Anda di halaman 1dari 8

Forecasting Palmer Index Using Neural Networks

and Climatic Indexes


P. Cutore, Ph.D.1; G. Di Mauro, Ph.D.2; and A. Cancelliere3
Downloaded from ascelibrary.org by Universidad Politecnica De Valencia on 06/05/19. Copyright ASCE. For personal use only; all rights reserved.

Abstract: In recent years, several drought monitoring indexes have found application to describe and compare droughts among different
time periods and regions as well as to forecast the evolution of ongoing droughts, in order to select appropriate mitigation measures and
policies for water resources management under shortage risk conditions. However, limited efforts have been made to investigate the
possibilities of using information conveyed by large-scale climatic indexes to improve the forecasting ability of drought indexes, provided
they exert some influence on the climatic variability in a region. The aim of the paper is to develop models for forecasting Palmer
Hydrological Drought Index series in Sicily 共Italy兲 based on artificial neural networks, and to extend such models in order to include
information from large-scale climatic indexes. First, the influence of North Atlantic Oscillation 共NAO兲 and European Blocking 共EB兲
indexes on Palmer index series, computed on areal monthly precipitation from 1955 until 1999 in Sicily, has been investigated by means
of a correlation analysis. Results indicate that NAO and EB series are significantly correlated with Palmer index series for winter and
autumn months, with special reference to the last decades. Then, forecasting models based on neural networks have been developed, using
different approaches. The comparison between the prediction for winter and autumn months obtained by either including or not including
the NAO and EB indexes within the forecasting model indicates some improvements in terms of R2 when NAO and especially EB are
considered. A different model behavior has been observed for the spring and summer predictions for which no significant improvements
in terms of model predictive capability, due to the introduction of the climatic indexes as input variables, have been observed.
DOI: 10.1061/共ASCE兲HE.1943-5584.0000028
CE Database subject headings: Forecasting; Climatology; Water resources; Droughts; Water supply; Neural networks.

Introduction Several drought indexes have been proposed for drought


monitoring 共Heim 2000兲, among which the Standardized Precipi-
Drought appears as a severe deficiency of water available in the tation Index 共SPI兲 共McKee et al. 1993兲 and the Palmer Drought
hydrologic cycle over an extended period of time and character- Severity Index 共PDSI兲 共Palmer 1965兲, or the slightly modified
ized by a large spatial extent. Like other climatic extreme events,
version of the index, the Palmer hydrological drought index
such as earthquakes, tornadoes, hurricanes, and floods, droughts
共PHDI兲, probably have found the most widespread application.
can cause significant damage both in natural environments and in
Although most of the proposed indexes have been developed with
human lives.
the intent to monitor current drought conditions, some of them
However, drought is a gradual phenomenon, slowly affecting
can be used to forecast the possible evolution of an ongoing
an area and tightening its grip with time. In severe cases, drought
can last for many years and can have devastating effects on agri- drought in order to adopt appropriate mitigation measures and
culture and water supplies. Nonetheless, the slow evolution of drought policies for water resources management under shortage
drought makes possible an effective mitigation of its impacts, risk conditions.
provided a timely monitoring and forecast of the phenomenon is Several authors have proposed methods that enable forecasts
available 共Wilhite 2000; Rossi 2000; Cancelliere et al. 2007兲. To or assessments of the probable evolution of SPI or Palmer in-
this end, the selection of adequate indexes for drought monitor- dex. Rao and Padmanabhan 共1984兲 investigated the stochastic
ing, which provide a synthetic and objective description of features of yearly and monthly PHDI and fitted stochastic models
drought conditions, can represent a key point for the implemen- to simulate and forecast PHDI series. Cancelliere et al. 共1996兲
tation of an efficient drought watch system. proposed a procedure for short- to midterm forecasting of PHDI
and tested its applicability to Mediterranean regions by com-
1
Dept. of Civil and Environmental Engineering, Univ. of Catania, puting the probability that an ongoing drought will end in the
Viale Andrea Doria 6, 95125 Catania, Italy. E-mail: pcutore@dica.unict.it following months. Other authors 共Lohani et al. 1998兲 proposed a
2
Dept. of Civil and Environmental Engineering, Univ. of Catania, forecasting procedure of the Palmer index based on the first-
Viale Andrea Doria 6, 95125 Catania, Italy. E-mail: gdimau@dica.unict.it order Markov chains, which enables the forecast of drought
3
Associate Professor, Dept. of Civil and Environmental Engineering, conditions for future months based on the current drought class
Univ. of Catania, Viale Andrea Doria 6, 95125 Catania, Italy. E-mail: described by PHDI values. Recently, Kim and Valdes 共2003兲
acance@dica.unict.it have used PDSI as a drought parameter to forecast drought using
Note. This manuscript was submitted on April 18, 2008; approved on
a conjunction of dyadic wavelet transforms and neural network.
September 18, 2008; published online on February 16, 2009. Discussion
period open until November 1, 2009; separate discussions must be sub- More recently, Moreira et al. 共2006兲 have analyzed SPI drought
mitted for individual papers. This paper is part of the Journal of Hydro- class transitions using loglinear models, whereas Cancelliere et al.
logic Engineering, Vol. 14, No. 6, June 1, 2009. ©ASCE, ISSN 1084- 共2007兲 have proposed a stochastic model to forecast SPI values
0699/2009/6-588–595/$25.00. at short to medium term, as well as to estimate transition prob-

