• Bastola, S., C. Murphy, and R. Fealy, 2012: Generating probabilistic estimates of hydrological response for Irish catchments using a weather generator and probabilistic climate change scenarios. Hydrol. Processes, 26, 23072321, doi:10.1002/hyp.8349.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • BoM, 1989: Climate of Australia. Bureau of Meteorology, 49 pp.

  • BoM, 2016: Climate data online. Accessed 22 April 2015. [Available online at www.bom.gov.au/climate/data/.]

  • Boughton, W. C., 1993: A hydrograph-based model for estimating the water yield of ungauged catchments. Proceedings of the 1993 Hydrology and Water Resources Conference. Institution of Engineers, National Conference Publication 93/14, 317–324.

  • Boughton, W. C., 2004: The Australian water balance model. Environ. Modell. Software, 19, 943956, doi:10.1016/j.envsoft.2003.10.007.

  • Boughton, W. C., and D. Carroll, 1993: A simple combined water balance/flood hydrograph model. Proceedings of the 1993 Hydrology and Water Resources Conference. Institution of Engineers, National Conference Publication 93/14, 299–304.

  • Chambers, L. E., 2001: Classifying rainfall districts: A southwestern Australian study. Aust. Meteor. Mag., 50, 91103.

  • Chen, J., and F. P. Brissette, 2014: Comparison of five stochastic weather generators in simulating daily precipitation and temperature for the Loess Plateau of China. Int. J. Climatol., 34, 30893105, doi:10.1002/joc.3896.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chen, J., and F. P. Brissette, 2015: Combining stochastic weather generation and ensemble weather forecasts for short-term streamflow prediction. Water Resour. Manage., 29, 33293342, doi:10.1007/s11269-015-1001-3.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chen, J., F. P. Brissette, R. Leconte, and A. Caron, 2012: A versatile weather generator for daily precipitation and temperature. Trans. ASABE, 55, 895906, doi:10.13031/2013.41522.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chen, J., F. P. Brissette, and X. J. Zhang, 2016: Hydrological modeling using a multisite stochastic weather generator. J. Hydrol. Eng., 21, 04015060, doi:10.1061/(ASCE)HE.1943-5584.0001288.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chiew, F. H. S., M. Peel, and A. Western, 2002: Application and testing of the simple rainfall–runoff model SimHyd. Mathematical Models of Small Watershed Hydrology and Applications, V. Singh and D. Frevert, Eds., Water Resources Publications, 335–367.

  • Chiew, F. H. S., J. Vaze, N. R. Viney, M. Perraud, J. Teng, and D. A. Post, 2008: Modelling runoff and climate change impact on runoff in 178 catchments in the Murray–Darling basin using Sacramento and SIMHYD rainfall–runoff models. Proceedings of Water Down Under 2008, M. Lambert, T. M. Daniell, and M. Leonard, Eds., Causal Productions, 87–97.

  • Cornish, P., 1977: Changes in seasonal and annual rainfall in New South Wales. Search, 8, 3840.

  • Dubrovský, M., J. Buchtele, and Z. Žalud, 2004: High-frequency and low-frequency variability in stochastic daily weather generator and its effect on agricultural and hydrologic modelling. Climatic Change, 63, 145179, doi:10.1023/B:CLIM.0000018504.99914.60.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Fan, J., C. Yang, C. Liu, and H. Huang, 2013: Assessment and validation of CLIGEN-simulated rainfall data for northern Taiwan. Paddy Water Environ., 11, 161173, doi:10.1007/s10333-011-0301-3.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Favis-Mortlock, D., and M. Savabi, 1996: Shifts in rates and spatial distributions of soil erosion and deposition under climate change. Advances in Hillslope Processes, Vol. 1, M. Anderson and S. Brooks, Eds., Wiley, 529–560.

  • Feng, J., J. Li, and Y. Li, 2010: A monsoon-like southwest Australian circulation and its relation with rainfall in southwest Western Australia. J. Climate, 23, 13341353, doi:10.1175/2009JCLI2837.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Foster, G. R., 1982: Modeling the erosion process. Hydrologic Modeling of Small Watersheds, ASAE Monograph, No. 5, American Society of Agricultural Engineers, 297–360.

  • Ghidey, F., and E. E. Alberts, 1996: Comparison of measured and WEPP predicted runoff and soil loss for Midwest claypan soil. Trans. ASAE, 39, 13951402, doi:10.13031/2013.27631.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Goodess, C., and J. Palutikof, 1998: Development of daily rainfall scenarios for southeast Spain using a circulation-type approach to downscaling. Int. J. Climatol., 18, 10511083, doi:10.1002/(SICI)1097-0088(199808)18:10<1051::AID-JOC304>3.0.CO;2-1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • IPCC, 2013: Climate Change 2013: The Physical Science Basis. Cambridge University Press, 1535 pp., doi:10.1017/CBO9781107415324.

    • Crossref
    • Export Citation
  • Johnson, G. L., C. L. Hanson, S. P. Hardegree, and E. B. Ballard, 1996: Stochastic weather simulation: Overview and analysis of two commonly used models. J. Appl. Meteor. Climatol., 35, 18781896, doi:10.1175/1520-0450(1996)035<1878:SWSOAA>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Jones, R., F. H. Chiew, F. Boughton, and L. Zhang, 2006: Estimating the sensitivity of mean annual runoff to climate change using selected hydrological models. Adv. Water Resour., 29, 14191429, doi:10.1016/j.advwatres.2005.11.001.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kilsby, C. P., and Coauthors, 2007: A daily weather generator for use in climate change studies. Environ. Modell. Software, 22, 17051719, doi:10.1016/j.envsoft.2007.02.005.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kou, X., J. Ge, Y. Wang, and C. Zhang, 2007: Validation of the weather generator CLIGEN with daily precipitation data from the Loess Plateau, China. J. Hydrol., 347, 347357, doi:10.1016/j.jhydrol.2007.09.051.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lavery, B., A. Kariko, and N. Nicholls, 1992: A high-quality historical rainfall data set for Australia. Aust. Meteor. Mag., 40, 3339.

  • Lavery, B., G. Joung, and N. Nicholls, 1997: An extended high-quality historical rainfall data set for Australia. Aust. Meteor. Mag., 46, 2738.

    • Search Google Scholar
    • Export Citation
  • Li, F., L. E. Chambers, and N. Nicholls, 2005: Relationships between rainfall in the southwest of Western Australia and near-global patterns of sea-surface temperature and mean sea-level pressure variability. Aust. Meteor. Mag., 54, 2333.

    • Search Google Scholar
    • Export Citation
  • Li, Y., W. Cai, and E. Campbell, 2005: Statistical modeling of extreme rainfall in southwest Western Australia. J. Climate, 18, 852863, doi:10.1175/JCLI-3296.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Li, Z., F. P. Brissette, and J. Chen, 2013: Finding the most appropriate precipitation probability distribution for stochastic weather generation and hydrological modelling in Nordic watersheds. Hydrol. Processes, 27, 37183729, doi:10.1002/hyp.9499.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Li, Z., F. P. Brissette, and J. Chen, 2014: Assessing the applicability of six precipitation probability distribution models on the Loess Plateau of China. Int. J. Climatol., 34, 462471, doi:10.1002/joc.3699.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lobo, G., J. Frankenbeger, D. Flanagan, and C. Bonilla, 2015: Evaluation and improvement of The CLIGEN model for storm and rainfall erosivity generation in central Chile. Catena, 127, 206213, doi:10.1016/j.catena.2015.01.002.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mavromatis, T., and J. Hansen, 2001: Interannual variability characteristics and simulated crop response of four stochastic weather generators. Agric. For. Meteor., 109, 283296, doi:10.1016/S0168-1923(01)00272-6.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • McFarlane, D., R. Stone, S. Martens, J. Thomas, R. Silberstein, R. Ali, and G. Hodgson, 2012: Climate change impacts on water yields and demands in southwestern Australia. J. Hydrol., 475, 488498, doi:10.1016/j.jhydrol.2012.05.038.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Min, Y., J. Kryjov, K. An, S. Hameed, S. Sohn, W. Lee, and J. Oh, 2011: Evaluation of the weather generator CLIGEN with daily precipitation characteristics in Korea. Asia-Pac. J. Atmos. Sci., 47, 255263, doi:10.1007/s13143-011-0014-y.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Nash, J. E., and J. V. Sutcliffe, 1970: River flow forecasting through conceptual models. Part 1: A discussion of principles. J. Hydrol., 10, 282290, doi:10.1016/0022-1694(70)90255-6.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Nearing, M. A., G. A. Foster, L. J. Lane, and S. C. Finkner, 1989: A process-based soil erosion model for USDA-Water Erosion Prediction Project technology. Trans. ASAE, 32, 15871593, doi:10.13031/2013.31195.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Nicholls, N., and B. Lavery, 1992: Australian rainfall trends during the twentieth century. Int. J. Climatol., 12, 153163, doi:10.1002/joc.3370120204.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Nicholls, N., and A. Kariko, 1993: East Australian rainfall events: Interannual variations, trends, and relationships with the Southern Oscillation. J. Climate, 6, 11411152, doi:10.1175/1520-0442(1993)006<1141:EAREIV>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Nicks, A., and G. Gardner, 1994: CLIGEN: A weather generator for climate inputs to water resource and other models. Proceedings of the 5th International Conference on Computers in Agriculture, D. G. Watson, F. S. Zazueta, and T. V. Harrison, Eds., American Society of Agricultural Engineers, 3–94.

  • Nicks, A., L. Lane, and G. Gardner, 1995: Chapter 2: Weather generator. In USDA Water Erosion Prediction Project: Hillslope Profile and Watershed Model Documentation. NSERL Rep. 10, USDA-ARS, 2.1–2.22.

