2. Cheval_SPI_Overview_IJOC_paper

Transcription

2. Cheval_SPI_Overview_IJOC_paper
 Volume 12
Issues 1-2
2015 Attribution CC BY-NC-ND
The Standardized Precipitation Index –
an overview
Sorin Cheval
National Meteorological Administration,
Sos. Bucuresti-Ploiesti 97, 013686 Bucharest, Romania.
E-mail: [email protected]
Abstract. At twenty years after its launch, the Standardized Precipitation
Index (SPI) is intensively used both in fundamental studies and practical
applications. It is implemented operationally in numerous Hydrological and
Meteorological Services, being considered a universal drought index. This
overview covers the interval 1993-2012, it focuses mainly on the
publications in international journals, and reports on the conceptual and
methodological developments (a), topics approached (b), operational status
(c), and perspectives (d). The overview demonstrates the reasons which
have made the SPI a powerful tool for drought studies, namely robust
concept, simple calculation and understanding, temporal flexibility, spatial
meaning, and wide applicability. Besides, the recent Lincoln Declaration on
Drought Indices urges for further advancements of the SPI, and this
overview makes up a support for such a growth.
Keywords: Standardized Precipitation Index; drought; wetness;
precipitation; drought indices; drought overview.
1. INTRODUCTION
The atmospheric precipitations turn out environmental changes,
damages and casualties either by extreme high or low amounts fallen over a
certain territory within a time interval. Scientists from various domains have
proposed numerous methodologies for assessing the magnitude of the
17 precipitation excess and deficit, and a few of them have been used more
intensively. Overviews of the drought concepts, definitions and
methodologies were tackled within scientific meetings and publications
(Heim, 2002; Redmond, 2002; Mishra and Singh, 2010; Vicente-Serrano et
al.., 2012), and many studies were devoted to promote drought indices or to
evaluate their performances (Bhalme and Mooley, 1979; Guttman, 1998,
1999; Vicente-Serrano et al.., 2010). This paper overviews the development
of the Standardized Precipitation Index (SPI) at 20 years after its launch.
In 2004, the SPI was considered a relatively new index and applied
occasionally, whereas nowadays it seems to get the universal applicability
predicted by some scholars (Tsakiris and Vangelis, 2004). The SPI is widely
used in research or operational activities in more than 70 countries (WMO,
2012), approaching both fundamental studies and applications (i.e.
agriculture, forestry, hydrology). In 2009, the participants at the InterRegional Workshop on Indices and Early Warning Systems for Drought
held at the University of Nebraska-Lincoln issued “The Lincoln Declaration
on Drought Indices” (Hayes et al., 2011). The document acknowledges the
value of drought monitoring, and recommends explicitly using the SPI for
characterizing meteorological droughts. Besides, the World Meteorological
Organization (WMO) is urged to promote the implementation of the SPI in
operational use at the National Meteorological and Hydrological Services
(NMHSs). As a consequence, WMO sponsored the recent publication of a
comprehensive SPI user manual (WMO, 2012). In its turn, the Joint
Research Centre (JRC) supported a study on predicting the SPI-based
drought (Singleton, 2012), while many stakeholders and public authorities
have chosen the index for practical needs (Carbone et al., 2008).
The summaries of the SPI development are limited to methodological
or regional aspects (Hayes, 2000; Lloyd-Hughes and Saunders, 2002;
Karavitis et al., 2011; Bonsal et al., 2011, 2012). At a decade after the first
SPI publication, Lloyd-Hughes and Saunders (2002) presented in-depth its
concept, calculation, advantages and limits, and the drought climatology for
Europe. Redmond (2002) supplied strong reasons for applying the SPI both
in practical scope and fundamental studies, i.e. solid theoretical background,
extensive and careful examination of its proprieties over almost 1200
stations. However, no comprehensive overview does exist so far. While a
vast bibliography refers to the SPI in various manners, this paper is based on
almost 300 articles published in journals within the main scientific
international stream of English language –with very few exceptions–
between 1993 and 2012.
The overview presents the significant aspects regarding the SPI’s
conceptual framework and calculation methodology (a), its strengths and
limitations (b), a summary of the topics approached (c), considerations on
the operational use (d), and future developments (e). Only the main
assumptions tackled in the SPI journal publications are summarized, aiming
18 to illustrate as widely as possible the methodology, the applications, the
geographical coverage, and the perspectives.
2. SPI CONCEPTUAL FRAMEWORK
AND CALCULATION METHODOLOGY
This section addresses the concept supporting the SPI functionality,
calculation procedure, time scales, significance of the thresholds, and
drought types, as derived from international journal publications.
2.1. Concept and calculation
At the beginning of the 1990s, Thomas B. McKee, Nolan J. Doesken,
and John Kleist, researchers at the Colorado State University, developed the
SPI with the intention to “propose an indicator and definition of drought
which could serve as a versatile tool in drought monitoring and analysis”
(McKee et al., 1993). At that time, the Palmer Drought Severity Index
(PDSI) was the most widely used drought index in the U.S.A. (Hayes et al.,
1999), despite caveats like inadequacy in mountain areas or spatial
inconsistency. The theoretical framework of the SPI described by McKee et
al., (1993, 1995) was substantially enhanced in the following years.
Edwards and McKee (1997) and Guttmann (1999) provided detailed
methodology for computation –including the Fortran source code– and
applications of the index. Furthermore, the concept and the calculation
procedure are tackled more or less detailed in most publications.
The SPI is a dimensionless probability index, and its computation for a
given location and period is conducted after the following sequence:
(a) Data sets are fitted to a probability density function (PDF). The
selection of the parametric distribution determines the accuracy of the SPI
values (Singleton, 2012). McKee et al. (1993) recommended gamma, but
other distributions perform at least similarly. Several options have been
tested for different applications, i.e. Pearson III (Guttman, 1999; VicenteSerrano, 2006b; Blain, 2011); Weibull (Livada and Assimakopoulos, 2007);
log-normal (Angelidis et al., 2012); geometric (Cebrián and Abaurrea,
2012), but no universal agreement has been reached. The gamma PDF is
defined as (Edwards and McKee 1997; Angelidis et al. 2012):
g ( x) =
1
x a −1e − x / β
α
β Γ(α )
(1)
where α>0 is a shape parameter, β>0 is a scale parameter, Γ is the gamma
function, and x>0 is the precipitation amount. Different formulas apply for
19 other PDFs.
(b) The results are then used to find the cumulative probability of a
precipitation event for a given month and time scale (Edwards and McKee,
1997).
(c) Further, the cumulative probability distribution is transformed into a
standard normal distribution, with mean of zero and variance of one, which
is the SPI value (Sönmez et al., 2005; Angelidis et al. 2012).
Parry et al. (2011) illustrate eloquently the methodological flowchart from
raw precipitation to SPI values showing the differences between two distinct
climates in Europe (Figure 1). The precipitation pattern in an area may
change over time, so that the shape and scale parameters of the PDFs
change accordingly. Ntale and Gan (2003) assert that even a reasonably
long calibration period provide results that would be affected significantly
by a prolongation of the period.
At present, the SPI executables and manuals are available for
Windows, Fortran, and R users (spi-support.blogspot.ro/p/software.html).
Turgu (2008) report on the SPI program developed within the Turkish State
Meteorological Service, capable to illustrate cumulative probability versus
precipitation, and to display drought categories at the selected time scales.
Figure 1. Standardized Precipitation Index applied to summer three-month accumulations
(JJA) for Madrid and London, 1901-2005 (Parry et al., 2011) 20 2.2. Time scale and length of data series
McKee et al. (1993, 1995) used the probability of the precipitation
occurrence for 3, 6, 12, 24, 48 months, and recommended at least 30 years
of data sets. Both the significance of the time steps and the length of the
records were extensively scrutinized (Sönmez et al., 2005; Wu et al., 2005;
WMO, 2012), and used in various ways. The short timescales proved their
utility in meteorological and agricultural drought, while the longer are
applied in hydrology (Guttman, 1998; Szalai and Szinell, 2000; Heim, 2002;
Seiler et al., 2002; Łabedzki 2007). Sub-monthly scales are infrequently
used. Wu et al. (2005) found that the length of the records becomes highly
significant for computation only in regions where the precipitation pattern
changes along time.
2.3. Thresholds
Beginning right with the original scale (McKee et al., 1993), the SPI
has been often used in a simplistic way by considering precipitation surplus
for SPI>0, and dry conditions for SPI<0, as Agnew (2000), Bhuiyan et al.,
(2006), or Ali and Lebel (2008) criticized. Most usual, the neutral
conditions are defined around zero SPI.
Some authors proposed drought categories different than the original
one, in terms of denominations (Hayes et al., 1999; Svoboda et al., 2002),
thresholds (Agnew, 2000; Lloyd-Hughes and Saunders, 2002), or both
(Michaelidis and Pashardis, 2008) (Figure 2). The neutral conditions are
named ‘no drought’ (Agnew, 2000; Li et al., 2012), ‘average’ (Zsákovics et
al., 2007), ‘normal’ (Vicente-Serrano et al., 2004; Liu et al., 2009; Zhai and
Feng 2009), or ‘near normal’ (Hayes et al., 1999; Lana et al., 2001; Bonsal
and Regier 2006; Zhang et al., 2009a). For some authors, the ‘mild drought’
begins immediately below zero (McKee et al., 1993, 1995; Liu et al., 2009;
Zhai and Feng, 2009). The drought severity and wetness are divided in 3 to
5 categories, namely ‘abnormally’, ‘moderate’, ‘severe’ or ‘very’,
‘extremely’ and ‘exceptionally’ (Svoboda et al. 2002; Michaelidis and
Pashardis, 2008). The precipitation deficit is named either ‘dry’ or
‘drought’, while the excess is generally named ‘wetness’.
Many studies provide the corresponding probability of occurrence for
each category (Lloyd-Hughes and Saunders, 2002; Vicente-Serrano et al.,
2004; Zhang et al., 2009a). Manatsa et al., (2008) propose a 3-digits SPI
scale, probably too finely tuned for practical purposes. There is no universal
recommendation for using the thresholds, and the operational needs are the
main regulator.
21 Figure 2. Various SPI denominations and thresholds. ExW = Exceptionally wet; ExD=
Exceptionally dry; 1. McKee et al. (1993, 1995); 2. Hwang and Carbone (2009); 3.
Svoboda et al. (2002); 4. Li et al. (2012); 5. Agnew (2000); 6. Vicente-Serrano et al.
(2004); 7. Manatsa et al. (2008); 8. Lloyd-Hughes and Saunders (2002); Sadat Noori et al.
(2012); 9. Liu et al. (2009), Zhai and Feng (2009); 10. Hayes et al. (1999); Lana et al.
(2001); 11. Zhang et al. (2009a); 12. Michaelidis and Pashardis (2008); 13. Bonsal and
Regier (2006); 14. Kangas and Brown (2007)
22 2.4. Drought types
The SPI was initially designed to assess the meteorological drought,
so that the first scale referred only to negative precipitation anomalies,
although McKee and his colleagues acknowledged that the index could
monitor both wet and dry periods (McKee et al., 1993). Nowadays, the SPI
has become the most powerful and widely used of the simple drought
indices (Bordi and Sutera, 2008; Angelidis et al., 2012), while publications
regarding both the water deficit and surplus are much more rare (Krepper
and Zucarelli, 2010; Yan et al., 2012), and no article dedicated to the excess
only has been identified along this study. Most studies refer to a single type
of drought (meteorological, agricultural or hydrological), but joint
approaches are also useful (Mo, 2008; Caparrini and Manzella, 2009;
Tabrizi et al., 2010; Vidal et al., 2010, 2012).
3. STRENGTHS AND LIMITATIONS
This section presents a succinct summary of the SPI’s benefits and
disadvantages, acknowledging that most publications tackle the topic, and a
few of them details it. One can remark that some characteristics may be
interpreted both as strength and limitations (i.e. simplicity).
3.1. Strengths
1. Simplicity. The computation uses only atmospheric precipitation.
However, despite its simplicity, Paulo and Pereira (2006) were capable to
define the main features of the meteorological drought, such as lead-time,
duration, severity, magnitude and intensity based on the SPI values, and the
comparisons with indices of higher complexity demonstrate its efficiency
(Bonsal and Regier, 2007; Sims et al., 2002; Vicente-Serrano et al., 2012).
2. Flexibility. It can be used to identify and assess both wet and dry periods
(Dai et al., 2011).
3. It has the same efficiency for solid or liquid precipitation.
4. The standardization of the index triggers meaningful comparison between
locations (Dai et al., 2011) and periods, undertaking consistency in
interpretation. Nevertheless, the dry climate should be treated cautiously
(see the SPI limitations).
5. Temporal versatility. SPI can be calculated over any timescale, depending
on the user’s interest (WMO, 2012). The conceptual edifice of the SPI holds
the idea that a region can be simultaneously in excess and in deficit,
depending on the time scale approached (Redmond, 2011). While Hayes et
al. (1999) advise avoiding the use of the 1, 2, or 3-month SPI in low
precipitation climates, Guttman (1999), Vicente-Serrano (2006b), and
23 Champagne et al. (2011) argues that the SPI provide reliable results only up
to 24-month periods. Sub-monthly periods are not recommended (Roudier
and Mahe, 2010; WMO, 2012), but several studies were performed. For
example, Anctil et al. (2002) found the 5-day SPI useful for predicting the
end of ongoing droughts, Sims et al. (2002) calculated the daily, weekly,
and bi-weekly SPI, while Wu and Wilhite (2004) pledged for the weekly
scale in agricultural drought risk studies.
6. Monitor/operational use. SPI is capable to identify the drought onset very
early. Bordi et al. (2001a, b) promoted the SPI for a monthly bulletin to
monitor drought in Italy, and Carbone et al. (2009) developed a regionalscale dynamic drought index tool for North and South Carolina.
7. The topography does not bias the results.
3.2. Limitations
1. The SPI uses a single meteorological element for describing a complex
event (Caparrini and Manzella, 2009). However, precipitation is the major
driver at least for the meteorological drought (Guttman, 1998; Heim, 2002;
Gebrehiwot et al., 2011).
2. The raw precipitation data are assumed to fit a theoretical probability
distribution before the standardization, which may induce errors in the
outputs. The results depend on the distribution used (Sienz et al., 2012), but
there is no agreement over the best one.
3. The SPI cannot differentiate the regions in relation with their proneness
to drought. As a consequence of its normal distribution, the index values
occur with similar frequencies at any location.
4. It can provide uncertain results in arid areas and dry times, with quasinormal distribution of the precipitation. Wu et al., (2007) documented
coherently this weakness, and recommended that the duration of the drought
should receive more focus than its severity. Naresh Kunar et al. (2009)
showed that the SPI under-estimates the intensity of dryness/wetness when
the rainfall is very low/very high, respectively. This limitation could make
arguable the SPI usage in climate change studies, since precipitations are
expected to decrease dramatically in certain spots (i.e. Southern Europe) in
the decades to come (Vasiliades et al., 2009).
5. It does not take into account pre-drought conditions and gives long-past
and recent precipitation the same weight; therefore, abrupt peaks may occur
within the series, although in reality the drought onset and recovery are
gradual. Paulo and Pereira (2006) proposed a backtracking procedure for set
up the initiation of a drought event, but its end remains sudden.
6. The initial drought categories were arbitrarily defined (McKee et al.,
1993, 1995). Turgu (2008) proposed an operational cumulative probability
correspondence between SPI thresholds and precipitation, and Quiring
(2009) proposed an objective location-specific method for defining
24 operational drought thresholds.
7. The use of SPI could be problematic in regions with less than 30 years of
data or with a sparse meteorological network. Reconstruction of the SPI
based on proxies could address this limitation, as shown by Touchan et al.
(2005) on tree rings. Shahid (2008) applied a neural network analysis to fill
in the gaps in data series, while Rhee and Carbone (2011) demonstrated that
the spatial interpolation of the missing daily data would be done previous to
SPI calculation.
4. TOPICS
A large number of publications approached various facets of the SPI,
such as methodology, causes and applications, e.g. Bonsal et al. (2011,
2012) reviewed the drought research in Canada, with a lot of references to
the SPI in relation with the causes, characteristics, prediction and
monitoring, convincing on the utility of the index in complex applications.
After considerations regarding the spatial coverage (a), the section
overviews, generally in a thematic and chronological order, the SPI
publications referring to methodology and concept (b), applications (c),
variability and forecast (d), drought and wetness causes (e), comparison and
combination with other indices (f).
Table 1 summarizes the topics from publications referring to the SPI
extended to regional, continental or global scale, with the following
remarks: the selection is not exhaustive; the term ‘variability’ may refer to
either spatial, temporal or both; the topics mentioned for each publication
are not necessarily exclusive, but dominant. Similar information for country
and local scale may be found at http://spi-support.blogspot.ro/p/topics.html.
4.1. Spatial coverage
The SPI has been used in applications across all the continents, in a
large variety of geographical conditions and climates, from local focus to
global exposure. Its performances do not depend on the size of the area, i.e.
Pai et al. (2011) achieved realistic outputs by analyzing the drought both
over India, and its 458 districts, respectively.
The majority of the publications are oriented to administrative,
geographical or hydrological basins, i.e. Bonaccorso et al. (2003) – Sicily
(Italy); Bordi et al. (2004) – Elbe Basin (Germany) and Sicily (Italy);
Vicente-Serrano and López-Moreno (2005) – Ebro’s mountainous basin;
Zsákovics et al. (2007) – Danube-Tisza interfluves; Khadr et al. (2009) –
Ruhr river basin (Germany); Zhang et al. (2012b) – Xinjang (China);
Champagne et al. (2011) – Alberta (Canada).
