Abstract
This paper develops and applies the minimum relative entropy (MRE) theory with spectral power as a random variable for streamflow forecasting. The MRE theory consists of (1) hypothesizing a prior probability distribution for the random variable, (2) determining the spectral power distribution, (3) extending the autocorrelation function, and (4) doing forecasting. The MRE theory was verified using streamflow data from the Mississippi River watershed. The exponential distribution was chosen as a prior probability in applying the MRE theory by evaluating the historical data of the Mississippi River. If no prior information is given, the MRE theory is equivalent to the Burg entropy (BE) theory. The spectral density obtained by the MRE theory led to higher resolution than did the BE theory. The MRE theory did not miss the largest peak at 1/12th frequency, which is the main periodicity of streamflow of the Mississippi River, but the BE theory sometimes did. The MRE theory was found to be capable of forecasting monthly streamflow with a lead time from 12 to 48 months. The coefficient of determination (r 2) between observed and forecasted stream flows was 0.912 for Upper Mississippi River and was 0.855 for Lower Mississippi River. Both MRE and BE theories were generally more reliable and had longer forecasting lead times than the autoregressive (AR) method. The forecasting lead time for MRE and BE could be as long as 48–60 months, while it was less than 48 months for the AR method. However, BE was comparable to MRE only when observations fitted the AR process well. The MRE theory provided more reliable forecasts than did the BE theory, and the advantage of using MRE is more significant for downstream flows with irregular flow patterns or where the periodicity information is limited. The reliability of monthly streamflow forecasting was the highest for MRE, followed by BE followed by AR.
Similar content being viewed by others
References
Adamowski JF (2008) River flow forecasting using wavelet and cross-wavelet transform models. Hydrol Process 22:4877–4891. doi:10.1002/hyp.7107
Adamowski J, Prokoph A (2014) Determining the amplitude and timing of streamflow discontinuities: A cross wavelet analysis approach. Hydrol Process 28:2782–2793
Andreo B, Jimenez P, Duran JJ, Carrasco F, Vadillo I, Mangin A (2006) Climatic and hydrological variations during the last 117-166 years in the south of the Iberian Peninsula, from spectral and correlation analyses and continuous wavelet analyses. J Hydrol 324:24–39
Baldwin CK, Lall U (1999) Seasonality of streamflow: the upper Mississippi River. Water Resour Res 35:1143–1154
Berger A, Melice JL, Van Der Mooren I, Beran A, Provenzale A, Stanford JL (1990) Evolutive spectral analysis of sunspot data over the past 300 years [and discussion]. Phil Trans R Soc A 330:529–541
Boshnakov GN, Lambert-Lacroix S (2012) A periodic Levinson-Durbin algorithm for entropy maximization. Comput Stat Data Anal 56:15–24. doi:10.1016/j.csda.2011.07.001
Box GEP, Jenkins GM (1970) Time series analysis, forecasting and control. Holden-Day series in time series analysis. Holden-Day, San Francisco
Burg JP (1967) Maximum entropy spectral analysis. In: Proceedings of 37th meeting of society exploration geophysics, pp 34–41
Burg JP (1975) Maximum entropy spectral analysis. Ph.D., Stanford University
Cengiz TM (2011) Periodic structures of Great Lakes levels using wavelet analysis. J Hydrol Hydromech 59:24–35
Chiew FHS, McMahon TA (2002) Global ENSO-streamflow teleconnection, streamflow forecasting and interannual variability. Hydrolog Sci J 47:505–522
Cui H, Singh VP (2015) Configurational entropy theory for streamflow forecasting. J Hydrol 521:1–17. doi:10.1016/j.jhydrol.2014.11.065
Cui H, Singh VP (2016) Maximum entropy spectral analysis for streamflow forecasting. Physica A 442:91–99. doi:10.1016/j.physa.2015.08.060
