Calculation of Climate indices using CMIP6 climate data over
Rajasthan State of India
Joshal K Bansal, Akansha Keshariya, Pooja Patle & Ajanta Goswami
Indian Institute of Technology Roorkee, India
Akansha_k@[Link]
KEYWORDS
Dry spell, Wet spell, SPI, CMIP6 and Python
ABSTRACT:
Climate change results in lower precipitation in some locations by altering the wind patterns and ocean currents
that govern the global climate system. Increased precipitation will not necessarily result in more water being
accessible for consumption, irrigation, and industry because warmer temperatures increase evaporation.
Therefore, information on rainfall characteristics such as dry-spell, wet-spell, Standard Precipitation Index (SPI)
and others is required for proper planning and management of regional water resources. For the prediction of
these indices, numerous GCM models have been created, and it is simple to retrieve the anticipated data from
numerous websites. The study performed here is on Rajasthan state and the climate data from CMIP6 model
has been used. A total of four climate models from CMIP6 (ACCESS-ESM-1, BCC-CSM2-MR, EC-Earth3 and
MRI-ESM2-0) and three scenarios from each model (Historical (1951-2014), SSP245 (2015-2100) and SSP585
(2015-2100)) is considered. Based on the availability of data, historical scenario is considered from 1985-2015
where as SSP245 and SSP585 is divided into two parts i.e. Near Term (2020-2050) and Far term (2060-2090).
The results show that Dry days are increasing and wet days are decreasing in the future. The results clearly
show that precipitation is decreasing in the future. All the results presented here are computed using python.
1. Introduction
The greatest threat to humanity is believed to be global climate change, which will have a negative
impact on the environment in the twenty-first century (Agarwal et al. 2014; Immerzeel et al. 2012).
Climate change is having an impact on several industries around the world, including water resources,
agriculture, energy, and tourism. Changes in the precipitation, hydrological cycle, and atmospheric water
content are all directly related to the climate (Pradhan et al., 2018). The average temperature of the
planet will rise due to global warming, which also affects the hydrological cycle, patterns of precipitation,
extreme weather, melting of ice, soil moisture, and the frequency and severity of extreme occurrences
(Nicholls and Tol 2006; Xu et al. 2011; Menon et al. 2013).
In India, the monsoon season lasts from June to September, and the monsoon months see 80% of the
country's yearly precipitation (Rao 1976; Reddy et al. 2014). Numerous scholars have researched the
variations in rainfall around India (Shukla et al. 2002; Sontakke et al., 2008). Based on observed gridded
datasets, Dash et al. (2013) examined trends in temperature and rainfall, severe events, and predictions
for four Indian towns. Using the results of various climate models, the potential effects of global warming
on the Indian summer monsoon are investigated. However, due to biases in the global climate models,
there are uncertainties in the projections of the regional climate (Kripalani et al. 2005; Kumar et al. 2006;
Rajendran and Kitoh 2008). IMD's daily gridded rainfall dataset was used by Jain and Kumar (2012) to
study the trend of precipitation amount and wet days in Indian River basins.
Rajasthan is one of India's most vulnerable states to climate change because of the ongoing arid
conditions, the lack of water resources, and the population's reliance on agriculture. Rajasthan has a
high likelihood of experiencing a drought due to its low monsoon rainfall (Mall et al. 2006). Between
1901 and 2002, 48 drought years of varying intensity were recorded (Rathore 2004). There aren't
enough study studies available to fully comprehend precipitation indexes. Knowing how rainfall indices
have varied historically and in the future is crucial for managing water supplies and agricultural
production (Pradhan et al., 2018). Knowing current precipitation changes and predicting future changes
in the study area's extreme precipitation indices are the main goals of the current investigation. The
analysis takes into account two future time periods: the Near Term (NT) scenario of 2020-2050 and the
Far Term (FT) period of 2060-2090.