588 / JOURNAL OF HYDROLOGIC ENGINEERING © ASCE / JUNE 2009

J. Hydrol. Eng., 2009, 14(6): 588-595


abilities ofSPI classes corresponding to drought of different Table 1. Dry and Wet Condition Classifications Based on the PHDI
severities. Values 共Karl 1986兲
Despite the fact that forecasting when a drought is likely to Index values Classification
begin or to come to an end is still extremely difficult 共Cordery
PHDI艌 4.0 Extremely wet
and McCall 2000兲, important progress is being made in relation
to the possibilities of using information provided by large- 3.0⬍ PHDI⬍ 4.0 Very wet
scale climatic indexes, such as the North Atlantic Oscillation 2.0⬍ PHDI⬍ 3.0 Moderately wet
共NAO兲 共Walker and Bliss 1932兲 and the European Blocking 共EB兲 1.0⬍ PHDI⬍ 2.0 Slightly wet
共Wallace and Gutzler 1981兲, as a support to forecasting droughts. 0.5⬍ PHDI⬍ 1.0 Incipient wet spell
Indeed, use of such indexes could allow for an improved forecast- −0.5⬍ PHDI⬍ 0.5 Near normal
ing ability of the models, as well as for a longer time horizon of −1.0⬍ PHDI⬍ −0.5 Incipient drought
Downloaded from ascelibrary.org by Universidad Politecnica De Valencia on 06/05/19. Copyright ASCE. For personal use only; all rights reserved.

forecasting, provided they exert some influence on the climatic −2.0⬍ PHDI⬍ −1.5 Mild drought
variability in a region. −3.0⬍ PHDI⬍ −2.0 Moderate drought
The aim of the paper is to develop models for forecasting −4.0⬍ PHDI⬍ −3.0 Severe drought
PHDI series in Sicily 共Italy兲 based on artificial neural networks PHDI⬍ −4.0 Extreme drought
共ANNs兲, and extending such models in order to include informa-
tion deriving from large scale climatic indexes. In particular, an
evaluation of whether considering NAO and EB indexes informa-
tion leads to improvements in the forecasting PHDI values is A drought classification proposed by Karl 共1986兲 for the
carried out by means of Monte Carlo analysis. United States and reported in Table 1, indicates extreme drought
The PHDI was selected due to its capability to assess water for PHDI艋 −4.0 with an approximate cumulative frequency
deficiencies at large geographical scale allowing a reasonable less than 4%, severe drought for −3.9艋 PHDI艋 −3.0 共frequency
comparison of drought conditions both in space and time, 5–10%兲, and mild to moderate drought for −2.9艋 PHDI艋 −1.5
whereas the adoption of ANNs is justified by their ability in mod- 共frequency 11–27%兲. The index can also express wet conditions;
eling and forecasting nonlinear and nonstationary time series in i.e., 0 is normal, +2 is moderate rainfall, etc.
hydrology, due to their intrinsic nonlinear property and flexibility The advantage of the Palmer index is that it is standardized to
for modeling 共ASCE 2000a,b兲. local climate, so it can be applied to different countries to dem-
onstrate relative dry or wet conditions.
Although the Palmer index is mainly used in United States
Palmer Drought Index 共Lohani and Loganathan 1997; Kim and Valdes 2003兲, it has
been applied also to other areas such as Australia and South
The original PDSI was developed as a descriptor of meteorologi- Africa 共Smith et al. 1993兲, Europe 共Briffa et al. 1994兲, and Medi-
cal drought and its computation is based on the comparison be- terranean areas 共Cancelliere et al. 1996兲 confirming that, despite
tween the actual precipitation over an area with the amount of its limitations, PHDI can be considered an appropriate index for
precipitation required to maintain the average supply of that area drought monitoring.
共climatically appropriate for existing conditions兲 共Palmer 1965兲.
The method is based on the estimation of a soil water balance,
usually on a monthly basis, over a calibration period using his- North Atlantic Oscillation and European Blocking
torical records of hydrological variables to derive climatic coeffi- Effects on Precipitation and Droughts in Europe
cients needed for the computation of the monthly values of the
index. The NAO and EB have long been recognized as two of the pat-
As the evaluation of the original PDSI relies on a backtracking terns of atmospheric variability in the Northern Hemisphere.
procedure to determine the beginning and the end of the drought, However, they have become the subject of a wider interest only
it cannot be used for real time monitoring. Therefore, the slightly recently.
modified version of the index, the PHDI, is generally used for Historically, NAO has been defined as an index that measures
operational purposes and has been applied here. the difference of normalized surface pressure between two areas
Computation of PHDI is based on the application of the recur- over the North Atlantic Ocean, Ponta Delgada in the Azores and
sive formula Stykkisholmur in Iceland, but recently, researchers have shown
that, during the winter season, stations located in Iberia can be
Hi used with some advantages over Ponta Delgada 共Hurrel 1995;
PHi = 0.897PHi−1 + 共1兲 Jones et al. 1997兲.
3
Several studies have established links between the NAO
where i = 12共y − 1兲 + t; y = 1 , . . . , n indicates the year; t = 1 , . . . , 12 phases and precipitation in western Europe and the Mediterranean
the month; and basin 共Hurrell 1995; Qian et al. 2000兲 showing a significant de-
pendence between these two variables. Such control exerted by
Hi = 共Pi − P̂i兲Kt 共2兲 NAO on the precipitation regime is related to corresponding
changes in the associated activity of North-Atlantic storm tracks
Pi being the precipitation on interval i, P̂i the climatologically that affect the western European border 共Goodess and Jones
appropriate precipitation for existing conditions at the same inter- 2002兲. Over the past decade NAO has remained in a positive
val, and Kt a monthly standardization factor. phase during winter, contributing significantly to the recent
PHDI values are centered around zero and range approxi- wintertime warmth across Europe and to cold conditions in the
mately between −5.0 and 5.0 with negative values indicating northwestern Atlantic 共Hurrell et al. 2001兲. During the positive
drought of different severities. phases of NAO, the North Atlantic westerlies, which provide

