the prediction of sea level rise in Guangzhou. The sea level rise was then directly added to the historical tidal level to predict the future tidal level under different climate scenarios, by assuming that the impact of other factors (e.g., heavy rainfall intervals, storm surges, and river flooding) on the tidal level in Guangzhou does not change in the future.

4. Results

a. Trends in future temperature and precipitation

The Mann–Kendall (MK) trend test and the Sen’s slope (Mann 1945; Kendall 1975) were conducted to detect statistical significance of trends and estimate the trends’ magnitude in future temperature and precipitation of climate models. Figure 3 shows the uncertainty range of trend magnitudes in monthly temperature, monthly precipitation, and annual maximum 1D and 3D rainfall (AM1R and AM3R) for the period 2020–50 under all the emissions scenarios of the eight climate models. As shown, there is a large difference in predicted trends of temperature and precipitation from the different climate models and emissions scenarios. However, almost all the models predict a warming trend in all 12 months, and most of them pass the significance test at the 0.05 level (not shown). Particularly, the largest warming trend for the period 2020–50 is projected to occur in winter (from December to February). For monthly precipitation, the large uncertainty range of the trends’ magnitude is mainly found in May–August, in which the precipitation is likely to show increasing trends in 2020–50. However, most of the trends are not significant at the 0.05 level (not shown). Similar to monthly precipitation, the predicted trends of AM1R and AM3R show relatively large variability under different future scenarios in the eight climate models, and all the trends are not significant in the study area.

Fig. 3.
Fig. 3.

Uncertainty range of trend magnitudes in (a) monthly temperature, (b) monthly precipitation, and (c) AM1R and AM3R for the period 2020–50 under all the emissions scenarios of the eight climate models. The red dotted lines indicate the average values of trend magnitudes in monthly temperature and precipitation, and the blue dashed lines indicate 95% confidence interval for the average values of trend magnitudes.

Citation: Journal of Hydrometeorology 18, 6; 10.1175/JHM-D-16-0157.1

b. Projected changes in monthly temperature and precipitation

Figure 4 shows the predicted changes (i.e., the long-term mean monthly values for the period 2020–50 minus those for the period 1970–2000) in monthly temperature under the A1B, RCP2.6, RCP4.5, and RCP8.5 scenarios. As shown, almost all the models predict an increase in monthly temperature, but individual models and emissions scenarios differ significantly. This highlights the large uncertainty inherent in projections of climate change. In particular, the increase of temperature is relatively small under scenario A1B with the largest increase in December (<0.35°C). In the RCP scenarios, the most significant warming is found for RCP8.5. A significant warming is predicted by the models BCC_CSM1.1, CanESM2, CSIRO Mk3.6.0, and MPI-ESM-LR, in which the largest increases for the RCP2.6, RCP4.5, and RCP8.5 scenarios can be up to 2°, 2.3°, and 2.5°C, respectively. However, the CMIP5 ensemble mean and GISS-E2-R predict relatively small increases in temperature, with the maximum increase no more than 1.5°C. Overall, the average values of the projected changes for all the emissions scenarios of eight climate models are in the range of 1.04°–1.36°C (Fig. 4e). Note that the predicted increases for individual models are generally greater than those for the CMIP3 and CMIP5 ensemble means. The reason for this phenomenon is probably that the averaging of combined ensemble models reduces the maximum weight, which weakens the effects of extreme values (Wu et al. 2014b).

Fig. 4.
Fig. 4.

Predicted changes in monthly temperature under the (a) A1B, (b) RCP2.6, (c) RCP4.5, and (d) RCP8.5 scenario for the future period 2020–50 relative to the historical period 1970–2000. The projected changes are computed as the long-term mean monthly values for the period 2020–50 minus those for the period 1970–2000. (e) The average values of the projected changes for all the emissions scenarios of eight climate models (red dotted line) and the corresponding 95% confidence interval (blue dashed lines).