25 Table 1. Regional and global journal articles using the SPI.
Publication
Geographical
extension
Time scale
(months)
2, 6, 9, 12,
16, 24, 36
Phenomena*
Ntale and Gan (2003)
Africa (E)
D
drought assessment; comparison with PDSI
and BMI; improvement of SPI
Rouault and Richard
(2005)
Africa (S)
3, 6, 12, 24
D
variability; assessment; correlation with ENSO
Manatsa et al (2008)
Africa (S)
6
D
Min et al (2006)
Kurnik et al (2011)
Asia (E)
Africa (Horn)
Asia (Central
and SW)
12
3, 6, 9, 12
D
D
relationship with ENSO and Darwin Sea level
pressure anomalies
variability
drought assessment; drought monitoring
8**
D
comparison with PDI
Zhang et al (2012a)
Asia (E)
1, 3
B
Bothe et al (2012b)
Asia (Central)
South
America
(Amazon
River)
South
America
1
B
paleoclimate; variability; causes; correlation
with ENSO and volcanic activity
patterns; relationship with circulation
13**
D
vegetation response to drought
6
B
variability; correlations with ENSO and NTA
D
variability; comparison with PDSI
D
variability
Shahabfar et al (2012)
Walker et al (2009)
Mo and Berbery (2011)
Lloyd-Hughes and
Saunders (2002)
Bordi and Sutera (2008)
López-Moreno and
Vicente-Serrano (2008)
Bordi et al (2009)
Peled et al (2010)
Europe
3, 6, 9, 12,
24
3, 24
Europe
1-12
D
variability; relationship with NAO
Europe
Europe
3, 24
3
1, 3, 6, 9,
12
B
D
variability
comparison with NDVI, PDSI, SC-PDSI, NSM
hydrological response to meteorological
drought; RSPI
relationships between dryness and
temperatures
relationships between dryness and
temperatures
Europe
Hannaford et al (2011)
Europe
Hirschi et al (2011)
Europe (SE)
3, 6
D
Beniston (2012)
Europe
1, 3, 6
D
Maule et al (2012)
Europe
Singleton (2012)
Heinrich and Gobiet
(2012)
Europe
Bordi and Sutera (2001)
AghaKouchak and
Nakhjiri (2012)
Burke and Brown (2008)
Lotsch et al (2003)
Mueller and Seneviratne
(2012)
Orlowsky and
Seneviratne (2012)
Europe
D
1, 3, 6, 9,
12
1, 3
1, 3, 6, 9,
12, 18, 24
D
validation of RCM
D
drought forecast
B
climate change
Europe,
Northern
Hemisphere
3, 6, 12, 24
B
patterns; teleconnections; comparison with
PDSI
Global
6, 12
D
drought forecast; remote sensing application
Global
12
D
Global
1-12
B
Global
3, 6, 9
B
drought forecast
remote sensing application; comparison with
NDVI
relationship surface moisture - temperature
extremes
Global
12
D
*) D: drought; B: both drought and excedent.
**) Days.
26 Key topics
variability; climate change
Other studies refer to national scale, even if a large or complex
territory, i.e. Domonkos (2003) – Hungary; Brázdil et al. (2009) – Czech
Republic; Costa (2011) – Portugal; Păltineanu et al. (2009) – Romania;
Dash et al. (2012) – Bangladesh.
Obviously, less publications employ the SPI in regional, continental or
global analysis, i.e. Ntale and Gan (2003) – East Africa; Vicente-Serrano et
al. (2011a) – Iberian Peninsula; Zhang et al. (2012a) – East Asia; Bothe et
al. (2012b) – Central Asia; Lloyd-Hughes and Saunders (2002) – Europe;
Mo and Berbery (2011) – South America; Bordi and Sutera (2001) –
Northern Hemisphere; Lotsch et al. (2003) – Globe; AghaKouchak and
Nakhjiri (2012) – Globe.
4.2. Methodology and concept
McKee et al. (1993, 1995) published the original papers that
introduced the SPI with limited methodological guidelines, so that in the
following years many studies approached the topic. The methodological and
conceptual developments have targeted mainly the PDFs, the time scales,
and the areal relevance. Edwards and McKee (1997) provided a detailed
methodology, clarifying the conceptual basis and the computation process.
Hayes (2000) remarked that relatively few publications had explained the
SPI, the term was occasionally misused in confusion with the Standardized
Anomaly Index (Jones and Hulme, 1996), alternatives to gamma functionbased normalization had been proposed, and different terminology for the
SPI could be considered.
(a) PDFs
The gamma distribution was employed by most studies (Ji and Peters,
2003; Capra and Scicolone, 2012), but other PDFs have been equally
applicable. Lloyd-Hughes and Saunders (2002) found that the gamma
distribution provided the best representation of the precipitation data,
excepting the arid regions of Europe for which the Cauchy distribution is
mentioned as a possible solution. Zhang et al. (2009a) considered gamma
and log-normal distributions performed similarly at 95% confidence level,
and used the latter one because of its simplicity. Angelidis et al. (2012)
suggest that the log-normal probability distribution can be more convenient
to use for 12- and 24-months, while the gamma distribution is the best
choice for time scales up to 6 months.
Guttman (1999) advocated the use of the Pearson type III distribution,
and a few applications have supported his assumption. Thus, VicenteSerrano (2006b) used Pearson Type III for analyzing the spatial patterns of
drought on different time scales in the Iberian Peninsula, Naresh Kumar
recommended it as an alternative to gamma, while Blain (2011) found out
that it performs better than gamma in describing the long term rainfall series
27 for the State of São Paulo (Brazil).
Ntale and Gan (2003) tried to improve the SPI by using an unbiased
plotting-position formula that can estimate the non-exceeding probability of
precipitation (applicable for long data sets) (a), and by addressing the
distortions in the distribution tails for skewed precipitation with a different
transformation procedure (b). Cebrián and Abaurrea (2012) tested eight
distributions (exponential, Weibull, gamma, log-normal, generalized Pareto,
geometric, negative binomial and Poisson), and found out that the geometric
distribution fitted best the durations of the dry spells in Huesca (Spain).
Sienz et al. (2012) claim for reconsider the adequacy of the gamma
distribution, and pledge for using the Weibull type distribution.
While any transformation may provide reasonable results, the PDF
should be selected cautiously, according to the climate background in the
area of interest. It is recommended that several distributions are tested to fit
the local precipitation series, before calculating the SPI.
(b) Time scales
There is no general recommendation regarding the appropriate scale,
so that the selection depends mainly on the user’s expertise, and
applications entailed (i.e. agriculture, surface or ground waters). Guttman
(1999) limited the SPI’s reliability to 24 months. However, the performance
remains remarkable even if accepting this appreciation, taking into account
that drought events within temperate climate (Touchan et al., 2005), or
severe and excessive droughts in arid climates (Al-Qinna et al., 2011) rarely
exceed two years.
Mishra et al. (2009) used a probabilistic approach to characterize the
SPI-based drought and emphasized the influence of the time scales on the
results. Some other examples are provided below, grouped by their
reference to meteorological, hydrological or agricultural drought.
For the southern Italy, Bordi et al. (2001a), and Piccarreta et al.
(2004) show that the 6 month scale reveals intense rainfall with short
periods of water deficit, the 12 month scale discloses an increase in dry
conditions, whereas the 24 month scale is able to record moderate to severe
droughts. Sönmez et al. (2005) revealed situations when shorter steps, i.e. 3month, capture the moderate and severe drought, while longer steps, i.e. 12month, fail. In a study combining remote sensing techniques and drought
indices, Champagne et al. (2011) determined that most suitable time scale
of precipitation extremes to capture the soil moisture anomalies is less than
9 months. On the other hand, Abarghouei et al. (2011) pledged for the
relevance of 24-month rather than shorter scales in dry climates.
For groundwater concerns, Carbone and Dow (2005) emphasized the
relevance of the 24-month SPI, and Mendicino and Versace (2007)
confirmed the capability of the time scales longer than 12 months to monitor
water resources, but shorter scales have been also tested. Shahib and
28 Hazarika (2010) used 6-, and 12-month SPI for groundwater drought in
Bangladesh.
Tabrizi et al. (2010) investigated the relationships between time-based
meteorological and streamflow droughts in the southwestern Iran, and
Somorowska (2011) confirmed that the substantial drying of the Lasica river
basin (Poland) was caused by an increase of the number of 3-, 6- and 12month SPI droughts. Du et al. (2012a) explored the suitability of different
SPI-temporal scales for hydrological applications, and found out that the 2
month-time scale was appropriate for detecting flood events.
Sub-monthly time scales were used in several agricultural drought
studies. Anctil et al. (2002) studied the capability of the 5-day SPI to predict
the end of an ongoing drought. For an application exploring the dryness
impact in the Amazon Basin, Walker et al. (2009) used a 13-day moving
average to smooth the SPI data, and marked the beginning of dry seasons as
the falling of the moving averages below 1.0 standard deviation of the mean
daily rainfall. Lu (2011) investigated the physical basis for developing a
daily flood and drought index based on the conceptual methodology of the
SPI. Zhang et al. (2012c) used a 10-day SPI (from 10 to 110 days) in China.
Since many studies have demonstrated that the time step can bias the
results, it is recommended to use multiple scales in SPI analysis. Time
scales from 1- to 24-months are the most common (http://spisupport.blogspot.ro/p/topics.html), but longer and shorter intervals may be
considered in some applications.
(c) Areal relevance
Some scholars aimed to adjust the SPI towards a better response to the
local conditions in the area of interest. Thus, Quiring (2009) proposed an
objective methodology to derive the drought thresholds according to the
local conditions, conditioned by the length of the records and by an
appropriate PDF, and tested the gamma, normal, log-normal, and
exponential distributions. Türkeş and Tatl (2009) computed the SPI values
by considering the local-time means of the precipitation series by fitting an
upper and a lower envelope to precipitation data, and claimed the proposed
method guarantees better results mainly for dry and wet conditions.
4.3. Applications
This section presents the main assumptions and findings regarding the
SPI use in different applications. The most common domains are agriculture
(a), and hydrology (b). Besides, very many studies analyse the vulnerability
of vegetation and water resources to precipitation deficit (c), mapping issues
(d), or case studies (e). The SPI’s assessment and monitoring capabilities
are likewise covered by journal publications, while the topics sometimes
interfere with each other. Mishra and Singh (2011) reviewed the state-of 29 the-art in drought modelling, and remarked the constant involvement of the
SPI in such applications. The SPI was also used in non-conventional
applications, i.e. atmospheric CO2 seasonal cycle and ecosystem (Angert et
al., 2005; Buermann et al., 2007; Wharton et al., 2012), drought social
perception (Sinoga and Gross, 2012).
(a) Agriculture
Yamoah et al. (2000) published the first SPI application in agriculture,
studying the combined repercussions of the wetness and nitrogen on the
maize crop yields.
After about a decade of systematic investigations, it is common to
merge the SPI values, other drought indices, and remote sensing products,
mainly the Normalized Difference Vegetation Index (NDVI). Ji and Peters
(2003) conducted a thorough analysis of the relationship between the
NDVI-retrieved condition of the vegetation, and the SPI-based moisture
availability in the U.S. Great Plains. They found that the correlation is
significant in both grasslands and croplands, it varies over the growing
season, and it has the highest values for the 3-month SPI. In 2003, Lotsch et
al., (2003) performed a similar analysis for terrestrial ecosystems, and
reported a high degree of correlation for 5-month time scale. VicenteSerrano et al. (2006) and Vicente-Serrano (2007) combined the SPI and
high temporal resolution remote sensing images for tackling the vegetation
response to drought in the middle Ebro valley. Peled et al. (2010) compared
the changes in vegetation over Europe as captured by NDVI and various
drought indices, and found better correlations for the Normalized total depth
Soil Moisture (NSM; Dutra et al., 2008), PDSI and PDSI-SC (PDSI-Self
Calibrated; Wells et al., 2004) than for the 3-month SPI, because of the soil
surface and temperature information included in the algorithms.
Derived from the NDVI, the Vegetation Condition Index (VCI;
Kogan, 1995) is highly correlated with PDSI and with 6- and 9-month SPI,
being less sensitive to short-term precipitation deficiencies (Quiring and
Ganesh, 2010). While the SPI remains a useful tool, Gebrehiwot et al.
(2011) pledged for the superiority of the VCI in vegetation drought
applications. For sub-humid and humid areas of NW India, Jain et al.
(2011) conclude that the 1 to 9-month SPI fits best the vegetation drought
severity and duration retrieved by NDVI and VCI. Champagne et al. (2011)
combined the passive microwave remote sensing with the SPI in analyzing
the soil moisture in Canada.
Scientific efforts have been dedicated to determine the most
appropriate time scale for various agricultural purposes, but there is no
general agreement reached. Łabedzki (2007) showed that 1–3 month SPI
reflects the agricultural drought development in Poland. The seasonal
drought patterns in the arid and semi-arid regions of Gujarat (India) let Patel
et al. (2007) conclude that the 3-month SPI of September is a good indicator
30 of anomalies for grain yields in drought-prone areas. Iglesias and Quiroga
(2007) proposed a methodology for measuring the climatic risk to cereal
production based on the 12-month SPI as climate indicator. Mkhabela et al.
(2010) found poor correlation between SPI and wheat yields in the
Canadian Prairies. Subash and Ram Mohan (2011) examined the
performance of the SPI for detection and monitoring of drought, with
reference to the rice-wheat crop in the monsoon India. Pasho et al. (2011)
and Pasho et al. (2012) documented the SPI utility in quantifying the
drought impacts on tree growth in NE Spain. Zhang et al. (2012c) used 10day SPI timescale for predicting the maize drought disaster in China.
Eventually, Sadat Noori et al. (2012) argue that the prediction of wheat
production in Iran is possible with an anticipation of several months before
harvest based on 1-, 3-, and 6 months SPI.
(b) Hydrology
Du et al. (2012a) reported that the SPI is a valuable tool for
hydrological monitoring, both in respect of dryness and wetness. Despite
most SPI applications in hydrology refer to water deficit, the earliest
publication on the topic focused on floods. Thus, Seiler et al. (2002)
claimed the utility of the SPI in assessing the flood risk, indicating that 3 to
6-month time scales should be used for quantifying the superficial water,
while 12 to 24 months are suitable for the subsoil moisture. Further, the
SPI’s potential to be used in hydrology has been validated by researches on
streams, groundwater, and lakes.
Within different river basins, Cancelliere and Salas (2004), VicenteSerrano and López-Moreno (2005), Edossa et al. (2010), and Ganguli and
Janga Reddy (2012) analysed multiple facets of the river response to
meteorological drought (i.e. duration, severity, return period) with 1 to 24month time scales SPI. These studies have validated Seiler et al. (2002),
revealing that the 3-6 months time scales are suitable for surface flows, the
reservoirs respond better to 7-12 months, and longer time scales should be
avoided in surface hydrology.
For hydrogeological applications, Bhuiyan et al. (2006) combined the
SPI with the groundwater recharge deficit and NDVI information.
Mendicino and Versace (2007) demonstrated its usefulness for monitoring
the resources from Southern Italy, Khan et al. (2008) employed it to track
the influence of the rainfall on groundwater and the impact on the irrigation
needs in Australia, and Di Matteo et al. (2011) pledge for its utility in
hydrogeology of the Mediterranean limestone areas.
Some studies compared the SPI and other indices in term of
hydrological performances. Based on the results retrieved by SPI and SPEI
(Standardized Precipitation Evapotranspiration Index; Vicente-Serrano,
2010) in the Tagus River Basin (Spain), Lorenzo-Lacruz et al. (2010) argue
that the influence of temperature on hydrological drought is not negligible.
31 In a study extended over 1950-2099 across the Blue River Basin
(Oklahoma), Liu L et al. (2012) acquired significant correlations between
SPI and SRI (Standardized Runoff Index; Shukla and Wood 2008) with 2month lag time.
Last, but not least, Myronidis et al. (2012) explored the long-term
effect of drought on a Mediterranean shallow lake in Greece, and found that
3- and 24-month can be used for such correlations.
(c) Risk and vulnerability
In the recent decade, the SPI has become a common tool for assessing
the drought risk and the vulnerability of certain areas and resources, and
numerous publications report on the magnitude, intensity, frequency or
spatial extension of drought hazards in agriculture and hydrology. Most of
them describe the performances of multivariate techniques and models (Wu
and Wilhite 2004), probabilistic approaches (González and Valdés, 2004;
Liu L et al., 2011), and copulas (Ganguli and Janga Reddy, 2012; Shiau,
2006; Shiau et al., 2012; Zin et al., 2012).
Local and country scales prevail. Sönmez et al. (2005) investigated
the drought vulnerability in Turkey, and Bordi et al. (2007) focused on the
analysis of extreme wet and dry periods in Sicily. Pandey et al. (2010)
scrutinized the temporal variability of the SPI-based drought severity within
an effort to integrate hydro-meteorological and physiographic factors for
assessing the vulnerability to drought in the Sonar basin (India). The SPI is
useful for studying the country-wide vulnerability and risk to drought both
under dry and wet climates, as demonstrated with the applications in
Botswana (Batisani 2011), Greece (Karavitis et al., 2011), Iran (Mansouri
Daneshvar et al., 2012), and Malaysia (Zin et al., 2012). Hirschi et al.
(2011), Beniston (2012), and Mueller and Seneviratne (2012) approached
the relationships between hot temperature extremes and dryness at regional
(SE Europe), continental (Europe) and global scale, defining the surface
moisture deficit with the SPI.
Hydrological risks have been widely approached. Roudier and Mahe
(2010) evaluated the drought hazards within the Bani River Basin (Mali)
using the Climatic Moisture Index (CMMI; Vörösmarty et al., 2005),
demonstrated that the 10-day time scale SPI was inadequate for estimating
the water deficit, and recommended the utilization of EDI (Effective
Drought Index; Byun and Wilhite, 1996). Ganguli and Janga Reddy (2012)
assessed the risk of hydrologic droughts in India.
The potential of the SPI to be used in forest fire and insurance is
underexploited, despite some published reports. Caccamo et al. (2011)
assessed the relationship between SPI and MODIS-based spectral indices at
different time scales in Australia, as a tool for monitor the vegetation
drought in the fire prone Sydney Basin Bioregion. Dimitrakopoulos et al.
(2011) and Drobyshev et al. (2012) reported on the SPI performance in
32 analyzing and detecting the forest fire in Greece and, respectively, in
Sweden, while Hofer et al. (2012) selected the SPI to define the drought
conditions in a multi-disciplinary forest fire risk assessment in the Iberian
Peninsula.
Lyon et al. (2009) stressed the relationships between drought relief
payments and meteorological indices like SPI and monthly rainfall
anomalies, as a valuable tool for the insurance business.
Several studies have resulted in risk measures for drought analysis
(Cebrián and Abaurrea, 2012), or Drought Hazard, Exposure, and
Vulnerability Indices (DHIk, DEIk, DVIk) (Shiau and Hsiao, 2012). Sun et
al. (2012) combined the SPI with two versions of PDSI to develop a multiindex drought (MID) model for improving the agricultural drought risk
assessment and prediction.