2. Study area and Datasets used
2.1 Study Area
The analysis is done for the Indian state of Rajasthan, which, as indicated in Fig. 1, spans from 69° 30′
to 78° 17′ East longitude and 23° 30′ to 30° 12′ North latitude. Rajasthan, which covers an area of
342,239 km2, is the largest state in western India (Roy 2015). 10.4% of the country's land area and
5.67% of the people are covered by the state. With only 1.2% of the nation's water resources, Rajasthan
is seriously water resource short (Swain et al. 2012). The state is primarily agrarian because roughly
65% of its citizens depend on agriculture-related activities for their living. Arid and semi-arid climate are
characteristics of Rajasthan. Winter temperatures range from 8 °C to 28 °C, while summer temperatures
typically range from 26 °C to 46 °C (Roy 2015). The average annual rainfall of Rajasthan is 574 mm,
which is lower than the 1100 mm average for all of India. The mean annual rainfall in Rajasthan varies
greatly, with the Jaisalmer district receiving less than 100 mm of precipitation and the Jhalawar and
Banswara districts receiving more than 900 mm. Approximately 91% of all annual precipitation falls
during the south-west monsoon.
Figure 1. Study Area of Rajasthan
2.2 Dataset used
Daily bias corrected High resolution gridded (0.25° × 0.25° Lat/Long) CMIP6 based daily datasets for
rainfall for various models, comprising of 150 years of data (1951–2100) covering the whole India region
was collected from the Zenodo website ([Link] The daily time series
datasets for total of 13 GCM models has been uploaded on the website. The data is fully bias corrected
using Quantile mapping techniques and the data is corrected using station data presented by Indian
Meterological department (IMD), India (Mishra et al., 2020).
The CMIP6 models presents the Shared Socio Economic Pathways (SSPs) Scenarios which are
projected for the global change upto 2100. They are used to derive greenhouse gas scenarios with
different climate policies. For the present study, a total of 2 GCM models has been considered and total
of 3 scenarios from each model has been used. The models name ACCESS-ESM1-5 and EC-EARTH3
is used with climate scenarios of Historical (1951-2014), SSP245 (2015-2100) and SSP585 (2015-2100)
(Table 1).
Table 1. List of selected Global Circulation Models (GCMs) used in study
Time
S. NO Datasets Resolution Series
Availability
ACCESS-
1. 0.25⁰ × 0.25⁰ 1951-2100
ESM1-5
EC-
2. 0.25⁰ × 0.25⁰ 1951-2100
EARTH3
3. Methodology
In order to calculate climate indices, Historical, NT and FT period is considered fir 3 models and 5 climate
indices namely Dry Days (DD), Wet Days (WD), Dry Spell Frequency (DSF), Wet Spell Frequency
(WSF) and Standard Precipitation Index (SPI) is considered. Figure 2 represents the flow chart of the
methodology used.
3.1 Extreme Climate Indices
The Expert Team on Climate Change Detection and Indices (ETCCDMI) takes into account a total of 27
temperature and precipitation indices (Peterson and Manton 2008). All of the indices are based on daily
precipitation and temperature measurements, while some are also based on predefined thresholds that
are important for particular applications (Wu and Huang 2016). In this study, a subset of five precipitation
indices— Dry Days (DD), annual Wet Day (WD), Wet Spell Frequency (WSF), Dry Spell Frequency
(DSF) and Standard Precipitation Index (SPI)—were used. These indices were chosen for the studies
Alexander et al. (2006), Pal and Al-Tabbaa (2009), and Agarwal et al. (2014) employed some of these
indices previously, both at the global and regional levels. The list of selected indices, which are
calculated annually for the historical (1984–2014) period, Near Term period (2020-2050) and Farm Term
Period (2060-2090), is shown in Table 2. The indices are calculated using the XCLIM package available
at [Link]
Figure 2. Flow chart of Methodology
Table 2. List of Precipitation Indices Used
Indicator
Index ID Units Application
Name
Rainfall less than 2.5 mm
DD Dry Days Days received within a period of
5 days.
Rainfall more than or equal
WD Wet Days Days to 2.5 mm received within a
period of 5 days.
Largest number of
Dry Spell consecutive days with less
DSF Days
Frequency than 2.5 mm for minimum
of 5 days
Largest number of
Wet Spell consecutive days with
WSF Days
Frequency more than or equal to 2.5
mm for minimum of 5 days
Standard Characterize
SPI Precipitation NA meteorological drought on
Index a range of timescales
5. Results and Discussion
Variations in each of the five precipitation indices are examined for the Rajasthan state. The change in
precipitation is assumed to be represented by the average value of each index for total of 30 years at
numerous randomly selected grid points. Three periods (Historical, NT and FT) are calculated to
determine how the average values of five indices have changed relative to the historical period for each
model. Although, total number of 500 grids has been selected and climate indices are calculated on it
but here in this paper only the combination of 11 randomly selected grids has been shown. Table 3 and
Table 4 lists the selected indices and the indices' outcomes.
