OPEN ACCESS
Much of Australia regularly experiences extremes of drought and flooding, with high variability in rainfall in many regions of the continent. Development of reliable and accurate medium-term rainfall forecasts is important, particularly for agriculture. Monthly rainfall forecasts 12 months in advance were made with artificial neural networks (ANNs), a form of artificial intelligence, for the locations of Bathurst Deniliquin, and Miles, which are agricultural hubs in the Murray Darling Basin, in south-eastern Australia. Two different approaches were used for the optimisation of the ANN models. In the first, all months in each calendar year were optimised together, while in the second approach, rainfall forecasts for each month of the year were made individually. For each of the three locations for most months, higher forecast skill scores were achieved using single-month optimizations. In the case of Bathurst, however, for the months of November and December, the root mean square error (RMSE) for all-month optimisation was lower than for single-month optimisation. The best overall rainfall forecasts for each site were obtained by generating a composite of the two approaches, selecting the forecast for each month with the lowest forecast errors. Composite model skill score levels of at least 40% above that of climatology were achieved for all three locations, whereas skill level derived from forecasts using general circulation models is generally only comparable to climatology at the long-lead time of 8 months.
Artificial neural network, flooding, forecast, Murray Darling Basin, rainfall, skill score
[1] Australian Bureau of Statistics, Water and the Murray–Darling Basin – a statistical profile, 2008, available at http://www.abs.gov.au/ausstats/abs@.nsf/mf/4610.0.55.007
[2] Ashton, D. & Oliver, M., Irrigation technology and water use on farms in the Murray-Darling Basin 2006–2007 to 2011–2012. ABARES Research Report 14.03, 2014.
[3] Gell, P., Mills, K. & Grundell, R., Challenges to management of environment and agriculture: A legacy of climate and catchment change: the real challenge for wetland management. Hydrobiologia, 708(1), pp. 133–144, 2013. https://doi.org/10.1007/s10750-012-1163-4
[4] Gallant, A.J.E., Reeder, M.J., Risbey, J.S. & Henessey, K.J., The characteristics of seasonal-scale droughts in Australia, 1911–2009. International Journal of Climatology, 33(7), pp. 1658–1672, 2013. https://doi.org/10.1002/joc.3540
[5] Pook, M., Lisson, S., Risbey, J., Ummenhofer, C.C., McIntosh, P. & Rebbeck, M., The autumn break for cropping in southeast Australia: trends, synoptic influences and impacts on wheat yield. International Journal of Climatology, 29, pp. 2012–2026, 2009. https://doi.org/10.1002/joc.1833
[6] Vizard, A.L., & Anderson, G.A., The resolution and potential value of Australian seasonal rainfall forecasts based on the five phases of the Southern Oscillation Index. Crop & Pasture Science, 60, pp. 230–239, 2009. https://doi.org/10.1071/CP08277
[7] Hammer, G.L, Holzworth, D.P. & Stone, R., The value of skill in seasonal climate forecasting to wheat crop management in a region with high climatic variability. Australian Journal of Agricultural Research, 47, pp. 717–737, 1996. https://doi.org/10.1071/AR9960717
[8] Ash, A., McIntosh, P., Cullen, B., Carberry, P. & Stafford Smith, M., Constraints and opportunities in applying seasonal climate forecasts in agriculture. Australian Journal of Agricultural Research, 58, pp. 952–965, 2007. https://doi.org/10.1071/AR06188
[9] Shao, Q. & Li, M., An improved statistical analogue downscaling procedure for seasonal precipitation forecast. Stochastic Environmental Research and Risk Assessment, 27, pp. 819–830, 2013. https://doi.org/10.1007/s00477-012-0610-0
[10] Hawthorne, S., Wang, Q.J., Schepen, A. & Robertson, D., Effective use of general circulation model outputs for forecasting monthly rainfalls to long lead times. Water Resources Research, 49, pp. 5427–5436, 2013. https://doi.org/10.1002/wrcr.20453
[11] Drosdowsky, W. & Chambers, L.E., Near-Global Sea Surface Temperature Anomalies as Predictors of Australian Seasonal Rainfall. Journal of Climate, 14, pp. 1677–1687, 2001. https://doi.org/10.1175/1520-0442(2001)014<1677:NACNGS>2.0.CO;2
[12] Fawcett, R.J.B. & Stone, R.C., A comparison of two seasonal rainfall forecasting systems for Australia. Australian Meteorological and Oceanographic Journal, 60, pp. 15–24, 2010. https://doi.org/10.22499/2.6001.002
[13] Schepen, A., Wang, Q.J. & Robertson, D.E., Seasonal forecasts of Australian rainfall through calibration and bridging of coupled GCM outputs. Monthly Weather Review, 142, pp. 1758–1777, 2014. https://doi.org/10.1175/MWR-D-13-00248.1