Citation: Journal of Hydrometeorology 18, 6; 10.1175/JHM-D-16-0157.1

Figure 5 shows the predicted percentage changes in monthly precipitation under the A1B, RCP2.6, RCP4.5, and RCP8.5 scenarios. Here the projected percentage changes represent the percentage ratio of the projected differences to the long-term mean monthly values for the period 1970–2000, where the projected differences are computed as the long-term mean monthly values for the period 2020–50 minus those for the period 1970–2000. As shown, a larger uncertainty is found for the projection of monthly precipitation than for monthly temperature. The A1B scenario of CMIP3 ensemble mean predicts an increase across all 12 months, especially for the flood season, in which the increases can be up to 30%. This scenario may cause an increased risk of flooding in Guangzhou during the flood season. In contrast, a large uncertainty in monthly precipitation exists for the RCP scenarios since different models show different results. Taking all the models as a whole, the monthly precipitation tends to increase in January, July, and November and decrease in February, March, October, and December (Fig. 5e).

Fig. 5.
Fig. 5.

Predicted changes (%) in monthly precipitation under the (a) A1B, (b) RCP2.6, (c) RCP4.5, and (d) RCP8.5 scenario for the future period 2020–50 relative to the historical period 1970–2000. The projected percentage changes represent the percentage ratio of the projected differences to the long-term mean monthly values for the period 1970–2000, where the projected differences are computed as the long-term mean monthly values for the period 2020–50 minus those for the period 1970–2000. (e) The average values of the projected percentage changes for all the emissions scenarios of eight climate models (red dotted line) and the corresponding 95% confidence interval (blue dashed lines).

Citation: Journal of Hydrometeorology 18, 6; 10.1175/JHM-D-16-0157.1

c. Projected changes in extreme precipitation

In general, extreme daily rainfall has a significant impact on flood control of Guangzhou (Chen et al. 2013b; Wu et al. 2014a). In this study, AM1R and AM3R were chosen for analysis of extreme precipitation projection. Three different return periods (i.e., 100, 50, and 20 years) were calculated using the P-III frequency distribution. As shown in Tables 3 and 4, although the projected ranges of AM1R and AM3R show relatively large variability with different future scenarios and models, most predict an increase in the future period (2020–50). The largest increases in AM1R and AM3R are found in the RCP4.5 scenario of CanESM2, which can be up to 148.3%. For AM1R, the largest decline is in the RCP2.6 scenario of CanESM2, while for AM3R the largest decline is in the RCP8.5 scenario of RegCM4. Taking all the models as a whole, the projected changes in AM1R for the 100-, 50-, and 20-yr return periods range from −16.1% to 111.2%, −13.0% to 101.5%, and −8.4% to 85.4%, respectively, and the corresponding average changes are 21.9%, 20.5%, and 18.3%, respectively. The 95% confidence intervals of the average changes in AM1R for the 100-, 50-, and 20-yr return periods are in the range of 9.51%–34.27%, 9.26%–31.72%, and 8.81%–27.68%, respectively. For AM3R, the projected changes in 100-, 50-, and 20-yr return periods range from −20.4% to 148.4%, −18.2% to 126.8%, and −14.6% to 94.4%, with the average changes being 18.8%, 17.1%, and 14.5%, respectively. The 95% confidence intervals of the average changes in AM3R for the 100-, 50-, and 20-yr return periods are in the range of 2.87%–34.66%, 3.36%–30.85%, and 3.95%–25.13%, respectively. The above analysis suggests that AM1R and AM3R will likely become more severe and frequent in Guangzhou during the future period of 2020–50.

Table 3.

Percentage changes (%) in AM1R for three return periods (100, 50, and 20 years) under different scenarios (relative to the historical 1970–2000). CI represents 95% confidence interval of the average changes.

Table 3.
Table 4.

