CHARACTERIZATION OF DROUGHT VULNERABILITY ASSESSMENT USING

STANDARDIZED PRECIPITATION INDEX(SPI): MERITS AND LIMITATIONS

Rakshith, R.K.1, and S. Vidhya Lakshmi2
1 B.E., (Civil Engineering), Final Year, Department of Civil Engineering,
Saveetha School of Engineering, Saveetha Institute of Medical and Technical
Sciences, Chennai – 602 105
Email- [email protected]
2 Associate Professor, Department of Civil Engineering, Saveetha School
Engineering, Saveetha Institute of Medical and Technical Sciences, Chennai –
602 105
Email- [email protected]
Abstract:
Drought considered as one of most devastating natural hazards can be monitored using
a variety of indices. After twenty-five years of its launch, the Standardized Precipitation Index
(SPI) is utilized both in central investigations and reasonable applications in drought related
studies. This study covers the fundamental concept of drought ranging from its classification
to its utility and the potential indices used for monitoring droughts. An elaborate review of the
SPI is given in which the definition, understanding the calculation of SPI, the applicability of
the index and its future scope are discussed. Various literature on the use of SPI is briefed to
augment the onset of drought using SPI.

Keywords: Drought, Indices, Standard precipitation Index, Time scales, Severity

Introduction:

Scarcity of water occurs frequently in the most part of the world. One of the reasons for the
scarcity is water demand owing to population growth. This water demand leads to scarcity
which in turn may lead to drought. Drought is simply called as a disaster which affects both
surface water and ground water and can be taken as deficiency in precipitation. However, there
are several parameters on which drought definition is based on(C, Restrepo, Brakenridge, &
Kettner, 2017). Parameters such as temperature, humidity, high winds, characteristics of rain,
intensity and duration of rain play a major role in occurrence of droughts and studies using
same parameters helps to evaluate drought risk assessment(Jeyaseelan, 2003). According to
National Drought mitigation center the conceptual definition of drought is “ Drought is a
protracted period of deficient precipitation resulting in extensive damage to crops, and a consequential
loss of yield.”(Belal, El-Ramady, Mohamed, & Saleh, 2014)
2. DROUGHT CHARACTERISTICS
2.1. Effects of drought:
Drought will further cause deteriorated water quality, crop failure etc. Droughts and
floods are water-related natural disasters which is mainly concerned with the shortage of water
which affect a wide range of environmental factors and activities related to agriculture,
vegetation, human and wild life and local economies. It mainly occurs when the demand for
water exceeds the supply.
General Categories of Drought

1. Meteorological Drought
2. Hydrological Drought
3. Agricultural Drought
4. Socioeconomic Drought
Meteorological drought: This drought result when there is insufficient precipitation for a
certain time period. Hence meteorological drought is an effect of precipitation and it is used
for the drought analysis (Belal et al., 2014)
Hydrological drought: When there is lack of surface and subsurface water, it will lead to
hydrological drought. Streamflow data is used for hydrological drought analysis (Belal et al.,
2014)
Agricultural drought: Droughts may be quantified due to their sudden and rapid onset and
disappearance in plants. Since plants need to be watered, a direct indicator of the drought is
soil moisture conditions. Agricultural drought is the result of soil moisture declination and
continuous crop failure. Several drought indices are considered to study agricultural drought
(Jeyaseelan, 2003)
Socio-economic drought: When the water resources system fails to meet the water demand it
leads to socioeconomic drought(Belal et al., 2014)

2.2 Characterisation of drought using drought indices

Being a natural hazard, drought, it is vital to understand and study the climatology of
the drought. Preparedness and mitigation measures must be developed that emphasise on risk
management of drought. This is called drought risk assessment. Droughts are usually
characterised using a set of influential parameters affecting the onset and disappearance in a
region with help of indices. Such indices are usually a number that signifies the severity of the
drought. Drought indices are used in variety of applications such as drought forecasting,
declaring drought levels, contingency planning and impact assessment.

2.2.1 Drought Indices:

Drought Indices records acclimatize a huge number of bits of information on
precipitation, snowpack, streamflow and other water supply pointers into an intelligible large
picture. A index value esteem is commonly a single number, undeniably more helpful than raw
data for decision making. (Ceglar, 1999) (Hayes, n.d.)

Fig 1. Natural and social dimensions of drought. (Source:

National Drought Mitigation Centre, University of Nebraska, Lincoln, Nebraska, USA.)