JOURNAL OF HYDROLOGIC ENGINEERING © ASCE / JUNE 2009 / 589

J. Hydrol. Eng., 2009, 14(6): 588-595


much of the atmospheric moisture to north Africa and Europe, fitting conditions. The number of ANN hidden neurons was cho-
shift northward. This, in turn, results in drier conditions over sen to provide the best predictive capability in the validation
southern Europe, the Mediterranean Sea, and northern Africa phase.
共Hurrell 1995兲.
The EB index characterizes the frequency of occurrence of
blocking events for European longitudes by using the 500 hPa Preliminary Correlation Analysis between Palmer
geopotential height 共Z500兲 data and the Tibaldi and Molteni Index Series in Sicily and Climatic Indexes Series
共1990兲 index.
During winter, the anticyclones usually are located in the A preliminary correlation analysis between Palmer series in Sicily
Greenland and Scandinavian polar areas, whereas in summer they and climatic indexes series 共NAO and EB兲 has been carried out
characterize the European East. Such events deserve particular by means of Pearson and Kendall’s ␶ correlation coefficients.
Downloaded from ascelibrary.org by Universidad Politecnica De Valencia on 06/05/19. Copyright ASCE. For personal use only; all rights reserved.

attention because they could cause climatic anomalies by block- Palmer series have been computed from areal averaged
ing the western currents for several weeks, so that the Atlantic monthly precipitation series from 1926 until 2003, obtained by
perturbations are forced to abandon their usual trajectories. applying the Thiessen polygons method on 40 precipitation
Therefore, together with the NAO index, the EB index consti- stations in Sicily. The selected stations are included in the
tutes a second atmospheric circulation pattern for the European drought monitoring bulletin published on the website of the
area 共Rogers 1990; Tibaldi and Molteni 1990; Pavan et al. Sicilian Regional Agency for Waste and Water—Water Obser-
2002a,b兲. As NAO and EB are strongly correlated, respectively, vatory, formerly the Hydrographic Service 关Rossi and Cancelliere
with the first and third principal components of the Atlantic vari- 共2002兲, http://www.uirsicilia.it兴 in which drought indexes
ability, they are uncorrelated. 共SPI and Palmer兲 are currently computed for real time drought
The seasonal temperature and precipitation anomalies over the monitoring.
European area are deeply connected to these two large-scale Regarding the NAO series, following Jones et al. 共1997兲, it
patterns and high correlation among these patterns and precipita- was decided to adopt the Gibraltar-Iceland NAO index developed
tion for the Mediterranean and the Alpine areas was proved by by the Climatic Research Unit of the University of East Anglia,
Quadrelli et al. 共2001a,b兲. U.K., for the same period 共1926–2003兲 共http://gcmd.nasa.gov/
records兲. Finally, the EB series has been computed as the third
principal component of the Z500 from the reanalysis data 共NCEP兲
ANN Forecasting Model for the European area from 90°W to 60°E and from 20°N to
90°N. The available EB data set covers the observed period
ANNs have been extensively used in the past two decades for the 1955–1999 共Cacciamani et al. 2007兲.
development of a large number of forecasting models in many The correlation between the PHDI index in a given month and
scientific areas. ANNs are able to mimic a large class of nonlinear the climatic index 共either NAO or EB兲, averaged over the previ-
functions, and therefore they are ideal candidates to develop em- ous h 共h = 1 , . . . , 4兲, months has been investigated, in agreement
pirical 共regression type兲 models. An ANN model is based on a with the procedure already applied by Vicente-Serrano 共2005兲 and
connectionist approach to computation involving several transfor- Lopez-Moreno et al. 共2007兲.
mation elements 共neurons兲, interconnected and distributed over Let PHDIyj+k be the value of the Palmer index at month t + k
different layers. 共t = 1 , . . . , 12; k = 0 , . . . , 3兲 and year y, and Z共h兲
y,t the value of the
In general, ANNs consist of an input layer, through which climatic index 共either NAO or EB兲 at month t for a given time
information reaches the network, one or more hidden layers, into averaging scale h 共h = 1 , . . . , 4兲.
which information is routed, and a layer of output, through which The Pearson correlation coefficient between the two variables,
the results are provided 共ASCE 2000a,b兲. for a given time averaging scale h, has been computed as
The advantages of ANNs over other regressive models are
that 共1兲 their application does not require a prior knowledge of
the process because ANNs have black-box properties; 共2兲 ANNs 兺ny=1共PHDIy,t+k − PHDI兲共Zy,t − Z̄兲
are inherently nonlinear as neurons activate a nonlinear filter rt,k = 共3兲
called an activation function; and 共3兲 the generalizing capabilities 关兺ny=1共PHDIy,t+k − PHDI兲2兺ny=1共Zy,t − Z̄兲2兴0.5
of ANNs make them attractive for a large class of problems in
hydrology. in which PHDI and Z̄ represent the mean values for the consid-
In the present case, ANNs have been adopted to model the ered variables.
relationship between future values of PHDI index and present In order to compute Kendall’s ␶, let us consider the vector of
values of both PHDI and large-scale climatic indexes. Thus, in- variables 共PHDIy,t+k , Zy,t兲. Two distinct pairs of observations
dependent variables, i.e., an input vector consisting of monthly 共PHDI j,t+k , Z j,t兲 and 共PHDIm,t+k , Zm,t兲 are said to be concordant if
PHDI and climatic indexes data are used to generate the depen- 共PHDI j,t+k − PHDIm,t+k兲 共Z j,t − Zm,t兲 ⬎ 0, and discordant if
dent variable values, i.e., the output vector of monthly Palmer 共PHDI j,t+k − PHDIm,t+k兲 共Z j,t − Zm,t兲 ⬍ 0. Kendall’s ␶ has been then
index predictions. computed as ␶ = 共c − d兲 / 共c + d兲, where c denotes the number of
Single hidden layer feed-forward networks have been used concordant pairs, and d is the number of discordant pairs.
in the present work. Based on experiments performed on this case Correlation analysis has been carried out for each month t.
study, the logistic activation function for the neurons of input As an example, in Tables 2 and 3, correlation values between
and hidden layers and the linear function for the neurons of output PHDI and the considered climatic series are shown for the month
layer were chosen. Moreover, following Swingler 共1996兲 and of March considering the three periods of simultaneous observa-
Berry and Linoff 共1997兲, the maximum number of neurons in tion of different length 共i.e., 1955–1999, 1970–1999, and 1980–
the hidden layer has been set equal to double the input layer 1999兲.
dimension in order to reduce calibration errors and to avoid over- It is evident how the correlation significantly increases as the