Percentage changes (%) in AM3R for three return periods (100, 50, and 20 years) under different scenarios (relative to the historical 1970–2000). CI represents 95% confidence interval of the average changes.

Table 4.

d. Projected changes in sea level

The projected changes in temperature driven by eight climate models were used to estimate future sea level rise with Eq. (3). Table 5 shows the projected changes in the sea level for Guangzhou during the period 2020–50. As shown, there are wide variations in the projection of sea level due to different temperatures predicted by different climate models.

Table 5.

Predicted increases in sea level for the future period 2020–50 compared with the historical period 1970–2000. CI represents 95% confidence interval of the average changes.

Table 5.

Notably, the smallest changes in sea level are found in the A1B scenario of CMIP3 ensemble mean, with an increase of 11.40 cm, while the largest increases are projected by the RCP8.5 scenario of CSIRO Mk3.6.0 (~23.37 cm). Furthermore, larger increases are projected by higher emissions scenarios. Taking all the emissions scenarios as a whole, the sea level rise in the period 2020–50 can be in the range of 11.40–23.37 cm, with an average value of 18.43 cm, the standard deviation value of 3.25, and the skewness value of −0.40. The 95% confidence intervals of the average increases in sea level are from 16.95 to 19.91 cm.

e. Waterlogging probability due to future climate change

This section quantifies the waterlogging risk at Guangzhou due to future climate change on the basis of the AM1R as predicted by climate models. Annual maximum tidal level (AMTL) at the Zhongda station from 1970 to 2000 is assumed as the historical simulation of climate models, and the AMTL for the future period 2020–50 is estimated by combining the historical AMTL and predicted increase of sea level. On the basis of the AM1R and AMTL data of eight climate models, the marginal distributions of rainfall and tide level [i.e., and ] for the future period 2020–50 and the historical period 1970–2000 are estimated by using the P-III frequency distribution. The AMH copula is then used to build the joint probability distributions of rainfall and tide level for the future period 2020–50 under different emissions scenarios and for the historical period 1970–2000. For the given values of rainfall and tide level, the corresponding simultaneous probability and waterlogging probability for 2020–50 and 1970–2000 are calculated using Eqs. (1) and (2). The projected changes in simultaneous probability and waterlogging probability are computed as the value of probability for the future period 2020–50 minus that for the historical period 1970–2000.

1) Projected changes in simultaneous probability

Figure 6 shows the projected changes in the simultaneous probability with different return periods under the A1B, RCP2.6, RCP4.5, and RCP8.5 scenarios. As shown, although there is a large uncertainty from both climate models and emissions scenarios in the projection of simultaneous probability, it is projected to show an upward trend in all scenarios relative to the historical period 1970–2000. Furthermore, the simultaneous probability tends to increase with a decreasing return period. For return periods of rainfall and tidal level within 100 years, the simultaneous probability is projected to increase by 0.12%–4.07% in scenario A1B, from −0.07% to 4.96% in scenario RCP2.6, from 0.02% to 7.42% in scenario RCP4.5, and from −0.07% to 7.76% in scenario RCP8.5. Table 6 shows the future (2020–50) average changes in simultaneous probability for all climate models with four different emissions scenarios. Overall, an average increase in the simultaneous probability is found for all emissions scenarios, with the largest increase in scenario RCP4.5 and the smallest increase in scenario RCP2.6. Notably, the probability of rainfall in a 100-yr return period and tidal level in a 50-yr return period would increase by 0.24%, 0.15%, 0.36%, and 0.24% in the A1B, RCP2.6, RCP4.5, and RCP8.5 scenarios, respectively. This suggests that the simultaneous probability of extreme rainfall and high tidal level in Guangzhou will become more frequent in the future.

Fig. 6.
Fig. 6.

Contours showing the projected changes (%) in simultaneous probability for the future period 2020–50 relative to the historical period 1970–2000, for four different emissions scenarios.