Durdu OF (2010) Application of linear stochastic models for drought forecasting in the Buyuk Menderes river basin, western Turkey. Stoch Environ Res Risk A 24:1145–1162. doi:10.1007/s00477-010-0366-3
Endo H, Randall RB (2007) Enhancement of autoregressive model based gear tooth fault detection technique by the use of minimum entropy deconvolution filter. Mech Syst Signal Process 21:906–919
Feng X, Porporato A, Rodriguez-Iturbe I (2013) Changes in rainfall seasonality in the tropics Nat. Clim Change 3:811–815
Fernandez C, Vega JA, Fonturbel T, Jimenez E (2009) Streamflow drought time series forecasting: a case study in a small watershed in North West Spain. Stoch Environ Res Risk A 23:1063–1070
Fleming SW, Lavenue AM, Aly AH, Adams A (2002) Practical applications of spectral analysis to hydrologic time series. Hydrol Process 16:565–574
Ghil M et al (2002) Advanced spectral methods for climatic time series. Rev Geophys 40:1003. doi:10.1029/2000rg000092
Gray RM, Gray AH, Rebolledo G, Shore JE (1981) Rate-distortion speech coding with a minimum discrimination information distortion measure. IEEE Trans Inf Theory 27:708–721
Hameed S (1984) Fourier-analysis of nile flood levels. Geophys Res Lett 11:843–845
Hipel KW, McLeod AI (1994) Time series modelling of water resources and environmental systems. In: Developments in water science, vol 45. Elsevier, New York
Hou ZS, Rubin Y (2005) On minimum relative entropy concepts and prior compatibility issues in vadose zone inverse and forward modeling. Water Resour Res 41
Jimenez C, Mcleod AI, Hipel KW (1989) Kalman filter estimation for periodic autoregressive-moving average models. Stoch Hydrol Hydraul 3:227–240
Kalman PE (1960) A new approach to linear filtering and prediction problems. J Basic Eng 83D:95–108
Kisi O (2009) Neural networks and wavelet conjunction model for intermittent streamflow forecasting. J Hydrol Eng 14:773–782. doi:10.1061/(asce)he.1943-5584.0000053
Kisi O (2010) Wavelet regression model for short-term streamflow forecasting. J Hydrol 389:344–353. doi:10.1016/j.jhydrol.2010.06.013
Krstanovic PF, Singh VP (1991a) A univariate model for long-term streamflow forecasting.1. Dev Stoch Hydrol Hydraul 5:173–188
Krstanovic PF, Singh VP (1991b) A univariate model for long-term streamflow forecasting.2. Appl Stoch Hydrol Hydraul 5:189–205
Kuhnel I, Mcmahon TA, Finlayson BL, Haines A, Whetton PH, Gibson TT (1990) Climatic influences on streamflow variability—a comparison between Southeastern Australia and Southeastern United-States-of-America. Water Resour Res 26:2483–2496
Kullback S (1959) Information theory and statistics. Wiley, New York
Labat D (2005) Recent advances in wavelet analyses: Part I. A review of concepts. J Hydrol 314:275–288
Labat D, Ronchail J, Guyot JL (2005) Recent advances in wavelet analyses: Part 2—Amazon, Parana, Orinoco and Congo discharges time scale variability. J Hydrol 314:289–311
Liefhebber F, Boekee DE (1987) Minimum information spectral-analysis. Signal Process 12:243–255
Marques CAF, Ferreira JA, Rocha A, Castanheira JM, Melo-Goncalves P, Vaz N, Dias JM (2006) Singular spectrum analysis and forecasting of hydrological time series. Phys Chem Earth 31:1172–1179. doi:10.1016/j.pce.2006.02.061
Mishra AK, Desai V (2005) Drought forecasting using stochastic models. Stoch Environ Res Risk A 19:326–339
Modarres R (2007) Streamflow drought time series forecasting. Stoch Environ Res Risk A 21:223–233
Molenat J, Davy P, Gascuel-Odoux C, Durand P (1999) Study of three subsurface hydrologic systems based on spectral and cross-spectral analysis of time series. J Hydrol 222:152–164. doi:10.1016/S0022-1694(99)00107-9
Moss ME, Bryson MC (1974) Autocorrelation structure of monthly streamflows. Water Resour Res 10:737–744
Nasrollahi N et al (2015) How well do CMIP5 climate simulations replicate historical trends and patterns of meteorological droughts? Water Resour Res 51:2847–2864. doi:10.1002/2014WR016318