[14] Schepen, A., Wang, Q.J. & Robertson, D., Evidence for using lagged climate indices to forecast Australian seasonal rainfall. Journal of Climate, 25, pp. 1230–1246, 2012. https://doi.org/10.1175/JCLI-D-11-00156.1
[15] Murphy, B.F. & Ribbe, J., Variability of Southeastern Queensland Rainfall and climate indices international. Journal of Climatology, 24, pp. 703–721, 2004. https://doi.org/10.1002/joc.1018
[16] Risbey, J.S., Pook, M.J., Mcintosh, P.C., Wheeler, M.C. & Hendon, H.H., On the remote drivers of rainfall variability in Australia. Monthly Weather Review, 137, pp. 3233–3253, 2009. https://doi.org/10.1175/2009MWR2861.1
[17] Ramseyer, C.A. & Mote, T.L., Atmospheric controls on Puerto Rico precipitation using artificial neural networks. Climate Dynamics, 47(7), pp. 2515–2526, 2016. https://doi.org/10.1007/s00382-016-2980-3
[18] Dahamsheha, A. & Aksoy, H., Artificial neural network models for forecasting intermittent monthly precipitation in arid regions. Applied Meteorology, 16, pp. 325–337, 2009. https://doi.org/10.1002/met.127
[19] Nayak, D.R., Mahapatra, A. & Mishra, P., A survey on rainfall prediction using artificial neural network. International Journal of Computer Applications, 72(16), pp. 32–40, 2013. https://doi.org/10.5120/12580-9217
[20] Abbot, J. & Marohasy J., Input selection and optimisation for monthly rainfall forecasting in Queensland, Australia, using artificial neural networks. Atmospheric Research, 138, pp. 166–178, 2014. https://doi.org/10.1016/j.atmosres.2013.11.002
[21] Abbot, J. & Marohasy, J., The potential benefits of using artificial intelligence for monthly rainfall forecasting for the Bowen Basin, Queensland, Australia. Water Resources Management VII. WIT Transactions on Ecology and the Environment, 171, pp. 287–297, 2013. https://doi.org/10.2495/WRM130261
[22] Abbot, J. & Marohasy, J., Using lagged and forecast climate indices with artificial intelligence to predict monthly rainfall in the Brisbane Catchment, Queensland, Australia. International Journal of Sustainable Development and Planning, 10(1), pp. 29–41, 2015. https://doi.org/10.2495/SDP-V10-N1-29-41
[23] Abbot, J. & Marohasy, J., Application of artificial neural networks to rainfall forecasting in Queensland, Australia. Advances in Atmospheric Sciences, 29(4), pp. 717–730, 2012. https://doi.org/10.1007/s00376-012-1259-9
[24] Vaze, J., Teng, J. & Chiew, F.H.S., Assessment of GCM simulations of annual and seasonal rainfall and daily rainfall distribution across south-east Australia. Hydrological Processes, 25, pp. 1486–1497, 2011. https://doi.org/10.1002/hyp.7916
[25] Langford, S. & Hendon, H.H., Improving reliability of coupled model forecasts of Australian Seasonal Rainfall. Monthly Weather Review, 141, pp. 728–741, 2013. https://doi.org/10.1175/MWR-D-11-00333.1
[26] Chowdhury, R.K. & Beecham, S., Australian rainfall trends and their relation to the southern oscillation index. Hydrological Processes, 24, pp. 504–514, 2010. https://doi.org/10.1029/2007GL030854
[27] Power, S.B. & Smith, I.M., Weakening of the walker circulation and apparent dominance of El Nino both reach record levels, but has ENSO really changed? Geophysical Research Letters, 34, L18702, 2007.
[28] Cai, W., Whetton, P.J. & Pittock, A.B., Fluctuations of the relationship between ENSO and northeast Australian rainfall. Climate Dynamics 17, pp. 421–432, 2001. https://doi.org/10.1007/PL00013738
[29] Cai, W., et al., Asymmetry in ENSO teleconnection with regional rainfall, its multidecadal variability, and impact. Journal of Climate, 23, pp. 4944–4955, 2010. https://doi.org/10.1175/2010JCLI3501.1
[30] Power, S.B., et al., Interdecadal modulation of the impact of ENSO on Australia. Climate Dynamics, 15, pp. 319–324, 1999. https://doi.org/10.1007/s003820050284
[31] Power, S.B., et al., The predictability of interdecadal changes in ENSO activity and ENSO teleconnections. Journal of Climate, 19, pp. 4755–4771, 2006. https://doi.org/10.1175/JCLI3868.1
[32] Verdon, D.C. & Franks, S.W., Long-term behaviour of ENSO: interactions with the PDO over the past 400 years inferred from paleoclimate records. Geophysical Research Letters, 33, L06712, 2006. https://doi.org/10.1029/2005GL025052
[33] Chiew, F.H.S., et al., El Nino/Southern oscillation and Australian Rain, streamflow and drought: links and potential for forecasting. Journal of Hydrology, 204, pp. 138–149, 1998. https://doi.org/10.1016/S0022-1694(97)00121-2
[34] Kirono, D.G.C., Chiew, F.H.S. & Kent, D.M., Identification of best predictors for forecasting seasonal rainfall and runoff in Australia. Hydrological Processes, 24, pp. 1237–1247, 2010. https://doi.org/10.1002/hyp.7585
[35] Abbot, J. & Marohasy, J., Forecasting of monthly rainfall in the Murray Darling Basin, Australia: Miles as a case study. River Basin Management VIII Book Series: WIT Transactions on Ecology and the Environment, 197, pp. 149–159, 2015.