(d) Mapping and drought regionalization
The standardization enclosed in the SPI sustains the spatial meaning,
while its ultimate normal distribution leads to similar frequencies at any
location, and complicates the interpretation. While very many publications
just use spatial representations of the values, some applications have strived
to bring more consistency to SPI interpolations, by approaching gridding
methods, deriving indices more oriented to spatiality, or implementing GISbased monitoring, and satellite imagery in regions with sparse
meteorological networks (Bhuiyan et al., 2006; Gebrehiwot et al., 2011).
Lloyd-Hughes and Saunders (2002) developed a grid-based (0.5°
resolution) drought climatology for Europe using the SPI and PDSI. Tonkaz
(2006) mapped the SPI in southeastern Anatolia based on a GIS application.
Rhee and Carbone (2008) used the SPI in a mapping exercise testing the
influence of the spatial interpolation and data aggregation in drought
assessment. Mishra and Nagarajan (2011) developed a GIS project to grid
the SPI-derived drought in India, while Mishra and Nagarajan (2012)
compared the SPI and irrigation demand maps, obtaining good correlations
for 2-, and 3-month. Akhtari et al. (2008, 2009) suggest that the
interpolation methods and SPI time-steps should be fitted to obtain the most
relevant results. They found that krigging is the best method for one-month
SPI, and the Thin Plate Smoothing Spline method is the best for 12-month
scale.
The introduction of indices and methodologies to support the better
spatial representation of the SPI has been developed in the recent years.
Loukas and Vasiliades (2004), and Loukas et al. (2008) introduced and
developed the Drought Severity-Areal Extent-Frequency (SAF), a method
to assess the spatial characteristics and the frequency of the drought over an
area, using GIS capabilities and SPI at various time scales.
Koutroulis et al. (2011) addressed the drought spatiality defining the
Spatially Normalized–Standardized Precipitation Index (SN-SPI) in an
33 application over the Island of Crete. In their turn, Liu X et al. (2012)
proposed the Areal Drought Magnitude (ADM), an aggregated index aimed
to assess both temporal variation of drought magnitude and spatial extent of
drought events. Recently, Mansouri Daneshvar et al. (2012) proposed a
Drought Hazard Index (DHI) produced by summing up the spatial
representation of the SPI values over a certain territory. In the same time,
aiming to add relevancy to the SPI-based comparison of the drought in
European regions, Hannaford et al. (2011) defined the Regional
Standardized Precipitation Index (RSPI), further used by Maule et al.
(2012).
Giddings et al. (2005) found the SPI analysis was relevant to justify
distinct, self-content and homogenous precipitation zones in Mexico, with
possible practical applications. Since then, other efforts to fundament the
delineation of homogenous regions using the SPI have been done. Inspired
by the SPI methodology, Reich et al. (2008) defined climatic zones in the
states of Jalisco and Colima (Mexico) combining monthly temperature,
precipitation and evaporation, and suggested the data transformation
whenever they are highly skewed. Based on SPI12 and PCA, Raziei et al.
(2009) were able to differentiate two climatic sub-regions in the western
Iran.
Dubrovsky et al. (2009) introduced the relative SPI (rSPI) and the
relative PDSI (rPDSI), obtained by using a reference series, either weather
station or period data, in order to allow for a better comparison of spatial
drought conditions. Further, Jung and Chang (2012) introduced the rSRI and
used it in conjunction with rSPI for studying the hydrological drought risk
under climate change perspectives in Oregon (USA).
Abolverdi and Khalili (2010) applied L-moments statistics to
investigate the drought characteristics, and they distinguished between the
different applicability of 3, 6, 12, and 24-month time steps SPI for extreme
drought studies. Thus, they derived four homogenous sub-regions using
SPI3, while longer intervals emphasized only one region in the
southwestern Iran. Santos et al. (2011) and Moreira et al. (2012b) also used
the SPI to differentiate homogenous regions relative to drought in Portugal.
(e) Case studies
Certain extreme events draw a special interest from researchers, and
some journal publications concentrate on case studies using the SPI as a
drought indicator: Szalai and Szinell (2000) – the 1983 drought in the
Hungarian Plain; Mihajlović (2006) – the 2003/2004 drought in the
Pannonian part of Croatia; Bonsal and Regier (2007) – the 2001/2002
drought in the Canadian Prairies; Michaelidis and Pashardis (2008) – the
2007/2008 drought in Cyprus; Bumbaco and Mote (2010) – the drought
events from 2001, 2003, and 2005 in the Northwestern Pacific U.S.;
Hanesiak et al. (2011) – the 1999–2005 drought in the Canadian Prairies.
34 4.4. Variability and forecast
The topics follow at least three directions, namely the past (a), and
future (b) long-term variability and changes, and the operational drought
forecast (c). Statistics and modelling approaches are abundant (Edwards and
McKee, 1997; Komuscu 1999; Feng and Hu 2004; Chen et al., 2009; Logan
et al., 2010; Nikolova et al., 2012), but tree-rings (Touchan et al., 2005;
Linderholm and Molin, 2005; Bonsal et al., 2012; Levanič et al., 2012), and
documentary archives (Min et al., 2003) were also utilized.
(a) Past variability
Investigating the past variability of the drought is probably the most
common approach exploiting the SPI, and the publications started at the
beginning of the 2000s. Thus, Lana et al. (2001) analyze the spatial and
temporal variability of the precipitation in Catalonia (Spain) and discuss the
SPI’s capacity to offer “the best picture of rainfall shortage and excess
patterns”. The majority of the articles cover decadal time resolution (Sirdas
and Şen, 2001; Bordi et al., 2009; Viste et al., 2012), but longer periods
extend to centuries. The non-parametric Mann-Kendall test seems to be the
most popular in decoding significant trends in precipitation and droughts.
Zhai and Feng (2009) applied it in the Gansu Province (China), and
Abarghouei et al. (2011) in Iran.
Min et al. (2003) compared the SPI-based drought temporal patterns
from South Korea and East Asia, including a “preliminary study” over the
pre-instrumental period. Touchan et al. (2005), Linderholm and Molin
(2005), Levanič et al. (2012), and Seftigen et al. (2012) developed longterm SPI reconstructions (centennial) in Turkey, Romania, and respectively,
Sweden using tree ring chronologies, and emphasized both the potential and
the limitations of such approaches. Bonsal et al. (2012) assessed the
variability of drought in the Canadian Prairie over pre-instrumental,
instrumental and future period, combining the SPI and PDSI benefits. Treerings analysis extended to the year 1350, and future droughts were derived
from statistically downscaled climate from several Atmosphere–Ocean
Global models with multiple emission scenarios. Zhang et al. (2012a)
selected the SPI as a precipitation deficit/surplus indicator in investigating
the conjugated influence of El Niño and volcanic activity on the East Asia
climate over the last 1,200 years.
The variations in precipitation may affect the duration-magnitude or
the intensity-duration-frequency relationships, and SPI is suitable for such
approaches (Sirdas and Şen, 2001; Hallack-Alegria and Watkins, 2007).
Kangas and Brown (2007) used thirteen SPI time steps from 1 to 120month scales in order to examine the long term characteristics (1895-2003)
of the precipitation regime in the U.S.A. with the help of the Parameter 35 elevation Regression on Independent Slopes Model (PRISM) (Daly et al.,
1994).
The studies tackle regions, i.e. NW Iberian Peninsula (VicenteSerrano et al., 2011b), Gujarat (Patel et al., 2007), Huaihe River Basin (Li
et al., 2012), countries, i.e Bangladesh (Shahib, 2010), Portugal (Paulo et
al., 2012), or continents (Bordi and Sutera, 2008; Bordi et al., 2009).
(b) Operational forecast
As regards the short- and medium-term hydrological and
meteorological drought and flood forecast, Hannaford et al. (2011), and
Estrela and Vargas (2012) consider that the SPI exploitation is modest, in
spite of the numerous outputs reported in international journals (Bacanli et
al., 2009; Durdu, 2010; Al-Qinna et al., 2011; Belayneh and Adamowski,
2012; Blain, 2012), based on modelling, teleconnections, joint and statistics
analysis.
The SPI and remote sensing products (NDVI) can be coupled for
predicting agricultural drought and cereal productions with months in
advance (Vicente-Serrano et al., 2006; Dutta et al., 2012).
Mishra and Desai (2006), Cancelliere et al. (2007), Paulo and Pereira
(2007), Paulo and Pereira (2008), and Durduk (2010) used stochastic
techniques, neural networks, and transition probabilities between drought
classes and different time horizons for methodologies addressing the
seasonal and non-seasonal forecast of the SPI with various periods ahead,
successfully validated with measurements.
In Portugal, Moreira et al. (2008), and Moreira et al. (2012a) used the
12-month SPI in log-linear models and ANOVA-like inference approach
aiming to predict drought categories, its initiating, development and
termination, as well as the temporal changes in frequency and severity. In
their turn, Santos et al. (2011) were concerned with the regional frequency
analysis of the drought derived from 1, 3, 6, and 12-month SPI. Xu et al.
(2011) noticed that the SPI of time scales larger than 9 months was effective
in identifying most extreme summer and summer-autumn drought in the
Han River Basin (China), so it may be used in the risk forecast.
Manatsa et al. (2008, 2010) documented the combination between El
Niño-Southern Oscillation (ENSO) and the Indian Ocean dipole/zonal mode
as a significant predictor of drought variability in the southern Africa, while
Zhang et al. (2009b) proved that SPI is a useful index in modeling both
dryness and wetness variations, and found similar periodicities between
ENSO and SPI24 for Shanghai.
It is certain that joint inputs may improve the results significantly.
Huang and Carbone (2009) developed a SPI-PDSI ensemble forecast for
South Carolina, while Sohn et al. (2012) used the SPI3 in an experimental
multi-model ensemble system designed to forecast drought and flood over
South Korea. Keskin et al. (2009, 2011) found a good potential of the
36 Adaptive Neural-based Fuzzy Inference System, fuzzy logic and artificial
neural networks methods to predict the regional meteorological drought
using the 12-month SPI values. Hannaford et al. (2011) used the correlation
between the Regional Deficiency Index (RgDI), a hydrological drought
indicator, and Regional Standardized Precipitation Index (RSPI), designed
as a meteorological drought indicator, for exploring the large-scale spatial
coherence of droughts over Europe and prospect for forecasting. RezaeianZadeh and Tabari (2012) applied a multilayer perceptrons (MLP) approach
including the antecedent SPI, NAOI and SOI values to forecast the SPI in
different climates of Iran, and reached the best results for 12 and 24-month
time scales.
(c) Long-term predictions
The long-term predictions refer to the 21st century and many are based
on IPCC gas emission scenarios (IPCC, 2007). One can remark the
considerable spatial coverage, the diversity of scenarios and models, and the
consistent use of the SPI, at different spatial and time-scales. The main topic
is the meteorological drought, but hydrology, agriculture, and adaptation as
well. The studies are published after 2007, and their main assumptions are
briefly summarized here.
Sienz et al. (2007) used the monthly SPI and pointed out the relative
stable trends in extreme dryness, and increasing wetness in winter and
spring for Iceland along the 21st century in the A1B IPCC scenario.
Mavromatis (2007) assessed the future wheat yields in Greece using the SPI
and PDSI projections coupled to the Hadley Centre regional climate model
(HadRM3) under the A2 IPCC scenario (IPCC, 2007). In their turn, Burke
and Brown (2008) used SPI12 to explore the drought projections under
single and double CO2 scenarios.
Standardized Precipitation Index; rSPI) were jointly used to project
drought conditions in the Czech Republic (Dubrovsky et al., 2009), or the
hydrological drought risk (Jung and Chang, 2012) in Oregon (USA).
In their turn, Li et al. (2008), Loukas et al. (2008), Mishra et al.
(2010), and Liu L et al. (2012) used various groups of models under
different IPCC scenarios to scrutinize the SPI variability over 20th and 21st
century in the Amazon region, Thessaly (Greece), Midwestern United States
and Blue River Basin (Oklahoma). Tölle et al. (2012) highlighted regional
differences in the future vulnerability to SPI-derived water surplus or deficit
risks between locations in the central part of Germany. Based on 21st
century runs of ARPEGE under three emission scenarios, Vidal et al. (2012)
advance spatial theoretical adaptation scenarios for France, with the spatial
consistency of the SPI and SSWI.
Apparently, the time scale can influence the outputs. For example,
Vasiliades et al. (2009) showed that the climate change is likely to affect the
drought severity in the Lake Karla watershed (Greece), but the uncertainty
37 is quite large and increases with the SPI timescale. Strzepek et al. (2010)
used 5- and 9-month SPI combined with the PDSI, to see how climate
change affects the frequency of SPI meteorological droughts in the U.S.A.
along the 21st century. Singleton (2012) analysed the performance of the
ECWMF monthly forecast for predicting the meteorological drought, and
found out that the reliability in forecasting SPI1 and SPI3 is weak,
suggesting calibration as a possible improvement.
Heinrich and Gobiet (2012) combined the results of the ENSEMBLES
project, the SPI and PDSI to analyse the dry and wet conditions in Europe in
the next decades. Orlowsky and Seneviratne (2012) used the outputs of the
Coupled Model Intercomparison Project (CMIP5) ensemble of GCM
simulations (Taylor et al., 2012) to compare the drought past variations and
future projections of the SPI and Soil Moisture Anomalies at planetary
scale. Within the IPCC-A2 scenario, the SPI12, Percent of Normal (PN),
and Erinç’s Aridity Index values show an increase in drought incidence over
the Aegean, southeastern Anatolia, western and central parts of the
Mediterranean region, and Sen et al. (2012) reflected the expected changes
in the biomass, yield and growth duration for corn.
4.5. Drought/wetness causes
The SPI-based investigations on the drought/wetness causes have
identified several triggering mechanisms: NAOI (Domonkos, 2003;
Vicente-Serrano and Cuadrat, 2007; López-Moreno and Vicente-Serrano,
2008); ENSO/SOI (Scian et al., 2006; Liu et al., 2009; Blain, 2012); North
Tropical Atlantic Index (NTA; Mo and Berbery, 2011); distance to water
bodies that govern the origin and direction of air masses and flows (VicenteSerrano and Cuadrat-Prats, 2007; Zhang et al., 2009a); Darwin Sea level
pressure anomalies (Manatsa et al., 2008); monsoon cycles (Subash and
Ram Mohan, 2011); El Niño and volcanic activity (Zhang et al., 2012a);
SST (Wu and Kinter, 2009); Europe-Greenland Index (Sienz et al., 2007).
The teleconnections have been considered one of the main influencing
factors. One of the earliest studies using the SPI linked the drought
occurrence in Europe with the Tropical Pacific signal (Bordi and Sutera,
2001). NAOI has been tackled more often than other teleconnections
(Vicente-Serrano and Cuadrat, 2007 for NE Spain; Di Mauro et al., 2008 for
Sicily). According to the homogeneity of the NAO signal in respect to
monthly precipitation, López-Moreno and Vicente-Serrano (2008)
differentiated 17 European regions and described each of them in terms of
SPI’s response at various time scales. They concluded that, generally,
positive NAO phases determine negative SPI averages in southern Europe
and positive average in northern areas, while the opposite situation occur
during negative phases. However, regional variations or seasonal deviations
must be carefully considered. Touchan et al. (2005) assumed that there are
38 similar spectral peaks in the SPI and NAO along 1251-1998, although
without a clear statistical relationship.
ENSO may induce drought conditions in Americas, southern and
western Asia, and southern Africa. Some studies combine the ENSO with
NAO and/or other indices. Tadesse et al. (2004) observed strong
relationships between the SPI and PDSI drought in Nebraska, and
teleconnections like the Southern Oscillation Index (SOI), the Multivariate
ENSO Index (MEI), and the Pacific Decadal Oscillation (PDO) Index.
Shahid (2008) found significant relationships between SPI and MEI in
Bangladesh. For the southern Africa, Rouault and Richard (2005) linked the
drought to ENSO variability, but Manatsa et al. (2008) asserted that an
earlier signal imposed by the Darwin Sea level pressure anomalies would be
a superior drought predictor, and defined a country-wide aggregated value
of the seasonal SPI, named Zimbabwe SPI, as a drought indicator, adjusting
the Agnew scale (Agnew, 2000) to their needs. In the U.S.A., Mo et al.
(2009) found that the Atlantic Multidecadal Oscillation (AMO) has a small
direct influence on the 6-month SPI droughts, but it modulates significantly
the ENSO signal. Keskin and Sorman (2010) found good correlations
between a hydrometeorological index aggregating the SPI, SRI and SEI (a),
NAO (b) and ENSO (c), for the vicinity of Ankara. ENSO and the Indian
Ocean dipole were identified as possible mesoscale triggers for the
agricultural drought in Zimbabawe (Manatsa et al. 2010), and Nepal (Sigdel
and Ikeda, 2010). Mo and Berberi (2011) found that ENSO and NTA have a
strong influence on the SPI-based drought and wet spells over South
America, mainly when they act in unison or antagonistic, and pledge for
using the findings in forecast. Lastly, Liu X et al. (2012) conducted a
summative publication on the relationships between the 12-month SPI and
several circulations relevant for China, including western Pacific
Subtropical High (WPSH), East Asian Summer Monsoon (EASM) and
ENSO.
Other approaches have focused on the atmospheric circulation,
anthropogenic forcing, soil moisture, and cloud cover. Scian et al. (2006)
referred to large scale atmospheric circulation to explain the extreme
positive and negative rainfall anomalies in Argentina, using the monthly SPI
in an EOF analysis. Brázdil et al. (2009) explored the synoptic climatology
of the drought in Czech Republic, while Seftigen et al. (2012) linked the
variability of the drought in Sweden to circulation patterns. Bothe et al.
(2012a) correlated the summer precipitation in Tibet with circulation
indices, and Bothe et al. (2012b) investigated the influence of the largescale atmospheric circulation on the extreme precipitation in the Central
Asia. In the Southern Amazon, Li et al. (2008) suggested that the
anthropogenic and external forcing could be responsible for increasing dry
conditions reflected by the 6-month SPI. Wu and Kinter (2009) correlated
statistically the SST forcing with the droughts in the Great Plains,
39 considering the role of the soil moisture as well. They highlighted the
importance of the time scale and, implicitly, the significant authority of the
SPI to address such studies. Greene et al. (2011) investigated the
relationships between drought derived from 1-month SPI and cloud cover
retrieved from satellite imagery in an experiment over the Canadian Prairies.