  • Pittock, A., 1983: Climate change in Australia: Implications for a CO2-warmed Earth. Climatic Change, 5, 321340, doi:10.1007/BF02423529.

  • Podger, G., 2004: Description of models. Rainfall Runoff Library, G. Podger, Ed., Cooperative Research Centre for Catchment Hydrology, 47–64. [Available online at http://www.toolkit.net.au/rrl.]

  • Porter, J., and T. A. McMahon, 1975: Application of a catchment model in southeastern Australia. J. Hydrol., 24, 121134, doi:10.1016/0022-1694(75)90146-8.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Prudhomme, C., N. Reynard, and S. Crroks, 2002: Downscaling of global climate models for flood frequency analysis: Where are we now? Hydrol. Processes, 16, 11371150, doi:10.1002/hyp.1054.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Pruski, F., and M. A. Nearing, 2002a: Runoff and soil-loss responses to changes in precipitation: A computer simulation study. J. Soil Water Conserv., 57, 716.

    • Search Google Scholar
    • Export Citation
  • Pruski, F., and M. A. Nearing, 2002b: Climate-induced changes in erosion during the 21st century for eight U.S. locations. Water Resour. Res., 38, 1298, doi:10.1029/2001WR000493.

    • Crossref
    • Export Citation
  • Richardson, C., and D. Wright, 1984: WGEN: A model for generating daily weather variables. Rep. ARS-8, 86 pp.

  • Richardson, C., 1981: Stochastic simulation of daily precipitation, temperature, and solar radiation. Water Resour. Res., 17, 182190, doi:10.1029/WR017i001p00182.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Semenov, M. A., and J. Porter, 1995: Climatic variability and the modelling of crop yields. Agric. For. Meteor., 73, 265283, doi:10.1016/0168-1923(94)05078-K.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Semenov, M. A., and E. M. Barrow, 1997: Use of a stochastic weather generator in the development of climate change scenarios. Climatic Change, 35, 397414, doi:10.1023/A:1005342632279.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Semenov, M. A., and E. M. Barrow, 2002: LARS-WG: A stochastic weather generator for use in climate impact studies. User manual, version 3.0, 28 pp. [Available online at http://resources.rothamsted.ac.uk/sites/default/files/groups/mas-models/download/LARS-WG-Manual.pdf.]

  • Silberstein, R., and Coauthors, 2012: Climate change and runoff in southwestern Australia. J. Hydrol., 475, 441455, doi:10.1016/j.jhydrol.2012.02.009.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Smith, I., 2004: An assessment of recent trends in Australian rainfall. Aust. Meteor. Mag., 53, 163173.

  • Stöckle, C., G. Campbell, and R. Nelson, 1999: ClimGen manual. Washington State University.

  • Timbal, B., 2004: Southwest Australia past and future rainfall trends. Climate Res., 26, 233249, doi:10.3354/cr026233.

  • Tiwari, A., L. M. Risse, and M. A. Nearing, 2000: Evaluation of WEPP and its comparison with USLE and RUSLE. Trans. ASAE, 43, 11291135, doi:10.13031/2013.3005.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Vaghefi, P., and B. Yu, 2011: Use of CLIGEN to simulate climate change in southeastern Australia. Trans. ASABE, 54, 857867, doi:10.13031/2013.37111.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Vaghefi, P., and B. Yu, 2016: Use of CLIGEN to simulate decreasing precipitation trends in the southwest of Western Australia. Trans. ASABE, 59, 4961, doi:10.13031/trans.59.10829.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wang, Q., T. Pagano, S. Zhou, H. Hapuarachchi, L. Zhang, and D. Robertson, 2011: Monthly versus daily water balance models in simulating monthly runoff. J. Hydrol., 404, 166175, doi:10.1016/j.jhydrol.2011.04.027.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wheater, H. S., and Coauthors, 2005: Spatial–temporal rainfall modelling for flood risk estimation. Stochastic Environ. Res. Risk Assess., 19, 403416, doi:10.1007/s00477-005-0011-8.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wilby, R., and T. Wigley, 1997: Downscaling general circulation model output: A review of methods and limitations. Prog. Phys. Geog., 21, 530–548, doi:10.1177/030913339702100403.

    • Crossref
    • Export Citation
  • Wilby, R., T. Wigley, D. Conway, P. Jones, B. Hewitson, J. Main, and D. Wilks, 1998: Statistical downscaling of general circulation model output: A comparison of methods. Water Resour. Res., 34, 29953008, doi:10.1029/98WR02577.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wilks, D., 1992: Adapting stochastic weather generation algorithms for climate change studies. Climatic Change, 22, 6784, doi:10.1007/BF00143344.

  • Wilks, D., 1999: Multisite downscaling of daily precipitation with a stochastic weather generator. Climate Res., 11, 125136, doi:10.3354/cr011125.

  • Xu, C., 1999: From GCMs to river flow: A review of downscaling methods and hydrologic modelling approaches. Prog. Phys. Geogr., 23, 229249, doi:10.1191/030913399667424608.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yu, B., 2000: Improvement and evaluation of CLIGEN for storm generation. Trans. ASAE, 43, 301307, doi:10.13031/2013.2705.

  • Yu, B., 2002: Using CLIGEN to generate RUSLE climate inputs. Trans. ASAE, 45, 9031001, doi:10.13031/2013.9952.

  • Yu, B., 2003: An assessment of uncalibrated CLIGEN in Australia. Agric. For. Meteor., 119, 131148, doi:10.1016/S0168-1923(03)00141-2.

  • Yu, B., 2005: Adjustment of CLIGEN parameters to generate precipitation change scenarios in southeastern Australia. Catena, 61, 196209, doi:10.1016/j.catena.2005.03.004.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yu, B., and D. T. Neil, 1991: Global warming and regional rainfall: The difference between average and high-intensity rainfalls. Int. J. Climatol., 11, 653661, doi:10.1002/joc.3370110607.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yu, B., and Z. Zhu, 2015: A comparative assessment of AWBM and SimHyd for forested watersheds. Hydrol. Sci. J., 60, 12001212, doi:10.1080/02626667.2014.961924.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yu, B., C. A. Ciesiolka, C. W. Rose, and K. J. Coughlan, 2000: A validation test of WEPP to predict runoff and soil loss from a pineapple farm on a sandy soil in subtropical Queensland, Australia. Aust. J. Soil Res., 38, 537554, doi:10.1071/SR99104.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, X.-C., 2005a: Generating correlative storm variables for CLIGEN using a distribution-free approach. Trans. ASAE, 48, 567575, doi:10.13031/2013.18331.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, X.-C., 2005b: Spatial downscaling of global climate model output for site-specific assessment of crop production and soil erosion. Agric. For. Meteor., 135, 215229, doi:10.1016/j.agrformet.2005.11.016.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, X.-C., 2007: A comparison of explicit and implicit spatial downscaling of GCM output for soil erosion and crop production assessments. Climatic Change, 84, 337363, doi:10.1007/s10584-007-9256-1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, X.-C., and J. Garbrecht, 2003: Evaluation of CLIGEN precipitation parameters and their implication on WEPP runoff and erosion prediction. Trans. ASAE, 46, 311320, doi:10.13031/2013.12982.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, X.-C., and W.-Z. Liu, 2005: Simulating potential response of hydrology, soil erosion, and crop productivity to climate change in Changwu tableland region on the Loess Plateau of China. Agric. For. Meteor., 131, 127142, doi:10.1016/j.agrformet.2005.05.005.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, X.-C., M. A. Nearing, L. M. Risse, and K. C. McGregor, 1996: Evaluation of WEPP runoff and soil loss predictions using natural runoff plot data. Trans. ASAE, 39, 855863, doi:10.13031/2013.27570.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, Y., B. Liu, Z. Wang, and Q. Zhu, 2008: Evaluation of CLIGEN for storm generation on the semiarid Loess Plateau in China. Catena, 73, 19, doi:10.1016/j.catena.2007.08.001.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, Y.-G., M. A. Nearing, X.-C. Zhang, Y. Xie, and H. Wei, 2010: Projected rainfall erosivity changes under climate change from multimodel and multiscenario projections in northeast China. J. Hydrol., 384, 97106, doi:10.1016/j.jhydrol.2010.01.013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhuang, X. W., Y. P. Li, G. H. Huang, and J. Liu, 2016: Assessment of climate change impacts on watershed in cold-arid region: An integrated multi-GCM-based stochastic weather generator and stepwise cluster analysis method. Climate Dyn., 47, 191209, doi:10.1007/s00382-015-2831-7.

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

    Precipitation and streamflow sites in SEA.

  • View in gallery

    Precipitation and streamflow sites in SWWA.

  • View in gallery

    Flowchart on how methods for parameter adjustment were evaluated.

  • View in gallery

    Comparison of the observed mean annual precipitation against CLIGEN-generated data for SEA and SWWA. The period was 30 years for sites in SEA and 90 years in SWWA.

  • View in gallery

    Annual time series of simulated streamflow with AWBM based on the observed (i.e., A) and CLIGEN-generated (i.e., B) precipitation at Cataract Dam in SEA.

  • View in gallery

    Mean annual flows predicted using observed and CLIGEN-generated daily precipitation for the two contrasting periods with (a) AWBM and (b) SimHyd for the three sites in SEA.

  • View in gallery

    Annual time series of calculated streamflow by SimHyd based on the observed (i.e., A) and CLIGEN-generated (i.e., B) precipitation at Nannup in SWWA.

  • View in gallery

    Long-term (1923–2012) mean annual flows predicted using observed and CLIGEN-generated daily precipitation with AWBM and SimHyd for three sites in SWWA. The error bars represent one standard error of the mean.

  • View in gallery

    Ratios of calculated annual streamflow for two contrasting periods in SEA.