2.2.1.1 Palmer Drought Severity Index (PDSI):

PDSI is an empirical relationship developed by Palmer(Cheval, 2015). It is a relation
between the water demand and current moisture supply for a water balance equation. It is a
mainly an hydrological index and referred to as Palmer Drought Hydrological Index
(PDHI)(Ceglar, 1999). It is used to identify abnormality of drought in a region and show the
historical aspects of current conditions.
2.2.1.2 Crop Moisture Index (CMI):
CMI is a derivative of PDSI. It is based on the weekly mean temperature and
precipitation. It is mainly designed for short-term soil moisture of the crops (Belal et al., 2014).
Long term drought monitoring is not much effective. It can detect drought sooner than PDSI
and PDHI.
2.2.1.3 Surface Water Supply Index (SWSI):
It is the analysis of the long-term data such as precipitation, stream flow, snow pack
and reservoir level. It is useful in mountain areas to measure the water supplied in the
area.(Sahoo, Dutta, Khanna, Kumar, & Bandyopadhyay, 2015)

2.2.1.4 Normalized Difference Vegetation Index (NDVI):

It is the measure of vegetation in an area. It uses visible (RED) and near infra-red (NIR)
radiation.

2.2.1.5 Soil Moisture Index (SMI):

The contrast between the present soil moisture and the permanent wilting point to the field
capacity will give the remaining soil moisture. The index values range from 0 to 1 where 0
indicates extreme dry conditions and 1 indicates extreme wet conditions.

2.3 Standard Precipitation Index (SPI):

It is the most common drought index used for drought monitoring. It is simply calculated from
long term record of precipitation for a certain location. It can be used for early warning signal
about drought occurrence and also its impact can be measured. Lack of precipitation will have
a greater impact on groundwater, soil moisture and streamflow. This impact led to the
development of Standard Precipitation Index.
SPI seems to get the universal applicability predicted by some scholars. The SPI is widely
used in research or operational activities in more than 70 countries(Cheval, 2015), approaching
both fundamental studies and applications.

2.3.1 Steps to Compute SPI Values for a location:

1. The accuracy of the SPI is determined by fitting the datasets to the probability density
function. The recommended parametric distribution is gamma (McKee et al. 1993).
 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.
2. Cumulative probability is computed using the results above for given month and time
scale.
3. The Cumulative probability distribution is transformed into Standard normal
distribution which determines the SPI value.

Fig 2. Standardized Precipitation Index applied to summer three-month accumulations for

London, 1901-2005

2.3.2 Stability of Standard Precipitation Index (SPI):

Use of Standard precipitation Index is much simpler when compared to other indices as it
requires only rainfall data for calculation. Severity of any drought in any area can be
computed using variable time scales of SPI. It is considered to be more standardized as the
frequency of extreme drought conditions at any locations and time scale are consistent. It
possess same efficiency for both solid and liquid precipitation. It can be used to asses both
dry and wet periods. Hence its usage is more flexible.
2.3.3 Time Scales of SPI
The SPI value is calculated on different time scales such as 3-, 6-, 12-, 24-, 48- months with
at least 30 years of data sets. The utility in meteorological and agricultural drought is proved
by shorter time scales whereas longer timescales is used in hydrology(McKee, 2012) The trade-
off between length of the precipitation records being used and the significance of the outcome
must be understood by the user in order to interpret SPI values and results. Hence, SPI is
developed by (Mckee, Doesken, & Kleist, 1993)which operates only on precipitation and
based on the fact that precipitation deficits over varying periods or time scales influence
groundwater, reservoir storage, soil moisture, snowpack and streamflow. The longer the length
of record used in the SPI calculation, the more reliable the SPI values will be, especially for
long-time-scale SPI values. Therefore, sound decisions probably need the longer length of
record, because the shorter one is likely not to capture the ‘signals’ of climate variability.
Based on various time scales of standard precipitation index the severity of drought is
classified as below:

Table I: Drought severity values

SPI Range Drought Category

>2 Extremely wet

1.5 to 1.99 Very wet

1.0 to 1.49 Moderately wet

-0.99 to 0.99 Near Normal

-1 to -1.49 Moderately dry

-1.5 to -1.99 Severely dry

<-2 Extremely dry

2.3.4 Limitations of SPI:

During the time of extreme droughts which occur at the same frequency on all locations, it
is difficult to identify drought prone regions using SPI. In areas where the seasonal precipitation
is small, large positive or negative values may result that may mislead drought analysis. Pre-
drought conditions are not taken into account therefore abrupt peaks may occur within the
series. The SPI cannot differentiate the regions in relation with their proneness to drought. In
the areas with less than 30 years of data use of SPI could be problematic.
2.4 Utility of Standard Precipitation Index:
2.4.1 National Scenario & Global Scenario
In analysing the spatial patterns of drought in Turkey (Komuscu, 2001), SPI values
have been registered for 40 stations, however just 7 stations speaking to the distinctive
climatic locales the nation over will be exhibited here for 3-, 6-, 12-, and two year scales,
covering 1940-97.The most striking attribute of the dry season is the adjustment in dry spell
recurrence as the time scale changes. On longer time scales, dry season turns out to be less
regular however keeps going longer. At the 3-month scale, dry spell recurrence increments
yet its length diminish. As such, on shorter time scales, dry spell turns out to be increasingly
visit however goes on for shorter periods. In assessing meteorological drought in
Maharashtra (Aswathi, Nikam, Chouksey, & Aggarwal, 2018), the SPI was determined for
the years 1901 to 2015, while the SPI based dry season region mapping was completed for
the years 2000-2015, by producing 192 SPI maps utilizing the Inverse Distance Weightage
(IDW) strategy. In study of different indices in Semi-arid river basin of India(Wable, Jha,
& Shekhar, 2019) five meteorological indices were selected. In the examination zone
region – Ananthpur (McKee, 2012), 2002 and 2006 are the most noticeably awful dry
season years and 2000 is a typical year as pronounced by the State organization. Peanut is
the chief yield with in excess of 80 percent of developed territory. The power of dry spell
circumstance is comprehended from the measurements distributed by the Government
which peruses that the yield of groundnut crop was 67 kg/ha in 2006, 355 kg/ha in 2002
and 1118 kg/ha in 2000. Correlation of SPI and precipitation deviations, relating to dry
spell years and typical year, was concentrated to comprehend the affectability of SPI and
its concurrence with precipitation deviations. Both SPI and precipitation deviations display
a similar pattern with typical year at more elevated level and two dry spell years falling a
lot of lower to ordinary. The precipitation deviations are huge, extending from - 40 to -
80% mirroring the inadequacy in the greater part of the months in 2002 and 2006 and
connoting serious dry spell circumstance. Positive precipitation deviations showing
abundance precipitation in a large portion of the months mean the typical season in the year
2000.
2.5 Future Scope of SPI:
SPI and remote sensing can be used as unique toolset in making proactive decisions
before the drought occurs. More research in this area could lead to the early drought
warning signals which prevent the region from disaster.

2.6. Conclusions
The definitions of droughts by various organisations have been discussed and it is
known that drought is a natural hazard that can have a negative impact on human lives,
society and economy. Various types and effects have been studied and parameters
influencing various droughts are concluded.
 Table 2 : The Drought risk assessment of various regions are studied using SPI timescales and results are obtained:

SL. no Data used Study area SPI scale Results year References

1. Average Rainfall Ebro River Basin 1,2,3,6 month Summer Drought 1998 (Cheval, 2015)

2. Average Monthly Rainfall Turkey- Konya Station 3-month Scale Frequent droughts with shorter 1999 Komuscu, A. U. (1999).
duration
6 & 12-month scale Drought occurs at higher
frequencies at the coastal stations
(Seasonal Droughts) whereas the
inner stations experience longer-
duration droughts at lower
frequencies (Prolonged Droughts).
3. Monthly Precipitation Greece 3-& 6- month scale Moderate droughts reduce from 2007 Livada, I., & Assimakopoulos, V. D.
(51 years) north to south and from west to east (2007).
12-month scale Occurrence of severe and extreme
drought conditions is very low over
the whole Greek territory on.
4. Monthly Rainfall data Anantapur 2-,3-,4-month Dryness and extreme droughts were 2009 He, Y., Wetterhall, F., Cloke, H. L.,
(39 years) and observed. Pappenberger, F., Wilson, M., &
Khammam Freer, J. (2009).
5. Monthly Precipitation Romania 6-month Few moderate and extreme 2009 Belal, A. A., El-Ramady, H. R.,
droughts Mohamed, E. S., & Saleh, A. M.
(2014).
6. Average Precipitation Portugal 12-month Summer and Summer-autumn 2008 (Cheval, 2015)
drought.
References:
[1] Cheval, Sorin. (2015). The Standardized Precipitation Index – an overview. Romanian
Journal of Meteorology. 12. 17-64.

[2] Belal, Abdel Aziz El-Ramady, Hassan R.Mohamed, Elsayed S.Saleh, Ahmed M(2014).
Drought risk assessment using remote sensing and GIS techniques. Arabian Journal of
Geosciences, 7, 35-53.