4.6. Comparison and combination with other indices
The previous sections have already illustrated that the SPI has been
critically compared and jointly used with several indices in a considerable
number of publications, tackling mainly the PDSI, Reconnaissance Drought
Index (RDI; Tsakiris and Vangelis 2004, 2005), SPEI, SRI, Percent of
Normal (PN), and Rainfall Deciles (RD; Gibbs and Maher, 1965). A few
syntheses referring to multiple indices have been also published. Moreover,
the combination of more indices is a key issue for validating the results in
specific applications. Within a condensed comparison of drought indices,
Mishra and Singh (2010) concluded that their application should be defined
according to the local conditions and utilization.
(a) Multi-indices approaches
Byun and Wilhite (1999) discussed the weaknesses of the major
drought indices, including the SPI, and introduced the concept of effective
precipitation as a basic tool for quantifying the drought duration and
severity, and considered the daily scale as more appropriate than the
monthly one. Keyantash and Dracup (2002) compared six indices (RD; SPI;
Cumulative Precipitation Anomaly - CPA, Foley 1957; Rainfall Anomaly
Index – RAI, van Rooy, 1965; Drought Area Index - DAI, Bhlame and
Mooley, 1980; PDSI) in terms of their robustness, tractability, transparency,
sophistication, extendibility and dimensionality. SPI ranked the second as
weighted total, very close to RD, and scored maximum points as robustness,
sophistication, and extendibility. Morid et al. (2006) included seven indices
(RD; PN; SPI; China Z-Score Index – CZSI, Wu et al., 2001; Modified
China Z-Score Index – MCZSI, Wu et al., 2001; Z-Score Index – ZSI, Wu
et al., 2001; Effective Drought Index – EDI; Byun and Wilhite, 1996) in a
comparison focused on their performances for drought monitoring in Iran.
SPI and EDI were able to detect more accurately the onset of drought, its
spatial and temporal variation. Quiring and Pradesh (2010) found low
correlations between the VCI and PDSI, ZSI, SPI, RD, and PN. In a study
aiming to evaluate the uncertainties in the projection of future drought,
Burke and Brown (2008) reported that the SPI supplied much smaller global
changes than indices considering the atmospheric demand for moisture
(Potential Evapotranspiration Anomaly – PPEA, PDSI, and Soil Moisture
Anomaly – SMA). The SPI seems to perform better than the Selianinov
hydrothermic coefficient (CHT; Selianinov, 1928), Si and PN in identifying
40 mild and moderate drought, while CHT is more accurate in the spatial
analysis of extreme drought events (Potop and Soukup, 2009).
(b) PDSI
Since the PDSI was considered the most powerful drought index at the
time SPI was launched quite many studies refer to the two indices. McKee
et al. (1993, 1995), and Edwards and McKee (1997) compared the SPI and
the PDSI, and Guttman (1998) performed a spectral comparative analysis
concluding that “the SPI may be a better indicator of dryness and wetness
than the PDSI”. Hayes et al. (1999) emphasized how SPI addressed the
PDSI’s drawbacks, and pledged for the combined use of the two indices.
The performance and monitoring capabilities of SPI and PDSI were
compared by Szalai and Szinell (2000) on the 1983 drought event in
Hungary. Sims et al. (2002) suggested that the SPI is more representative of
short-term precipitation and soil moisture variations than the PDSI, and
propose a regression equation that uses SPI for soil moisture. Redmond
(2002) asserted that the highest correlations occur between 6 to 12-month
SPI, and other subsequent studies confirmed. Thus, Lloyd-Hughes and
Saunders (2002) found a closer correspondence between SPI12 and PDSI,
and Paulo and Pereira (2006) reached similar coherency in characterizing
local and regional drought in Alentejo (Portugal) using SPI12, PDSI, and
the Theory of Runs (TR) (Guerrero-Salazar and Yevjievich, 1975). Brázdil
et al. (2009) used the same temporal scale for correlations over the droughts
in the Czech Republic. Kangas and Brown (2007) used various time scales
SPI and PDSI in analyzing the drought and pluvial characteristics over the
U.S. and pledged for the comparison of the two indices to be performed in
applications, with no a priori recommendation. In their turn, Ceglar et al.
(2008) identified good correlations between SPI9 or SPI12 and PDSI for
Ljubljana series (Slovenia).
Goodrich and Ellis (2006) commented on the conceptual and
computational differences between SPI and PDSI, and recommended to use
frequency distributions for comparative analysis. In analyzing the Canadian
Prairies 2001/2002 drought, Bonsal and Regier (2007) found that SPI and
PDSI retrieved comparable results, and emphasized the later index might be
a better option for drought events associated with extreme temperatures.
Vasiliades et al. (2011) demonstrated the utility of the 3- and 6-month SPI
along with a PDSI-derived index for operational monitoring of the
hydrological drought in Pinios river basin from Greece. Tunalioğlu and
Durdu (2012) found that the PDSI-based drought indices performed better
than the SPI in assessing the olive yield in western Turkey. Dash et al.
(2012) found that the PDSI results for RegCM3 model and observed data
are nearly the same to SPI calculations in Bangladesh.
Krysanova et al. (2008) introduced the DS, an index built from
average daily temperature and precipitation, in a comparative study with
41 SPI and PDSI, and noticed that the former method was more sensitive to
trends in drought frequency than the other two indices.
Sun et al. (2012) demonstrated that the prediction accuracy of the
MID model, which exploits jointly the SPI and PDSI, is at least as accurate
to using a single drought index.
Ntale and Gan (2003) analyzed the properties for BMI and modified
versions of SPI and PDSI performances in East Africa, and found that the
SPI was the most appropriate for drought monitoring due to characteristics
already mentioned here (i.e. modest data requirements, temporal flexibility).
As a result of a research developed over Sweden, Drobyshev et al. (2012)
showed that the SPI’s performance in detecting the forest fire is comparable
to other indices, like PDSI or the ratio between actual and potential
evapotranspiration.
(c) RDI
Tsakiris and Vangelis (2004), and Tsakiris et al. (2005) introduced the
RDI performing an extended comparison with the SPI. Tsakiris et al. (2007)
presented the indices used for application in Mediterranean countries within
the project Medroplan, including the SPI, PDSI, the method of deciles and
RDI. Pashardis and Michaelides (2008) found a very good correlation
between the two indices in Cyprus and pledged for the common
implementation in drought assessment. Bazrafshan et al. (2010) compared
the SPI with the RDI in coastal areas of Iran and indicated the SPI as better
in detecting meteorological droughts. Asadi Zarch et al. (2011) performed a
similar comparison for the whole Iran and found that the two indices
provided comparable results, excepting the humid and sub-humid regions,
and they were better correlated for 3, 6, and 9-month time scales than for the
larger ones. In their turn, Khalili et al. (2011) concluded that for agricultural
purposes the RDI can be more appropriate than the SPI, at least for 3, 6, and
12-month values.
(d) SPEI
There are many meteorological, hydrological, and agricultural drought
studies that compare or combine the SPI and the recently developed SPEI,
as the latter index supplies crucial information for drought analysis, namely
the evapotranspiration. Lorenzo-Lacruz et al. (2010) considered that the
hydrological drought response is eventually slightly higher for SPEI than for
SPI-based analysis, because of the thermal influences. The SPI, SPEI and
SC-PDSI reflect the meteorological drought to river discharge, but Joetzjer
et al. (2012) found differences in their performances in two extreme large
world basins (Amazon and Mississippi). Potop (2011) remarked that SPEI
and an adjusted version of the Si provide better results than SPI in detecting
agricultural drought, due to enclosure of specific information such as
temperature or soil moisture. Combining the three indices was considered a
42 useful strategy for drought monitoring. Potop et al. (2012) remarked large
differences between SPI and SPEI whenever the adversely trend in
temperature and precipitation occurs, such as 1900-1920 or 2000-2010 in
the lowlands of the Czech Republic. Besides, for short time-scales (1-2
months) the two indices provided comparable results for average drought
durations, while for mid- and long-terms (3-24 months) the SPEI
determined notably higher values than SPI. Paulo et al. (2012) demonstrated
that 9 and 12-month SPI and SPEI identify the drought occurrence earlier
than the PDSI and the PDSI modified for Mediterranean conditions.
Besides, they found better correlation between SPI and SPEI for semiarid
locations than for humid ones in Portugal. Recently, Vicente-Serrano et al.
(2012) used data from 151 worldwide hydrological basins to evaluate
comparatively the SPI, SPEI and PDSI for meteorological, agricultural or
hydrological drought. They showed that the SPI and SPEI captured the
droughts more accurately than the PDSI, and they argued that SPEI is
superior to SPI for assessing summer drought. Telesca et al. (2012) found
out structural similarities between SPI and SPEI in analyzing the drought in
the Ebro river basin at 1, 3, 6, 9, and 12-month scale along 1950-2006.
(e) Other indices
Combinations of SPI with other indices were tackled occasionally and
they are summarized in this section. It has to be mentioned that the term SPI
have been sometimes misused for the Standardized Anomaly Index (SAI),
and several new indices have been developed from the SPI concept.
Jones and Hulme (1996) defined the precipitation SAI as the
difference between the amounts over a certain time scale and the
precipitation average, divided by the standard deviation of the series, and
Seiler et al. (2002) emphasized the differences between SAI and SPI.
Nevertheless, a number of studies used the SAI and call it SPI, while in
others it remains unclear if the precipitation series are fitted to a cumulative
probability distribution prior to transformation into a standard normal
distribution (Agnew, 2000; Ali and Lebel, 2008; Imanov et al., 2012; Kasei
et al., 2010; Ali et al., 2011; Masoudi and Afrough, 2011; Mokhtari et al.,
2011; Shamsipour et al., 2011).
Mo (2008) addressed the meteorological, agricultural and
hydrological droughts over the U.S. in a joint analysis based on modelderived SPI, SRI and Soil Moisture (SM). Keskin and Sorman (2010)
applied a PCA procedure in order to aggregate the SPI, SRI, and SEI into a
compacted hydrometeorological index capable to assess the water resources
in the vicinity of Ankara (Turkey).
Byun and Kim (2010), and Kim et al. (2010) compared the SPI with
the EDI, and expressed the superior performance of the latter in several
aspects, such as better detection of long and short term droughts and better
representation of the gradual development of droughts. In their turn, Akhtari
43 et al. (2009) evaluate the performance of several geostatistical methods in
interpolating SPI and EDI.
Quiring (2010) developed an objective methodology for operational
drought definition based on thresholds calculated for the SPI, PDSI and PN.
Pai et al. (2011) showed that SPI and PN provided similarly results for
identifying drought occurrences over India, whereas SPI performs better at
district scale, at least during the southwest monsoon season.
Sušnik et al. (2012) compared the SPI with the Net Irrigation
Requirements (NIR) in terms of detecting the agricultural drought.
The CZSI and statistical ZSI are versatile indices that provide very similar
results to SPI (Krepper and Zucarelli, 2010), but they could perform better
for wet conditions, and they can be calculated on datasets with missing
values (Wu et al., 2001).
Wu et al. (2004) combined the SPI and CSDI in assessing the drought
risk of crops. Vergni and Todisco (2011) put the SPI in a comparative
analysis with other climatic and agro-climatic indices (i.e. SDfI) in a
drought variability study in Umbria (Italy).
Cancelliere and Salas (2004) comprise the 3-month SPI with other
indices (i.e. PHDI) in analyzing the hydrologic drought length. Livada and
Assimakopoulos (2007), and Zhang et al. (2009a) found exponential
correlations between SPI and MAI (de Martonne, 1926), reflecting similarly
well the dry conditions over Greece and over the Pearl River basin (China).
Păltineanu et al. (2009) consider that the combination of the climatic water
deficit index and 12-month SPI may be useful for water management in
agriculture. Somorowska (2011) jointly used the SPI and Standardized
Cumulative Annual Deviation (SCAD) to assess the precipitation impact on
the groundwater and stream flow in the central Poland. Further, in a study
covering the continental U.S.A., Anderson et al. (2011) report good spatial
and temporal correlations between the Evaporative Stress Index (ESI)
retrieved from satellite imagery and the SPI at various timescales.
Barua et al. (2009) found out that the Aggregated Drought Index
(ADI; Keyantash and Dracup, 2004) and the Surface Water Supply Index
(SWSI) (Shafer and Dezman, 1982) perform better than SPI1 in identifying
the historical droughts. Norouzi et al. (2012) argued that, for all types of
droughts, ADI provided more reliable results comparing to SPI.
Ghasemi et al. (2011) pledged for monitoring the drought based on
the method of the RD and on the SPI, considering that the later index could
detect the onset and variations of the drought more adequately than the
former.
The methodological concept behind the SPI computation inspired the
development of analogous indices with applications in meteorology,
climatology, soil science or hydrology: Standardized Water-Level Index
(SWI; Bhuiyan, 2004); Standardized Streamflow Index (SSFI; Modarres,
2007); Standardized Runoff Index (SRI; Shukla and Wood, 2008);
44 Streamflow Drought Index (SDI; Nalbantis and Tsakiris, 2009); Relative
Standardized Precipitation Index (rSPI; Dubrovsky et al., 2009);
Standardized Precipitation Evapotranspiration Index (SPEI; VicenteSerrano et al., 2010); Standardized Flow Index (SFI) and Standardized Soil
Wetness Index (SSWI; Vidal et al., 2010); Cumulation of Reduced
Precipitation (CRP) and Weighted Average of Precipitation (WAP; Lu,
2011); Standardized Deficit Index (SDfI; Vergni and Todisco, 2011);
Standardized Hydrological Index (SHI; Sharma and Panu, 2012).
The SPI was often used for comparing the performances and
validating other indices, i.e. Dryness/Wetness Index (DWI; Shen et al.,
2008); Evapotranspiration Deficit Index (ETDI) and Soil Moisture Deficit
Index (SMDI; Narasimhan and Srinivasan, 2009); Scaled Drought
Condition Index (SDCI; Rhee et al., 2010); Synthesized Drought Index
(SynDI; Du et al., 2012b); Perpendicular Drought Index (PDI; Shahabfar et
al., 2012), or the performance of regional climate model simulations to
retrieve precipitation deficit (Maule et al., 2012).
5. STATUS OF THE OPERATIONAL USE
Several organizations have provided SPI-based interpolation products
in near-real time, i.e. National Drought Mitigation Center (NDMC),
Drought Management Centre for Southeastern Europe (DMCSEE),
European Drought Observatory (EDO), many journal articles report on the
SPI efficiency in monitoring activities, and one can expect an increasing
usage in the years to come (Hayes et al. 2011; WMO 2012). NDMC was
probably the first institution that implemented the SPI for operational use,
creating national US maps since 1996 (Hayes et al., 1999). At ten years
since its launch, the SPI was already one of the six key indicators combined
within the Drought Monitor, a product designed to describe the drought
conditions on a weekly basis (Svoboda et al., 2002). Goodrich and Ellis
(2006) provided a comparative analysis between SPI and PDSI specifically
to serve for a better implementation of a drought task force in Arizona.
Carbone et al. (2008) described a drought monitoring system designed and
implemented in the Carolinas (USA) that incorporates the PDSI, Palmer
Hydrological Drought Index (PHDI; , Moisture Anomaly Index (Z-index;
Palmer, 1965; Quiring and Ganesh, 2010), PMDI, Keetch-Byram Drought
Index (KBDI), CMI and SPI, while McRoberts and Nielsen-Gammon
(2012) spur for using the SPI as a high-resolution monitoring tool to assess
drought on multiple time scales for the areas where precipitation is the most
important triggering factor. However, Quiring (2009) reported that only 6 of
the 33 state drought plans used the SPI for monitoring, in favour of other
indices like reservoir levels, PDSI or precipitation amounts.
In Europe, Tsakiris and Vangelis (2004) proposed a procedure to
45 interpolate the SPI values over a complex terrain that could support a
Drought Watch System for an area at mesoscale extent. Mendicino and
Versace (2007) coupled SPI-based drought, satellite products and water
balance modeling into a GIS designed to serve the implementation of an
Integrated Drought Watch System in the Southern Italy. Further, Mendicino
et al. (2008) provided an example of water resources management in
agriculture, in which SPI thresholds indicate the pre-alert and alert state
about the drought. Caparrini and Manzella (2009) present the concept of an
integrated system for drought monitoring and water resources assessment in
Tuscany (Italy), based on the cross-evaluation of the SPI, vegetation indices
and hydrological model outputs.
Ceglar et al. (2012) describe the interoperability between DMCSEE
and EDO and observe that even low resolution precipitation datasets could
reproduce quite well the general drought temporal variability at shorter scale
in Slovenia, with a possible application in estimating the impact of the
meteorological drought on maize yield. Further, Medved-Cvikl et al. (2012)
evaluated the utility of the integrated monitoring system developed within
the DMCSEE.
Rouault and Richard (2005), and Kurnik et al. (2011) observed some
spatial characteristics of the drought in the southern Africa, demonstrated
the SPI’s superiority for drought monitoring at regional scales, and pointed
out the necessity to identify and build suitable data sets. Morid et al. (2006)
recommended SPI and EDI for operational drought monitoring in the
Tehran province of Iran, after comparing seven indices. Based on a case
study in South Asia, Smakhtin and Hughes (2007) promoted the operational
use of the SPI with a software package capable to estimate, display and
analyze multiple drought indices automatically. AghaKouchak and Nakhjiri
(2012) demonstrated that the characteristics of the SPI, the present
technological development and data availability favour the near real-time
drought monitoring even in ungauged areas. They promote a drought
monitoring tool that incorporates three large datasets of ground-based and
satellite retrieved information, namely Global Precipitation Climatology
Project (GPCP), Precipitation Estimation from Remotely Sensed
Information using Artificial Neural Networks (PERSIANN) and Tropical
Rainfall Measuring Mission (TRMM).
A summary list of the institutions providing operational SPI
information at different scales may be found in the at http://spisupport.blogspot.ro/p/operational-use.html/
6. CONCLUSIONS AND PERSPECTIVES
The SPI is characterized by technical credibility, policy relevance, and
technical adequacy (Karavitis et al., 2011), it brings important benefits to
46 the drought investigators, like simplicity and flexibility, and its drawbacks
can be overcome with reasonable efforts. Therefore, at two decades after the
first publication, the SPI has become one of the most known and used
drought indices in the world, getting to be a “universal meteorological
drought index” (Hayes et al., 2011).
The scientific literature referring to the SPI is abundant, and it tackles
aspects like methodology, suitability for different purposes and
geographical circumstances, performance, comparison with other indices,
and monitoring. In the first decade, the publications were fewer and
addressed mainly methodological and conceptual matters, but the number of
published articles and the diversity of applications have increased
significantly in the last years (2010-2012). The geographical coverage is
remarkable, reaching all the continents. While at least 70 countries have
already implemented the SPI in drought monitoring and assessment, the
operational applications can still be extended.