  • View in gallery

    Rate of changes in simulated flow per decade from 1923 to 2012 in SWWA.

  • View in gallery

    Monthly distribution of average flow and standard error of the mean at Coonabarabran for 30-yr period of 1919–48 in SEA.

  • View in gallery

    Monthly distribution of average flow and standard error of the mean at Coonabarabran for 30-yr period of 1949–78 in SEA.

  • View in gallery

    Monthly distribution of average flow and standard error of the mean at Manjimup for 90-yr period of 1923–2012 in SWWA.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 5 5 2
PDF Downloads 9 9 4

Validation of CLIGEN Parameter Adjustment Methods for Southeastern Australia and Southwestern Western Australia

View More View Less
  • 1 School of Engineering, Griffith University, Nathan, Queensland, Australia
© Get Permissions
Full access

Abstract

Global climate models (GCMs) are usually used for future climate projections. Model output from GCMs needs to be downscaled and stochastic weather generators such as Climate Generator (CLIGEN) are tools to downscale GCM output and to produce synthetic weather sequences that are statistically similar to the observed weather data. Two methods of adjusting CLIGEN parameters were developed to reproduce precipitation sequences for southeastern Australia (SEA), where significant changes in annual precipitation had occurred, and for southwestern Western Australia (SWWA), where the precipitation has shown a significant decreasing trend since the 1920s. The adjustment methods have been validated using observed precipitation data for these regions. However, CLIGEN outputs ultimately will be used as input to other simulation models. The objective of this research was to further validate the methods of CLIGEN parameter adjustment using conceptual hydrological models to simulate streamflow and to compare the streamflow using observed and CLIGEN-generated precipitation data. Six precipitation sites from SEA and SWWA were selected and synthetic time series of daily precipitation were generated for these sites. Conceptual hydrological models, namely, the Australian Water Balance Model and SimHyd, were used for flow simulation and were calibrated using recorded daily streamflow data from six gauging stations in SEA and SWWA. Both monthly and annual streamflow show statistically similar patterns using observed and CLIGEN-generated precipitation data. The adjustment methods for CLIGEN parameters are further validated and can be used to reproduce the significant changes, both abrupt and gradually decreasing, in streamflow for these two climatically contrasting regions of Australia.

© 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: Parshin Vaghefi, p.vaghefi@griffith.edu.au

Abstract

Global climate models (GCMs) are usually used for future climate projections. Model output from GCMs needs to be downscaled and stochastic weather generators such as Climate Generator (CLIGEN) are tools to downscale GCM output and to produce synthetic weather sequences that are statistically similar to the observed weather data. Two methods of adjusting CLIGEN parameters were developed to reproduce precipitation sequences for southeastern Australia (SEA), where significant changes in annual precipitation had occurred, and for southwestern Western Australia (SWWA), where the precipitation has shown a significant decreasing trend since the 1920s. The adjustment methods have been validated using observed precipitation data for these regions. However, CLIGEN outputs ultimately will be used as input to other simulation models. The objective of this research was to further validate the methods of CLIGEN parameter adjustment using conceptual hydrological models to simulate streamflow and to compare the streamflow using observed and CLIGEN-generated precipitation data. Six precipitation sites from SEA and SWWA were selected and synthetic time series of daily precipitation were generated for these sites. Conceptual hydrological models, namely, the Australian Water Balance Model and SimHyd, were used for flow simulation and were calibrated using recorded daily streamflow data from six gauging stations in SEA and SWWA. Both monthly and annual streamflow show statistically similar patterns using observed and CLIGEN-generated precipitation data. The adjustment methods for CLIGEN parameters are further validated and can be used to reproduce the significant changes, both abrupt and gradually decreasing, in streamflow for these two climatically contrasting regions of Australia.

© 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: Parshin Vaghefi, p.vaghefi@griffith.edu.au

1. Introduction

Hydrologic cycles and processes have been affected by climate change over recent decades. Future climates are usually projected using numerical models such as global climate models (GCMs; IPCC 2013). Model output from GCMs needs to be downscaled to suitable spatial and temporal resolutions to drive hydrologic and biomass production models because climate model output is typically a much coarser spatial resolution (Zhang 2005b, 2007). Stochastic weather generators (SWGs) are statistical models for simulating weather sequences, which are expected to be similar to observations statistically. These SWGs are usually used in combination with hydrologic and other environmental models for water resource and environmental management (Semenov and Porter 1995; Mavromatis and Hansen 2001; Wheater et al. 2005). SWGs have also been used as downscaling tools to produce high-resolution climate change projections by linking their parameters to climate model outputs (Semenov and Barrow 1997; Wilby and Wigley 1997; Goodess and Palutikof 1998; Wilby et al. 1998; Zhang and Garbrecht 2003; Yu 2005; Zhang and Liu 2005; Zhang 2005b, 2007; Chen and Brissette 2014, 2015).

Several SWGs have been developed over the past few decades, such as the Weather Generator (WGEN; Richardson 1981; Richardson and Wright 1984), the Climate Generator (CLIMGEN; Stöckle et al. 1999), the Climate Generator (CLIGEN; Nicks and Gardner 1994; Nicks et al. 1995), and the Long Ashton Research Station Weather Generator (LARS-WG) (Semenov and Barrow 2002). They have been widely used to simulate daily weather sequences for impact studies (Semenov and Barrow 1997; Wilks 1992, 1999; Zhang 2005b; Chen et al. 2012).

CLIGEN is designed to provide climate input to the Water Erosion Prediction Project (WEPP) and has been used for climate change impact studies (Nicks et al. 1995; Xu 1999; Favis-Mortlock and Savabi 1996; Prudhomme et al. 2002; Pruski and Nearing 2002a; Zhang 2005b; Yu 2005; Zhang et al. 2010). WEPP is a process-based daily runoff and erosion simulation model built on the fundamentals of hydrology, plant science, hydraulics, and erosion mechanics (Foster 1982; Nearing et al. 1989). For each day of a simulation period, 10 weather variables are generated to provide input to simulate daily runoff, biomass production, and soil losses with WEPP (Nicks et al. 1995; Yu 2000).

There were earlier attempts to validate CLIGEN in the United States and Australia (Johnson et al. 1996; Yu 2002, 2003; Zhang and Garbrecht 2003). More recently, Fan et al. (2013) tested CLIGEN to determine whether CLIGEN is able to produce daily precipitation for the subtropical monsoon region of northern Taiwan. They attempted to reproduce 30 years of daily precipitation using CLIGEN, and they needed to recalibrate the parameters to generate storm patterns to reduce the discrepancy between the observed and simulated rainfall to acceptable levels. Min et al. (2011) also assessed CLIGEN’s ability to reproduce 55 years of daily precipitation for eight sites on the Korean Peninsula. They found that CLIGEN reproduces most of the daily precipitation characteristics satisfactorily, but CLIGEN tends to underestimate the mean and variability of daily precipitation slightly. Lobo et al. (2015) evaluated CLIGEN storm durations for 30 sites in Chile and noted that the storm duration was consistently overestimated, while the maximum intensity was underestimated, resulting in reduced rainfall erosivity for these sites.

Precipitation has decreased in southeastern Australia (SEA) since the 1970s, but the level of significance associated with this decrease has not yet been quantified. On the other hand, the abrupt increase of the annual precipitation in SEA for the three decades since the late 1940s is well known and widely documented (Cornish 1977; Pittock 1983; Yu and Neil 1991; Nicholls and Kariko 1993). In contrast, southwestern Western Australia (SWWA) has experienced steady and significant decline in annual precipitation since the 1920s (Nicholls and Lavery 1992; Chambers 2001; Timbal 2004; Smith 2004; F. Li et al. 2005; Y. Li et al. 2005; Feng et al. 2010; Silberstein et al. 2012; McFarlane et al. 2012).

While these secular variations and changes in precipitation during the last 100 years in Australia are well documented (Cornish 1977; Pittock 1983; Yu and Neil 1991; Nicholls and Kariko 1993; Nicholls and Lavery 1992; Chambers 2001; Timbal 2004; Smith 2004; F. Li et al. 2005; Y. Li et al. 2005; Feng et al. 2010; Silberstein et al. 2012; McFarlane et al. 2012), methods to adjust parameter values for CLIGEN to simulate these observed climate variations and changes have not been established. To represent climate change scenarios, the input parameter values for CLIGEN are usually modified (Pruski and Nearing 2002a; Zhang 2005a). In particular, mean precipitation amounts on wet days are altered to simulate the likely change in precipitation predicted by GCMs (Pruski and Nearing 2002b; Zhang 2005a). Zhang (2005b) used an empirical relationship between the mean monthly precipitation and transitional probabilities to estimate the likely changes in the number of wet days and used an analytical expression to adjust the standard deviation of daily precipitation. However, previous studies have not attempted to use CLIGEN to simulate precipitation series where observed precipitation has significantly changed over time (Zhang and Garbrecht 2003; Yu 2005).

To use CLIGEN to simulate the observed significant changes in precipitation for these two regions of Australia, Vaghefi and Yu (2011) developed a method based on the changes in mean monthly precipitation to adjust CLIGEN parameters for SEA where annual precipitation has abruptly changed among three contrasting 30-yr periods between 1919 and 2008. They have also developed a site-specific method to adjust CLIGEN parameters for SWWA where annual precipitation amount has shown a declining trend since the 1920s (Vaghefi and Yu 2016).