Citation: Journal of Hydrometeorology 18, 6; 10.1175/JHM-D-16-0157.1

Table 6.

Average changes (%) in simultaneous probability for the future period 2020–50 relative to the historical period 1970–2000.

Table 6.

2) Projected changes in waterlogging probability

Contours of the projected changes in waterlogging probability for the future period 2020–50 using different climate models are shown in Fig. 7. Compared with Fig. 6, it can be seen that the waterlogging probability is much higher than the simultaneous probability. All models predict a rising trend in waterlogging probability. The changes in waterlogging probability within return periods of 100 years will be in the range of 7.98%–25.97% in scenario A1B, from −1.00% to 26.81% in scenario RCP2.6, from 1.49% to 27.99% in scenario RCP4.5, and from −0.71% to 25.33% in scenario RCP8.5. Table 7 shows the future (2020–50) average changes in waterlogging probability for four different emissions scenarios, which highlights the increasing trend of the simultaneous probability in the future period. The probability of rainfall in a 100-yr return period and tidal level in a 50-yr return period is likely to increase by 8.65%, 5.89%, 8.52%, and 6.81% in the A1B, RCP2.6, RCP4.5, and RCP8.5 scenarios, respectively.

Fig. 7.
Fig. 7.

Contours showing the projected changes (%) in waterlogging probability for the future period 2020–50 relative to the historical period 1970–2000 for four different emissions scenarios of eight climate models.

Citation: Journal of Hydrometeorology 18, 6; 10.1175/JHM-D-16-0157.1

Table 7.

Average changes (%) in waterlogging probability for the future period 2020–50 relative to the historical period 1970–2000.

Table 7.

5. Discussion

This study applied eight climate models and four emissions scenarios to discuss the uncertainty in predictions of future temperature, extreme rainfall, sea level, and waterlogging probability in Guangzhou city. The modeling results suggest an intensified hydrological cycle and a rising risk of waterlogging in Guangzhou with the predicted temperature increase in the future scenarios, which is consistent with many climate impact assessment studies worldwide, including central Asia (Panday et al. 2015), America (Wuebbles et al. 2014), China (Liu et al. 2012), Zhujiang River (Fischer et al. 2010), and Canada (Mladjic et al. 2011). However, assessments of extreme risk are not robust and depend on the climate models and emissions scenarios used, which highlight the uncertainty in climate change projections, especially for extreme precipitation. Some previous studies have indicated that the GCMs and emissions scenarios are a large source of uncertainty in predictions of extreme precipitation (Wu et al. 2014b, 2015). It should also be noted that, although there is a consistent future waterlogging probability for the study area with both CMIP3 and CMIP5 multimodel combinations, there are important differences between CMIP3 and CMIP5 (e.g., in monthly precipitation) due to distinct driving mechanisms between the models (Lutz et al. 2013; Yao et al. 2013; Wu et al. 2014b). In addition, although the outputs of climate models are downscaled and bias-corrected to high-resolution grids (~0.25°), the spatial resolution is still a bit coarse for the city of Guangzhou at a fine spatial scale. The finer-resolution climate data need to be developed for reducing the impact of uncertainty in the spatial resolution of climate model outputs in future work.

In this study, a semiempirical method, connecting sea level rise to surface temperature, was used for the prediction of sea level rise. Particularly, we found that the rises in sea level under the RCP8.5 and RCP4.5 scenarios of the CMIP5 ensemble mean are generally lower than the estimates of the process-model-based sea level rise reported in the AR5 (IPCC 2013) and Carson et al. (2016). One of the main reasons is that the CMIP5 ensemble mean in this study is driven by the averaging of combined ensemble models, which reduces the maximum weight and weakens the effects of extreme values. This would result in a lower temperature rise for the CMIP5 ensemble mean than for individual climate models (as shown in Table 5). Another major reason is probably due to the limitations of using semiempirical method, because some driving forces of regional sea level rise, such as ice melt, terrestrial water storage changes, and glacial isostatic adjustment (Carson et al. 2016), are neglected in this method. This would also lead to an underestimation of the sea level rise and the waterlogging risk in the study region.