590 / JOURNAL OF HYDROLOGIC ENGINEERING © ASCE / JUNE 2009

J. Hydrol. Eng., 2009, 14(6): 588-595


Table 2. Pearson and Kendall’s ␶ 共in Parentheses兲 Correlation Coefficient between NAO Series 共Averaged over 1 – 4 Months兲 and Palmer Series
Simultaneous and Shifted 1, 2, or 3 Months Ahead
Palmer
NAO averaging scale h
Period Month 共month兲 Same month 1 month ahead 2 months ahead 3 months ahead
1955–1999 March 1 −0.31 共−0.21兲 −0.22 共−0.15兲 −0.17 共−0.13兲 −0.17 共−0.11兲
2 −0.37 共−0.24兲 −0.33 共−0.18兲 −0.29 共−0.16兲 −0.29 共−0.16兲
3 −0.59 (−0.39) −0.55 (−0.34) −0.52 (−0.34) −0.50 (−0.34)
4 −0.56 (−0.40) −0.56 (−0.38) −0.51 (−0.34) −0.49 (−0.32)
1970–1999 March 1 −0.33 共−0.25兲 −0.24 共−0.18兲 −0.24 共−0.15兲 −0.24 共−0.15兲
2 −0.39 共−0.25兲 −0.33 共−0.17兲 −0.31 共−0.15兲 −0.33 共−0.17兲
Downloaded from ascelibrary.org by Universidad Politecnica De Valencia on 06/05/19. Copyright ASCE. For personal use only; all rights reserved.

3 −0.60 (−0.37) −0.58 (−0.34) −0.56 (−0.35) −0.55 (−0.36)


4 −0.61 (−0.43) −0.66 (−0.46) −0.62 (−0.45) −0.60 (−0.38)
1980–1999 March 1 −0.52 (−0.30) −0.51 (−0.31) −0.47 共−0.26兲 −0.48 共−0.28兲
2 −0.55 (−0.42) −0.45 (−0.32) −0.38 共−0.27兲 −0.39 (−0.32)
3 −0.79 (−0.46) −0.76 (−0.45) −0.69 (−0.38) −0.66 (−0.42)
4 −0.81 (−0.52) −0.82 (−0.61) −0.73 (−0.56) −0.69 (−0.49)
Note: Bold type= significant values for a significance level ␣ = 5%.

Table 3. Pearson and Kendall’s ␶ 共in Brackets兲 Correlation Coefficient between EB Series 共Averaged over 1 – 4 months兲 and Palmer Series Simultaneous
and Shifted 1, 2, or 3 Months Ahead
Palmer
EB averaging scale h
Period Month 共month兲 Same month 1 month ahead 2 months ahead 3 months ahead
1955–1999 March 1 0.40 共0.26兲 0.39 共0.25兲 0.39 共0.28兲 0.40 共0.28兲
2 0.25 共0.14兲 0.21 共0.12兲 0.19 共0.13兲 0.19 共0.14兲
3 0.37 共0.24兲 0.35 共0.25兲 0.37 共0.27兲 0.38 共0.29兲
4 0.44 (0.30) 0.43 (0.34) 0.44 (0.36) 0.45 (0.37)
1970–1999 March 1 0.44 (0.30) 0.41 共0.27兲 0.45 (0.30) 0.45 (0.30)
2 0.26 共0.16兲 0.24 共0.17兲 0.28 共0.18兲 0.32 共0.23兲
3 0.42 共0.28兲 0.42 (0.30) 0.49 (0.32) 0.52 (0.36)
4 0.51 (0.37) 0.52 (0.41) 0.60 (0.43) 0.62 (0.47)
1980–1999 March 1 0.59 (0.38) 0.60 (0.36) 0.63 (0.40) 0.65 (0.40)
2 0.42 (0.31) 0.43 (0.35) 0.42 (0.37) 0.45 (0.37)
3 0.62 (0.37) 0.64 (0.42) 0.68 (0.43) 0.72 (0.47)
4 0.63 (0.44) 0.65 (0.47) 0.71 (0.51) 0.74 (0.53)
Note: Bold type= significant values for a significance level ␣ = 5%.