When validating the generated daily precipitation data using CLIGEN, typically simulated and observed precipitation data were compared in terms of the mean, standard deviation, and their seasonal variations for various regions of the world (Yu 2000, 2003, 2005; Dubrovský et al. 2004; Kilsby et al. 2007; Kou et al. 2007; Zhang et al. 2008; Bastola et al. 2012; Min et al. 2011; Chen et al. 2012, 2016; Fan et al. 2013; Chen and Brissette 2014, 2015; Lobo et al. 2015; Zhuang et al. 2016). While definitely needed, validated SWGs may not be adequate in terms of the output from other applications to which CLIGEN provides the required weather data. Thus, it is highly pertinent to assess the quality of SWGs in terms of simulated flow when CLIGEN is combined with hydrologic models because, ultimately, CLIGEN-generated precipitation data need to be used as input to other programs, tools, and models. Although some studies evaluated flow generated based on precipitation generated by CLIGEN (Ghidey and Alberts 1996; Zhang et al. 1996; Tiwari et al. 2000; Yu et al. 2000; Li et al. 2013, 2014), previous studies essentially assumed that the daily precipitation generated using CLIGEN could be used in other applications so long as the generated precipitation was statistically similar to observations. Li et al. (2014) assessed the applicability of six precipitation probability distribution models in China and, while they found that the skewed normal distribution used in CLIGEN is the best among the six models at reproducing extreme precipitation events, they suggested further investigation would be required to assess the ability of these distribution models to simulate daily flow. Li et al. (2013) also assessed the ability of these probability distribution models to simulate daily flow for 24 catchments in Canada. They found that three-parameter probability distribution models (such as the skewed normal distribution) perform better than the other distributions for simulating precipitation and discharge. However, to our knowledge, CLIGEN has never been evaluated in terms of simulated daily flow when CLIGEN parameters were adjusted to represent climate variation and change. In other words, little is known about how flow estimated with observed precipitation compares with that estimated with CLIGEN-generated precipitation for regions where precipitation has significantly changed.

Therefore, the objectives of the paper are to evaluate the quality of precipitation data generated by CLIGEN in terms of flow estimated using observed and CLIGEN-generated daily precipitation data and to further assess methods to adjust CLIGEN parameters to represent significant changes in precipitation in regions where precipitation is known to have changed significantly on the time scale of climatology, that is, 30 years. This was achieved by applying two conceptual hydrologic models, namely, the Australian Water Balance Model (AWBM) and SimHyd, to selected sites in SEA and SWWA where methods to adjust CLIGEN parameters have been developed (Vaghefi and Yu 2011, 2016).

2. Methodology and data selection

a. Site selection

Two regions of Australia were the focus of this study. SEA is a region that has experienced abrupt changes in annual precipitation during the last 100 years. The other region is SWWA, where annual precipitation has decreased steadily since the 1920s. To evaluate parameter adjustment methods proposed for CLIGEN in these two regions, three precipitation sites from each of the two regions were selected from the lists of high-quality precipitation sites in Australia (Lavery et al. 1992, 1997).

The sites in SEA are located in New South Wales (NSW) and Australian Capital Territory (ACT). The spatial extent was defined by 31°16′16″–34°16′52″S and 149°16′17″–151°47′53″E. The region has a temperate climate with essentially uniform monthly precipitation throughout the year (BoM 1989). Among the 30 sites that were considered in Vaghefi and Yu (2011), two sites with the most and the least significant changes in annual precipitation (Cataract Dam and Clarence Town, respectively) have been selected for validation purposes. As these two sites are considered to be coastal (within 18 km from the coastline), a third site from inland of NSW (Coonabarabran) was also selected (>300 km from the coastline).

Lavery et al. (1992, 1997) identified 379 sites in Australia where the precipitation data are of high quality. There were seven such high-quality sites in SWWA that were previously used to develop the CLIGEN parameter adjustment method (Vaghefi and Yu 2016). Among them, three sites (namely, Manjimup, Wilgarrup, and Nannup) were selected based on their geographical location and proximity to high-quality streamflow sites nearby in SWWA. The spatial extent of a previous study (Vaghefi and Yu 2016) was from 33°58′47″ to 34°09′07″S and from 115°45′56″ to 116°12′15″E. The region has a temperate climate with a marked wet season in winter (BoM 1989). Table 1 and Figs. 1 and 2 show the six selected sites for this study, and Table 1 presents site information on these sites.

Table 1.

Selected precipitation sites from SEA and SWWA.

Table 1.
Fig. 1.
Fig. 1.

Precipitation and streamflow sites in SEA.

Citation: Journal of Hydrometeorology 18, 7; 10.1175/JHM-D-16-0237.1

Fig. 2.
Fig. 2.

Precipitation and streamflow sites in SWWA.

Citation: Journal of Hydrometeorology 18, 7; 10.1175/JHM-D-16-0237.1

To validate CLIGEN parameter adjustment methods, streamflow data were required to calibrate hydrologic models and to estimate parameter values for these models. The estimated parameter values, coupled with hydrologic models, were assumed to represent the rainfall–runoff relationship for the selected regions to evaluate daily flow simulated using CLIGEN-generated precipitation data. Recorded precipitation data are point based, while streamflow data were recorded at the gauging station for a catchment over which precipitation is inherently nonuniform in space. Ideally, a streamflow gauge for a small catchment with essentially uniform precipitation would be the best candidate to be associated with a single precipitation station in that catchment. However, a streamflow site that meets this requirement is unavailable for the two regions considered in this paper. Therefore, in this research streamflow sites were selected according to the proximity between the location of the gauging station and selected precipitation sites, followed by the catchment area (smaller being more suitable). Selected streamflow sites for SEA and SWWA are presented in Table 2 and Figs. 1 and 2.

Table 2.

Selected streamflow sites SEA and SWWA.

Table 2.

b. Methodology

Daily precipitation retrieved from BoM (2016) was used to calculate CLIGEN parameters related to daily precipitation using the standard methods described in detail elsewhere (Vaghefi and Yu 2011, 2016), and the daily data were accumulated to monthly and annual amounts for the selected sites. These parameter values were adjusted using the methods developed by Vaghefi and Yu (2011, 2016) to simulate the observed changes in precipitation in SEA and SWWA. These adjustment methods are briefly described below.

Vaghefi and Yu (2011) suggested that changes to CLIGEN parameter values are related to the underlying changes to the mean monthly precipitation, and they applied this method for CLIGEN parameter adjustment in SEA. The annual precipitation totals were tested for significant differences in the means between two nonoverlapping 30-yr periods of 1919–48 and 1949–78 and two nonoverlapping 30-yr periods of 1949–78 and 1979–2008. A standard Student’s t test for two samples with equal variance was used for the contrasting periods and for all 30 sites. To test the adjustment method for SEA, changes in the mean monthly precipitation between period 1 (1919–48) and period 2 (1949–78) and regression equations for changes in CLIGEN parameter values were used to adjust CLIGEN input parameter values for period 1 to derive parameter values for period 2. The calculated and the adjusted parameter values were then used to simulate the climate for periods 1 and 2, respectively, with CLIGEN. The simulated precipitation changes between the first two periods were then compared with the changes in the observed precipitation. To determine how CLIGEN parameter values could be adjusted to represent observed precipitation patterns, Vaghefi and Yu (2011) calculated the ratio for precipitation-related CLIGEN parameter values for the three contrasting 30-yr periods for all stations. These ratios were then related to the ratios for the mean monthly precipitation for the same contrasting periods. Vaghefi and Yu (2011) found there are strong positive correlations between changes in the mean monthly precipitation and changes in mean daily precipitation, standard deviation of daily precipitation, and the probability of wet-following-dry sequences, while there is little evidence to suggest ways of adjusting the skewness coefficient or wet-following-wet probabilities to simulate changes in the mean monthly precipitation for this region. Vaghefi and Yu (2011) developed a set of regression equations to allow easy adjustment of CLIGEN parameter values to simulate monthly precipitation change for scenarios of increases or decreases in precipitation in SEA, and it is potentially applicable in any similar regions in the world. Changes in the mean monthly precipitation between 1919–48 and 1949–78 and regression equations for changes in CLIGEN parameter values were used to adjust CLIGEN input parameter values. The calculated and the adjusted parameter values were then used to simulate the weather sequences for these contrasting periods with CLIGEN.

In SWWA, trend analysis was undertaken by Vaghefi and Yu (2016) for each of the five precipitation-related CLIGEN parameters in addition to the monthly and annual precipitation amounts. The trend analysis was based on linear regression between monthly or annual precipitation amount and time. A slope of the time series that is significantly different from zero indicates a significant trend. The standard Student’s t test for two samples with equal variance was used to identify significance level of the changes in the monthly or annual precipitation for the seven selected sites. Relationship among daily, monthly, and annual precipitation amounts was developed to identify how CLIGEN parameter values could be adjusted to simulate the significant decreasing trends in precipitation. Regression equations were developed for the wet months using the rates of change using three regionalization methods. These methods, namely, the site-specific method, the average method using the average parameter values, and the Wilgarrup method using parameter values from the site with the most significant change in precipitation, were tested and evaluated. Based on the rates of change, regression equations were developed for the wet months as follows:
e1
In this equation, R is CLIGEN parameter’s slope of the trend line and Rm is the slope of the trend line for monthly precipitation (mm month−1 yr−1).

Once the values for α and β are known using observed monthly and daily precipitation data, CLIGEN parameters can be adjusted from the rate of change in monthly precipitation using Eq. (1). Each site would have a regression equation for each CLIGEN parameter. Because of the large number of equations, three regionalization methods were evaluated. These methods were the “site specific” method, which simply uses the α and β values for each CLIGEN parameter and for each site; the “average” method, which is based on the average values of α and β from all seven sites in SWWA for a given CLIGEN parameter; and the “Wilgarrup” method, which uses Wilgarrup’s α and β values, as Wilgarrup has the largest and the most significant change in monthly precipitation among all selected sites. Results indicated that the site-specific method is the most appropriate approach for adjusting CLIGEN parameter values and for reproducing the declining trend in precipitation for SWWA. Therefore, in this study, CLIGEN parameters were adjusted using the site-specific regionalization method of adjustment (Vaghefi and Yu 2016), followed by running CLIGEN to produce 90 years of daily precipitation (1923–2012). Then, slope of the changes, mean annual precipitation, standard deviation, and coefficient of variance were calculated based on the observed and CLIGEN-generated precipitation.