The debates have revealed that the SPI calculation should carefully
consider the PDF and time scale appropriate for the application and area
envisaged, as they can influence the results. There is no general agreement
on the use of thresholds and categories, and many studies have addressed
the SPI’s limitations, such as the performances in low precipitation areas
(Zhai and Feng, 2009), the focus on a single variable to characterize
complex phenomena (Gebrehiwot et al., 2011), or the deficiencies in
considering the conditions previous to the period of interest (Paulo and
Pereira, 2006). Therefore, one can assume that methodological adjustments
in the SPI computation and application should be still tackled in future
studies.
The original concept (McKee et al., 1993) has inspired the
computation of new indices (i.e. SRI, SDI, SPEI), while some
methodological developments refer to better spatial representativeness
(Dubrovsky et al., 2009), or introduce other variables (Vicente-Serrano et
al., 2010). Despite the numerous reports published in the recent years, the
combination of the SPI with other indices and the use of remote sensing
information are constantly encouraged as a way likely to improve the results
(Szalai and Szinell, 2000; Wu and Wilhite, 2004; Gebrehiwot et al., 2011;
Potop, 2011; WMO, 2012).
At present, the applications are oriented mainly to agriculture and
hydrology, but the potential is considerable higher (i.e. ecosystem research,
forest fire and insurance business). The SPI has clearly proved its utility in
assessing the drought variability, or for short- and long-term predictions.
Estrela and Vargas (2012) suggest that the SPI has still an unused potential
for drought forecasting, and Singleton (2012) considers that better
calibration, superior models, and right adjustment of the SPI calculation to
the purpose could add more accuracy to the drought forecasting. At the
same time, the wetness usage in wetness analysis should be reinforced.
47 While simplicity doubled with accuracy of the results sustain the SPI’s
authority over other drought indicators, the overview of the studies and
applications have revealed that the combined use of more indices is the best
strategy for securing the relevance of the outputs.
Acknowledgments
The work has been developed within the research project Changes in
climate extremes and associated impact in hydrological events in Romania
(CLIMHYDEX), code PN II-ID-2011-2-0073, sponsored by the National
Authority for Scientific Research.
References
Abarghouei HB, Zarch MAA, Dastorani MT, Kousari MR, Zarch MS. 2011. The survey of
climatic drought trend in Iran. Stochastic Environmental Research and Risk Assessment 25:
851-863, doi: 10.1007/s00477-011-0491-7.
Abolverdi J, Khalili D. 2010. Probabilistic analysis of extreme regional meteorological
droughts by L-moments in a semi-arid environment. Theoretical and Applied Climatology
102: 351-366.
AghaKouchak A, Nakhjiri N. 2012. A near real-time satellite-based global drought climate
data record. Environmental Research Letters 7: 044037, 8 pp, doi:10.1088/17489326/7/4/044037.
Agnew CT. 2000. Using the SPI to identify drought. Drought Network News. Vol. 12(1): 6-12.
Akhtari R, Bandarabdi SR, Saghafian B. 2008. Spatio-temporal pattern of drought in
Northeast of Iran. Options Méditerranéennes Série B. Etudes et Recherches, Series A 80:
71-77.
Akhtari R, Morid S, Mahdian MH, Smakhtin V. 2009. Assessment of areal interpolation
methods for spatial analysis of SPI and EDI drought indices. International Journal of
Climatology 29: 135–145, doi: 10.1002/joc.1691.
Al-Qinna MI, Hammouri NA, Obeidat ME, Ahmad FY. 2011. Drought analysis in Jordan
under current and future climates. Climatic Change 106, 421–440, doi 10.1007/s10584010-9954-y.
Ali A, Lebel T. 2008. The Sahelian standardized rainfall index revisited. International
Journal of Climatology 29: 1705-1714, doi: 10.1002/joc.1832.
Ali MG, Younes K, Esmaeil A, Fatemeh T. 2011. Assessment of Geostatistical Methods for
Spatial Analysis of SPI and EDI Drought Indices. World Applied Sciences Journal 15 (4):
474-482.
Anctil F, Larouche W, Viau AA, Parent LE. 2002. Exploration of the standardized precipitation index with regional analysis. Canadian Journal of Soil Science 82(1): 115-125.
Anderson MC, Hain C, Wardlow B, Pimstein A, Mecikalski JR. 2011. Evaluation of
Drought Indices Based on Thermal Remote Sensing of Evapotranspiration over the
Continental United States. Journal of Climate 24(8): 2025-2044.
Angelidis P, Maris F, Kotsovinos N, Hrissanthou V. 2012. Computation of Drought Index
SPI with Alternative Distribution Functions. Water Resources Management 26: 2453-2473,
doi: 10.1007/s11269-012-0026-0.
Angert A, Biraud S, Bonfils C, Henning CC, Buermann W, Pinzon J, Tucker CJ, Fung I.
2005. Drier summers cancel out the CO2 uptake enhancement induced by warmer springs.
Proceedings of the National Academy of Sciences of the United States of America 102(31):
10823-10827.
Asadi Zarch MA, Mobin MH, Malekinezhad H, Dastorani MT. 2011. Drought Monitoring
by Reconnaissance Drought Index (RDI) in Iran. Water Resourources Management
25:3485–3504, doi: 10.1007/s11269-011-9867-1.
48 Bacanli UG, Firat M, Dikbas F. 2009. Adaptive Neuro-Fuzzy Inference System for drought
forecasting. Stochastic Environmental Research and Risk Assessment 23:1143–1154, doi:
10.1007/s00477-008-0288-5.
Barua S, Perera BJC, Ng AWM. 2009. A comparative drought assessment of Yarra River
Catchment in Victoria, Australia. 18th World IMACS / MODSIM Congress, Cairns,
Australia 13-17 July 2009: 3245-3251.
Batisani N. 2011. The Spatio-Temporal-Severity Dynamics of Drought in Botswana.
Journal of Environmental Protection 2: 803-816, doi: 10.4236/jep.2011.26092.
Bazrafshan J, Hejabi S, Nokhandan MH. 2010. Is the SPI sufficient for monitoring
meteorological droughts in extreme coastal climates of Iran? Advances in Natural and
Applied Sciences 4(3): 345-351.
Belayneh A, Adamowski J. 2012. Standard Precipitation Index drought forecasting using
neural networks, wavelet neural networks, and support vector regression. Applied
Computational Intelligence and Soft Computing, Article ID 794061, 13 pp,
doi:10.1155/2012/794061.
Beniston M. 2012. Exploring the behaviour of atmospheric temperatures under dry
conditions in Europe: evolution since the mid-20th century and projections for the end of
the 21st century. International Journal of Climatology 10.1002/joc.3436.
Bhalme HN, Mooley DA. 1979. On the performance of modified Palmer index. Proceedings
of International Symposium: Hydrological Aspects of Droughts, vol. 1, New Delhi, India:
373–383.
Bhalme HN, Mooley DA. 1980: Large-scale droughts/floods and monsoon circulation.
Monthly Weather Review 108: 1197–1211.
Bhuiyan C. 2004. Various drought indices for monitoring drought condition in Aravalli
terrain of India. In: Proceedings of the XXth ISPRS Conference, Istanbul: 1283-1288.
Bhuiyan C, Singh RP, Kogan FN. 2006. Monitoring drought dynamics in the Aravalli
region (India) using different indices based on ground and remote sensing data.
International Journal of Applied Earth Observation and Geoinformation 8(4): 289-302,
doi: 10.1016/j.jag.2006.03.002.
Blain GC. 2011. Standardized precipitation index based on Pearson type III distribution.
Revista Brasileira de Meteorologia 26(2): 167-180.
Blain GC. 2012. Monthly values of the standardized precipitation index in the State of São
Paulo, Brazil: trends and spectral features under the normality assumption. Bragantia,
Campinas 71(1): 122-131.
Bonaccorso B, Bordi I, Cancelliere A, Rossi G, Sutera A. 2003. Spatial Variability of
Drought: An Analysis of the SPI in Sicily. Water Resources Management 17: 273–296.
Bonsal B, Regier M. 2007. Historical comparison of the 2001/2002 drought in the Canadian
Prairies. Climate Research 33: 229-242.
Bonsal BR, Wheaton EE, Chipanshi AC, Lin C, Sauchyn DJ, Wen L. 2011. Drought
Research in Canada: A Review. Atmosphere-Ocean 49(4): 303-319, doi:
10.1080/07055900.2011.555103.
Bonsal BR, Aider R, Gachon P, Lapp S. 2012. An assessment of Canadian prairie drought:
past, present, and future. Climate Dynamics doi: 10.1007/s00382-012-1422-0.
Bordi I, Sutera A. 2001. Fifty Years of Precipitation: Some Spatially Remote
Teleconnnections. Water Resources Management 15: 247–280.
Bordi I, Frigio S, Parenti P, Speranza A, Sutera A. 2001a. The analysis of the Standardized
Precipitation Index in the Mediterranean area: large-scale patterns. Annali di Geofisica 44
(5/6): 965-978.
Bordi I, Frigio S, Parenti P, Speranza A, Sutera A. 2001b. The analysis of the Standardized
Precipitation Index in the Mediterranean area: regional patterns. Annali di Geofisica 44
(5/6): 979-993.
Bordi I, Fraedrich K, Gerstengarbe F-W, Werner A, Sutera A. 2004. Potential predictability
of dry and wet periods: Sicily and Elbe-Basin (Germany). Theoretical and Applied
49 Climatology 77: 125-138, doi: 10.1007/s00704-003-0029-0.
Bordi I, Fraedrich K, Petitta M, Sutera A. 2007. Extreme value analysis of wet and dry
periods in Sicily. Theoretical and Applied Climatology 87: 61-71, doi: 10.1007/s00704005-0195-3.
Bordi I, Sutera A. 2008. Drought over Europe in recent years. Options Méditerranéennes,
Series A 80: 63-68.
Bordi I, Fraedrich K, Sutera A. 2009. Observed drought and wetness trends in Europe: an
update. Hydrology and Earth System Sciences Discussions 13: 1519-1530.
Bothe O, Fraedrich K, Zhu X. 2012a. Precipitation Tibetan Plateau summer precipitation:
covariability with circulation indices. Theoretical and Applied Climatology 108: 293-300,
doi: 10.1007/s00704-011-0538-1.
Bothe O, Fraedrich K, Zhu X. 2012b. Precipitation climate of Central Asia and the largescale atmospheric circulation. Theoretical and Applied Climatology 108: 345-354, doi:
10.1007/s00704-011-0537-2.
Brázdil R, Trnka M, Dobrovolný P, Chromá K, Hlavinka P, Žalud Z. 2009. Variability of
droughts in the Czech Republic, 1881–2006. Theoretical and Applied Climatology 97: 297315, doi: 10.1007/s00704-008-0065-x.
Buermann W, Lintner BR, Koven CD, Angert A, Pinzon JE, Tucker CJ, Fung IY. 2007. The
changing carbon cycle at Mauna Loa Observatory. Proceedings of the National Academy of
Sciences of the United States of America 104(11): 4249-4254,
doi:10.1073/pnas.0611224104.
Bumbaco KA, Mote PW. 2010. Three Recent Flavors of Drought in the Pacific Northwest.
Journal of Applied Meteorology and Climatology 49: 2058-2068.
Burke EJ, Brown SJ. 2008. Evaluating Uncertainties in the Projection of Future Drought.
Journal of Hydrometeorology 9: 292-299, doi: 10.1175/2007JHM929.1.
Byun HR, Wilhite DA. 1996. Daily quantification of drought severity and duration. Journal
of Climate 5: 1181–1201.
Byun HR, Wilhite DA. 1999. Objective Quantification of Drought Severity and Duration.
Journal of Climate 12: 2747-2756.
Byun HR, Kim DW. 2010. Comparing the Effective Drought Index and the Standardized
Precipitation Index. Options Méditerranéennes, Series A 95: 85-89.
Caccamo G, Chisholm LA, Bradstock RA, Puotinen ML. 2011. Assessing the sensitivity of
MODIS to monitor drought in high biomass ecosystems. Remote Sensing of Environment
115: 2626–2639, doi:10.1016/j.rse.2011.05.018.
Cancelliere A, Salas JD. 2004. Drought length properties for periodic-stochastic hydrologic
data. Water Resources Research 40 W02503, doi:10.1029/2002WR001750
Cancelliere A, Di Mauro G, Bonaccorso B, Rossi G. 2007. Drought forecasting using the
Standardized Precipitation Index. Water Resources Management 21: 801-819, doi:
10.1007/s11269-006-9062-y.
Caparrini F, Manzella F. 2009. Hydrometeorological and vegetation indices for the drought
monitoring system in Tuscany Region, Italy. Advances in Geosciences 17, 105-110.
Capra A, Scicolone B. 2012. Spatiotemporal variability of drought on a short–medium time
scale in the Calabria Region (Southern Italy). Theoretical and Applied Climatology 110:
471-488, doi: 10.1007/s00704-012-0720-0.
Carbone GJ, Dow K. 2005. Water Resource Management and Drought Forecasts in South
Carolina. Journal of the American Water Resources Association 41(1): 145-155, doi:
10.1111/j.1752-1688.2005.tb03724.x.
Carbone GJ, Rhee J, Mizzell HP, Boyles R. 2008. A Regional-Scale Drought Monitoring
Tool for the Carolinas. Bulletin of the American Meteorological Society 89(1): 20-28.
Cebrián AC, Abaurrea J. 2012. Risk measures for events with a stochastic duration: an
application to drought analysis. Stochastic Environmental Research and Risk Assessment
26: 971-981, doi: 10.1007/s00477-011-0521-5.
Ceglar A, Črepinšek Z, Kajfež-Bogataj L. 2008. Analysis of meteorological drought in
50 Slovenia with two drought indices. BALWOIS 2008 Proceedings, Ohrid, Republic of
Macedonia, 27- 31 May 2008, 10pp.
Ceglar A, Medved-Cvikl B, Moran-Tejeda E, VicenteSerrano S, Kajfež-Bogataj L. 2012.
Assessment of multi-scale drought datasets to quantify drought severity and impacts in
agriculture: a case study for Slovenia. International Journal of Spatial Data Infrastructures
Research 7: 464-487, doi: 10.2902/1725-0463.2012.07.art21.
Champagne C, McNairn H, Berg AA. 2011. Monitoring agricultural soil moisture extremes
in Canada using passive microwave remote sensing. Remote Sensing of Environment 115:
2434–2444.
Chen S-T, Kuo C-C, Yu P-S. 2009. Historical trends and variability of meteorological
droughts in Taiwan. Hydrological Sciences Journal 54(3): 430-441.
CNMA (Chinese National Meteorological Administration). 1981. Yearly charts of
dryness/wetness in China for the last 500-year period. Chinese Cartographic Publishing
House, Beijing.
Costa AC. 2011. Local patterns and trends of the Standard Precipitation Index in southern
Portugal (1940–1999). Advances in Geosciences 30: 11-16, doi: 10.5194/adgeo-30-11-2011.
Daly C, Neilson RP, Phillips DL. 1994. A statistical-topographic model for mapping
climatological precipitation over mountainous terrain. Journal of Applied Meteorology 33:
140–158.
Dash BK, Rafiuddin M, Khanam F, Nazrul Islam M. 2012. Characteristics of
meteorological drought in Bangladesh. Natural Hazards 64: 1461-1474, doi:
10.1007/s11069-012-0307-1.
De Martonne E. 1926. Une nouvelle fonction climatologique: l’indice d’aridité.
Météorologie 2, 449–458.
Di Matteo L, Valigi D, Cambi C. 2011. Climatic characterization and response of water
resources to climate change in limestone areas: some considerations on the importance of
geological setting. Symposium on Data-Driven Approaches to Droughts. Paper 41.
Di Mauro G, Bonaccorso B, Cancelliere A, Rossi G. 2008. Use of NAO index to improve
drought forecasting in the Mediterranean area: Application to Sicily region. Options
Méditerranéennes Série B. Etudes et Recherches, Series A 80: 311-318.
Dimitrakopoulos AP, Vlahou M, Anagnostopoulou CG, Mitsopoulos ID. 2011. Impact of
drought on wildland fires in Greece: implications of climatic change? Climatic Change
109:331–347 doi: 10.1007/s10584-011-0026-8.
Domonkos P. 2003. Recent Precipitation Trends in Hungary in the Context of Larger Scale
Climatic Changes. Natural Hazards 29: 255-271.
Drobyshev I, Niklasson M, Linderholm HW. 2012. Forest fire activity in Sweden: climatic
controls and geographical patterns in 20th century. Agricultural and Forest Meteorology
154-155: 174-186, doi: 10.1016/j.agrformet.2011.11.002.
Du J, Fang J, Xu W, Shi P. 2012a. Analysis of dry/wet conditions using the standardized
precipitation index and its potential usefulness for drought/flood monitoring in Hunan
Province, China. Stochastic Environmental Research and Risk Assessment doi:
10.1007/s00477-012-0589-6.
Du L, Tian Q, Yu T, Meng Q, Jancso T, Udvardy P, Huang Y. 2012b. A comprehensive
drought monitoring method integrating MODIS and TRMM data. International Journal of
Applied Earth Observation and Geoinformation, doi: 10.1016/j.jag.2012.09.010.
Dubrovsky M, Svoboda MD, Trnka M, Hayes MJ, Wilhite DA, Zalud Z, Hlavinka P. 2009.
Application of relative drought indices in assessing climate-change impacts on drought
conditions in Czechia. Theoretical and Applied Climatology 96: 155-171, 10.1007/s00704008-0020-x.
Durdu ÖF. 2010. Application of linear stochastic models for drought in the Büyük
Menderes river basin, western Turkey. Stochastic Environmental Research and Risk
Assessment 24:1145–1162, doi: 10.1007/s00477-010-0366-3.
Dutra E, Viterbo P, Miranda PMA. 2008. ERA-40 reanalysis hydrological applications in
51 the characterization of regional drought. Geophysical Research Letter 35, L19402,
doi:10.1029/2008GL035381.
Dutta D, Kundu A, Patel NR. 2012. Predicting agricultural drought in eastern Rajasthan of
India using NDVI and standardized precipitation index. Geocarto International
doi:10.1080/10106049.2012.679975.