Table 4. Pearson and Kendall’s ␶ 共in Brackets兲 Correlation Coefficient between EB Series 共Averaged over 1 – 4 Months兲 and Palmer Series Simultaneous
and Shifted 1, 2, or 3 Months Ahead
Palmer
EB averaging scale h
Period Month 共month兲 Same month 1 month ahead 2 months ahead 3 months ahead
1955–1999 June 1 −0.23 共−0.06兲 −0.17 共−0.02兲 −0.22 共−0.09兲 −0.21 共−0.16兲
2 −0.08 共−0.05兲 −0.06 共−0.04兲 −0.13 共−0.13兲 −0.16 共−0.19兲
3 −0.04 共0.02兲 0.02 共0.07兲 −0.06 共−0.01兲 −0.03 共−0.04兲
4 −0.16 共−0.02兲 −0.14 共−0.01兲 −0.19 共−0.06兲 −0.13 共−0.07兲
1970–1999 June 1 −0.29 共−0.04兲 −0.23 共−0.05兲 −0.28 共−0.11兲 −0.24 共−0.15兲
2 −0.09 共−0.01兲 −0.11 共−0.07兲 −0.17 共−0.17兲 −0.14 共−0.18兲
3 −0.13 共0.04兲 −0.10 共0.01兲 −0.13 共−0.07兲 −0.05 共−0.04兲
4 −0.27 共0.01兲 −0.30 共−0.05兲 −0.29 共−0.07兲 −0.17 共−0.05兲
1980-1999 June 1 −0.19 共0.00兲 −0.13 共−0.02兲 −0.20 共−0.01兲 −0.25 共−0.12兲
2 −0.05 共0.08兲 −0.10 共−0.07兲 −0.20 共−0.13兲 −0.24 共−0.18兲
3 −0.06 共0.09兲 −0.04 共−0.01兲 −0.11 共−0.08兲 −0.06 共−0.08兲
4 −0.11 共0.11兲 −0.12 共0.00兲 −0.18 共−0.04兲 −0.11 共−0.02兲

JOURNAL OF HYDROLOGIC ENGINEERING © ASCE / JUNE 2009 / 591

J. Hydrol. Eng., 2009, 14(6): 588-595


Downloaded from ascelibrary.org by Universidad Politecnica De Valencia on 06/05/19. Copyright ASCE. For personal use only; all rights reserved.

Fig. 1. R2 between observed and predicted PHDI by using the Fig. 2. R2 between observed and predicted PHDI by using the
two proposed approaches for the months of January–March 共time two proposed approaches for the months of July–September 共time
horizon= 1 , 2 , 3 , 4 months ahead兲 horizon= 1 , 2 , 3 , 4 months ahead兲

period considered coincides with the last three or two decades


and, generally, as the averaged period coincides with 3 or considered, and forecasted values are then used as input to fore-
4 months. The null hypothesis of no correlation is rejected at cast at k ⬎ 1. Thus, in the first scheme the model predictions are
␣ = 5% significance level using Student’s t-test or the Kendall’s ␶ based on the observed values of the input variables even if
test for most of the values, especially those corresponding to the far away in time, whereas, in the second scheme, predictions for
last decades. k ⬎ 1 are based on the Palmer index values estimated for the
Nevertheless, different results have been obtained for different previous time step.
months of the year. For example, in Table 4 Pearson and Ken- The simultaneous observation period for all the adopted in-
dall’s correlation coefficients obtained through the correlation dexes is 1955–1999. In order to take into account the higher cor-
analysis between PHDI and the NAO index series for the month relation observed in the last decades and, at the same time, to
of June are shown. It can be inferred that the correlation in this ensure a sufficient number of available data for model calibration,
case is rather poor. the period 1970–1999 has been chosen for further analyses. The
Higher values of correlation generally correspond to the choice of such a period could represent a limitation for the long-
months from January to March for the NAO index and January term forecasting capability of the model, namely for cases when
to March and September to November for the EB. climatic conditions are similar to those observed on the period
Such a kind of correlation can be justified by the fact that over 1955–1969.
the past three decades during the winters the Oscillation has re- The selected period has been preliminarily divided into a cali-
mained in one extreme phase and both precipitation and tempera- bration set 共1970–1989兲 and a validation set 共1990–1999兲.
ture values have revealed a negative trend over Southern Europe After training the networks, the Nash–Sutcliffe R2 criteria
and the Mediterranean 共Hurrel, 1995兲. 共Nash and Sutcliffe 1970兲 for assessing the agreement between
From a qualitative point of view, the significant correlations observed and predicted Palmer index values has been computed
shown in Tables 2 and 3 allow one to consider the NAO and EB with reference to the calibration and validation sets.
indexes as potential candidates to be included as exogenous vari- The obtained R2 for the months of January–March considering
ables within a forecasting model for Palmer series. the two proposed forecasting approaches are shown in Fig. 1.
Obviously, the two approaches coincide for Lag 1 and there-
fore the corresponding R2 are the same. However, as the lag in-
Application of the Model creases, the Iterative model seems to improve the predictive
capability with respect to the Direct model, in particular for Lags
The prediction of monthly Palmer index values has been per- 3 and 4. Similar results, not shown here, related to the predictive
formed for up to 4 months into the future by using ANN models. capability of the two adopted forecasting approaches, have been
In particular, first the prediction of the Palmer index at month t obtained for the months of October–December.
+ k, PHDI共t + k兲, 共k = 1 , . . . , 4兲, has been carried out by selecting as Different results have been observed for the months of July–
predictor variable the value of the index in the current month, September as shown in Fig. 2.
PHDI共t兲. The adopted ANNs have one output neuron correspond- For this period of the year the Iterative model, in fact, exhibits
ing to the predicted Palmer index values. a slightly worse predictive capability than the Direct model for
In order to take into account seasonal effects and, at the same Lags 3 and 4. The same results, showing the similar predictive
time, to have a calibration set of adequate length, four different capability of the two adopted forecasting approaches, have been
ANNs have been considered, each valid for a 3 month period obtained for the months of April–June.
within the year 共namely January–March, April–June, etc.兲. Fur- The comparison of the two forecasting approaches reveals that
ther, two forecasting schemes have been adopted to forecast at none of them appears preferable to the other. Although, in prin-
different time horizons. In the first scheme, named the Direct ciple, the Iterative approach should be preferred due to its parsi-
approach, different ANNs have been developed for each forecast- mony of parameters, it appears inapplicable to cases in which
ing horizon k. In the second scheme, named the Iterative ap- different predictors, besides past values of PHDI index, are
proach, ANNs able to predict 1 month ahead have been adopted, such as climatic indexes.