Table 3. Different climate indices for 11 grids of ACCESS-ESM1-5 Model
Table 4. Different climate indices for 11 grids of EARTH-ESM3 Model
All three periods are predicted to see a reduction in the duration of the Dry Days (DD) (in both models),
with the exception of Grid 8 in ACCESS-ESM1-5 model which anticipates a minor rise. Additionally,
during the 2020–2050 and 2060–2090 time periods the state will experience the largest reduction in Dry
days. The Wet Days, on the other hand, is anticipated to rise throughout the course of the three future
periods in all the Grids and for all the 2 models. Figure 3 and Figure 4 represents the plots for Dry days
and Wet days with figure 3(a) and 3(b) represents dry days and wet days for ACCESS-ESM1-5 model
and figure 4(a) and 4(b) represents dry days and wet days for EC-EARTH3 model.
Figure 3. Represents (a) Dry days (b) Wet days for ACCESS-ESM1-5 model
Figure 4. Represents (a) Dry days (b) Wet days for EC-EARTH3 model
The frequency of the Dry Spell (DSF) is expected to increase for all the periods in all the grids, except
for the Grid 4 and Grid 5 for the model ACCESS-ESM1-5, which project slightly decrease. On the other
hand, the Wet Spell frequency (WSF) for the Rajasthan state is also increasing during all the periods in
all the grids. Figure 5 and Figure 6 represents the plots for DSF and WSF with figure 5(a) and 5(b)
represents dry spell frequency and wet spell frequency for ACCESS-ESM1-5 model and figure 6(a) and
6(b) represents dry spell frequency and wet spell frequency for EC-EARTH3 model.
Figure 5. Represents (a) Dry Spell Frequency (DSF) (b) Wet Spell Frequency (WSF) for ACCESS-ESM1-5 model
Figure 6. Represents (a) Dry Spell Frequency (DSF) (b) Wet Spell Frequency (WSF) for EC-EARTH3 model
For the Standard Precipitation Index (SPI), it represents the severity of drought. The value of SPI lies
between -3 to +3 with region having value -3 represents the most drought affected region and +3 value
represents least drought affected region. In this perticular study, most of the values are closer to zero
or negative (Figure 7 and Figure 8) which clearly represents the region to be less drought affected for
the future period for the ACCESS-ESM1-5 model whereas for model EC-EARTH3, most of the values
are on negative side represents the areas to be drought affected except Grid 1 and Grid 8 being the
exceptions. Figure 7 and Figure 8 represents the plots for SPI for ACCESS-ESM1-5 model and EC-
EARTH3 model.
Figure 7. SPI for ACCESS-ESM1-5 Figure 8. SPI for EC-EARTH3
6. Conclusion
The results of the study conclude that there will be changes in the future. According to different models
presented, the Dry days (DD) for the state is decreasing in various parts whereas at Wet Days (WD) are
increasing for the future with the exception of several grids. The Dry Spell Frequency as well as the Wet
Spell Frequency is increasing in various areas of the Rajasthan. The Standard Precipitation Index (SPI)
varies with the each model data. For ACCESS-ESM1-5 model, it is on the positive side which reprsents
less droughts in the future whereas for EC-EARTH3, the severity is more as most of the values are on
the negative side.
Although the study done here represents the comparison of different GCM models and it may differ from
model to model. As mentioned by Pradhan et al., (2018), the variability of the climate in the future varies
spatially and temporally. So, a much more in-depth research is required to access the potential change
in the climate for near future and more precise mitigation measures should be designed to overcome it.
7. References
Agarwal A, Babel MS, Maskey S (2014) Analysis of future precipitation in the Koshi river basin, Nepal. J Hydrol
513:422–434.
Alexander LV, Zhang X, Peterson TC, Caesar J, Gleason B, Klein Tank AMG, Tagipour A (2006) Global observed
changes in daily climate extremes of temperature and precipitation. J Geophys Res Atmos 111:D5
Dash SK, Saraswat V, Panda SK, Sharma N (2013) A study of changes in rainfall and temperature patterns at four
cities and corresponding meteorological subdivisions over coastal regions of India. Glob Planet Chang 108:175–
194.