Noakes DJ, McLeod AI, Hipel KW (1985) Forecasting monthly riverflow time series. Int J Forecast 1:179–190
Papademetriou RC (1998) Experimental comparison of two information-theoretic spectral estimators. In: Signal processing proceedings, vol 141, pp 141–144. ICSP ‘98. Fourth international conference, 1998. doi:10.1109/ICOSP.1998.770170
Pramanik N, Panda RK, Singh A (2011) Daily river flow forecasting using wavelet ANN hybrid models. J Hydroinf 13:49–63. doi:10.2166/hydro.2010.040
Prokoph A, Adamowski J, Adamowski K (2012) Influence of the 11 year solar cycle on annual streamflow maxima in Southern Canada. J Hydrol 442:55–62
Rohr T, Manzoni S, Feng X, Menezes RSC, Porporato A (2013) Effect of rainfall seasonality on carbon storage in tropical dry ecosystems. J Geophys Res 118:1156–1167. doi:10.1002/jgrg.20091
Salas JD, Boes DC, Smith RA (1982) Estimation of ARMA models with seasonal parameters. Water Resour Res 18:1006–1010. doi:10.1029/Wr018i004p01006
Sang YF, Wang D, Wu JC, Zhu QP, Wang L (2009) The relation between periods’ identification and noises in hydrologic series data. J Hydrol 368:165–177. doi:10.1016/j.jhydrol.2009.01.042
Sang YF, Wang ZG, Liu CM (2012) Period identification in hydrologic time series using empirical mode decomposition and maximum entropy spectral analysis. J Hydrol 424:154–164. doi:10.1016/j.jhydrol.2011.12.044
Schroeder MR (1982) Linear prediction, extermal entropy and prior information in speech signal analysis and synthesis. Speech Commun 1:9–20. doi:10.1016/0167-6393(82)90004-8
Shannon CE (1948) A mathematical theory of communication. At&T Tech J 27(4):623–656
Sharma A (2000) Seasonal to interannual rainfall probabilistic forecasts for improved water supply management: Part 3—a nonparametric probabilistic forecast model. J Hydrol 239:249–258
Sharma A, O’Neill R (2002) A nonparametric approach for representing interannual dependence in monthly streamflow sequences. Water Resour Res 38:1100. doi:10.1029/2001WR000953
Sharma A, Tarboton D, Lall U (1997) Streamflow simulation: a nonparametric approach. Water Resour Res 33:291–308. doi:10.1029/96WR02839
Sharma A, Luk KC, Cordery I, Lall U (2000) Seasonal to interannual rainfall probabilistic forecasts for improved water supply management: Part 2—Predictor identification of quarterly rainfall using ocean-atmosphere information. J Hydrol 239:240–248
Shiri J, Kisi O (2010) Short-term and long-term streamflow forecasting using a wavelet and neuro-fuzzy conjunction model. J Hydrol 394:486–493. doi:10.1016/j.jhydrol.2010.10.008
Shore JE (1979) Minimum cross-entropy spectral analysis. Naval Research Laboratory, Washington, DC
Shore JE (1981) Minimum cross-entropy spectral-analysis. IEEE Trans Acoust Speech 29:230–237
Shore JE, Gray RM (1982) Minimum cross-entropy pattern-classification and cluster-analysis. IEEE Trans Pattern Anal 4:11–17
Singh VP (1997) The use of entropy in hydrology and water resources. Hydrol Process 11:587–626
Smith LC, Turcotte DL, Isacks BL (1998) Stream flow characterization and feature detection using a discrete wavelet transform. Hydrol Process 12:233–249
Wang QJ, Robertson DE, Chiew FHS (2009) A Bayesian joint probability modeling approach for seasonal forecasting of streamflows at multiple sites. Water Resour Res 45
Westra S, Sharma A, Brown C, Lall U (2008) Multivariate streamflow forecasting using independent component analysis. Water Resour Res 44
Woodbury AD (2004) A FORTRAN program to produce minimum relative entropy distributions. Comput Geosci-Uk 30:131–138. doi:10.1016/j.cageo.2003.09.001
Woodbury AD (2011) Minimum relative entropy, Bayes and Kapur. Geophys J Int 185:181–189. doi:10.1111/j.1365-246X.2011.04932.x
Woodbury AD, Ulrych TJ (1993) Minimum relative entropy—forward probabilistic modeling. Water Resour Res 29:2847–2860
