Central Punjab includes over 40% of the state of Punjab. In Central Punjab, groundwater levels are dropping by more than 1 m every year. The primary cause of groundwater depletion in Central Punjab between 1998 and 2019 is presented in this paper. In order to detect any sudden changes in the area under paddy and the depth of groundwater, Pettitt's test was used. Based on change points and trends, the entire duration was split into two time periods: 1998–2008 (T1) and 2009–2019 (T2); rainfall, potential evapotranspiration, and paddy area were assessed in both periods with 1998–2019 (T3). The slope methodology developed by Mann–Kendall and Sen demonstrated the trend analysis. The findings showed that changes in climate were caused by humans during the 1998–2008 era, changes in meteorological variables were caused by humans during the 2008–2019 period, and changes in climate were caused by humans during the T3 period. The major growing trend of groundwater depth loss is thought to be caused by the combined effects of rainfall variations, an increase in the maximum temperature, cropping intensity, and the number of pumping units.

  • Paddy area has been increasing in Punjab for the last 20 years, and this increase is causing a huge decrease in groundwater levels.

  • Pettitt's test was applied to see in which year an abrupt change happened.

  • Mann–Kendall and Sen's slope were applied to analyze the trend.

  • Rainfall was not enough for groundwater recharge as compared to declination due to the paddy area.

  • Central Punjab must evolve decision-making tools for arresting groundwater (GW) decline.

Among all Indian states, Punjab occupies more than 1.5% of the total geographical area of India (CGWB 2018). Punjab's agriculture is very intensive and has a high water requirement that cannot be satisfied by rainfall. Additionally, existing surface water resources cannot satisfy agricultural needs because canal irrigation systems completely utilize surface water resources (Kaur et al. 2012). Punjab contributes 25.5% of paddy and 35.5% of wheat to the Central pool (Singla et al. 2022). The major source of irrigation water for crops like paddy and wheat is groundwater. As a result, from 1966–1967 to 2017–2018, the area irrigated by tube wells grew from 24.5 to 71.5% (Singla et al. 2022). Over the past 50 years, the state's groundwater resources have suffered due to the rapid development of tube wells and the area under them. The availability of groundwater for agricultural purposes is a significant concern, particularly in the region of Indian Punjab. This is due to the fact that agriculture is considered the cornerstone of the livelihoods of farmers and the overall economy of the state.

Given that precipitation is a fundamental element of the worldwide hydrological cycle, the implications of anthropogenic climate change on precipitation patterns have substantial ramifications for agricultural practices (Guhathakurta & Rajeevan 2008). Seasonal monsoon rains, which fall between June and September and contribute about 75% of the country's yearly rainfall, are essential to India's agriculture and economy (Kumar & Jain 2011). The average annual rainfall across India did not show any trend between 1901 and 2015 (Guhathakurta & Rajeevan 2008). Nonetheless, over the most recent years, 1951–2015 as well as 1986–2015, the annual rainfall series exhibits a diminishing tendency (Kishtawal et al. 2010). Complicated relationships between the space-time distribution of monsoon rainfall and socio-economic demand are well established, and meteorological droughts in India typically begin with a precipitation deficit. Groundwater is under pressure as a result. Analyzing meteorological parameter trend changes and their causes accurately is one of the most difficult tasks (Kumar & Jain 2011).

In the region of Punjab, there exists a total of 150 blocks. However, it is worth noting that within Central Punjab, all blocks have been subjected to excessive exploitation (Singla et al. 2022). The potential impact of climate change on freshwater availability is a significant issue of concern. Changes in rainfall patterns brought about by climate change might negatively impact the replenishment of surface and groundwater resources. The purpose of this research is to (1) identify the precise causes of groundwater depletion and (2) assess and discuss the patterns of a few selected variables over a long- and short-term timeframe.