Immerzeel WW, Van Beek LPH, Konz M, Shrestha AB, Bierkens MFP (2012) Hydrological response to climate
change in a glacierized catchment in the Himalayas. Clim Chang 110:721–736.
IPCC (2014) Climate Change 2014: synthesis report. In: Core Writing Team, Pachauri RK, Meyer LA (eds)
Contribution of working groups I, II and III to the 5th AR of the Intergovernmental Panel on Climate Change. IPCC,
Geneva, Switzerland, page no. 57.
Jain SK, Kumar V (2012) Trend analysis of rainfall and temperature data for India. Curr Sci (Bangalore) 102:37–49
Kripalani RH, Kulkarni A, Sabade SS (2005) Are intra-seasonal Oscillations ‘Speed-breakers’ to seasonal
predictions. Clivar Exchanges 10:27–29.
Kumar KR, Sahai AK, Kumar KK, Patwardhan SK, Mishra PK, Revadekar JV, Kamala K, PantGB (2006) High-
resolution climate change scenarios for India for the 21st century. Curr Sci 90:334–345.
Kumar D, Arya DS, Murumkar AR, Rahman MM (2014) Impact of climate change on rainfall in Northwestern
Bangladesh using multi GCM ensembles. Int J Climatol 34:1395–1404.
Ma C, Pan S, Wang G, Liao Y, Xu YP (2016) Changes in precipitation and temperature in Xiangjiang River Basin,
China. Theor Appl Climatol 123:859–871.
Mall RK, Gupta A, Singh R, Singh RS, Rathore LS (2006) Water resources and climate change: an Indian
perspective. Curr Sci 90:1610–1626.
Menon A, Levermann A, Schewe J, Lehmann J, Frieler K (2013) Consistent increase in Indian monsoon rainfall
and its variability across CMIP-5 models. Earth Sys Dyn 4:287–300.
Nicholls RJ, Tol RS (2006) Impacts and responses to sea-level rise: a global analysis of the SRES scenarios over
the twenty-first century. Philos Trans R Soc Lond A: Math Phys Eng Sci 364:1073–1095.
Pal I, Al-Tabbaa A (2009) Trends in seasonal precipitation extremes–An indicator of ‘climate change’ in Kerala.
India J Hydrol 367:62–69.
Peterson TC, Manton MJ (2008) Monitoring changes in climate extremes: a tale of international collaboration. Bull
Am Meteorol Soc 89:1266–1271.
Pradhan, R. K., Sharma, D., Panda, S. K., Dubey, S. K., & Sharma, A. (2019). Changes of precipitation regime and
its indices over Rajasthan state of India: impact of climate change scenarios experiments. Climate Dynamics, 52(5),
3405-3420.
Rajendran K, Kitoh A (2008) Indian summer monsoon in future climate projection by a super high-resolution global
model. Curr Sci 10:1560–1569 Rao YP (1976) Southwest monsoon (Vol.1). India Meteorological Department India
No.1/1976 1:366.
Rathore MS (2004) State level analysis of drought policies and impacts in Rajasthan, India, International Water
Management Institute, Working paper, Colombo, Sri Lanka. Drought Series Paper No.6:93.
Reddy KS, Kumar M, Maruthi V, Umesha B, Vijayalaxmi, Rao CN (2014) Climate change analysis in southern
Telangana region, Andhra Pradesh using LARS-WG model. Curr Sci 107:54–62.
Roy AD (2015) Trend detection in Temperature and Rainfall over Rajasthan during the Last Century. Asian J Res
Soc Sci Hum 5:12–26.
Semenov MA, Brooks RJ (1999) Spatial interpolation of the LARSWG stochastic weather generator in Great Britain.
Clim Res 11:137–148.
Shukla PR, Sharma SK, Ramana PV (2002) Climate change and India: issues, concerns and opportunities. Tata
McGraw-Hill Publishing Company, New Delhi.
Sontakke NA, Singh N, Singh HN (2008) Instrumental period rainfall series of the Indian region (AD 1813–2005):
revised reconstruction, update and analysis. The Holocene 18:1055–1066.
Wu C, Huang G (2016) Projection of climate extremes in the Zhujiang River basin using a regional climate model.
Int J Climatol 36:1184–1196.
Xu H, Taylor RG, Xu Y (2011) Quantifying uncertainty in the impacts of climate change on river discharge in sub-
catchments of the Yangtze and Yellow River Basins, China. Hydrol Earth Syst Sci 15:333s–344s.