Edwards DC, McKee TB. 1997. Characteristics of 20th century drought in the United States
at multiple time scales. Atmospheric Science Paper No. 634, Climatology Report 97–2,
Department of Atmospheric Science, Colorado State University, Fort Collins, Colorado,
155 pp.
Edossa DC, Babel MS, Gupta AD. 2010. Drought Analysis in the Awash River Basin,
Ethiopia. Water Resources Management 24: 1441–1460, doi: 10.1007/s11269-009-9508-0.
Erinç S. 1965. Climatology and its methods. Istanbul University, Institute of Geography
Press, Istanbul (in Turkish).
Estrela T, Vargas E. 2012. Drought Management Plans in the European Union. The Case of
Spain. Water Resources Management 26: 1537-1553, doi: 10.1007/s11269-011-9971-2.
Feng Su, Hu Q. 2004. Changes in agro-meteorological indicators in the contiguous United
States: 1951–2000. Theoretical and Applied Climatology 78: 246-264, doi:
10.1007/s00704-004-0061-8.
Foley JC. 1957. Droughts in Australia: Review of Records from Earliest Years of
Settlement to 1955. Australian Bureau of Meteorology Bulletin 43, 281 pp.
Ganguli P, Janga Reddy M. 2012. Risk Assessment of Droughts in Gujarat Using Bivariate
Copulas. Water Resources Management 26: 3301–3327, doi: 10.1007/s11269-012-0073-6.
Gebrehiwot T, van der Veena A, Maathuisb B. 2011. Spatial and temporal assessment of
drought in the Northern highlands of Ethiopia. International Journal of Applied Earth
Observation and Geoinformation 13: 309-321.
Ghasemi H, Gholami AA, Hadarbadi G, Ebrahimi H. 2011. Drought Monitoring Using
Climatic Indices and Geostatistic Technique (Case Study: Hossein Abad Plain, Sarbisheh,
Iran). Journal of Water Sciences Research 3(1): 37-43.
Gibbs WJ, Maher JV. 1967. Rainfall Deciles as Drought Indicators. Bureau of Meteorology
Bulletin No. 48. Commonwealth of Australia, Melbourne, 28 pp.
Giddings LM, Soto B, Rutherford M, Maarouf A. 2005. Standardized Precipitation Index
zones for México. Atmósfera Vol. 18(1): 33-56.
González J, Valdés JB. 2004. The mean frequency of recurrence of in-timemultidimensional events for drought analyses. Natural Hazards and Earth System Sciences
4: 17–28.
Goodrich GB, Ellis AW. 2006. Climatological Drought in Arizona: An Analysis of
Indicators for Guiding the Governor's Drought Task Force. The Professional Geographer
58(4): 460-469, doi: 10.1111/j.1467-9272.2006.00582.x.
Greene H, Leighton H, Stewart RE. 2011. Drought and associated cloud fields over the
Canadian Prairie Provinces. Atmosphere-Ocean 49(4): 356-365,
doi:10.1080/07055900.2011.559771.
Guerrero-Salazar P, Yevjievich V. 1975. Analysis of Drought Characteristics by the Theory
of Runs. Hydrology Paper 80, Fort Collins: Colorado State University, 44 pp.
Guttman NB. 1998. Comparing the Palmer Drought Index and the Standardized
Precipitation Index. Journal of The American Water Resources Association 34(1): 113-121,
doi: 10.1111/j.1752-1688.1998.tb05964.x.
Guttman NB. 1999. Accepting the Standardized Precipitation Index: A calculation
algorithm. Journal of the American Water Resources Association 35(2): 311–22.
Hallack-Alegria M, Watkins Jr DW. 2007. Annual and Warm Season Drought Intensity–
Duration–Frequency Analysis for Sonora, Mexico. Journal of Climate 20: 1897-1909.
Hanesiak JM, Stewart RE, Bonsal BR, Harder P, Lawford R, Aider R, Amiro BD, Atallah E,
Barr AG, Black TA, Bullock P, Brimelow JC, Brown R, Carmichael H, Derksen C,
Flanagan LB, Gachon P, Greene H, Gyakum J, Henson W, Hogg EH, Kochtubajda B,
52 Leighton H, Lin C, Luo Y, McCaughey JH, Meinert A, Shabbar A, Snelgrove K, Szeto K,
Trishchenko A, van der Kamp G, Wang S, Wen L, Wheaton E, Wielki C, Yang Y, Yirdaw
S, Zha T. 2011. Characterization and Summary of the 1999–2005 Canadian Prairie Drought.
Atmosphere-Ocean 49(4): 421-452.
Hannaford J, Lloyd-Hughes B, Keef C, Parry S, Prudhomme C. 2011. Examining the large
scale spatial coherence of European drought using regional indicators of precipitation and
streamflow deficit, Hydrological Processes 25, 1146-1162.
Hayes MJ, Svoboda MD, Wilhite DA, Vanyarkho OV. 1999. Monitoring the 1996 drought
using the Standardized Precipitation Index. Bulletin of the American Meteorological
Society 80(3): 429-438.
Hayes MJ. 2000. Revisiting the SPI: clarifying the process. Drought Network News Vol.
12(1): 13-14.
Hayes M, Svoboda M, Wall N, Widham M. 2011. The Lincoln Declaration on Drought
Indices: Universal meteorological Drought Index Recommended. Bulletin of the American
Meteorological Society 92(4): 485-488, doi: 10.1175/2010BAMS3103.1.
Heim Jr RR. 2002. A Review of Twentieth Century Drought Indices Used in the United
States. Bulletin of the American Meteorological Society 83(8): 1149-1165.
Heinrich G, Gobiet A. 2012. The future of dry and wet spells in Europe: A comprehensive
study based on the ENSEMBLES regional climate models. International Journal of
Climatology 32(13): 1951–1970, doi: 10.1002/joc.2421.
Hirschi M, Seneviratne SI, Alexandrov V, Boberg F, Boroneant C, Christensen OB,
Formayer H, Orlowsky B, Stepanek P. 2011. Observational evidence for soil-moisture
impact on hot extremes in southeastern Europe. Nature Geoscience 4: 17–21, doi:
10.1038/NGEO1032.
Hofer B, Carrão H, Mcinerney D. 2012. Multi-Disciplinary Forest Fire Danger Assessment
in Europe: The Potential to Integrate Long-Term Drought Information. International
Journal of Spatial Data Infrastructures Research 7, 300-322, doi: 10.2902/17250463.2012.07.arty.
Hwang Y, Carbone GJ. 2009. Ensemble Forecasts of Drought Indices Using a Conditional
Residual Resampling Technique. Journal of Applied and Climatology 48: 1289-1301, doi:
10.1175/2009JAMC2071.1.
Iglesias A, Quiroga S. 2007. Measuring the risk of climate variability to production at five
sites in Spain. Climate Research 34: 47-57.
Imanov FA, Mammadov AS, Hasanova NI. 2012. Investigation of droughts in the Lankaran
region of Azerbaijan. Journal of Water and Land Development 16(I-VI): 11-15, doi:
10.2478/v10025-012-0019-7.
IPCC. 2007. Climate Change 2007: The Physical Science Basis. Contribution of Working
Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate
Change [Solomon, S., D. Qin, M. Manning, Z. Chen M. Marquis, K.B. Averyt, M.Tignor
and H.L. Miller (eds.)]. Cambridge University Press, Cambridge, United Kingdom and
New York, NY, USA.
Jain SK, Keshri R, Goswami A, Sarkar A. 2010. Application of meteorological and
vegetation indices for evaluation of drought impact: a case study for Rajasthan, India.
Natural Hazards 54: 643-656, doi: 10.1007/s11069-009-9493-x.
Ji L, Peters AJ. 2003. Assessing vegetation response to drought in the northern Great Plains
using vegetation and drought indices. Remote Sensing of Environment 87: 85–98,
doi:10.1016/S0034-4257(03)00174-3.
Joetzjer E, Douville H, Delire C, Ciais P, Decharme B, Tyteca S. 2012. Evaluation of
drought indices at interannual to climate change timescales: a case study over the Amazon
and Mississippi river basins. Hydrol. Earth Syst. Sci. Discuss. 9: 13231–13249, doi:
10.5194/hessd-9-13231-2012.
Jones PD, Hulme M. 1996. Calculating regional climatic time series for temperature and
precipitation: Methods and illustrations. International Journal of Climatology 16: 361–377.
53 Jung IW, Chang H. 2012. Climate change impacts on spatial patterns in drought risk in the
Willamette River Basin, Oregon, USA. Theoretical and Applied Climatology 108: 355-371,
doi: 10.1007/s00704-011-0531-8.
Kangas RS, Brown TJ. 2007. Characteristics of US drought and pluvials from a highresolution spatial dataset. International Journal of Climatology 27: 1303-1325, doi:
10.1002/joc.1473.
Karavitis CA, Alexandris S, Tsesmelis DE, Athanasopoulos G. 2011. Application of the
Standardized Precipitation Index (SPI) in Greece. Water 3: 787-805,
doi:10.3390/w3030787.
Karl TR, Knight RW. 1985, Atlas of Monthly Palmer Hydrological Drought Indices (19311983) for the Contiguous United States. Historical Climatology Series 3–7, Natl. Clim.
Data Cent., Asheville, N.C.
Kasei R, Diekkrüger B, Leemhuis C. 2010. Drought frequency in the Volta Basin of West
Africa. Sustain. Sci.: 5: 89-97, doi: 10.1007/s11625-009-0101-5.
Keetch JJ, Byram GM. 1968. A drought index for forest fire control. Forest Service
Research Paper SE-38, U.S. Dept. of Agriculture, 32 pp.
Keskin F, Sorman AU. 2010. Assessment of the dry and wet period severity with
hydrometeorological index. International Journal of Water Resources and Environmental
Engineering 2(2): 29-39.
Keskin ME, Terzi Ö, Taylan ED, Küçükyaman D. 2009. Meteorological drought analysis
using data-driven models for the Lakes District, Turkey. Hydrological Sciences Journal
54(6): 1114-1124, doi: 10.1623/hysj.54.6.1114.
Keskin ME, Terzi Ö, Taylan ED, Küçükyaman D. 2011. Meteorological drought analysis
using artificial neural networks. Scientific Research and Essays 6(21): 4469-4477, doi:
10.5897/SRE10.1022
Keyantash JA, Dracup JA. 2002. The Quantification of Drought: An Evaluation of Drought
Indices. Bulletin of the American Meteorological Society 83: 1167-1180.
Keyantash JA Dracup JA. 2004. An aggregate drought index: Assesing drought severity
based on fluctuations in the hydrologic cycle & surface water storage. Water Resources
Research 40 W09304, doi:10.1029/2003WR002610.
Kim D-W, Byun H-R, Choi K-S. 2009. Evaluation, modification, and application of the
Effective Drought Index to 200-Year drought climatology of Seoul, Korea. Journal of
Hydrology 378: 1-12, doi: 10.1016/j.jhydrol.2009.08.021.
Khadr M, Morgenschweis G, Schlenkhoff A. 2009. Analysis of Meteorological Drought in
the Ruhr Basin by Using the Standardized Precipitation Index. World Academy of Science,
Engineering and Technology 33: 607-616.
Khalili D, Jamshidi TFH, Haghighi AAK, Zand-Parsa S. 2011. Comparability Analyses of
the SPI and RDI Meteorological Drought Indices in Different Climatic Zones. Water
Resources Management 25: 1737–1757, doi: 10.1007/s11269-010-9772-z.
Khan S, Gabriel HF, Rana T. 2008. Standard precipitation index to track drought and assess
impact of rainfall on watertables in irrigation areas. Irrigation Drainage Systems 22: 159–
177, doi: 10.1007/s10795-008-9049-3.
Kogan FN. 1990. Remote sensing of weather impacts on vegetation in nonhomgeneous
areas. International Journal of Remote Sensing 11: 1405–1419.
Komuscu AU. 1999. Using the SPI to Analyze Spatial and Temporal Patterns of Drought in
Turkey. Drought Network News Vol. 11(1): 7-13.
Koutroulis AG, Vrohidou A-EK, Tsanis IK. 2011. Spatiotemporal Characteristics of
Meteorological Drought for the Island of Crete. Journal of Hydrometeorology 12: 206-226,
doi:10.1175/2010JHM1252.1.
Krepper CM, Zucarelli GV. 2010. Climatology of water excesses and shortages in the La
Plata Basin. Theoretical and Applied Climatology 102: 13–27, doi: 10.1007/s00704-0090234-6.
Krysanova V, Vetter T, Hattermann F. 2008. Detection of change in drought frequency in
54 the Elbe basin: comparison of three methods. Hydrological Sciences Journal 53(3), 519537, doi: 10.1623/hysj.53.3.519.
Kurnik B, Barbosa P, Vogt J. 2011. Testing two different precipitation datasets to compute
the standardized precipitation index over the Horn of Africa. International Journal of
Remote Sensing 32(21): 5947–5964, doi: 10.1080/01431161.2010.499380.
Lana X, Serra C, Burgueno A. 2001. Patterns of Monthly Rainfall Shortage and Excess in
Terms of the Standardized Precipitation Index for Catalonia (NE Spain). International
Journal of Climatology 21: 1669–1691, doi: 10.1002/joc.697.
Levanič T, Popa I, Poljanšek S, Nechita C. 2012. A 323-year long reconstruction of drought
for SW Romania based on black pine (Pinus Nigra) tree-ring widths. International Journal
of Biometeorology, doi: 10.1007/s00484-012-0596-9.
Li W, Fu R, Juárez RIN, Fernandes K. 2008. Observed change of the standardized
precipitation index, its potential cause and implications to future climate change in the
Amazon region. Philosphical Transactions of the Royal Society B 363: 1767–1772,
doi:10.1098/rstb.2007.0022.
Li Y, Zheng X, Lu F, Ma J. 2012. Analysis of Drought Evolvement Characteristics Based
on Standardized Precipitation Index in the Huaihe River Basin. Procedia Engineering 28:
434-437, doi:10.1016/j.proeng.2012.01.746.
Linderholm HW, Molin T. 2005. Early nineteenth century drought in east central Sweden
inferred from dendrochronological and historical archives. Climate Research 29: 63-72.
Liu C-L, Zhang Q, Singh VP, Cui Y. 2011. Copula-based evaluations of drought variations
in Guangdong, South China. Natural Hazards 59: 1533-1546, doi: 10.1007/s11069-0119850-4.
Liu L, Hong Y, Bednarczyk CN, Yong B, Shafer MA, Riley R, Hocker JE. 2012. HydroClimatological Drought Analyses and Projections Using Meteorological and Hydrological
Drought Indices: A Case Study in Blue River Basin, Oklahoma. Water Resources
Management 26:2761–2779, doi: 10.1007/s11269-012-0044-y.
Liu X, Xu Z, Liu B. 2009. Spatio-Temporal Characteristics of Standardized Precipitation
Index in the Taihu Basin during 1951-2000. Wuhan University Journal of Natural Sciences
14(6): 518-524, doi: 10.1007/s11859-009-0611-0.
Liu X, Xu Z, Yu R. 2012. Spatiotemporal variability of drought and the potential
climatological driving factors in the Liao River basin. Hydrological Processes 26: 1-14,
doi: 10.1002/hyp.8104.
Livada I, Assimakopoulos VD. 2007. Spatial and temporal analysis of drought in Greece
using the Standardized Precipitation Index (SPI). Theoretical and Applied Climatology 89:
143-153, doi: 10.1007/s00704-005-0227-z.
Lloyd-Hughes B, Saunders MA. 2002. Seasonal prediction of European spring precipitation
from El Niño-southern oscillation and local sea-surface temperatures. International Journal
of Climatology 22: 1–14.
Logan KE, Brunsell NA, Jones AR, Feddema JJ. 2010. Assessing spatiotemporal variability
of drought in the U.S. central plains. Journal of Arid Environments 74: 247–255, doi:
10.1016/j.jaridenv.2009.08.008.
Lorenzo-Lacruz J, Vicente SM, López-Moreno JI, Beguería S, García-Ruiz JM, Cuadrat
JM. 2010. The impact of droughts and water management on various hydrological systems
in the headwaters of the Tagus River (central Spain). Journal of Hydrology 386: 13-26.
Lotsch A, Friedl MA, Anderson BT. 2003. Coupled vegetation-precipitation variability
observed from satellite and climate records. Geophysical Research Letters 30(14), 1774,
doi:10.1029/2003GL017506.
Loukas A, Vasiliades L. 2004. Probabilistic analysis of drought spatiotemporal
characteristics in Thessaly region, Greece. Natural Hazards and Earth System Sciences 4:
719-731.
Loukas A, Vasiliades L, Tzabiras J. 2008. Climate change effects on drought severity.
Advances in Geosciences 17: 23-29.
55 López-Moreno JI, Vicente-Serrano SM. 2008. Positive and Negative Phases of the
Wintertime North Atlantic Oscillation and Drought Occurrence over Europe: A
Multitemporal-Scale Approach. Journal of Climate 21: 1220-1243, doi:
10.1175/2007JCLI1739.1.
Lu E. 2011. A Method for Monitoring Meteorological Drought at Daily Scale. Science and
Technology Infusion Climate Bulletin. 36th NOAA Annual Climate Diagnostics and
Prediction Workshop Fort Worth, TX, 3-6 October 2011: 33-39.
Lyon B, Zubair L, Ralapanawe V, Yahiya Z. 2009. Finescale Evaluation of Drought in a
Tropical Setting: Case Study in Sri Lanka. Journal of Applied Meteorology and
Climatology 48: 77-88, doi: 10.1175/2008JAMC1767.1.
Łabedzki L. 2007. Estimation of local drought frequency in central Poland using the
standardized precipitation index SPI. Irrigation and Drainage 56(1): 67–77, doi:
10.1002/ird.285.
Manatsa D, Chingombe W, Matsikwa H, Matarira CH. 2008.The superior influence of
Darwin Sea level pressure anomalies over ENSO as a simple drought predictor for
Southern Africa. Theoretical and Applied Climatology 92: 1-14, doi: 10.1007/s00704-0070315-3.
Manatsa D, Mukwada G, Siziba E, Chinyanganya T. 2010. Analysis of multidimensional aspects of agricultural droughts in Zimbabwe using the Standardized Precipitation Index (SPI).
Theoretical and Applied Climatology 102: 287–305, doi: 10.1007/s00704-010-0262-2.
Mansouri Daneshvar RM, Bagherzadeh A, Khosravi M. 2012. Assessment of drought
hazard impact on wheat cultivation using standardized precipitation index in Iran. Arabian
Journal of Geosciences, doi: 10.1007/s12517-012-0695-2.