Study areas and data collection

Indian Punjab can be broadly divided into three zones: the North-East zone, the Central zone, and the South-Western zone (Singla et al. 2022). Figure 1 shows a map of the research area's location. Since all blocks in the study region are overused, Central Punjab has been taken into consideration. In the past 20 years, tubewell irrigation has significantly replaced canal irrigation in Central Punjab (CGWB 2018). Ten districts of Central Punjab have been chosen for the study. Central Punjab has a combined area of 18,000 km2 or about 36% of the state's total territory (Table 1).
Table 1

Geographic characteristics of sites used in the study

Station nameLongitude (E)Latitude (N)Elevation (m.a.s.l.)
Amritsar 74°52′ 31°38′ 234 
Tarantaran 74°55′ 31°27′ 227 
Kapurthala 75°22′ 31°22′ 229 
Ludhiana 75°51′ 30°54′ 262 
Sangrur 75° 49′ 30°13′ 237 
Moga 75°10′ 30°48′ 226 
Barnala 75°32′ 30°22′ 238 
Fatehgarh Sahib 76°22′ 30°38′ 246 
Jalandhar 75°34′ 31°19′ 243 
Patiala 76°22′ 30°20′ 351 
Station nameLongitude (E)Latitude (N)Elevation (m.a.s.l.)
Amritsar 74°52′ 31°38′ 234 
Tarantaran 74°55′ 31°27′ 227 
Kapurthala 75°22′ 31°22′ 229 
Ludhiana 75°51′ 30°54′ 262 
Sangrur 75° 49′ 30°13′ 237 
Moga 75°10′ 30°48′ 226 
Barnala 75°32′ 30°22′ 238 
Fatehgarh Sahib 76°22′ 30°38′ 246 
Jalandhar 75°34′ 31°19′ 243 
Patiala 76°22′ 30°20′ 351 
Figure 1

Location map of the study area.

Figure 1

Location map of the study area.

Close modal

Data collection

The daily gridded data for the maximum temperature (Tmax) and the minimum temperature (Tmin) for Punjab have been acquired from the India Meteorological Department (IMD) at a resolution of 1.0° × 1.0° for the duration of 1998–2019. District-wise daily rainfall data were collected for the study period 1998–2019 from the IMD at a resolution of 0.25° × 0.25°. In order to examine the behavior of groundwater over a 22-year span (1998–2019), pre-monsoon (June) groundwater level data were acquired from the Directorate of Water Resource & Environment and the Directorate of Agriculture of Punjab. A groundwater surface for different years was constructed using the water table data from observation wells and the kriging interpolation technique in Arc GIS 9.3. The groundwater level at each community in relation to the closest known groundwater depth was made easier with this technique. District-by-district data analysis was conducted (Kaur et al. 2012). The Papadakis method (Papadakis 1965) was used to compute potential evapotranspiration (PET).

Trend analysis methods

In this study, two non-parametric methods (Mann–Kendall (MK) and Sen's slope estimator) were applied to notice meteorological variables' trends, and Pettitt's test was used for the detection of inhomogeneity.

MK trend test

The non-parametric MK test is frequently utilized to identify positive or negative or non-null or null trends in a series of observed values (Tabari & Marofi 2011). The null hypothesis (H0) shows null variation in the selected sequence of data, and there is identical distribution among the data values, whereas the alternative hypothesis represents an upward or downward trend (Helsel & Hirsh 2002; Kisi & Ay 2014; Irannezhad et al. 2016; Pohlert 2016). The MK test (Mann 1945; Kendall 1975) is estimated as follows:
formula
(1)
The number of observations is represented by m, while ya and yb are the observed values in the given series a and b (b > a), respectively (Subash et al. 2011).
formula
(2)
formula
(3)
Similar value groups are defined by Q, and sa is the integer of similar values in the Qth group. A similar value group is a set of observations having an identical value. The standard normal test statistic ZS (Silva et al. 2013) is computed as follows:
formula
(4)

The MK test was used to examine variations in annual rainfall, Tmax, Tmin, and groundwater depth during the period of 1998–2019 as shown in Table 2.