In this study, the future tidal level in Guangzhou is predicted by directly combining historical tidal level and sea level increases, without considering the impact of other factors, such as heavy rainfall intervals, storm surges, and river flooding over the study area. These may contribute to further uncertainty in projection of the waterlogging risk. Moreover, other sources of uncertainty, such as statistical downscaling methods and parameter estimates in copulas, are not discussed in this study. Therefore, a thorough investigation of the uncertainty involved in the projection of waterlogging probability needs to be considered in future work.

6. Conclusions

This study applied a copula-based probability model to evaluate the impact of climate change on the waterlogging probability in Guangzhou using eight climate models with four emissions scenarios. The probability model combining simultaneous probability and waterlogging probability was developed using the Archimedean copulas. Uncertainty in the projection of future temperature, extreme rainfall, sea level, and the combined probability of extreme rainfall and tidal level in Guangzhou were discussed.

Results indicated a large uncertainty driven by climate models and emissions scenarios in the projection of climate change, and a larger uncertainty was found in the projection of precipitation than of temperature. For the period of 2020–50, significant warming trends were predicted by all the emissions scenarios of the climate models. Consistent with climate warming, the sea level tended to increase by between 11.40 and 23.37 cm during the period 2020–50 (compared to the historical period 1970–2000). Furthermore, larger increases were projected for higher emissions scenarios. Most models indicated an increase in extreme precipitation for the period of 2020–50, in which the projected changes in annual maximum 1-day rainfall with 100-, 50-, and 20-yr return periods ranged from −16.1% to 111.2%, −13.0% to 101.5%, and −8.4% to 85.4%, respectively.

Both simultaneous probability and waterlogging probability were projected to show an upward trend in the future period 2020–50 (relative to the historical period 1970–2000), with the largest increase in scenario RCP4.5 and the smallest increase in scenario RCP2.6. The average changes of the simultaneous probability for rainfall in a 100-yr return period and tidal level in a 50-yr return period will likely increase by 0.24%, 0.15%, 0.36%, and 0.24% in the A1B, RCP2.6, RCP4.5, and RCP8.5 scenarios, respectively. In contrast, the average changes of the waterlogging probability for rainfall in a 100-yr return period and tidal level in a 50-yr return period will increase by 8.65%, 5.89%, 8.52%, and 6.81% in the A1B, RCP2.6, RCP4.5, and RCP8.5 scenarios, respectively. Overall, the increased waterlogging probability suggests that Guangzhou will suffer more severe waterlogging in the future.

Acknowledgments

This study was supported by the Innovation Fund of Guangdong province water conservancy science and technology (2016-3) and the Innovation Fund of Guangzhou city water science and technology (GZSW-201401).

REFERENCES

  • Carson, M., A. Köhl, D. Stammer, A. B. A. Slangen, C. A. Katsman, R. S. W. van de Wal, J. Church, and N. White, 2016: Coastal sea level changes, observed and projected during the 20th and 21st century. Climatic Change, 134, 269281, doi:10.1007/s10584-015-1520-1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chen, H.-P., J.-Q. Sun, and C. Xiao-Li, 2013a: Future changes of drought and flood events in China under a global warming scenario. Atmos. Ocean. Sci. Lett., 6, 813.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chen, Q., C. Wu, G. Huang, and S. Wu, 2013b: Study on characteristics of short duration rainstorm in Guangzhou City (in Chinese). Water Res. Power, 31, 46.