592 / JOURNAL OF HYDROLOGIC ENGINEERING © ASCE / JUNE 2009

J. Hydrol. Eng., 2009, 14(6): 588-595


Downloaded from ascelibrary.org by Universidad Politecnica De Valencia on 06/05/19. Copyright ASCE. For personal use only; all rights reserved.

Fig. 3. R2 between observed and predicted PHDI by using the Fig. 4. R2 between observed and predicted PHDI by using the
ANN model for the months of January–March 共time horizon ANN model for the months of July–September 共time horizon
= 1 , 2 , 3 , 4 months ahead, Direct approach兲 = 1 , 2 , 3 , 4 months ahead, Direct approach兲

Thus, an extension of the Direct model is proposed here in when climatic indexes are included in the model is significant.
order to include information deriving from large scale climatic Similar results 共here not shown for brevity兲 have been obtained
indexes and to assess whether considering the information of for the months of October–December.
NAO and EB indexes leads to improvements in forecasting PHDI Nonetheless, different results have been observed for the
values. A similar extension of the Iterative model has not been months of July–September as shown in Fig. 4.
implemented as it would require the use of other models to fore- For this period of the year, considering the climatic indexes as
cast future NAO and EB values. predictor variables does not improve significantly the predictive
Next, prediction of the PHDI共t + k兲 was carried out by select- capabilities of the model with respect to the case when only cur-
ing as predictor variables PHDI共t兲 and NAO共t兲 or PHDI共t兲 and rent value of Palmer index is taken into account.
EB共t兲. On the basis of the preliminary correlation analysis, both Indeed, although some small differences in R2 can be ob-
NAO and EB have been averaged over the previous 4 months. served, the R2 values obtained through the models having EB or
Further, in order to assess whether increases in R2 obtained NAO as predictor variable generally lie within the confidence
when NAO and/or EB are considered as exogenous variables are intervals. This was somehow expected, due to the limited corre-
due to the higher number of model parameters or to an effective lation observed between Palmer index and climatic indexes in
dependence between the Palmer values in Sicily and the climatic such months.
indexes, R2 confidence intervals have been computed through a Similar results, indicating no influence of climatic indexes
Monte Carlo procedure. More specifically, 100 NAO series and on forecasting PHDI, have been obtained for the months of April–
100 EB series uncorrelated with the observed Palmer series have June.
been generated and the corresponding 100 ANN predictions and Finally, Fig. 5 illustrates the observed and forecasted PHDI
the associated R2 values have been computed for the proposed time series at Lag 3 related to the months of January–March. The
models. Confidence intervals at 10% significance level have then plot confirms from a qualitative point of view the better predictive
been estimated by considering the lower and upper 5% sample capability obtained through the application of the model based on
quantiles of the R2 series. Thus, if R2 computed on observed the EB index with respect to the model based only on the PHDI
series lie above the upper confidence limits, the better predictive index or the one based on NAO index.
capabilities of the models are presumably due to an effective in- Fig. 6 illustrates the observed and forecasted PHDI time series
fluence of the NAO and EB indexes and not to the higher number at Lag 2 related to the months of July–September 共validation
of model parameters. period兲. From Fig. 6 the very similar predictive capability of all
The computed R2 corresponding to the ANNs for the months investigated models is confirmed.
of January–March considering different input variables are pre-
sented in Fig. 3. Also in Fig. 3, the values of R2 obtained through
application of ANNs having only PHDI共t兲 as the predictor vari- Conclusions
able, following the Direct approach, are also shown.
For each model, observed R2 values are reported as bars, Advances in understanding the influence of large-scale climatic
whereas the intervals between the lower and upper 5% quantiles patterns on the climate in a region can potentially contribute to a
of generated R2 values are indicated as black lines. better forecast of drought conditions. In the present paper an at-
From Fig. 3 the following considerations can be made. First, tempt is made to verify whether the use of the North Atalantic
including climatic indexes as predictor variables leads to an im- Oscillation and European Blocking indexes could improve fore-
provement of the predictive capability for the months of January– casting of Palmer PHDI index, with reference to the Sicily region.
March. This is particularly evident for the case when the EB First, the influence of NAO and EB indexes on Palmer index
index is taken into account and for increasing forecasting time series, computed on areal monthly precipitation from 1955 until
horizon. 1999 in Sicily, has been investigated by means of a correla-
Further, the R2 values obtained through the models with EB or tion analysis. Results indicate that NAO series are significantly
NAO as predictor variables always lie above the upper confidence anticorrelated with Palmer index series for winter months and
limit thus indicating that the improvement in prediction ability especially in the last decades. The EB series are significantly

JOURNAL OF HYDROLOGIC ENGINEERING © ASCE / JUNE 2009 / 593

J. Hydrol. Eng., 2009, 14(6): 588-595


Downloaded from ascelibrary.org by Universidad Politecnica De Valencia on 06/05/19. Copyright ASCE. For personal use only; all rights reserved.