Masoudi M, Afrough E. 2011. Analyzing trends of precipitation for humid, normal and
drought classes using Standardized Precipitation Index (SPI), a case of study: Fars
Province, Iran. International Journal of AgriScience 1(2): 58-69.
Maule CF, Thejll P, Christensen JH, Svendsen SH, Hannaford J. 2012. Improved
confidence in regional climate model simulations of precipitation evaluated using drought
statistics from the ENSEMBLES models. Climate Dynamics, doi: 10.1007/s00382-0121355-7.
Mavromatis T. 2007. Drought index evaluation for assessing future wheat production in
Greece. International Journal of Climatology 27: 911–924, doi: 10.1002/joc.1444.
McRoberts DB, Nielsen-Gammon JW. 2012. The Use of a High-Resolution Standardized
Precipitation Index for Drought Monitoring and Assessment. Journal of Applied
Meteorology and Climatology 51(1): 68-83.
McKee TB, Doesken NJ, Kleist J. 1993. The relationship of drought frequency and duration
to time scales. Proceedings of the 8th Conference on Applied Climatology, American
Meteorological Society, Boston: 179-184.
McKee TB, Doesken NJ, Kleist J. 1995. Drought monitoring with multiple time scales.
Preprints, 9th Conference on Applied Climatology, 15-20 January, Dallas, Texas, American
Meteorological Society: 233-236.
Medved-Cvikl B, Ceglar A, Hofer B, Kajfež-Bogataj L. 2012. Benefits of an integrated
drought monitoring system in Southeastern Europe. Article under Review for the
International Journal of Spatial Data Infrastructures Research, submitted 2012-02-14.
http://ijsdir.jrc.ec.europa.eu/index.php/ijsdir/article/viewFile/275/310
Mendicino G, Versace P. 2007. Integrated Drought Watch System: A Case Study in
Southern Italy. Water Resources Management 21:1409–1428, doi: 10.1007/s11269-0069091-6.
Mendicino G, Senatore A, Versace P. 2008. Water Resources Management in Agriculture
under Drought and Water Shortage Conditions: A Case Study in Southern Italy. European
Water 23/24: 41-56.
Meyer SJ, Hubbard KG, Wilhite DA. 1993 A crop-specific drought index for corn: I. Model
development and validation. Agronomy Journal 86: 388–395.
56 Michaelides S, Pashardis S. 2008. Monitoring Drought in Cyprus during the 2007-2008
Hydrometeorological Year by using the Standardized Precipitation Index (SPI). European
Water 23/24: 123-131.
Min S-K, Kwon W-T, Park E-H, Choi Y. 2003. Spatial and Temporal Comparisons of
Droughts over Korea with East Asia. International Journal of Climatology 23: 223–233,
doi: 10.1002/joc.872.
Mishra AK, Desai VR. 2005. Drought forecasting using stochastic models. Stochastic
Environmental Research and Risk Assessment 19, 326–339, doi: 10.1007/s00477-0050238-4.
Mishra AK, Singh VP, Desai VR. 2009. Drought characterization: a probabilistic approach.
Stochastic Environmental Research and Risk Assessment 23 (1), 41–55, doi:
10.1007/s00477-007-0194-2.
Mishra AK, Singh VP. 2010. A review of drought concepts. Journal of Hydrology 391:
202–216, doi:10.1016/j.jhydrol.2010.07.012.
Mishra AK, Singh VP. 2011. Drought modeling – A review. Journal of Hydrology 403:
157-175, doi: 10.1016/j.jhydrol.2011.03.049.
Mishra SS, Nagarajan R. 2011. Spatio-temporal drought assessment in Tel river basin using
Standardized Precipitation Index (SPI) and GIS. Geomatics, Natural Hazards and Risk
2(1): 79-93, doi: 10.1080/19475705.2010.533703.
Mishra SS, Nagarajan R. 2012. Correlation of SPI with Irrigation Demand for Assessment
of Agricultural Drought in Tel River Basin of Odisha,India. Journal of Agricultural Science
and Applications 1(3): 74-78.
Mishra V, Cherkauer KA, Shukla S. 2010. Assessment of Drought due to Historic Climate
Variability and Projected Future Climate Change in the Midwestern United States. Journal
of Hydrometeorology 11: 46-68, doi: 10.1175/2009JHM1156.1.
Mihajlović D. 2006. Monitoring the 2003–2004 Meteorological Drought over Pannonian
Part of Croatia. International Journal of Climatology 26: 2213-2225, doi:
10.1002/joc.1366.
Mkhabela M, Bullock P, Gervais M, Finlay G, Sapirstein H. 2010. Assessing indicators of
agricultural drought impacts on spring wheat yield and quality on the Canadian prairies.
Agricultural and Forest Meteorology 150: 399–410, doi: 10.1016/j.agrformet.2010.01.001.
Mo KC. 2008. Model-Based Drought Indices over the United States. Journal of
Hydrometeorology 9 1212-1230.
Mo KC, Schemm J-KE, Yoo S-H. 2009. Influence of ENSO and the Atlantic Multidecadal
Oscillation on Drought over the United States. Journal of Climate 22: 5962-5982, doi:
10.1175/2009JCLI2966.1.
Mo KC, Berbery EH. 2011. Drought and Persistent Wet Spells over South America Based
on Observations and the U.S. CLIVAR Drought Experiments. Journal of Climate 24(6):
1801-1820.
Modarres R. 2007. Streamflow drought time series forecasting. Stochastic Environmental
Research and Risk Assessment 21: 223-233, doi: 10.1007/s00477-006-0058-1.
Mokhtari A, Mansor SB, Mahmud AR, Helmi ZM. 2011. Monitoring the Impacts of
Drought on Land Use/Cover: A Developed Object-based Algorithm for NOAA AVHRR
Time Series Data. Journal of Applied Sciences 11(17): 3089-3103, doi:
10.3923/jas.2011.3089.3103.
Moreira EE, Coelho CA, Paulo AA, Pereira LS, Mexia JT. 2008. SPI-based drought
category prediction using loglinear models. Journal of Hydrology 354: 116– 130, doi:
10.1016/j.jhydrol.2008.03.002.
Moreira EE, Mexia JT, Pereira LS. 2012a. Are drought occurrence and severity
aggravating? A study on SPI drought class transitions using log-linear models and
ANOVA-like inference. Hydrology and Earth System Sciences, 16: 3011–3028,
doi:10.5194/hess-16-3011-2012.
Moreira EE, Mexia JT, Pereira LS. 2012b. Assessing homogeneous regions relative to
57 drought class transitions using an ANOVA-like inference. Application to Alentejo,
Portugal. Stochastic Environmental Research and Risk Assessment, doi:10.1007/s00477012-0575-z.
Morid S, Smakhtin V, Moghaddasi M. 2006. Comparison of Seven Meteorological Indices
for Drought Monitoring in Iran. International Journal of Climatology. 26: 971–985, doi:
10.1002/joc.1264.
Mueller B, Seneviratne SI. 2012. Hot days induced by precipitation deficits at the global
scale. Proceedings of the National Academy of Sciences of the United States of America
109(31): 12398-12403, doi:10.1073/pnas.1204330109/-/DCSupplemental.
Myronidis D, Stathis D, Ioannou K, Fotakis D. 2012. An Integration of Statistics Temporal
Methods to Track the Effect of Drought in a Shallow Mediterranean Lake. Water
Resources Management 26: 4587–4605 doi: 10.1007/s11269-012-0169-z.
Nalbantis I, Tsakiris G. 2009. Assessment of Hydrological Drought Revisited. Water
Resources Management 23:881–897, doi: 10.1007/s11269-008-9305-1.
Narasimhan B, Srinivasan R. 2005. Development and evaluation of Soil Moisture Deficit
Index (SMDI) and Evapotranspiration Deficit Index (ETDI) for agricultural drought
monitoring. Agricultural and Forest Meteorology 133: 69–88, doi:
10.1016/j.agrformet.2005.07.012.
Naresh Kumar M, Murthy CS, Sesha Saib MVR, Roy PS. 2009. On the use of Standardized
Precipitation Index (SPI) for drought intensity assessment. Meteorological Applications 16:
381–389, doi: 10.1002/met.136.
Nikolova N, Alieva G, Voislavova I. 2012. Drought Periods in Non-Mountainous Part of
South Bulgaria on the Background of Climate Change. Geographica Pannonica 16: 18-25.
Norouzi A, Nohegar A, Ghorbani A. 2012. Comparison of the suitability of standardized
precipitation index (SPI) and aggregated drought index (ADI) in Minab Watershed
(Hormozgan Province/South of Iran). African Journal of Agricultural Research 7(44):
5905-5911, doi: 10.5897/AJAR12.1521.
Ntale HK, Gan TY. 2003. Drought Indices and their Application to East Africa.
International Journal of Climatology 23: 1335-1357, doi: 10.1002/joc.931.
Orlowsky B, Seneviratne SI. 2012. Elusive drought: uncertainty in observed trends and
short- and long-term CMIP5 projections. Hydrol. Earth Syst. Sci. Discuss., 9: 13773–
13803, doi:10.5194/hessd-9-13773-2012.
Pai DS, Sridhar L, Guhathakurta P, Hatwar HR. 2011. District-wide drought climatology of
the southwest monsoon season over India based on standardized precipitation index (SPI).
Natural Hazards 59: 1797-1813, doi: 10.1007/s11069-011-9867-8.
Palmer WC. 1965. Meteorological drought. Research Paper 45, U.S. Dept. of Commerce.
Weather Bureau, Washington D.C. 58 pp.
Pandey RP, Pandey A, Galkate RV, Byun H-R, Mal BC. 2010. Integrating HydroMeteorological and Physiographic Factors for Assessment of Vulnerability to Drought.
Water Resources Management 24: 4199–4217, doi: 10.1007/s11269-010-9653-5.
Parry S, Hannaford J, Prudhomme C, Lloyd-Hughes B, Williamson. 2011. Objective
Drought and High Flow Catalogues for Europe. Water and Global Change Technical
Report No 33, 57 pp.
Pashardis S, Michaelides S. 2008. Implementation of the Standardized Precipitation Index
(SPI) and the Reconnaissance Drought Index (RDI) for Regional Drought Assessment A
Case Study. European Water 23/24: 57-64.
Pasho E, Camarero JJ, de Luis M, Sergio M. Vicente-Serrano SM. 2011. Impacts of drought
at different time scales on forest growth across a wide climatic gradient in north-eastern
Spain. Agricultural and Forest Meteorology 151: 1800– 1811.
Pasho E, Camarero JJ, Vicente-Serrano SM. 2012. Climatic impacts and drought control of
radial growth and seasonal wood formation in Pinus halepensis. Trees 26: 1875-1886.
Patel NR, Chopra P, Dadhwala VK. 2007. Analyzing spatial patterns of meteorological
drought using standardized precipitation index. Meteorological Applications 14: 329-336,
58 doi: 10.1002/met.33.
Paulo AA, Pereira LS. 2006. Drought Concepts and Characterization: Comparing Drought
Indices Applied at Local and Regional Scales. Water International 31(1): 37-49, doi:
10.1080/02508060608691913.
Paulo AA, Pereira LS. 2007. Prediction of SPI Drought Class Transitions Using Markov
Chains. Water Resources Management 21: 1813-1827, doi: 10.1007/s11269-006-9129-9.
Paulo AA, Pereira LS. 2008. Stochastic Prediction of Drought Class Transitions. Water
Resources Management 22: 1277-1296, doi: 10.1007/s11269-007-9225-5.
Paulo AA, Rosa RD,, Pereira LS. 2012. Climate trends and behaviour of drought indices
based on precipitation and evapotranspiration in Portugal. Natural Hazards Earth System
Sciences 12, 1481–1491, doi: 10.5194/nhess-12-1481-2012.
Păltineanu C, Mihăilescu IF, Prefac Z, Dragotă C, Vasenciuc F. 2009. Combining the
standardized precipitation index and climatic water deficit in characterizing droughts: a
case study in Romania. Theoretical and Applied Climatology 97: 219-233.
Peled E, Dutra E, Viterbo P, Angert A. 2010. Technical Note: Comparing and ranking soil
drought indices performance over Europe, through remote-sensing of vegetation.
Hydrological Hazards Earth System Sciences 14: 271-277.
Ped DA. 1975. On indicators of droughts and wet conditions (in Russian). Proceedings of
the USSR Hydrometeorological Centre 156: 19-39.
Piccarreta M, Capolongo D, Boenzi F. 2004. Trend analysis of precipitation and drought in
Basilicata from 1923 to 2000 within a southern Italy context. International Journal of
Climatology 24(7): 907-922, doi: 10.1002/joc.1038.
Potop V, Soukup J. 2009. Spatiotemporal characteristics of dryness and drought in the
Republic of Moldova. Theoretical and Applied Climatology 96: 305-318, doi:
10.1007/s00704-008-0041-5.
Potop V. 2011. Evolution of drought severity and its impact on corn in the Republic of
Moldova. Theoretical and Applied Climatology 105: 469-483, doi: 10.1007/s00704-0110403-2.
Potop V, Možný M, Soukup J. 2012. Drought evolution at various time scales in the
lowland regions and their impact on vegetable crops in the Czech Republic. Agricultural
and Forest Meteorology 156: 121-133, doi: 10.1016/j.agrformet.2012.01.002.
Qin Q, Ghulam A, Zhu L, Wang L, Li J, Nan P. 2008. Evaluation of MODIS derived
perpendicular drought index for estimation of surface dryness over northwestern China.
International Journal of Remote Sensing 29(7): 1983-1995.
Quiring SM. 2009. Developing Objective Operational Definitions for Monitoring Drought.
Journal of Applied Meteorology and Climatology 48: 1217-1229, doi:
10.1175/2009JAMC2088.1.
Quiring SM, Ganesh S. 2010. Evaluating the utility of the Vegetation Condition Index
(VCI) for monitoring meteorological drought in Texas. Agricultural and Forest
Meteorology 150: 330–339, doi:10.1016/j.agrformet.2009.11.015.
Raziei T, Saghafian B, Paulo AA, Pereira LS, Bordi I. 2009. Spatial Patterns and Temporal
Variability of Drought in Western Iran. Water Resources Management 23: 439–455,
doi:10.1007/s11269-008-9282-4.
Redmond KT. 2002. The depiction of drought: A commentary. Bulletin of the American
Meteorological Society, 83: 1143-1147.
Redmond K. 2011. Whither U.S. Drought: Much Ado About Nothing. Science and
Technology Infusion Climate Bulletin NOAA’s National Weather Service. 36th NOAA
Annual Climate Diagnostics and Prediction Workshop Fort Worth, TX, 3-6 October 2011:
19-24.
Reich RM, Aguirre-Bravo C, Bravo VA. 2008. New approach for modeling climatic data
with applications in modeling tree species distributions in the states of Jalisco and Colima,
Mexico. Journal of Arid Environments 72: 1343-1357, doi: 10.1016/j.jaridenv.2008.02.004.
Rezaeian-Zadeh M, Tabari H. 2012. MLP-based drought forecasting in different climatic
59 regions. Theoretical and Applied Climatology 109: 407–414, doi: 10.1007/s00704-0120592-3.
Rhee J, Carbone GJ. 2008. Drought Index Mapping at Different Spatial Units. Journal of
Hydrometeorology 9: 1523-1534, doi: 10.1175/2008JHM983.1.
Rhee J, Im J, Carbone GJ. 2010. Monitoring agricultural drought for arid and humid regions
using multi-sensor remote sensing data. Remote Sensing of Environment 114: 2875-2887,
doi:10.1016/j.rse.2010.07.005.
Rhee J, Carbone GJ. 2011. Estimating Drought Conditions for Regions with Limited
Precipitation Data. Journal of Applied Meteorology and Climatology 50(3): 548-559.
Rouault M, Richard Y. 2005. Intensity and spatial extent of droughts in southern Africa.
Geophysical Research Letters 32 L15702, doi:10.1029/2005GL022436.
Roudier P, Mahe G. 2010. Study of water stress and droughts with indicators using daily
data on the Bani River (Niger basin, Mali). International Journal of Climatology 30: 16891705, doi : 10.1002/joc.2013.
Sadat Noori SM, Liaghat AM, Ebrahimi K. 2012. Prediction of crop production using
drought indices at different time scales and climatic factors to manage drought risk. Journal
of the American Water Resources Association 48(1): 1-9.
Santos JF, Portela MM, Pulido-Calvo I. 2011. Regional Frequency Analysis of Droughts in
Portugal. Water Resources Management 25: 3537-3558, doi: 10.1007/s11269-011-9869-z.
Scian B, Labraga JC, Reimers W, Frumento O. 2006. Characteristics of large-scale
atmospheric circulation related to extreme monthly rainfall anomalies in the Pampa Region,
Argentina, under non-ENSO conditions. Theoretical and Applied Climatology 85: 89-106,
doi: 10.1007/s00704-005-0182-8.
Seiler RA, Hayes M, Bressan L. 2002. Using the standardized precipitation index for flood
risk monitoring. International Journal of Climatology 22: 1365-1376, doi: 10.1002/joc.799.
Seftigen K, Linderholm HW, Drobyshev I, Niklasson M. 2012. Reconstructed drought
variability in southeastern Sweden since the 1650s. International Journal of Climatology
doi: 10.1002/joc.3592.
Selianinov GT. 1928. On agricultural climate valuation (in Russian). Proceedings of
Agricultural Meteorology 20:165-177.
Sen B, Topcu S, Türkeş M, Sen B, Warner JF. 2012. Projecting climate change, drought
conditions and crop productivity in Turkey. Climate Research 52: 175–191, doi:
10.3354/cr01074.
Shahabfar A, Reinwand M, Conrad C, Schorcht G. 2012b. A re-examination of
perpendicular drought indices over Central and South-West Asia. Proceedings of SPIE
8531, Remote Sensing for Agriculture, Ecosystems, and Hydrology XIV, 85311O, doi:
10.1117/12.974561.
Shafer BA, Dezman LE. 1982. Development of a surface water supply index (SWSI) to
assess the severity of drought conditions in snowpack runoff areas. Proceedings of the
Western Snow Conference, 164-175.
Shahid S. 2008. Spatial and temporal characteristics of droughts in the western part of
Bangladesh. Hydrological Processes 22(13): 2235-2247, doi: 10.1002/hyp.6820.
Shahid S. 2010. Rainfall variability and the trends of wet and dry periods in Bangladesh.