    • Search Google Scholar
    • Export Citation
  • Cuo, L., T. K. Beyene, N. Voisin, F. Su, D. P. Lettenmaier, M. Alberti, and J. E. Richey, 2011: Effects of mid-twenty-first century climate and land cover change on the hydrology of the Puget Sound basin, Washington. Hydrol. Processes, 25, 17291753, doi:10.1002/hyp.7932.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Elshamy, M. E., I. A. Seierstad, and A. Sorteberg, 2009: Impacts of climate change on Blue Nile flows using bias-corrected GCM scenarios. Hydrol. Earth Syst. Sci., 13, 551565, doi:10.5194/hess-13-551-2009.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Favre, A.-C., S. El Adlouni, L. Perreault, N. Thiémonge, and B. Bobée, 2004: Multivariate hydrological frequency analysis using copulas. Water Resour. Res., 40, W01101, doi:10.1029/2003WR002456.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Fischer, T., M. Gemmer, L. Liu, and B. Su, 2010: Trends in monthly temperature and precipitation extremes in the Zhujiang River basin, south China (1961–2007). Adv. Clim. Change Res., 1, 6370, doi:10.3724/SP.J.1248.2010.00063.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Fischer, T., B. Su, Y. Luo, and T. Scholten, 2012: Probability distribution of precipitation extremes for weather index–based insurance in the Zhujiang River basin, south China. J. Hydrometeor., 13, 10231037, doi:10.1175/JHM-D-11-041.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gemmer, M., T. Fischer, T. Jiang, B. Su, and L. L. Liu, 2011: Trends in precipitation extremes in the Zhujiang River basin, south China. J. Climate, 24, 750761, doi:10.1175/2010JCLI3717.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Harding, B. L., A. W. Wood, and J. R. Prairie, 2012: The implications of climate change scenario selection for future streamflow projection in the upper Colorado River basin. Hydrol. Earth Syst. Sci., 16, 39894007, doi:10.5194/hess-16-3989-2012.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hay, L. E., R. L. Wilby, and G. H. Leavesley, 2000: A comparison of delta change and downscaled GCM scenarios for three mountainous basins in the United States. J. Amer. Water Resour. Assoc., 36, 387397, doi:10.1111/j.1752-1688.2000.tb04276.x.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hosking, J. R. M., and J. R. Wallis, 2005: Regional Frequency Analysis: An Approach Based on L-Moments. Cambridge University Press, 244 pp.

  • IPCC, 2007: Summary for policymakers. Climate Change 2007: The Physical Science Basis, S. Solomon et al., Eds., Cambridge University Press, 1–18.

    • Search Google Scholar
    • Export Citation
  • IPCC, 2013: Summary for policymakers. Climate Change 2013: The Physical Science Basis, T. F. Stocker et al., Eds., Cambridge University Press, 3–29.

  • Kay, A. L., and D. A. Jones, 2012: Transient changes in flood frequency and timing in Britain under potential projections of climate change. Int. J. Climatol., 32, 489502, doi:10.1002/joc.2288.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kendall, M. G., 1975. Rank Correlation Methods. 4th ed. Charles Griffin, 272 pp.

  • Kendon, E. J., D. P. Rowell, R. G. Jones, and E. Buonomo, 2008: Robustness of future changes in local precipitation extremes. J. Climate, 21, 42804297, doi:10.1175/2008JCLI2082.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Li, J., Y. D. Chen, L. Zhang, Q. Zhang, and F. H. S. Chiew, 2016: Future changes in floods and water availability across China: Linkage with changing climate and uncertainties. J. Hydrometeor., 17, 12951314, doi:10.1175/JHM-D-15-0074.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Li, X., J. L. Zhang, and Z. G. Gao, 2011: Discussion on semiempirical prediction method for sea level change of China. Mar. Sci. Bull., 30, 540543.