Table 2

Trend classification according to the value of Zs for α = 0.05 (5%)

CategoriesScales
Significant trends of increase Zs > 1.96 
Non-significant trends of increase 0 < Zs > 1.96 
No trend Zs = 0 
Non-significant trends of reduction −1.96 < Zs < 0 
Significant trends of reduction Zs < −1.96 
CategoriesScales
Significant trends of increase Zs > 1.96 
Non-significant trends of increase 0 < Zs > 1.96 
No trend Zs = 0 
Non-significant trends of reduction −1.96 < Zs < 0 
Significant trends of reduction Zs < −1.96 

Sen's slope estimator

Sen's slope is also a widely used non-parametric method for the identification of variations in data values. It also calculates the rate of change (Sen 1968). First, the rate of change is measured as follows:
formula
(5)
where yb and yc are the observed values at series b and c (b > c), respectively.
Then, the median of slope (Qmed) is measured after arranging Qi values from the lowest to the highest:
formula
(6)
formula
where M is defined as slope observations (Hollander & Wolfe 1973), and m is the number of observed values in a selected sequence.
The positive or negative trend is represented by the sign of Qmed, and the value of the rate of change is represented by its value. The confidence interval around the slope (Gilbert 1987) is measured as follows:
formula
(7)

Var(K) is measured as estimated in the MK test (Equation (3)) and Z1−α/2 is attained from the standard normal distribution table. In the present analysis, a 5% significance level is chosen, and A1 = (MPα)/2 and A2 = (M + Pα)/2 are measured. Qmin and Qmax are the A1th maximum and the (A2 + 1)th maximum of the M-ordered slope estimates (Gilbert 1987). Qmed is positive if Qmin and Qmax are both positive; if they are both negative, then Qmed is negative. In meteorological time series, a Sen's slope estimator is hugely prevalent for the rate change measurement of variables for a short or long time series. The MK test and Sen's slope methods were applied using XLSTAT.

Pettitt's test

Extreme occurrences may be incorrectly interpreted due to the observed dataset's inhomogeneity (Jiang et al. 2011). Accordingly, Pettitt's test is a crucial technique that has been discovered by numerous researchers to detect inhomogeneity (Liu et al. 2012). A dataset's sudden change point or break point is indicated by its inhomogeneity (Pettitt 1979). As a result, Pettitt's test is regarded as a very reliable indicator for determining the cause of changes in the dataset's chosen sequence. XLSTAT 2019 v.5 was used to analyze the test, which was reported at a significance level of 5% in the current analysis. In the Pettitt test, two subsamples are examined in N samples of a chosen data sequence, as indicated by the formula below:
formula
(8)
where
formula
(9)
Pettitt's statistics is measured as follows:
formula
(10)
The W value is measured as given in Equation (11) after calculating the largest value of Vr,M. The dataset's chosen sequence will only change if the computed W value is less than 0.05. There will not be a change point in the data if not. The year resultant to the biggest value of Vr,M is identified as the shift year.
formula
(11)

Abrupt change in paddy area and groundwater depth

To see the impact of paddy area on groundwater declination in Central Punjab, Pettitt's test was utilized in the time series 1998–2019 on the paddy area and groundwater depth for any abrupt change.

Paddy area

Central Punjab exhibited inhomogeneity in the yearly paddy area series, with 2008 serving as the change point that represents the notable rise in the paddy area between 2009 and 2019. 2008 was the breakpoint in Central Punjab's increasing trend (Figure 2). The mean annual rice acreage in Central Punjab was 1,660 (‘000ha) in the interval preceding the shift year (shown by the red line in Figure 2) and 1,782 (‘000ha) in the interval following the abrupt change point year.
Figure 2

Pettitt homogeneity test for the paddy area in Central Punjab (mu1 and mu2 denote the average paddy area before and after the change point).