With adjusted parameter values, CLIGEN was run to produce daily precipitation series for the six sites considered in the paper, and these daily series were subsequently accumulated to produce monthly and annual precipitation totals. To validate these adjustment methods for CLIGEN parameters, two 30-yr (60 years) periods of observed and CLIGEN-generated precipitation data in SEA and 90 years of observed and CLIGEN-generated precipitation data in SWWA were used to simulate daily flows for each of the two regions.

Two conceptual hydrological models were selected to simulate daily flows for each site to validate the proposed adjustment methods for CLIGEN parameters. The hydrological models, AWBM and SimHyd, were used to estimate the daily, monthly, and annual flow sequences. These models have been widely used and accepted in Australia (Chiew et al. 2002; Boughton 2004; Jones et al. 2006; Chiew et al. 2008; Wang et al. 2011; Yu and Zhu 2015). Input requirements are identical, including daily precipitation and daily potential evapotranspiration (Yu and Zhu 2015).

AWBM was developed in the early 1990s (Boughton 1993; Boughton and Carroll 1993) and is now one of the most widely used rainfall–runoff models in Australia. AWBM uses three surface stores (surface storage capacity 1 C1, surface storage capacity 2 C2, and surface storage capacity 3 C3) to simulate partial areas of runoff (partial area 1 A1 and partial area 2 A2) considering a baseflow index (BFI) and two constants for surface store and base flow (Podger 2004). SimHyd is also a conceptual rainfall–runoff model that estimates daily streamflow from daily precipitation and areal potential evapotranspiration data.

The SimHyd model is actually a simplified version of another daily rainfall–runoff model developed in the 1970s (Porter and McMahon 1975). Unlike AWBM with unconnected surface stores, SimHyd conceptualizes the runoff from four different sources, namely, direct runoff from impervious areas, runoff due to infiltration excess, interflow, and base flow from a groundwater store. Apart from the infiltration capacity, which is modeled as a nonlinear function of the soil moisture, all other relationships are linear and constrained by store capacities.

In total AWBM and SimHyd have eight and seven parameters for catchments (without impervious surfaces in SimHyd). The parameters of AWBM and SimHyd were calibrated using observed daily potential evapotranspiration and catchment precipitation and streamflow data for three calibration catchments in each region. Then, the calibrated parameter values were utilized to simulate daily flow series A using observed precipitation data and series B using the CLIGEN-generated precipitation data. For both calibration and simulation, the Rainfall Runoff Library (RRL) software was used (Podger 2004). A genetic algorithm was used to minimize the sum of squares of errors in observed and simulated flows for parameter calibration. Finally, statistical analysis was undertaken to compare flow series A and B to test whether CLIGEN could be used to reproduce the change and/or trend in precipitation and streamflow (Fig. 3).

Fig. 3.
Fig. 3.

Flowchart on how methods for parameter adjustment were evaluated.

Citation: Journal of Hydrometeorology 18, 7; 10.1175/JHM-D-16-0237.1

Tables 3 and 4 present calibrated parameter values for these selected streamflow gauging stations. Nash–Sutcliffe criteria (Nash and Sutcliffe 1970) show that these conceptual hydrologic models are able to fit the observed streamflow data reasonably well for these gauging stations in SEA and SWWA.

Table 3.

Calibrated AWBM parameter values and model performance indicators for six selected streamflow gauges in SEA and SWWA.

Table 3.
Table 4.

Calibrated SimHyd parameter values and model performance indicators for six selected streamflow gauges in SEA and SWWA.

Table 4.

Daily streamflow based on the observed and CLIGEN-generated precipitation data for each station was calculated using calibrated parameters of AWBM and SimHyd for these three streamflow gauges. Therefore, three precipitation sites, three sets of parameter values, two regions, and two conceptual hydrological models result in 18 combinations of daily streamflow using observed precipitation (series A) and CLIGEN-generated precipitation (series B) for comparative analysis.

Simulated flow series A and B were compared using monthly and annual flows based on the observed and CLIGEN-generated precipitation data. The simulated flows were statistically tested using a Student’s t test to determine whether streamflows generated using the observed and CLIGEN-generated precipitation data are significantly different from each other. The t-test result with a p value of less than 0.01 was considered to indicate significant difference in estimated flows between the two data sources.

3. Results

a. Annual precipitation

Prior to comparing simulated streamflow using AWBM and SimHyd, the observed and CLIGEN-generated precipitation data were compared for the six sites from the two regions. The results showed that the observed and CLIGEN-generated precipitation data are quite similar in the mean. Figure 4 shows a comparison of the mean annual precipitation between the observed and CLIGEN-generated data for SEA and SWWA. The average mean annual precipitation using the observed data was 900 mm, and that using the CLIGEN-generated data was 874 mm, a difference of <3% for the six sites.

Fig. 4.
Fig. 4.

Comparison of the observed mean annual precipitation against CLIGEN-generated data for SEA and SWWA. The period was 30 years for sites in SEA and 90 years in SWWA.

Citation: Journal of Hydrometeorology 18, 7; 10.1175/JHM-D-16-0237.1

b. Annual flow

The time series of annual streamflow for each region was plotted and compared graphically as well as tested statistically. In all 18 scenarios, for AWBM or SimHyd, the difference between annual streamflow series A and B was found to be insignificant. However, in general, the performance of SimHyd in terms of flow simulation using observed and CLIGEN-generated precipitation data was better than AWBM.

In SEA, time series of annual streamflow for all 18 scenarios follow a similar pattern. As an example, Fig. 5 presents the annual time series of calculated streamflow using AWBM based on the observed, adjusted CLIGEN-generated precipitation, and CLIGEN-generated precipitation without adjustment at Cataract Dam for the two contrasting period in SEA. In Fig. 5, annual streamflow C calculated using CLIGEN-generated precipitation without parameter adjustment shows the 30-yr average flow of period 2 is slightly lower than flows from period 1, and the difference in the mean is not statistically different for streamflow C. This demonstrates the importance of adjusting CLIGEN parameter values prior to using generated precipitation for flow simulation if the significant increase in streamflow is to be reproduced.

Fig. 5.
Fig. 5.

Annual time series of simulated streamflow with AWBM based on the observed (i.e., A) and CLIGEN-generated (i.e., B) precipitation at Cataract Dam in SEA.

Citation: Journal of Hydrometeorology 18, 7; 10.1175/JHM-D-16-0237.1

Figure 6 presents the simulated mean annual flow predicted using the observed and CLIGEN-generated daily precipitation data for the two contrasting periods for the three sites in SEA. The error bars represent one standard error of the mean. It can be seen from Fig. 6 that the mean annual streamflow is broadly similar using the observed and CLIGEN-generated precipitation data for both models; the streamflow estimated using the CLIGEN-generated precipitation data is systematically less than that estimated using the observed data. The mean annual streamflow estimated using CLIGEN-generated precipitation data (series B) is smaller than streamflow generated using observed precipitation data by 9–94 mm, and the difference varies from −7% to −44% between the two streamflow series (Table 5). The average mean streamflow was 141 mm yr−1 for the 18 scenarios using the CLIGEN-generated data and AWBM. This is 24% less than 186 mm yr−1 using the observed data. For SimHyd, the average mean streamflow was 199 mm yr−1 for the 18 scenarios using the CLIGEN-generated data, which is 13% less than that using the observed data (228 mm yr−1).

Fig. 6.
Fig. 6.

Mean annual flows predicted using observed and CLIGEN-generated daily precipitation for the two contrasting periods with (a) AWBM and (b) SimHyd for the three sites in SEA.

Citation: Journal of Hydrometeorology 18, 7; 10.1175/JHM-D-16-0237.1

Table 5.

Differences (%) in the mean annual streamflow between observed and CLIGEN-generated precipitation data for selected sites in SEA.

Table 5.

In SWWA, where the annual precipitation shows a significant decreasing trend since the 1920s, the estimated streamflow follows a similar trend irrespective of which precipitation dataset or hydrological model was used. Figure 7 shows, as an example, the annual time series of calculated flow with SimHyd using the observed, CLIGEN-generated precipitation with adjusted parameter values and CLIGEN-generated precipitation without parameter adjustment at Nannup in SWWA. Figure 7 shows that annual streamflow C using CLIGEN-generated precipitation without parameter adjustment has a small positive trend (2.3 mm decade−1) that is not significantly different from zero. Simulated flow A using the observed precipitation data and simulated flow B with CLIGEN-generated precipitation with parameter adjustment show the significantly decreasing trends in flow as expected (Fig. 7). This once again demonstrates the importance of adjusting CLIGEN parameter values prior to flow simulation if a significant trend in flow is to be reproduced.

Fig. 7.
Fig. 7.

Annual time series of calculated streamflow by SimHyd based on the observed (i.e., A) and CLIGEN-generated (i.e., B) precipitation at Nannup in SWWA.

Citation: Journal of Hydrometeorology 18, 7; 10.1175/JHM-D-16-0237.1

In SWWA, the observed and CLIGEN-generated precipitation data were used to simulate streamflow for a 90-yr period from 1923 to 2012. The results show that, using the observed and CLIGEN-generated precipitation data, streamflow was quite similar for all scenarios and for both models. Figure 8 shows the long-term (1923–2012) mean annual streamflow predicted using the observed and CLIGEN-generated daily precipitation data with AWBM and SimHyd for three sites in SWWA. The maximum difference between estimated annual streamflow (series A) and (series B) varies from −8.1 to 7.3 mm. The average mean flow was 163 mm yr−1 for the 18 scenarios using the CLIGEN-generated data and was 162 mm yr−1 using the observed data for the same 18 scenarios. The discrepancy was less than 0.3% in the mean.