[3] Mohapatra, Ambarish G Keswani, Bright Kumar Lenka, Saroj(2018). ICT Specific
Technological Changes in Precision Agriculture Environment. International Journal of
Computer Science and Mobile Applications, 6, 1-16.

[4] Sahoo, R. N.Dutta, Dipanwita Khanna, M. Kumar, N. Bandyopadhyay, S. K(2015).

Drought assessment in the Dhar and Mewat Districts of India using meteorological,
hydrological and remote-sensing derived indices. Natural Hazards, 77, 733-751.

[5] Chandrasekar, K Murthy, C.S. Seshai, Mullapudi Roy, P.S(2018). Agricultural Drought
Assessment and Monitoring Using Geospatial Information.

[6] Wable, Pawan S. Jha, Madan K. Shekhar, Ankit (2019). Comparison of Drought Indices
in a Semi-Arid River Basin of India. Water Resources Management, 33, 75-102.

[7] Wankhede, S. S., Gandhi, N., & Armstrong, L. (2014). Role of ICTs in improving
drought scenario management in India. In Asian Federation for Information Technology in
Agriculture (AFITA, 2014) (pp. 521–530). Australian Society of Information and
Communication Technologies in Agriculture.

[8] Popova Z., Ivanova M., Martins D., Pereira L.S., K. Doneva, Alexandrov V., Kercheva
M., 2014, Vulnerability of Bulgarian agriculture to drought and climate variability with focus
on rainfed maize systems, Natural Hazards, Vol.74 (2):865-886 Springer Science+Business
Media Dordrecht, DOI 10.1007/s11069-014-1215-3, published on line on 09/05 2014
(IF=1.72)

[9] C, R. E., Restrepo, J. D., Brakenridge, G. R., & Kettner, A. J. (2017). Remote Sensing
of Hydrological Extremes. Remote Sensing of Hydrological Extremes, (February 2019), 3–19.

[10] Dalezios, N. R., Blanta, A., Spyropoulos, N. V., & Tarquis, A. M. (2014). Risk
identification of agricultural drought for sustainable Agroecosystems. Natural Hazards and
Earth System Sciences, 14(9), 2435–2448.

[11] Aswathi, P. V., Nikam, B. R., Chouksey, A., & Aggarwal, S. P. (2018). Assessment and
monitoring of agricultural droughts in Maharashtra using meteorological and remote sensing
based indices. In ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial
Information Sciences (Vol. 4, pp. 253–264).

[12] Hazaymeh, K., & K. Hassan, Q. (2016). Remote sensing of agricultural drought
monitoring: A state of art review. AIMS Environmental Science, 3(4), 604–630.
[13] Komuscu, A. U. (1999). Using the SPI to Analyze Spatial and Temporal Patterns of
Drought in Turkey Using the SPI to Analyze Spatial and Temporal Patterns of Drought in
Turkey. Drought Network News, 11(1), 6–13.

[14] Nagarajan, R., & Nagarajan, R. (2009). Drought Indices. In Drought Assessment (pp.
160–204). Springer Netherlands.

[15] Aghakouchak, A., Farahmand, A., Melton, F. S., Teixeira, J., Anderson, M. C.,
Wardlow, B. D., & Hain, C. R. (2015). Reviews of geophysics remote sensing of drought:
Progress, challenges. Reviews of Geophysics, 53, 1–29.

[16] Zargar, A., Sadiq, R., Naser, B., & Khan, F. I. (2011). A review of drought
indices. Environmental Reviews. National Research Council of Canada.

[17] Pandey, V., & Srivastava, P. K. (2018). Integration of satellite, global reanalysis data
and macroscale hydrological model for drought assessment in sub-tropical region of India.
In International Archives of the Photogrammetry, Remote Sensing and Spatial Information
Sciences - ISPRS Archives (Vol. 42, pp. 1347–1351). International Society for
Photogrammetry and Remote Sensing.

[18] McKee, T. B., Nolan, J., & Kleist, J. (1993). The relationship of drought frequency and
duration to time scales. Preprints, Eighth Conf. on Applied Climatology, Amer. Meteor, Soc.,
(January), 179–184.

[19] Thomas B. McKee, N. J. D. and J. K. (1993). Analysis of Standardized Precipitation

Index (SPI) data for drought assessment. Water (Switzerland), 26(2), 1–72.

[20] Khan, S., Gabriel, H. F., & Rana, T. (2008). Standard precipitation index to track drought
and assess impact of rainfall on water tables in irrigation areas. Irrigation and Drainage
Systems, 22(2), 159–177.