Figure 2

Pettitt homogeneity test for the paddy area in Central Punjab (mu1 and mu2 denote the average paddy area before and after the change point).

Close modal

Groundwater depth

This test indicated that an abrupt change in groundwater depth occurred in 2008 in Central Punjab (Figure 3). The average groundwater depth was 11.9 m in the period before the abrupt change point year (the red line in Figure 3), while it was 19.1 m in the period after the abrupt change point year (the green line in Figure 3).
Figure 3

Pettitt homogeneity test for groundwater depth in Central Punjab (mu1 and mu2 denote the average paddy area before and after the change point).

Figure 3

Pettitt homogeneity test for groundwater depth in Central Punjab (mu1 and mu2 denote the average paddy area before and after the change point).

Close modal

Since there was a noticeable shift in the series in 2008, it was split into three time periods: 1998–2008 (T1), 2009–2019 (T2), and 1998–2019 (T3). The MK test and Sen's slope were used for paddy area, rainfall, and PET during selected times.

Rainfall trends

Figure 4 illustrates the evaluation of the annual rainfall statistics for Central Punjab between 1998 and 2008, as well as for 2009 and 2019. According to the results of the MK test, Central Punjab had a substantial upward trend in rainfall from 1998 to 2008 (Zs = 2.13) but not from 2009 to 2019 (Zs = 0). The quantity of rainfall in Central Punjab increased by 43.6 mm/year between 1998 and 2008. The current findings demonstrated that rainfall was noticeably rising for the period T1. T2, on the other hand, did not reveal any upward or downward pattern in Central Punjab.
Figure 4

Spatial distribution of rainfall by MK during different periods.

Figure 4

Spatial distribution of rainfall by MK during different periods.

Close modal

Analysis of PET

Central Punjab (Zs = 3.27) showed a significant increase in the PET trend in the annual time series for only the T3 era according to both tests. In Central Punjab, the annual PET data show a rise of 15.53 mm/year between 1998 and 2019 (Figure 5). PET has increased over the past 22 years, but no trend could be found in T1 and T2, respectively.
Figure 5

Spatial distribution of PET by MK during different periods.

Figure 5

Spatial distribution of PET by MK during different periods.

Close modal

Analysis of the paddy area

The MK results indicated that there has been a considerable increase in paddy area trends throughout all periods (Figure 6). Sen's slope analysis showed that the T1 period (Qmed = 14.68 thousand ha/year) had the highest rate of paddy area growth, followed by the T3 period (Qmed = 9.94 thousand ha/year) and the T2 period (Qmed = 3.88 thousand ha/year). According to these findings, the paddy area grew largely annually during the first phase of 1998–2008, compared to the last 10 years of the study and the whole period (T3).
Figure 6

Spatial distribution of the paddy area by Sen's slope in different periods.

Figure 6

Spatial distribution of the paddy area by Sen's slope in different periods.

Close modal

Groundwater depth

The groundwater depth declined in all the selected time periods signficantly (Figure 7). The greatest decline in groundwater depth was noticed in the T2 period (Qmed = 1.01 m/year) followed by the T3 period (Qmed = 0.80 m/year) and the T1 period (Qmed = 0.73 m/year).
Figure 7

Spatial distribution of groundwater depth by Sen's slope in different periods.

Figure 7

Spatial distribution of groundwater depth by Sen's slope in different periods.