International Journal of Climatology 30(15): 2299–2313.
Shahid S, Hazarika MK. 2010. Groundwater Droughts in the Northwestern Districts of
Bangladesh. Water Resources Management 24(10): 1989-2006.
Sharma TC, Panu US. 2010. Analytical procedures for weekly hydrological droughts: a case
of Canadian rivers. Hydrological Sciences Journal 55(1):79–92.
Sharma TC, Panu US. 2012. Predicting Drought Magnitudes: A Parsimonious Model for
Canadian Hydrological Droughts. Water Resources Management, doi: 10.1007/s11269012-0207-x.
Shen C, Wang W-C, Hao Z, Gong W. 2008. Characteristics of anomalous precipitation
events over eastern China during the past five centuries. Climate Dynamics 31: 463-476,
60 doi: 10.1007/s00382-007-0323-0.
Shiau JT. 2006. Fitting Drought Duration and Severity with Two-Dimensional Copulas.
Water Resources Management 20: 795–815, doi:10.1007/s11269-005-9008-9.
Shiau JT, Hsiao Y-Y. 2012. Water-deficit-based drought risk assessments in Taiwan.
Natural Hazards 64:237–257, doi: 10.1007/s11069-012-0239-9.
Shiau JT, Modarres R, Nadarajah S. 2012. Assessing Multi-site Drought Connections in
Iran Using Empirical Copula. Environmental Modeling and Assessment 17:469–482,
doi:10.1007/s10666-012-9318-2.
Shukla S, Wood AW. 2008. Use of a standardized runoff index for characterizing
hydrologic drought. Geophysical Research Letters 35, L02405, doi:
10.1029/2007GL032487.
Sienz F, Bordi I, Fraedrich K, Schneidereit A. 2007. Extreme dry and wet events in Iceland:
Observations, simulations and scenarios. Meteorologische Zeitschrift 16: 9-16.
Sienz F, Bothe O, Fraedrich K. 2012. Monitoring and quantifying future climate projections
of dryness and wetness extremes: SPI bias. Hydrology and Earth System Sciences 16,
2143–2157, doi:10.5194/hess-16-2143-2012.
Sigdel M, Ikeda M.2010. Spatial and Temporal Analysis of Drought in Nepal using
Standardized Precipitation Index and its Relationship with Climate Indices. Journal of
Hydrology and Meteorology 7(1): 59-74.
Sims AP, Niyogi DS, Raman S. 2002. Adopting drought indices for estimating soil
moisture: A North Carolina case study. Geophysical Research Letters 29.
10.1029/2001GL013343.
Singleton A. 2012. Forecasting Drought in Europe with the Standardized Precipitation
Index. JRC Scientific and Technical Reports, EUR 25254 EN – Joint Research Centre –
Institute for Environment and Sustainability, 68 pp, doi:10.2788/16522.
Sinoga JDR, Gross TL. 2012. Droughts and their social perception in the mass media
(southern Spain). International Journal of Climatology doi: 10.1002/joc.3465.
Sirdas S, Şen Z. 2001. Application of the standardized precipitation index (SPI) to the
Marmara region, Turkey. Integrated Water Resources Management . IAHS Publications no.
272: 291-296.
Sivakumar MVK, Das HP, Brunini O. 2005. Impacts of Present and Future Climate
Variability and Change on Agriculture and Forestry in the Arid and Semi-Arid Tropics.
Climatic Change 70: 31-72.
Sohn S-J, Tam C-Y, Ahn J-B. 2012. Development of a multimodel-based seasonal
prediction system for extreme droughts and floods: a case study for South Korea.
International Journal of Climatology doi: 10.1002/joc.3464.
Somorowska U. 2011. Temporal changes in precipitation impacting groundwater and stream
flow. Die Bodenkultur 62(1-4): 115-121.
Sönmez FK, Kömüscü AÜ, Erkan A., Turgu E. 2005. An Analysis of Spatial and Temporal
Dimension of Drought Vulnerability in Turkey Using the Standardized Precipitation Index.
Natural Hazards 35: 243–264, doi: 10.1007/s11069-004-5704-7.
Smakhtin VU, Hughes DA. 2007. Automated estimation and analyses of meteorological
drought characteristics from monthly rainfall data. Environmental Modelling & Software
22(6): 880-890, doi: 10.1016/j.envsoft.2006.05.013.
Strzepek K, Yohe G, Neumann J., Boehlert B. 2010. Characterizing changes in drought risk
for the United States from climate change. Environmental Research Letters 5 044012,
doi:10.1088/1748-9326/5/4/044012.
Subash N, Ram Mohan HS. 2011. Trend detection in rainfall and evaluation of standardized
precipitation index as a drought assessment index for rice–wheat productivity over IGR in
India. International Journal of Climatology 31: 1694-1709 doi: 10.1002/joc.2188.
Sušnik A, Valher A, Gregorič G, Trošt M. 2012. Tools For Agricultural Drought Detection
in the Frame of Drought Management Centre for Southeastern Europe – DMCSEE. Acta
Agriculturae Slovenica 99(2): 235-253.
61 Sun L, Mitchell SW, Davidson A. 2012. Multiple drought indices for agricultural drought
risk assessment on the Canadian prairies. International Journal of Climatology 32(11):
1628-1639, doi: 10.1002/joc.2385.
Svoboda M, LeComte D, Hayes M, Heim R, Gleason K, Angel J, Rippey B, Tinker R,
Palecki M, Stooksbury D, Miskus D, Stephens S. 2002. The Drought Monitor. Bulletin of
the American Meteorological Society 83: 1181-1189.
Szalai S, Szinell C. 2000. Comparison of two drought indices for drought monitoring in
Hungary – a case study. In: J.V. Vogt and F. Somma (eds), Drought and Drought
Mitigation in Europe, Kluwer, Dordrecht, pp. 161-166.
Tabrizi AA, Khalili D, Kamgar-Haghighi AA, Zand-Parsa S. 2010. Utilization of TimeBased Meteorological Droughts to Investigate Occurrence of Streamflow Drought. Water
Resources Management 24: 4287–4306, doi: 10.1007/s11269-010-9659-z.
Tadesse T, Wilhite DA, Harms SK, Hayes MJ, Goddard S. 2004. Drought Monitoring Using
Data Mining Techniques: A Case Study for Nebraska, U.S.A. Natural Hazards 33(1): 137159.
Taylor KE, Stouffer RJ, Meehl GA. 2012. An Overview of CMIP5 and the experiment
design. Bulletin of the American Meteorological Society, 93: 485-498, doi:10.1175/BAMSD-11-00094.1.
Telesca L, Vicente-Serrano SM, Lόpez-Moreno JI. 2012. Power spectral characteristics of
drought indices in the Ebro river basin at different temporal scales. Stochastic
Environmental Research and Risk Assessment doi: 10.1007/s00477-012-0651-4.
Tonkaz T. 2006. Spatio-Temporal Assessment of Historical Droughts using SPI with GIS in
GAP Region, Turkey. Journal of Applied Sciences 6(12): 2565-2571.
Touchan R, Funkhouser G, Hughes MK, Erkan N. 2005. Standardized precipitation index
reconstructed from Turkish tree-ring widths. Climatic Change 72: 339–353 doi:
10.1007/s10584-005-5358-9.
Tölle MH, Moseley C, Panferov O, Busch G, A. Knohl A. 2012. Water supply patterns in
two agricultural areas of Central Germany under climate change conditions. Biogeosciences
Discussions 9, 5153–5176, doi:10.5194/bgd-9-5153-2012.
Tsakiris G, Vangelis H. 2004. Towards a Drought Watch System based on Spatial SPI.
Water Resources Management 18: 1–12.
Tsakiris G, Vangelis H. 2005. Establishing a drought index incorporating
evapotranspiration. European Water 9/10: 3–11.
Tsakiris G, Pangalou D, Vangelis H. 2007. Regional Drought Assessment Based on the
Reconnaissance Drought Index (RDI). Water Resources Management 21:821-833, doi:
10.1007/s11269-006-9105-4.
Tunalioğlu R, Durdu ÖF. 2012. Assessment of future olive crop yield by a comparative
evaluation of drought indices: a case study in western Turkey. Theoretical and Applied
Climatology 108: 397–410, doi: 10.1007/s00704-011-0535-4.
Turgu E. 2008. Using Standardized Precipitation Index for Monitoring Drought and
Analyzing Drought. BALWOIS 2008 Proceedings, Ohrid, Republic of Macedonia, 27-31
May 2008, 9 pp.
Türkeş M, Tatl H. 2009. Use of the standardized precipitation index (SPI) and a modified
SPI for shaping the drought probabilities over Turkey. International Journal of
Climatology 29(15): 2270-2282, doi: 10.1002/joc.1862.
van Rooy MP. 1965. A rainfall anomaly index independent of time and space. Notos 14, 43-48.
Vasiliades L, Loukas A, Patsonas G. 2009. Evaluation of a statistical downscaling procedure
for the estimation of climate change impacts on droughts. Natural Hazards and Earth
System Sciences 9: 879-894.
Vasiliades L, Loukas A, Liberis N. 2011. AWater Balance Derived Drought Index for
Pinios River Basin, Greece. Water Resources Management 25:1087–1101, doi:
10.1007/s11269-010-9665-1.
Vergni L, Todisco F. 2011. Spatio-temporal variability of precipitation, temperature and
62 agricultural drought indices in Central Italy. Agricultural and Forest Meteorology 151:
301–313, doi:10.1016/j.agrformet.2010.11.005.
Vicente-Serrano SM, González-Hidalgo JC, de Luis M, Raventós J. 2004. Drought patterns
in the Mediterranean area: the Valencia region (eastern Spain). Climate Research 26: 5-15.
Vicente-Serrano SM, López-Moreno JI. 2005. Hydrological response to different time
scales of climatological drought: an evaluation of the Standardized Precipitation Index in a
mountainous Mediterranean basin. Hydrology and Earth System Sciences 9: 523-533.
Vicente-Serrano SM. 2006a. Spatial and temporal analysis of droughts in the Iberian
Peninsula (1910–2000). Hydrological Sciences Journal 51(1): 83-97, doi:
10.1623/hysj.51.1.83.
Vicente-Serrano SM. 2006b. Differences in Spatial Patterns of Drought on Different Time
Scales: An Analysis of the Iberian Peninsula. Water Resources Management 20: 37–60,
doi: 10.1007/s11269-006-2974-8.
Vicente-Serrano SM, Cuadrat-Prats JM, Romo A. 2006. Early prediction of crop production
using drought indices at different time-scales and remote sensing data: application in the
Ebro Valley (north-east Spain). International Journal of Remote Sensing 27: 511–518, doi:
10.1080/01431160500296032.
Vicente-Serrano SM. 2007. Evaluating the Impact of Drought Using Remote Sensing in a
Mediterranean, Semi-arid Region. Natural Hazards 40:173-208, doi: 10.1007/s11069-0060009-7.
Vicente-Serrano SM, Cuadrat JM. 2007. North Atlantic oscillation control of droughts in
north-east Spain: evaluation since 1600 A.D. Climatic Change 85: 357-379.
Vicente-Serrano SM, Cuadrat-Prats JM. 2007. Trends in drought intensity and variability in the
middle Ebro valley (NE of the Iberian peninsula) during the second half of the twentieth century. Theoretical and Applied Climatology 88: 247-258, doi: 10.1007/s00704-006-0236-6.
Vicente-Serrano SM, Beguería S, A López-Moreno JI. 2010. Multiscalar Drought Index
Sensitive to Global Warming: The Standardized Precipitation Evapotranspiration Index.
Journal of Climate 23: 1696-1718.
Vicente-Serrano SM, Trigo RM, López-Moreno JI, Liberato MLR, Lorenzo-Lacruz J,
Beguería J, Morán-Tejeda E, El Kenawy A. 2011a. Extreme winter precipitation in the
Iberian Peninsula in 2010: anomalies, driving mechanisms and future projections. Climate
Research 46: 51-65, doi: 10.3354/cr00977.
Vicente-Serrano SM, López-Moreno JI, Drumond A, Gimeno L, Nieto R, Morán-Tejeda1 E,
Lorenzo-Lacruz J, Beguería S, Zabalza J. 2011b. Effects of warming processes on droughts
and water resources in the NW Iberian Peninsula (1930−2006). Climate Research 48: 203212, doi: 10.3354/cr01002.
Vicente-Serrano SM, Beguería S, Lorenzo-Lacruz J, Camarero JJ, López-Moreno JI,
Azorin-Molina C, Revuelto J, Morán-Tejeda E, Sanchez-Lorenzo. 2012. Performance of
Drought Indices for Ecological, Agricultural, and Hydrological Applications. Earth
Interactions 16: 1-27.
Vidal J-P, Martin E, Franchistéguy L, Habets F, Soubeyroux J-M, Blanchard M, Baillon M.
2010. Multilevel and multiscale drought reanalysis over France with the Safran-IsbaModcou hydrometeorological suite. Hydrology and Earth System Sciences 14: 459-478.
Vidal J-P, Martin E, Kitova N, Najac J, Soubeyroux J-M. 2012. Evolution of spatiotemporal drought characteristics: validation, projections and effect of adaptation scenarios.
Hydrology and Earth System Sciences 16: 2935–2955, doi: 10.5194/hess-16-2935-2012.
Viste E, Korecha D, Sorteberg A. 2012. Recent drought and precipitation tendencies in
Ethiopia. Theoretical and Applied Climatology doi: 10.1007/s00704-012-0746-3.
Wharton S, Falk M, Bible KM, Schroeder M, Paw UKT. 2012. Old-growth CO2 flux
measurements reveal high sensitivity to climate anomalies across seasonal, annual and
decadal time scales. Agricultural and Forest Meteorology 161: 1-14.
Vörösmarty CJ, Douglas EM, Green PA, Revenga C. 2005. Geospatial indicators of
emerging water stress: an application to Africa. Ambio 34 (3): 230-237.
63 Walker R, Moore NJ, Arima E, Perz S, Simmons C, Caldas M, Vergara D, Bohrer C. 2009.
Protecting the Amazon with protected areas. Proceedings of the National Academy of
Sciences of the United States of America 106(26): 10582-10586.
Wells N, Goddard S, Hayes MJ. 2004. A self-calibrating Palmer Drought Severity Index.
Journal of Climate 17: 2335-2351.
WMO. 2012. Standardized Precipitation Index (M. Svoboda, M. Hayes and D. Wood). User
guide. WMO-1090, Geneva, 16 pp.
Wu H, Hayes MJ, Weiss A, Hu Q. 2001. An evaluation of the Standardized Precipitation
Index, the China-Z Index and the statistical Z-score. International Journal of Climatology
21: 745-758, doi: 10.1002/joc.658.
Wu H, Wilhite DA. 2004. An Operational Agricultural Drought Risk Assessment Model for
Nebraska, USA. Natural Hazards 33: 1-21.
Wu H, Hayes MJ, Wilhite DA, Svoboda MD. 2005. The effect of the length of record on the
Standardized Precipitation Index calculation. International Journal of Climatology 25:
505–520, doi: 10.1002/joc.1142.
Wu H, Svoboda MD, Hayes MJ, Wilhite DA, Wen F. 2007. Appropriate application of the
Standardized Precipitation Index in arid locations and dry seasons. International Journal of
Climatology 27: 65-79, doi: 10.1002/joc.1371.
Wu R, Kinter JL III. 2009. Analysis of the Relationship of U.S. Droughts with SST and Soil
Moisture: Distinguishing the Time Scale of Droughts. Journal of Climate 22: 4520-4538.
Xu Y-P, Lin S-J, Huang Y, Zhang Q-Q, Ran Q-H. 2011. Drought analysis using multi-scale
standardized precipitation index in the Han River Basin, China. Journal of Zhejiang
University-Science A (Applied Physics & Engineering) 12(6): 483-494, doi:
10.1631/jzus.A1000450.
Yamoah CF, Walters DT, Shapiro CA, Francis CA, Hayes MJ. 2000. Standardized
precipitation index and nitrogen rate effects on crop yields and risk distribution in maize.
Agriculture, Ecosystems and Environment 80: 113–120.
Yan D-H, Feng J, Li C-Z, Qin T-L, Weng B-S, Jin X. 2012. General framework and key
issues concerning integrated strategies for coping with drought and flood in China in a
changing environment. Natural Hazards 64: 577-592, doi: 10.1007/s11069-012-0258-6.
Zhai L, Feng Q. 2009. Spatial and temporal pattern of precipitation and drought in Gansu
Province, Northwest China. Natural Hazards 49: 1-24, doi: 10.1007/s11069-008-9274-y.
Zhang Q, Xu C-Y, Zhang Z. 2009a. Observed changes of drought/wetness episodes in the
Pearl River basin, China, using the standardized precipitation index and aridity index.
Theoretical and Applied Climatology 98: 89-99, doi: 10.1007/s00704-008-0095-4.
Zhang Z, Zhang Q, Zhang J, Zou L, Jiang J. 2009b. Observed dryness and wetness
variability in Shanghai during 1873-2005. Journal of Geographical Sciences 19: 143-52,
doi: 10.1007/s11442-009-0143-8.
Zhang D, Blender R, Fraedrich K. 2012a.Volcanoes and ENSO in millennium simulations:
global impacts and regional reconstructions in East Asia. Theoretical and Applied
Climatology doi: 10.1007/s00704-012-0670-6.
Zhang Q, Li J, Singh VP, Bai Y. 2012b. SPI-based evaluation of drought events in Xinjiang,
China. Natural Hazards 64: 481-492, doi: 10.1007/s11069-012-0251-0.
Zhang Q, Zhang J, Yan D, Bao Y. 2012c. Dynamic risk prediction based on discriminant
analysis for maize drought disaster. Natural Hazards doi: 10.1007/s11069-012-0406-z.
Zhu X, Liang S, Pan YZ. 2012. Observational evidence of the cooling effect of agricultural
irrigation in Jilin, China. Climatic Change 114:799-811, doi: 10.1007/s10584-012-0435-3.
Zin WZW, Jemain AA, Ibrahim K. 2012. Analysis of drought condition and risk in
Peninsular Malaysia using Standardised Precipitation Index. Theoretical and Applied
Climatology doi: 10.1007/s00704-012-0682-2.
Zsákovics G, Kovács F, Kiss A, Pócsik E. 2007. Risk Analysis of the AridificationEndangered Sand-Ridge Area in the Danube-Tisza Interfluve. Acta Climatologica et
Chorologica 40-41: 169-178.
64