    • Search Google Scholar
    • Export Citation
  • Liu, L.-L., T. Jiang, J.-G. Xu, J. Q. Zhai, and Y. Luo, 2012: Responses of hydrological processes to climate change in the Zhujiang River basin in the 21st century. Adv. Climate Change Res., 3, 8491, doi:10.3724/SP.J.1248.2012.00084.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lutz, A. F., W. W. Immerzeel, A. Gobiet, F. Pellicciotti, and M. F. P. Bierkens, 2013: Comparison of climate change signals in CMIP3 and CMIP5 multi-model ensembles and implications for central Asian glaciers. Hydrol. Earth Syst. Sci., 17, 36613677, doi:10.5194/hess-17-3661-2013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ma, M., S. Song, L. Ren, S. Jiang, and J. Song, 2013: Multivariate drought characteristics using trivariate Gaussian and Student t copulas. Hydrol. Processes, 27, 11751190, doi:10.1002/hyp.8432.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mann, H. B., 1945: Non-parametric tests against trend. Econometrica, 13, 245259, doi:10.2307/1907187.

  • McIntosh, P. C., J. A. Church, E. R. Miles, K. Ridgway, and C. M. Spillman, 2015: Seasonal coastal sea level prediction using a dynamical model. Geophys. Res. Lett., 42, 67476753, doi:10.1002/2015GL065091.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Meehl, G. A., C. Covey, K. E. Taylor, T. Delworth, R. J. Stouffer, M. Latif, B. McAvaney, and J. B. Mitchell, 2007: The WCRP CMIP3 multimodel dataset: A new era in climate change research. Bull. Amer. Meteor. Soc., 88, 13831394, doi:10.1175/BAMS-88-9-1383.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Min, S.-K., X. Zhang, F. Zwiers, H. Shiogama, Y.-S. Tung, and M. Wehner, 2013: Multimodel detection and attribution of extreme temperature changes. J. Climate, 26, 74307451, doi:10.1175/JCLI-D-12-00551.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mladjic, B., L. Sushama, M. N. Khaliq, R. Laprise, D. Caya, and R. Roy, 2011: Canadian RCM projected changes to extreme precipitation characteristics over Canada. J. Climate, 24, 25652584, doi:10.1175/2010JCLI3937.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Nelsen, R. B., 2006: An Introduction to Copulas. 2nd ed. Springer, 272 pp.

  • O’Gorman, P. A., and T. Schneider, 2009: Scaling of precipitation extremes over a wide range of climates simulated with an idealized GCM. J. Climate, 22, 56765685, doi:10.1175/2009JCLI2701.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Panday, P. K., J. Thibeault, and K. E. Frey, 2015: Changing temperature and precipitation extremes in the Hindu Kush–Himalayan region: An analysis of CMIP3 and CMIP5 simulations and projections. Int. J. Climatol., 35, 30583077, doi:10.1002/joc.4192.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Petrow, T., and B. Merz, 2009: Trends in flood magnitude, frequency and seasonality in Germany in the period 1951–2002. J. Hydrol., 371, 129141, doi:10.1016/j.jhydrol.2009.03.024.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Raff, D. A., T. Pruitt, and L. D. Brekke, 2009: A framework for assessing flood frequency based on climate projection information. Hydrol. Earth Syst. Sci., 13, 21192136, doi:10.5194/hess-13-2119-2009.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rahmstorf, S., 2007: A semi-empirical approach to projecting future sea-level rise. Science, 315, 368370, doi:10.1126/science.1135456.

  • Sansom, J., and J. A. Renwick, 2007: Climate change scenarios for New Zealand rainfall. J. Appl. Meteor. Climatol., 46, 573590, doi:10.1175/JAM2491.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Shiau, J.-T., S. Feng, and S. Nadarajah, 2007: Assessment of hydrological droughts for the Yellow River, China, using copulas. Hydrol. Processes, 21, 21572163, doi:10.1002/hyp.6400.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sklar, A., 1959: Fonctions de répartition à n dimensions et leurs marges. Publ. Inst. Stat. Univ. Paris, 8, 229231.