Close modal

The identification of trends is necessary for the management of water resources, particularly in the semiarid regions of the world (Gocic & Trajkovic 2013). Determining the precise cause of groundwater exploitation also requires an understanding of the trends that are altered by human activity. In the current study, during T1, groundwater depleted, paddy area rose, rainfall increased, and PET remained unchanged. It means that the increase in groundwater demand due to the increase in paddy areas might be greater than the amount of groundwater recharge from rainfall. However, in T2, there were no changes in PET and rainfall but an increase in paddy area and groundwater depletion. So, in T2, the only reason behind groundwater depletions is human activity. The paddy area increased significantly after 2008 due to 100% procurement of paddy at the centre's Minimum Support Price (MSP) and consistent economic returns for the farmers. In the T3 period, PET and paddy area increased and groundwater depleted with no changes in rainfall. So, in T3, the reason behind groundwater depletion is human activity and changes in the weather parameter (only PET). Overall, PET indicated a considerable growing trend (1998–2019), but smaller time periods showed no difference. It might be a sign that PET is exhibiting the effects of a significant, long-term fluctuation in both the highest and lowest temperatures. While maximum temperatures have climbed, minimum temperatures have decreased, which is likely responsible for the notable increase in PET (Basso et al. 2021). These results are consistent with those of Zhang et al. (2014) and Piao et al. (2010), indicating that PET has increased in recent years in line with the general warming of most of China.

The period T1 showed the largest increase in the paddy area, signifying the maximum annual increase in the paddy area. This indicates that the high water-consuming crop in the paddy region in period T1 had a considerable impact on the overall groundwater depth. Even though period T2's paddy area increased at the slowest rate annually, period T2 had an impact of period T1. All of these contributed to the groundwater decrease in period T2.

The study findings indicated that period T2 can be classified as a time influenced by human activities, while period T3 is characterized by both human activities and changes in meteorological factors. The concurrent factors of precipitation, rising maximum temperature, intensified farming practices, and increased number of pumping units are regarded as the primary causes for the notable upward trajectory of groundwater depletion.

In the region of Central Punjab, groundwater levels have been negatively affected by the impacts of climate change and rotational water withdrawal. Consequently, there exists a pressing necessity for comprehensive research and the formulation of policies aimed at enhancing agricultural resilience in the area. Water conservation and rainwater collection, groundwater control, as well as enhancing the efficiency of agricultural water usage and crop variety can help in the restoration of groundwater level.

The phenomenon of altering precipitation patterns and temperature has aggravated the issue of groundwater overexploitation. The present study employed a method of dividing the extended temporal duration into two shorter intervals, each spanning a decade, in order to discern the distinct factors contributing to groundwater exploitation in Central Punjab over these times. It has been revealed that Central Punjab is currently facing challenges arising from both anthropogenic activities and climatic variations. The imperative for the implementation of sustainable agriculture is evident. Central Punjab should develop appropriate strategies for crop contingency planning in response to climate change. The recognition of climate change as an integral element within a multifaceted system holds significant importance. To effectively operationalize knowledge, it is imperative to engage in a continuous and enduring endeavor of research, foster collaborative efforts across many disciplines, and establish strong partnerships between scholars and policy-makers. Developing a comprehensive research and application agenda is of utmost importance in order to effectively bridge the gap between academic research and practical decision-making tools for addressing the challenges posed by climate change and the resource waste cycle.

C.S. made contributions to the conceptualization and design of the study. C.S. was responsible for the production of materials, as well as the collection and analysis of data. The initial version of the manuscript was authored by C.S., R.A., and S.K. The text has been reviewed and edited by R.A. and S.K. The authors assert that they did not get any financial assistance, grants, or other forms of support while preparing the manuscript.

The authors would like to express their gratitude for the support provided by the All India Coordinated Research Project on ‘Irrigation Water Management’. We would also like to express our gratitude to the India Meteorological Department (IMD) for their provision of meteorological data.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.