Fig. 5. Observed and predicted PHDI varying the ANNs models for the months of January–March 共validation phase, Direct approach兲

correlated with Palmer index series both for winter and autumn casting model for the evaluation of the PHDI future values.
months and especially in the last decades. The comparison between the prediction for winter and autumn
The prediction of the PHDI index for up to 4 months ahead months 共for which an high correlation between Palmer index and
was, at first, carried out by adopting as predictor variable the climatic indexes has been shown兲 obtained by either including or
observed index in the previous months. not including the NAO and EB indexes within the model, indi-
Two forecasting schemes have been adopted to forecast at cates some improvements in terms of R2 when the NAO and
different time horizons. In the first scheme, named the Direct especially EB are considered.
approach, different ANNs have been developed for each forecast- A different model behavior has been observed for the spring
ing horizon. In the second scheme, named the Iterative approach, and summer predictions, for which no significant improvements
ANNs able to predict 1 month ahead have been considered, and in terms of model predictive capability due to the introduction of
forecasted values are then used as input to forecast at lag ⬎1. the climatic indexes as input variables have been observed.
The comparison between the two forecasting schemes has The overall conclusion of the paper is that NAO and EB in-
shown some improvements in terms of predictive capability dexes exert some influence on drought evolution in Sicily. Thus,
when the Iterative approach for the winter and autumn months information provided by such indexes can be usefully adopted to
is applied. No significant differences in predictive quality for improve the forecast of drought events in terms of PHDI values
the other months have been obtained varying the forecasting both in autumn and winter period.
approaches. Further research is needed to extend the Iterative model in
Therefore, the information provided by NAO and EB indexes order to include information deriving from large scale climatic
has been included, following the Direct approach, within a fore- indexes.

Fig. 6. Observed and predicted PHDI varying the ANNs models for the months of July–September months 共validation phase, Direct
approach兲

594 / JOURNAL OF HYDROLOGIC ENGINEERING © ASCE / JUNE 2009

J. Hydrol. Eng., 2009, 14(6): 588-595


Acknowledgments Ruiz, J. M. 共2007兲. “Influence of the North Atlantic Oscillation on
water resources in central Iberia: Precipitation, streamflow anomalies
The present research has been developed with the partial financial and reservoir management strategies.” Water Resour. Res., 43共9兲,
support of the Italian project MIUR PRIN 2005 “Drought indexes W09411.
McKee, T. B., Doesken, N. J., and Kleist, J. 共1993兲. “The relationship of
forecasting and definition of water supply systems operating
drought frequency and duration to time scales.” Proc., 8th Conf.
rules.” In particular, the writers acknowledgement Dr. Carlo Cac-
on Applied Climatology, American Meteorol. Soc., Anaheim, Calif.,
ciamani 共ARPA Emilia Romagna兲 for his contributions and for
179–184.
the information provided about climatic indexes.
Moreira, E. E., Paulo, A. A., Pereira, L. S., and Mexia, J. T. 共2006兲.
“Analysis of SPI drought class transitions using loglinear models.”
J. Hydrol., 331共1–2兲, 349–359.
References Nash, J. F., and Sutcliffe, J. 共1970兲 “River flow forecasting through con-
Downloaded from ascelibrary.org by Universidad Politecnica De Valencia on 06/05/19. Copyright ASCE. For personal use only; all rights reserved.