https://doi.org/10.2495/rm150141
[36] Mekanik, F., Imteaz, M.A., Gato-Trinidad, S. & Elmahdi, A., Multiple regression and Artificial Neural Network for long-term rainfall forecasting using large scale climate modes. Journal of Hydrology, 503, pp. 11–21, 2013. https://doi.org/10.1016/j.jhydrol.2013.08.035
[37] Zimmerman, B.G., Vimont, D. J. & Block, P. J., Utilizing the state of ENSO as a means for season-ahead predictor selection. Water Resources Research, 52(5), pp. 3761–3774, 2016. https://doi.org/10.1002/2015WR017644
[38] Tiwari, P.R., Kar, S.C., Mohanty, U.C., et al., Skill of precipitation prediction with GCMs over north India during winter season. International Journal of Climatology, 34(12), pp. 3440–3455, 2014. https://doi.org/10.1002/joc.3921
[39] Schepen, A. & Wang, Q. J., Ensemble forecasts of monthly catchment rainfall out to long lead times by post-processing coupled general circulation model output. Journal of Hydrology, 519, pp. 2920–2931, 2014. https://doi.org/10.1016/j.jhydrol.2014.03.017
[40] Schepen, A. & Wang, Q. J., Toward accurate and reliable forecasts of Australian seasonal rainfall by calibrating and merging multiple coupled GCMs. Monthly Weather Review, 141(12), pp. 4554–4563, 2013. https://doi.org/10.1175/MWR-D-12-00253.1
[41] Acharya, N., Chattopadhyay, S., Mohanty, U.C. & Ghosh, K., Prediction of Indian summer monsoon rainfall: a weighted multi-model ensemble to enhance probabilistic forecast skills. Meteorological Applications, 21, pp. 724–732, 2014.
https://doi.org/10.1002/met.1400
[42] Lang, X. & Wang, H., Improving extraseasonal summer rainfall prediction by merging information from GCMs and observations. Weather and Forecasting, 25(4), pp. 1263–1274, 2010. https://doi.org/10.1175/2010WAF2222342.1
[43] Ying, L. & Ke, F., Improve the prediction of summer precipitation in the Southeastern China by a hybrid statistical downscaling model. Meteorology and Atmospheric Physics, 117(3–4), pp. 121–134, 2012. https://doi.org/10.1007/s00703-012-0201-0
[44] Sun, J. & Chen, H., A statistical downscaling scheme to improve global precipitation forecasting. Meteorology and Atmospheric Physics, 117(3), pp. 87–102, 2012. https://doi.org/10.1007/s00703-012-0195-7
[45] Wang, Q.J., Schepen, A. & Robertson, D.E., Merging seasonal rainfall forecasts from multiple statistical models through Bayesian model averaging. Journal of Climate, 25(16), pp. 5524–5537, 2012. https://doi.org/10.1175/JCLI-D-11-00386.1
[46] Halide, H. & Ridd, P., Complicated ENSO models do not significantly outperform very simple ENSO models. International Journal of Climatology, 28, pp. 219–233, 2008. https://doi.org/10.1002/joc.1519
[47] Abbot, J. & Marohasy, J., Forecasting monthly rainfall in the Bowen Basin of Queensland, Australia, using neural networks with Niño Indices for El Niño-Southern Oscillation. Lecture Notes in Artificial Intelligence, 9992, pp. 88–100, 2016.
[48] Duan, W. & Zhang, R., Is model parameter error related to a significant spring predictability barrier for El Nino events? Results from a theoretical model. Advances in Atmospheric Science, 27(5), pp. 1003–1013, 2010. https://doi.org/10.1007/s00376-009-9166-4
[49] Yan, L. & Yu, Y., The spring prediction barrier in ENSO hindcast experiments using the FGOALS-g model. Chinese Journal of Oceanology Limnology, 30(6), pp. 1093–1104, 2012. https://doi.org/10.1007/s00343-012-1271-z
[50] Coventry, W.L. & Dalgleish, L.I., Farmers’ accuracy interpreting seasonal climate forecast probability. International Journal of Climatology, 34(6), pp. 2097–2107, 2014. https://doi.org/10.1002/joc.3825
[51] http://www.bom.gov.au/jsp/sco/archive/index.jsp?map=rain&outlook=probmedian&period=season1&year=2016&month=05&day=26
[52] Woodburn, J., Rain wipes out NSW crops in third wettest winter on record. ABC News, available at
http://www.abc.net.au/news/2016-09-02/rain-wipes-out-nsw-crops-in-third-wettestwinter-on-record/7810722, September 2, 2016.
[53] Hoerr, K., NSW flooding leaves rural town of Deniliquin devastated, community set to feel impact for months. ABC News, available at http://www.abc.net.au/news/2016-10-16/nsw-floodingleaves-rural-town-of-deniliquin-devastated/7935136, October 16, 2016.