Fig. 8.
Fig. 8.

Long-term (1923–2012) mean annual flows predicted using observed and CLIGEN-generated daily precipitation with AWBM and SimHyd for three sites in SWWA. The error bars represent one standard error of the mean.

Citation: Journal of Hydrometeorology 18, 7; 10.1175/JHM-D-16-0237.1

c. Changes in annual streamflow

To compare streamflow using CLIGEN-generated and observed precipitation in SEA, the ratios of mean annual streamflow for two contrasting periods are presented in Fig. 9. Let Q1 be the mean annual streamflow for the period from 1929 to 1948 and let Q2 be the mean annual streamflow for the period from 1949 to 1978. The ratio of Q2/Q1 represents the change in streamflow over the two contrasting periods when precipitation has significantly increased.

Fig. 9.
Fig. 9.

Ratios of calculated annual streamflow for two contrasting periods in SEA.

Citation: Journal of Hydrometeorology 18, 7; 10.1175/JHM-D-16-0237.1

As expected, the mean annual streamflow for period 2 (i.e., Q2) was higher than that for period 1 (i.e., Q1). The ratio of Q2 over Q1 varies between 1.1 and 3.3, regardless of which precipitation dataset or conceptual hydrological model was used. In addition, the larger the increase in precipitation, the larger the increase in streamflow between the two periods. This means that the streamflow ratio over the two contrasting periods for Cataract Dam, which has recorded the most significant increase in precipitation, was the largest, while Clarence Town, with the least amount of increase in annual precipitation between two periods, has the smallest streamflow ratios. Figure 9 also shows the ratio of simulated streamflow (series B using CLIGEN-generated precipitation) is consistently higher than that using observed precipitation for all 18 scenarios. This occurred largely because streamflow was underestimated using the CLIGEN-generated precipitation as shown in Fig. 6. Lower estimated streamflow led to an increased ratio between the two contrasting periods. The ratio of Q2/Q1 varied from 1.1 to 2.6 using the observed precipitation data. This ratio was increased to 1.4–2.7 using SimHyd and increased even further to 1.5–3.3 using AWBM. Although there are noticeable differences in the streamflow ratio shown in Fig. 9, the standard Student’s t test indicated that there are no significant differences between the flow generated using observed or CLIGEN-generated precipitation data.

For SWWA, the rate of decrease in streamflow in millimeters per decade from 1923 to 2012 is presented in Fig. 10. The rate of decrease in streamflow using CLIGEN-generated precipitation (7–29 mm decade−1) is generally smaller than the rate of decrease in streamflow using the observed precipitation (11–31 mm decade−1). However, in this region, there is no noticeable difference between two conceptual hydrological models in terms of the simulated rate of decrease in streamflow (7–31 mm decade−1 for AWBM and 8–27 mm decade−1 for SimHyd).

Fig. 10.
Fig. 10.

Rate of changes in simulated flow per decade from 1923 to 2012 in SWWA.

Citation: Journal of Hydrometeorology 18, 7; 10.1175/JHM-D-16-0237.1

The Student’s t test was undertaken to evaluate the differences between annual flow amounts generated by different methods for SEA and SWWA. Table 6 presents the percentage differences in rate of change between AWBM and SimHyd for selected sites in SWWA. The t-test results for the annual streamflow indicate that the differences between annual streamflow using the observed and CLIGEN precipitation are statistically insignificant for all 18 scenarios considered, and these are unrelated to selected hydrological models.

Table 6.

Differences (%) in the rate of change between observed and CLIGEN-generated precipitation data for selected sites in SWWA.

Table 6.

d. Monthly streamflow

Similar to the annual streamflow, simulated monthly streamflow using the observed and CLIGEN precipitation for each region was plotted and compared graphically as well as statistically tested. Figures 11 and 12 show comparisons of monthly distribution of estimated streamflow in SEA using the observed and CLIGEN precipitation for Coonabarabran using SimHyd as an example.

Fig. 11.
Fig. 11.

Monthly distribution of average flow and standard error of the mean at Coonabarabran for 30-yr period of 1919–48 in SEA.

Citation: Journal of Hydrometeorology 18, 7; 10.1175/JHM-D-16-0237.1

Fig. 12.
Fig. 12.

Monthly distribution of average flow and standard error of the mean at Coonabarabran for 30-yr period of 1949–78 in SEA.

Citation: Journal of Hydrometeorology 18, 7; 10.1175/JHM-D-16-0237.1

As presented in these sample graphs, the monthly distributions of estimated flow values using AWBM or SimHyd are relatively similar in SEA and follow the same pattern. The range of changes found in simulated monthly flow is between −28 and 18 mm month−1. However, the t-test results showed that the p values for each month of 18 scenarios (216 values) are all greater than 0.01, and therefore the difference in simulated streamflow is considered to be statistically insignificant. In other words, while the range of changes is relatively large, the variability of monthly flow is so great that no significant differences between monthly streamflow were found. Table 7 presents the maximum difference in the simulated flow for each month using each parameter set as well as the simulated mean monthly streamflow with the largest difference on a monthly basis.

Table 7.

The max differences in the mean monthly streamflow Q using observed precipitation Pobs and CLIGEN-generated precipitation PCLG for selected sites in SEA.

Table 7.

The monthly distribution of simulated streamflow in SWWA based on the observed and CLIGEN-generated precipitation is presented in Fig. 13 using AWBM as an example. Similar to the SEA region, monthly distributions of estimated streamflow using AWBM or SimHyd are quite similar in SWWA and they follow similar seasonal patterns. However, in SWWA, the median of monthly streamflow during the dry months is mostly zero. The t-test results showed that approximately 37% of monthly flow p values were smaller than the significance threshold level of 0.01, mostly in the summer months (December–February).

Fig. 13.
Fig. 13.

Monthly distribution of average flow and standard error of the mean at Manjimup for 90-yr period of 1923–2012 in SWWA.

Citation: Journal of Hydrometeorology 18, 7; 10.1175/JHM-D-16-0237.1

Table 8 shows the mean monthly flow simulated using the observed (series A) and CLIGEN-generated (series B) values are quite similar. In fact, the statistical test shows that differences in all wet months are insignificant and only some of the dry months show significant differences between flow series A and B. The median of monthly streamflow for all these months is zero.

Table 8.

The max differences in changes of simulated mean monthly streamflow per decade using observed precipitation and CLIGEN-generated precipitation for selected sites in SWWA.

Table 8.

Tables 9 and 10 present the maximum differences in the rate of change in the simulated mean monthly streamflow. These tables show that both AWBM and SimHyd are able to reproduce the decreasing trend in monthly streamflow for SWWA. Not surprisingly, the maximum differences in the rate of decrease occur in winter months when streamflow is much higher than that in other months.

Table 9.

The max difference in the rate of decrease Rc between observed precipitation flows and CLIGEN-generated precipitation flows for SWWA using AWBM.

Table 9.
Table 10.

The max difference in the rate of decrease between observed precipitation flows and CLIGEN-generated precipitation flows for SWWA using SimHyd.

Table 10.

4. Discussion

To our knowledge, this is the first time that output from CLIGEN with adjusted parameter values was assessed in terms of simulated streamflow when significant changes in precipitation have occurred. The majority of statistical tests undertaken in this study showed that there were no significant differences between simulated streamflow using observed and CLIGEN-generated precipitation data. Despite the statistical tests indicating no significant differences, a comparison of the simulated flow as an end product of CLIGEN-generated precipitation with streamflow simulated with observed precipitation data showed some discrepancy, particularly for the simulated streamflow in drier months. The discrepancy between simulated flows could occur as a result of the unavoidable limitations such as conceptual hydrological model selection, parameter calibration, and sample size. In addition, using streamflow data at gauging stations to calibrate models adds to the uncertainty associated with this investigation. Generally, catchment precipitation (based on interpolation of point precipitation data) was used to calibrate the model. The larger the catchment, the more data points that would be required to represent the spatial distribution of precipitation. However, when simulating flow using CLIGEN-generated precipitation data, only point precipitation data are used.

Three streamflow gauges were used in this study to calibrate two hydrological models. This was considered to be sufficient for the validation purposes. However, using more streamflow gauges with similar catchment characteristics would allow proper averaging of parameter values. An averaging method could overcome some of the limitations and possibly reduce the uncertainty associated with simulated streamflow for these sites.

For the purpose of validating the adjustment methods for CLIGEN parameters on a monthly basis, precipitation distribution can have considerable impact on the statistical results. For example, in SWWA where summer precipitation is low and streamflow is close to zero, statistical tests showed that 37% of monthly simulated streamflow is significantly different using the observed data when compared to streamflow using CLIGEN-generated precipitation data.

5. Conclusions

CLIGEN is an SWG that can produce sequences of daily weather variables. Since precipitation has significantly changed in some regions of Australia, CLIGEN parameters need to be adjusted to simulate the changes in precipitation and the impact of such changes on catchment hydrology in these regions. A method of CLIGEN parameters adjustment was developed by Vaghefi and Yu (2011) based on the changes in the mean monthly precipitation for SEA, where there was an abrupt change in annual precipitation in 1950s. Vaghefi and Yu (2016) subsequently proposed a site-specific method of CLIGEN parameter adjustment for SWWA, where annual precipitation showed a clear decreasing trend over the last 90 years.