Basso
B.
,
Martinez-Feria
R. A.
,
Rill
L.
&
Richie
J.
2021
Contrasting long-term temperature trends reveal minor changes in projected potential evapotranspiration in the US Midwest
.
Nat. Commun.
12
,
1476
.
https://doi.org/10.1038/s41467-021-21763-7
.
Central Ground Water Board
.
2018
Ground Water Resources of Punjab State – 31st March, 2017
.
Water Resources & Environment Directorate, Water Resources Department
,
Mohali, Punjab
.
Gilbert
R. O.
1987
Statistical Methods for Environmental Pollution Monitoring
.
Wiley
,
New York
.
Guhathakurta
P.
&
Rajeevan
M.
2008
Trends in the rainfall pattern over India
.
Int. J. Climatol.
28
(
11
),
1453
1469
.
Helsel
D. R.
&
Hirsh
R. M.
2002
Trend analysis
. In:
Techniques of Water- Resources Investigations of the United States Geological Survey Book 4, Hydrologic Analysis and Interpretation Chapter A3: Statistical Methods in Water Resources
,
Chapter 12 (Helsel, D.R. & R. M. Hirsch, eds.). US Geological Survey, Reston, VA
, p.
327
.
Hollander
M.
&
Wolfe
D. A.
1973
Nonparametric Statistical Methods
.
John Wiley & Sons
,
New York
.
Irannezhad
M.
,
Marttila
H.
,
Chen
D.
&
Klove
B.
2016
Century-long variability and trends in daily precipitation characteristics at three Finnish stations
.
Adv. Clim. Chang. Res.
7
,
54
69
.
Jiang
S.
,
Ren
L.
,
Yong
B.
,
Singh
V.
,
Yang
X.
&
Yuan
F.
2011
Quantifying the effects of climate variability and human activities on runoff from the Laohahe Basin in Northern China using three different methods
.
Hydrol. Process.
25
,
2492
2505
.
doi:10.1002/hyp.8002
.
Kaur
S.
,
Lubana
P. P. S.
&
Aggarwal
R.
2012
Groundwater management for adaptation under changing climate conditions in Indian Punjab
.
J. Water Clim. Chang.
4
(
1
),
38
50
.
Kendall
M. G.
1975
Rank Correlation Methods
.
Griffin
,
London
.
Kishtawal
C. M.
,
Niyogi
D.
,
Tewari
M.
,
PielkeSr
R. A.
&
Shepherd
J. M.
2010
Urbanization signature in the observed heavy rainfall climatology over India
.
Int. J. Climatol.
30
(
13
),
1908
1916
.
Liu
L.
,
Xu
Z. X.
&
Huang
J. X.
2012
Spatio-temporal variation and abrupt changes for major climate variables in the Taihu Basin, China
.
Stoch. Environ. Res. Risk Assess.
26
,
777
791
.
Mann
H. B.
1945
Nonparametric tests against trend
.
Econometrica
13
,
245
259
.
Papadakis
J.
1965
Potential Evapotranspiration
.
Av Cordoba, Buenos Press
,
Argentina
, p.
54
.
Piao
S. L.
,
Ciais
P.
,
Huang
Y.
,
Shen
Z. H.
,
Peng
S. S.
&
Li
J. S.
2010
The impacts of climate change on water resources and agriculture in China
.
Nature
467
,
43
51
.
Pohlert
T.
2016
Non-parametric Trends and Change-Point Detection
.
Silva
R. M.
,
Santos
C. A. G.
,
Macedo
M. L. A.
,
Silva
L.
&
Freire
P. K. M. M.
2013
Space–time variability of rainfall and hydrological trends in the Alto São Francisco River basin
.
IAHS-AISH Publ.
359
,
48
54
.
Singla
C.
,
Aggarwal
R.
&
Kaur
S.
2022
Groundwater decline in Central Punjab – Is it a warning?
Groundw. Sustain. Dev.
16
,
100718
.
ISSN 2352-801X. https://doi.org/10.1016/j.gsd.2021.100718
.
Tabari
H.
&
Marofi
S.
2011
Changes of pan evaporation in the west of Iran
.
Water Resour. Manage.
25
,
97
111
.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY-NC-ND 4.0), which permits copying and redistribution for non-commercial purposes with no derivatives, provided the original work is properly cited (http://creativecommons.org/licenses/by-nc-nd/4.0/).