ceptual models. Part 1: A discussion of principles.” J. Hydrol., 10共3兲,


ASCE Task Committee on Application of Artificial Neural Networks in 282–290.
Hydrology. 共2000a兲. “Artificial neural networks in hydrology. I: Pre- Palmer, W. C. 共1965兲. “Meteorological drought.” Research Paper No. 45,
liminary concepts.” J. Hydrol. Eng., 5共2兲, 115–123. U.S. Weather Bureau, Washington, D.C.
ASCE Task Committee on Application of Artificial Neural Networks in Pavan, V., Molteni, F., and Bankovic, C. 共2000a兲. “Winter variability in
Hydrology. 共2000b兲. “Artificial neural networks in hydrology. II: Hy- the Euro-Atlantic region in observations and in ECMWF seasonal
drologic applications.” J. Hydrol. Eng., 5共2兲, 124–137. ensemble experiments.” Q. J. R. Meteorol. Soc., 126共567兲, 2143–
Berry, M. J. A., and Linoff, G. 共1997兲. Data mining techniques, Wiley, 2173.
New York. Pavan, V., Tibaldi, S., and Brankovic, C. 共2000b兲. “Seasonal prediction
Briffa, K. R., Jones, P. D., and Hulme, M. 共1994兲. “Summer moisture of blocking frequency: Results from winter ensemble experiments.”
variability across Europe, 1892–1991: An analysis based on the Q. J. R. Meteorol. Soc., 126共567兲, 2125–2142.
Palmer drought severity index.” J. Climatol., 14共5兲, 475–506. Qian, B., Corte-Real, J., and Xu, H. 共2000兲. “Is the North Atlantic Oscil-
Cacciamani, C., Morgillo, A., Marchesi, S., and Pavan, V. 共2007兲. “Moni- lation the most important atmospheric pattern for precipitation in Eu-
toring and forecasting drought on a regional scale: Emilia Romagna rope?” J. Geophys. Res., 105共D9兲, 11901–11910.
Region.” Methods and tools for drought analysis and management, Quadrelli, R., Lazzeri, M., Cacciamani, C., and Tibaldi, S. 共2001a兲. “Ob-
G. Rossi et al., eds., Springer, Dordrecht, The Netherlands. served winter Alpine precipitation variability and links with large-
Cancelliere, A., Di Mauro G., Bonaccorso, B., and Rossi, G. 共2007兲. scale circulation patterns.” Clim. Res., 17, 275–2854.
“Drought forecasting using the standardized precipitation index.” Quadrelli, R., Pavan, V., and Molteni, F. 共2001b兲. “Wintertime variability
Water Resour. Manage., 21共5兲, 801–819. of Mediterranean precipitation and its links with large-scale circula-
Cancelliere, A., Rossi, G., and Ancarani, A. 共1996兲. “Use of Palmer tion anomalies.” Climate Dyn., 17, 457–466.
Index as drought indicator in Mediterranean regions.” Proc., Int. Rao, A. R., and Padmanabhan, G. 共1984兲. “Analysis and modelling of
Association for Hydraulic Research Congress “From Flood to Palmer’s drought index series.” J. Hydrol., 68, 211–229.
Drought,” Sun City, South Africa, August 5–7, IAHR, 1–25. Rogers, J. C. 共1990兲. “Patterns of low-frequency monthly sea level pres-
Cordery, I., and McCall, M. 共2000兲. “A model for forecasting drought sure variability 共1899–1989兲 and associated wave cyclone frequen-
from teleconnections.” Water Resour. Res., 36共3兲, 763–768. cies.” J. Clim., 3共12兲, 1364–1379.
Goodess, C. M., and Jones, P. D. 共2002兲. “Links between circulation and Rossi, G. 共2000兲. “Drought mitigation measures: A comprehensive frame-
changes in the characteristics of Iberian rainfall.” Int. J. Climatol., work.” Drought and drought mitigation in Europe, J. V. Vogt and F.
22共13兲, 1593–1615. Somma, eds., Kluwer Academic, Dardreecht, The Netherlands, 233–
Heim, R. R. 共2000兲. “Drought indices: A review.” Drought: A global 246.
assessment, D. A. Wilhite, ed., Routledge, London. Rossi, G., and Cancelliere, A. 共2002兲 “Early warning of drought: Devel-
Hurrell, J. W. 共1995兲. “Decadal trends in the North Atlantic Oscillation: opment of a drought bullettin for Sicily.” Proc., 2nd Int. Conf. on New
Regional temperatures and precipitation.” Science, 269共5224兲, 676– Trends in Water and Environmental Engineering for Safety and Life:
679. Eco-Compatible Solutions for Aquatic Environments, CSDU, Capri,
Hurrell, J. W., Kushnir, Y., and Visbeck, M. 共2001兲. “The North Atlantic
Italy, 1–12.
Oscillation.” Science, 291共5504兲, 603–605.
Smith, D. I., Hutchinson, M. F., and McArthur, R. J. 共1993兲. “Australian
Jones, P. D., Jonsson, T., and Wheeler, D. 共1997兲. “Extension of the North
climatic and agricultural drought: Payments and policy.” Drought
Atlanctic Oscillation using early instrumental pressure observations
Network News, 5共3兲, 11–12.
from Gibraltar and southwest Iceland.” Int. J. Climatol., 17共13兲,
Swingler, K. 共1996兲. Applying neural networks: A practical guide, Aca-
1433–1450.
Karl, T. R. 共1986兲. “The sensitivity of the Palmer drought severity index demic, London.
and Palmer’s Z-index to their calibration coefficients including poten- Tibaldi, S., and Molteni, F. 共1990兲. “On the operational predictability of
tial evapotranspiration.” J. Clim. Appl. Meteorol., 25共1兲, 77–86. blocking.” Tellus, Ser. A, 42A, 343–365.
Kim, T., and Valdes, J. B. 共2003兲. “Nonlinear model for drought forecast- Vicente Serrano, S. M. 共2005兲. “El Niño and La Niña influence on
ing based on a conjunction of wavelet transforms and neural net- drought conditions at different time scales in the Iberian Peninsula.”
works.” J. Hydrol. Eng., 8共6兲, 319–328. Water Resour. Res., 41共12兲, W12415.
Lohani, V. K., and Loganathan, G. V. 共1997兲. “An early warning system Walker, G. T., and Bliss, E. W. 共1932兲. “World weather V.” Mem. R.
for drought management using the Palmer drought index.” J. Am. Meteorol. Soc., 4, 53–84.
Water Resour. Assoc., 33共6兲, 1375–1386. Wallace, J. M., and Gutzler, D. S. 共1981兲. “Teleconnections in the geo-
Lohani, V. K., Loganathan, G. V., and Mostaghimi, S. 共1998兲. “Long- potential height field during the Northern Hemisphere winter.” Mon.
term analysis and short-term forecasting of dry spells by Palmer Weather Rev., 109, 784–812.
Drought Severity Index.” Nord. Hydrol., 29共1兲, 21–40. Wilhite, D. A. 共2000兲 Drought: A global assessment, Vol. 2, Routledge,
Lopez-Moreno, J. I., Begueria, S., Vicente-Serrano, S. M., and Garcia- London.

JOURNAL OF HYDROLOGIC ENGINEERING © ASCE / JUNE 2009 / 595

J. Hydrol. Eng., 2009, 14(6): 588-595

Anda mungkin juga menyukai