The aim of this paper was to further evaluate these proposed methods by comparing simulated daily streamflow using CLIGEN-generated and observed precipitation. AWBM and SimHyd, two conceptual hydrological models commonly used in Australia were calibrated and utilized to test the quality of the CLIGEN-generated precipitation using above mentioned adjustment methods. Simulated monthly and annual flows were statistically tested for significant differences. The results showed monthly and annual simulated streamflow time series using observed and CLIGEN-generated precipitation all follow similar patterns. No significant difference was found between annual streamflow generated using observed or CLIGEN-generated precipitation. This provides additional support for the proposed method to adjust CLIGEN parameters. Changes in flows where flow has significantly changed could be slightly exaggerated due to the underestimation of streamflow using conceptual hydrological models. For regions where there is a decreasing trend in annual precipitation, the rate of decrease using CLIGEN-generated precipitation tends to be smaller than that in the streamflow estimated using observed precipitation data. The t test for monthly flow showed that differences in simulated flow using observed and CLIGEN-generated precipitation data can be significant for dry months in SWWA. This occurred largely because of the extremely high variability of streamflow during drier months when most streams cease to flow in SWWA.

In summary, using developed adjustment methods, CLIGEN parameters were adjusted to simulate flows using conceptual hydrological models. These flows were compared to simulated flows using observed precipitation data. Simulated flows using CLIGEN-generated precipitation was able to reproduce the change in the mean in SEA and the decreasing trend in SWWA. Therefore, the adjustment methods developed for CLIGEN parameters are further validated using CLIGEN-generated precipitation as input data for conceptual hydrological models.

REFERENCES

  • Bastola, S., C. Murphy, and R. Fealy, 2012: Generating probabilistic estimates of hydrological response for Irish catchments using a weather generator and probabilistic climate change scenarios. Hydrol. Processes, 26, 23072321, doi:10.1002/hyp.8349.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • BoM, 1989: Climate of Australia. Bureau of Meteorology, 49 pp.

  • BoM, 2016: Climate data online. Accessed 22 April 2015. [Available online at www.bom.gov.au/climate/data/.]

  • Boughton, W. C., 1993: A hydrograph-based model for estimating the water yield of ungauged catchments. Proceedings of the 1993 Hydrology and Water Resources Conference. Institution of Engineers, National Conference Publication 93/14, 317–324.

  • Boughton, W. C., 2004: The Australian water balance model. Environ. Modell. Software, 19, 943956, doi:10.1016/j.envsoft.2003.10.007.

  • Boughton, W. C., and D. Carroll, 1993: A simple combined water balance/flood hydrograph model. Proceedings of the 1993 Hydrology and Water Resources Conference. Institution of Engineers, National Conference Publication 93/14, 299–304.

  • Chambers, L. E., 2001: Classifying rainfall districts: A southwestern Australian study. Aust. Meteor. Mag., 50, 91103.

  • Chen, J., and F. P. Brissette, 2014: Comparison of five stochastic weather generators in simulating daily precipitation and temperature for the Loess Plateau of China. Int. J. Climatol., 34, 30893105, doi:10.1002/joc.3896.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chen, J., and F. P. Brissette, 2015: Combining stochastic weather generation and ensemble weather forecasts for short-term streamflow prediction. Water Resour. Manage., 29, 33293342, doi:10.1007/s11269-015-1001-3.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chen, J., F. P. Brissette, R. Leconte, and A. Caron, 2012: A versatile weather generator for daily precipitation and temperature. Trans. ASABE, 55, 895906, doi:10.13031/2013.41522.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chen, J., F. P. Brissette, and X. J. Zhang, 2016: Hydrological modeling using a multisite stochastic weather generator. J. Hydrol. Eng., 21, 04015060, doi:10.1061/(ASCE)HE.1943-5584.0001288.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chiew, F. H. S., M. Peel, and A. Western, 2002: Application and testing of the simple rainfall–runoff model SimHyd. Mathematical Models of Small Watershed Hydrology and Applications, V. Singh and D. Frevert, Eds., Water Resources Publications, 335–367.

  • Chiew, F. H. S., J. Vaze, N. R. Viney, M. Perraud, J. Teng, and D. A. Post, 2008: Modelling runoff and climate change impact on runoff in 178 catchments in the Murray–Darling basin using Sacramento and SIMHYD rainfall–runoff models. Proceedings of Water Down Under 2008, M. Lambert, T. M. Daniell, and M. Leonard, Eds., Causal Productions, 87–97.

  • Cornish, P., 1977: Changes in seasonal and annual rainfall in New South Wales. Search, 8, 3840.

  • Dubrovský, M., J. Buchtele, and Z. Žalud, 2004: High-frequency and low-frequency variability in stochastic daily weather generator and its effect on agricultural and hydrologic modelling. Climatic Change, 63, 145179, doi:10.1023/B:CLIM.0000018504.99914.60.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Fan, J., C. Yang, C. Liu, and H. Huang, 2013: Assessment and validation of CLIGEN-simulated rainfall data for northern Taiwan. Paddy Water Environ., 11, 161173, doi:10.1007/s10333-011-0301-3.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Favis-Mortlock, D., and M. Savabi, 1996: Shifts in rates and spatial distributions of soil erosion and deposition under climate change. Advances in Hillslope Processes, Vol. 1, M. Anderson and S. Brooks, Eds., Wiley, 529–560.

  • Feng, J., J. Li, and Y. Li, 2010: A monsoon-like southwest Australian circulation and its relation with rainfall in southwest Western Australia. J. Climate, 23, 13341353, doi:10.1175/2009JCLI2837.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Foster, G. R., 1982: Modeling the erosion process. Hydrologic Modeling of Small Watersheds, ASAE Monograph, No. 5, American Society of Agricultural Engineers, 297–360.

  • Ghidey, F., and E. E. Alberts, 1996: Comparison of measured and WEPP predicted runoff and soil loss for Midwest claypan soil. Trans. ASAE, 39, 13951402, doi:10.13031/2013.27631.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Goodess, C., and J. Palutikof, 1998: Development of daily rainfall scenarios for southeast Spain using a circulation-type approach to downscaling. Int. J. Climatol., 18, 10511083, doi:10.1002/(SICI)1097-0088(199808)18:10<1051::AID-JOC304>3.0.CO;2-1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • IPCC, 2013: Climate Change 2013: The Physical Science Basis. Cambridge University Press, 1535 pp., doi:10.1017/CBO9781107415324.

    • Crossref
    • Export Citation
  • Johnson, G. L., C. L. Hanson, S. P. Hardegree, and E. B. Ballard, 1996: Stochastic weather simulation: Overview and analysis of two commonly used models. J. Appl. Meteor. Climatol., 35, 18781896, doi:10.1175/1520-0450(1996)035<1878:SWSOAA>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Jones, R., F. H. Chiew, F. Boughton, and L. Zhang, 2006: Estimating the sensitivity of mean annual runoff to climate change using selected hydrological models. Adv. Water Resour., 29, 14191429, doi:10.1016/j.advwatres.2005.11.001.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kilsby, C. P., and Coauthors, 2007: A daily weather generator for use in climate change studies. Environ. Modell. Software, 22, 17051719, doi:10.1016/j.envsoft.2007.02.005.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kou, X., J. Ge, Y. Wang, and C. Zhang, 2007: Validation of the weather generator CLIGEN with daily precipitation data from the Loess Plateau, China. J. Hydrol., 347, 347357, doi:10.1016/j.jhydrol.2007.09.051.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lavery, B., A. Kariko, and N. Nicholls, 1992: A high-quality historical rainfall data set for Australia. Aust. Meteor. Mag., 40, 3339.

  • Lavery, B., G. Joung, and N. Nicholls, 1997: An extended high-quality historical rainfall data set for Australia. Aust. Meteor. Mag., 46, 2738.

    • Search Google Scholar
    • Export Citation
  • Li, F., L. E. Chambers, and N. Nicholls, 2005: Relationships between rainfall in the southwest of Western Australia and near-global patterns of sea-surface temperature and mean sea-level pressure variability. Aust. Meteor. Mag., 54, 2333.

    • Search Google Scholar
    • Export Citation
  • Li, Y., W. Cai, and E. Campbell, 2005: Statistical modeling of extreme rainfall in southwest Western Australia. J. Climate, 18, 852863, doi:10.1175/JCLI-3296.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Li, Z., F. P. Brissette, and J. Chen, 2013: Finding the most appropriate precipitation probability distribution for stochastic weather generation and hydrological modelling in Nordic watersheds. Hydrol. Processes, 27, 37183729, doi:10.1002/hyp.9499.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Li, Z., F. P. Brissette, and J. Chen, 2014: Assessing the applicability of six precipitation probability distribution models on the Loess Plateau of China. Int. J. Climatol., 34, 462471, doi:10.1002/joc.3699.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lobo, G., J. Frankenbeger, D. Flanagan, and C. Bonilla, 2015: Evaluation and improvement of The CLIGEN model for storm and rainfall erosivity generation in central Chile. Catena, 127, 206213, doi:10.1016/j.catena.2015.01.002.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mavromatis, T., and J. Hansen, 2001: Interannual variability characteristics and simulated crop response of four stochastic weather generators. Agric. For. Meteor., 109, 283296, doi:10.1016/S0168-1923(01)00272-6.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • McFarlane, D., R. Stone, S. Martens, J. Thomas, R. Silberstein, R. Ali, and G. Hodgson, 2012: Climate change impacts on water yields and demands in southwestern Australia. J. Hydrol., 475, 488498, doi:10.1016/j.jhydrol.2012.05.038.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Min, Y., J. Kryjov, K. An, S. Hameed, S. Sohn, W. Lee, and J. Oh, 2011: Evaluation of the weather generator CLIGEN with daily precipitation characteristics in Korea. Asia-Pac. J. Atmos. Sci., 47, 255263, doi:10.1007/s13143-011-0014-y.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Nash, J. E., and J. V. Sutcliffe, 1970: River flow forecasting through conceptual models. Part 1: A discussion of principles. J. Hydrol., 10, 282290, doi:10.1016/0022-1694(70)90255-6.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Nearing, M. A., G. A. Foster, L. J. Lane, and S. C. Finkner, 1989: A process-based soil erosion model for USDA-Water Erosion Prediction Project technology. Trans. ASAE, 32, 15871593, doi:10.13031/2013.31195.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Nicholls, N., and B. Lavery, 1992: Australian rainfall trends during the twentieth century. Int. J. Climatol., 12, 153163, doi:10.1002/joc.3370120204.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Nicholls, N., and A. Kariko, 1993: East Australian rainfall events: Interannual variations, trends, and relationships with the Southern Oscillation. J. Climate, 6, 11411152, doi:10.1175/1520-0442(1993)006<1141:EAREIV>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Nicks, A., and G. Gardner, 1994: CLIGEN: A weather generator for climate inputs to water resource and other models. Proceedings of the 5th International Conference on Computers in Agriculture, D. G. Watson, F. S. Zazueta, and T. V. Harrison, Eds., American Society of Agricultural Engineers, 3–94.