Woodbury AD, Ulrych TJ (1996) Minimum relative entropy inversion: theory and application to recovering the release history of a groundwater contaminant. Water Resour Res 32:2671–2681. doi:10.1029/95wr03818
Woodbury AD, Ulrych TJ (1998) Minimum relative entropy and probabilistic inversion in groundwater hydrology. Stoch Hydrol Hydraul 12:317–358
Wu CL, Chau KW (2010) Data-driven models for monthly streamflow time series prediction. Eng Appl Artif Intell 23:1350–1367
Wu CL, Chau KW, Li YS (2009) Predicting monthly streamflow using data-driven models coupled with data-preprocessing techniques. Water Resour Res 45:W08432. doi:10.1029/2007wr006737
Yule GU (1927) On a method of investigating periodicities in disturbed series, with special reference to Wofer’s sunspot numbers. Philos Trans R Soc Lond 226:267–298
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix 1: Posterior spectral density without prior information given
If no prior information is given, one can consider the uniform distribution as \(p(\mathop x\limits^{ \to } ) = C\)and let \(A = p(\mathop x\limits^{ \to } )\exp [ - \lambda_{0} ]\) in Eq. (14), then it becomes
To simplify Eq. (35), let \(u_{k} = \sum\limits_{r = - N}^{N} {\lambda_{r} c_{rk} }\). Then, Eq. (35) becomes
Integrating Eq. (36) over 0 to infinite yields
for \(\int {\exp ( - u_{k} x_{k} )} dx_{k} = - \frac{1}{{u_{k} }}\exp ( - u_{k} x)\left| {_{0}^{\infty } } \right. = \frac{1}{{u_{k} }}\). Rearranging Eq. (37) yields
Thus, inputting Eq. (38) into Eq. (35), one obtains
It is shown that Eq. (39) is a multivariate exponential distribution, where the probability of the spectral power at each frequency is \(q(x_{k} ) = u_{k} \exp [ - u_{k} x_{k} ]\). Thus, the expected spectral power at each frequency can be obtained as
which is equivalent to the Burg entropy.
Appendix 2: Posterior spectral density with exponential prior distribution
If prior information points to an exponential distribution, with the expected spectrum power S k for each frequency, then the prior becomes
Inputting Eq. (41) into the posterior distribution Eq. (14), the result becomes
where \(u_{k} = \sum\limits_{r = - N}^{N} {\lambda_{r} c_{rk} }\), the same as previously defined. It can be seen from Eq. (42) that the posterior distribution of spectral power is in the form that is transformable to the exponential distribution of \(\eta \exp ( - \eta x)\), where \(u_{k} + \frac{1}{{S_{k} }}\) can be considered as the exponential parameter of each distribution. Thus, λ0 needs to satisfy that \(\exp ( - \lambda_{0} )\prod\limits_{k} {\frac{1}{{S_{k} }}} = \prod\limits_{k} {u_{k} + \frac{1}{{S_{k} }}}\) so that Eq. (42) forms the multivariate exponential distribution as
and the posterior distribution of each frequency is
Thus, the expected posterior power spectral power at each frequency becomes
Appendix 3: Linear extension of autocovariance
Recall that in time series, the AR process or the linear prediction process has the spectral density in the form of
Taking the inverse Fourier transformation of Eq. (46), one obtains the autocovariance function like in Eq. (7) as
Multiplying Eq. (47) by a * r and summing up for n from 0 to N,
It is noted from Cauchy’s integral formula that
Thus, integrating of Eq. (49) using Cauchy’s formula, one obtains
Taking the conjugate of Eq. (50a, b) yields
which can be written in the matrix form as:
Equation (52) is the Yule-Walker equation (Yule 1927), with coefficient a n and gain σ 2, and according to the linear extension, the N + 1th autocovariance can be obtained by
Rights and permissions
About this article
Cite this article
Cui, H., Singh, V.P. Application of minimum relative entropy theory for streamflow forecasting. Stoch Environ Res Risk Assess 31, 587–608 (2017). https://doi.org/10.1007/s00477-016-1306-7
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00477-016-1306-7