  • Slangen, A. B. A., J. A. Church, X. Zhang, and D. P. Monselesan, 2015: The sea level response to external forcings in historical simulations of CMIP5 climate models. J. Climate, 28, 85218539, doi:10.1175/JCLI-D-15-0376.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Smith, A., P. Bates, J. Freer, and F. Wetterhall, 2014: Investigating the application of climate models in flood projection across the UK. Hydrol. Processes, 28, 28102823, doi:10.1002/hyp.9815.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Smith, I., J. Syktus, C. McAlpine, and K. Wong, 2013: Squeezing information from regional climate change projections—Results from a synthesis of CMIP5 results for south-east Queensland, Australia. Climatic Change, 121, 609619, doi:10.1007/s10584-013-0956-4.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Taylor, K. E., R. J. Stouffer, and G. A. Meehl, 2012: An overview of CMIP5 and the experiment design. Bull. Amer. Meteor. Soc., 93, 485498, doi:10.1175/BAMS-D-11-00094.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Vermeer, M., and S. Rahmstorf, 2009: Global sea level linked to global temperature. Proc. Natl. Acad. Sci. USA, 106, 21 52721 532, doi:10.1073/pnas.0907765106.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wu, C., G. Huang, and S. Wu, 2014a: Risk analysis of combinations of short duration rainstorm and tidal level in Guangzhou based on Copula function (in Chinese). J. Hydraul. Eng., 33, 3340.

    • Search Google Scholar
    • Export Citation
  • Wu, C., G. Huang, H. Yu, Z. Chen, and J. Ma, 2014b: Impact of climate change on reservoir flood control in the upstream area of the Beijiang River basin, south China. J. Hydrometeor., 15, 22032218, doi:10.1175/JHM-D-13-0181.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wu, C., G. Huang, and H. Yu, 2015: Prediction of extreme floods based on CMIP5 climate models: A case study in the Beijiang River basin, south China. Hydrol. Earth Syst. Sci., 19, 13851399, doi:10.5194/hess-19-1385-2015.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wu, C., Z. Xian, and G. Huang, 2016: Meteorological drought in the Beijiang River basin, south China: Current observations and future projections. Stochastic Environ. Res. Risk Assess., 30, 18211834, doi:10.1007/s00477-015-1157-7.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wuebbles, D., and et al. , 2014: CMIP5 climate model analyses: Climate extremes in the United States. Bull. Amer. Meteor. Soc., 95, 571583, doi:10.1175/BAMS-D-12-00172.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yang, T., Q. Shao, Z.-C. Hao, X. Chen, Z. Zhang, C.-Y. Xu, and L. Sun, 2010: Regional frequency analysis and spatio-temporal pattern characterization of rainfall extremes in the Pearl River basin, China. J. Hydrol., 380, 386405, doi:10.1016/j.jhydrol.2009.11.013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yao, Y., Y. Luo, J. Huang, and Z. Zhao, 2013: Comparison of monthly temperature extremes simulated by CMIP3 and CMIP5 models. J. Climate, 26, 76927707, doi:10.1175/JCLI-D-12-00560.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, D.-L., Y. Lin, P. Zhao, X. Yu, S. Wang, H. Kang, and Y. Ding, 2013: The Beijing extreme rainfall of 21 July 2012: “Right results” but for wrong reasons. Geophys. Res. Lett., 40, 14261431, doi:10.1002/grl.50304.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, H., K. Fraedrich, R. Blender, and X. Zhu, 2013: Precipitation extremes in CMIP5 simulations on different time scales. J. Hydrometeor., 14, 923928, doi:10.1175/JHM-D-12-0181.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhao, Y., X. Zou, L. Cao, and X. Xu, 2014: Changes in precipitation extremes over the Pearl River basin, southern China, during 1960–2012. Quat. Int., 333, 2639, doi:10.1016/j.quaint.2014.03.060.

    • Crossref
    • Search Google Scholar
    • Export Citation
Save