  • Nicks, A., L. Lane, and G. Gardner, 1995: Chapter 2: Weather generator. In USDA Water Erosion Prediction Project: Hillslope Profile and Watershed Model Documentation. NSERL Rep. 10, USDA-ARS, 2.1–2.22.

  • Pittock, A., 1983: Climate change in Australia: Implications for a CO2-warmed Earth. Climatic Change, 5, 321340, doi:10.1007/BF02423529.

  • Podger, G., 2004: Description of models. Rainfall Runoff Library, G. Podger, Ed., Cooperative Research Centre for Catchment Hydrology, 47–64. [Available online at http://www.toolkit.net.au/rrl.]

  • Porter, J., and T. A. McMahon, 1975: Application of a catchment model in southeastern Australia. J. Hydrol., 24, 121134, doi:10.1016/0022-1694(75)90146-8.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Prudhomme, C., N. Reynard, and S. Crroks, 2002: Downscaling of global climate models for flood frequency analysis: Where are we now? Hydrol. Processes, 16, 11371150, doi:10.1002/hyp.1054.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Pruski, F., and M. A. Nearing, 2002a: Runoff and soil-loss responses to changes in precipitation: A computer simulation study. J. Soil Water Conserv., 57, 716.

    • Search Google Scholar
    • Export Citation
  • Pruski, F., and M. A. Nearing, 2002b: Climate-induced changes in erosion during the 21st century for eight U.S. locations. Water Resour. Res., 38, 1298, doi:10.1029/2001WR000493.

    • Crossref
    • Export Citation
  • Richardson, C., and D. Wright, 1984: WGEN: A model for generating daily weather variables. Rep. ARS-8, 86 pp.

  • Richardson, C., 1981: Stochastic simulation of daily precipitation, temperature, and solar radiation. Water Resour. Res., 17, 182190, doi:10.1029/WR017i001p00182.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Semenov, M. A., and J. Porter, 1995: Climatic variability and the modelling of crop yields. Agric. For. Meteor., 73, 265283, doi:10.1016/0168-1923(94)05078-K.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Semenov, M. A., and E. M. Barrow, 1997: Use of a stochastic weather generator in the development of climate change scenarios. Climatic Change, 35, 397414, doi:10.1023/A:1005342632279.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Semenov, M. A., and E. M. Barrow, 2002: LARS-WG: A stochastic weather generator for use in climate impact studies. User manual, version 3.0, 28 pp. [Available online at http://resources.rothamsted.ac.uk/sites/default/files/groups/mas-models/download/LARS-WG-Manual.pdf.]

  • Silberstein, R., and Coauthors, 2012: Climate change and runoff in southwestern Australia. J. Hydrol., 475, 441455, doi:10.1016/j.jhydrol.2012.02.009.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Smith, I., 2004: An assessment of recent trends in Australian rainfall. Aust. Meteor. Mag., 53, 163173.

  • Stöckle, C., G. Campbell, and R. Nelson, 1999: ClimGen manual. Washington State University.

  • Timbal, B., 2004: Southwest Australia past and future rainfall trends. Climate Res., 26, 233249, doi:10.3354/cr026233.

  • Tiwari, A., L. M. Risse, and M. A. Nearing, 2000: Evaluation of WEPP and its comparison with USLE and RUSLE. Trans. ASAE, 43, 11291135, doi:10.13031/2013.3005.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Vaghefi, P., and B. Yu, 2011: Use of CLIGEN to simulate climate change in southeastern Australia. Trans. ASABE, 54, 857867, doi:10.13031/2013.37111.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Vaghefi, P., and B. Yu, 2016: Use of CLIGEN to simulate decreasing precipitation trends in the southwest of Western Australia. Trans. ASABE, 59, 4961, doi:10.13031/trans.59.10829.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wang, Q., T. Pagano, S. Zhou, H. Hapuarachchi, L. Zhang, and D. Robertson, 2011: Monthly versus daily water balance models in simulating monthly runoff. J. Hydrol., 404, 166175, doi:10.1016/j.jhydrol.2011.04.027.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wheater, H. S., and Coauthors, 2005: Spatial–temporal rainfall modelling for flood risk estimation. Stochastic Environ. Res. Risk Assess., 19, 403416, doi:10.1007/s00477-005-0011-8.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wilby, R., and T. Wigley, 1997: Downscaling general circulation model output: A review of methods and limitations. Prog. Phys. Geog., 21, 530–548, doi:10.1177/030913339702100403.

    • Crossref
    • Export Citation
  • Wilby, R., T. Wigley, D. Conway, P. Jones, B. Hewitson, J. Main, and D. Wilks, 1998: Statistical downscaling of general circulation model output: A comparison of methods. Water Resour. Res., 34, 29953008, doi:10.1029/98WR02577.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wilks, D., 1992: Adapting stochastic weather generation algorithms for climate change studies. Climatic Change, 22, 6784, doi:10.1007/BF00143344.

  • Wilks, D., 1999: Multisite downscaling of daily precipitation with a stochastic weather generator. Climate Res., 11, 125136, doi:10.3354/cr011125.

  • Xu, C., 1999: From GCMs to river flow: A review of downscaling methods and hydrologic modelling approaches. Prog. Phys. Geogr., 23, 229249, doi:10.1191/030913399667424608.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yu, B., 2000: Improvement and evaluation of CLIGEN for storm generation. Trans. ASAE, 43, 301307, doi:10.13031/2013.2705.

  • Yu, B., 2002: Using CLIGEN to generate RUSLE climate inputs. Trans. ASAE, 45, 9031001, doi:10.13031/2013.9952.

  • Yu, B., 2003: An assessment of uncalibrated CLIGEN in Australia. Agric. For. Meteor., 119, 131148, doi:10.1016/S0168-1923(03)00141-2.

  • Yu, B., 2005: Adjustment of CLIGEN parameters to generate precipitation change scenarios in southeastern Australia. Catena, 61, 196209, doi:10.1016/j.catena.2005.03.004.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yu, B., and D. T. Neil, 1991: Global warming and regional rainfall: The difference between average and high-intensity rainfalls. Int. J. Climatol., 11, 653661, doi:10.1002/joc.3370110607.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yu, B., and Z. Zhu, 2015: A comparative assessment of AWBM and SimHyd for forested watersheds. Hydrol. Sci. J., 60, 12001212, doi:10.1080/02626667.2014.961924.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yu, B., C. A. Ciesiolka, C. W. Rose, and K. J. Coughlan, 2000: A validation test of WEPP to predict runoff and soil loss from a pineapple farm on a sandy soil in subtropical Queensland, Australia. Aust. J. Soil Res., 38, 537554, doi:10.1071/SR99104.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, X.-C., 2005a: Generating correlative storm variables for CLIGEN using a distribution-free approach. Trans. ASAE, 48, 567575, doi:10.13031/2013.18331.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, X.-C., 2005b: Spatial downscaling of global climate model output for site-specific assessment of crop production and soil erosion. Agric. For. Meteor., 135, 215229, doi:10.1016/j.agrformet.2005.11.016.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, X.-C., 2007: A comparison of explicit and implicit spatial downscaling of GCM output for soil erosion and crop production assessments. Climatic Change, 84, 337363, doi:10.1007/s10584-007-9256-1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, X.-C., and J. Garbrecht, 2003: Evaluation of CLIGEN precipitation parameters and their implication on WEPP runoff and erosion prediction. Trans. ASAE, 46, 311320, doi:10.13031/2013.12982.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, X.-C., and W.-Z. Liu, 2005: Simulating potential response of hydrology, soil erosion, and crop productivity to climate change in Changwu tableland region on the Loess Plateau of China. Agric. For. Meteor., 131, 127142, doi:10.1016/j.agrformet.2005.05.005.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, X.-C., M. A. Nearing, L. M. Risse, and K. C. McGregor, 1996: Evaluation of WEPP runoff and soil loss predictions using natural runoff plot data. Trans. ASAE, 39, 855863, doi:10.13031/2013.27570.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, Y., B. Liu, Z. Wang, and Q. Zhu, 2008: Evaluation of CLIGEN for storm generation on the semiarid Loess Plateau in China. Catena, 73, 19, doi:10.1016/j.catena.2007.08.001.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, Y.-G., M. A. Nearing, X.-C. Zhang, Y. Xie, and H. Wei, 2010: Projected rainfall erosivity changes under climate change from multimodel and multiscenario projections in northeast China. J. Hydrol., 384, 97106, doi:10.1016/j.jhydrol.2010.01.013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhuang, X. W., Y. P. Li, G. H. Huang, and J. Liu, 2016: Assessment of climate change impacts on watershed in cold-arid region: An integrated multi-GCM-based stochastic weather generator and stepwise cluster analysis method. Climate Dyn., 47, 191209, doi:10.1007/s00382-015-2831-7.

    • Crossref
    • Search Google Scholar
    • Export Citation
Save