1 Introduction

Groundwater, which is highly dynamic in nature, is the source of free water for the entire world. It has become vulnerable as a result of rising demand brought on by population increase and rapid industrialization, which is especially concerning in developing nations where resources for sustainable groundwater management are short. Groundwater levels are lower due to over-utilization, inadequate irrigation management, and a lack of groundwater recharge (Bellingeri et al., 2017). Groundwater decline has severe environmental consequences, including degradation of the value of groundwater (Foster et al., 2018). As a result of the combined effect of groundwater depletion and water quality degradation, the future of water supply is bleak (Tabari et al., 2012).

In developing countries, groundwater decline is a prevalent concern, with India being no exception. Groundwater is the primary source of irrigation and domestic water supply in the country. It is declining at a rate of 0.1–0.5 m3 per year (m3/year) in the country (Dey et al., 2017). As a result, the country's groundwater supply requires careful management to ensure its long-term viability. Knowledge of the geographical and temporal variation of groundwater conditions over the affected area is critical for effective groundwater management (Varouchakis et al., 2012). However, an extensive groundwater monitoring network is required to develop reliable groundwater redistribution over a large area. In practise, due to inaccessibility and higher maintenance expenses, monitoring networks are not usually evenly dispersed. As a result, in making educated management decisions, groundwater level projections at ungauged areas are required.

In the past few years, climate change and global warming have resulted in a rainfall shortfall and an increase in evapotranspiration as a result of a temperature rise directly altering a region's recharge. Precipitation, temperature, and evapotranspiration variability, as expected by various climate change scenarios, will have an impact on aquifers that rely on the physical features of recharge sites. Minimal changes in precipitation volume can cause recharge in dry and semi-arid regions to vary (Green, 2016). An increase in mean global temperature will result in a 30% decrease in groundwater recharge for 4% of the land area and a 70% decrease in groundwater recharge for 1% of the land area (Portmann et al., 2013). This situation of groundwater scarcity is referred to as a “groundwater drought”. The problem of groundwater shortage is more in the semi-arid to arid regions of the world. Groundwater extraction has proliferated all over the world since the twentieth century. Groundwater extraction is over 1000 km3/year, according to a 2010 database, with the largest users coming from Asia, including India, China, Pakistan, and Bangladesh, with extraction rates having quadrupled in the previous 50 years and continuing to climb (Ven Der Gun, 2012).

In semi-arid climates, groundwater management is a significant issue. Excessive use of groundwater produces a significant drop in groundwater levels. Groundwater is restricted and rapidly disappearing around the world, despite its widespread availability. As a result, it justifies long-term and sustainable use for future generations, as well as the preservation of life on Earth. The demand for water has surpassed all previous records as a result of extravagant growth in world population, urbanisation, and industrialization, with groundwater being the most severely impacted. In India, a large portion of the population, both rural and urban (30%), relies on groundwater to meet their fresh water needs. Thus it is pertinent that studies on groundwater level trends be done for proper ground water management and sustainable use. Various scholars have conducted groundwater level analyses to estimate groundwater variability patterns (Patel et al., 2005; Halder et al., 2020). Both Indian and international academics use a variety of statistical tests and methods to measure groundwater fluctuation, depth change and trend analysis (Biswas et al., 2018; Duhan & Pandey, 2013). Mann–Kendal, Sen’s slope test and linear regression seems to be the most widely used method for assessing time series data for various hydrological and climatic indicators, such as ground water change, long-term climatic features, and so on (Abdullahi et al., 2015; Krogulec et al., 2021; Sivapragasam et al., 2015).

The study area (Churu district and Dungargarh tehsil of Bikaner) and its surroundings have an inherent problem of water resources, particularly groundwater resources, which has been prompting regular outmigration from this region. It is a semi-arid region having a staunch scarcity of water, prompting only dry farming irrigation to sustain it. Water scarcity in this region is worsening at a breakneck pace, rather than abating. The main causes of groundwater loss are increasing water demands due to different anthropogenic sources as well as climate change (lower rainfall and higher temperatures). The purpose of this study is to gain a better understanding of the groundwater level features of the study area as well as to estimate the severity of any groundwater depletion problems that may exist. The present study tries to understand the groundwater level characteristics of the study area on a spatial and temporal scale and assess the criticality of the problem of depleting groundwater, if any. Such studies are relevant for semi-arid, drought-prone regions around the world in developing countries on the brink of being transformed into arid and drought-affected regions.

2 Study area

The study area (Fig. 1) consists of two different political regions, namely, Churu district and Dungargarh tehsil of Bikaner. The district occupies approximately 16,830 km2. The population was approximately 2,039,547 according to the Census of India, 2011. The district comprises of 7 tehsils: Sardarshahar, Churu, Ratangarh, Taranagar, Rajgarh, Sujangarh, and Bidasar. The important cultivated crops are Bajra and Jowar. The average elevation of the region is 292 m above MSL. The region has a dry sub-tropical or semi-arid climate. There is scanty rainfall all through the 12 months. According to Köppen and Geiger, this weather is categorised as BWh. About 381 mm of precipitation falls yearly. At an average annual temperature of 25 °C, June is the hottest month of the year, with temperatures soaring from 40 to 44 °C. The lowest average temperatures of the year occur in January, when it is around 13 °C. The variation in temperatures throughout the year is 20.4–50 °C.

Fig. 1
figure 1

Location of study area and sampling points

Full size image

3 Methodology

3.1 Geospatial analysis of groundwater

A base map of Churu district and Dungargarh tehsil of Rajasthan was first prepared using the district boundaries. For the geographical study of groundwater levels, pre- and post-monsoon groundwater depth data from 65 observation wells (Fig. 1) were collected from 2006 to 2018. Rajasthan groundwater department, CGWB, Jaipur and through primary field data collection. For the purpose of analysing the area's groundwater situation, raster maps of pre-monsoon and post-monsoon groundwater levels for each of the sampled years were created using the IDW approach and station data.

3.2 Statistical tests for trend analysis

3.2.1 Mann–Kendall test

The Mann–Kendall is a nonparametric test which is extensively used to evaluate significant trends in climatological and hydrological time series analysis. This is based on sequence and rank correlation of a time series (Hamed & Rao, 1998; Ribeiro et al., 2015; Sayemuzzaman & Jha, 2014; Wang et al., 2020). For time series (Yi, i = 1, 2, 3, … x), assumes the null hypothesis (Ho) independently distributed and alternative hypothesis(H1) exist as a monotonic trend (Wang et al., 2020).

The Mann–Kendall test (S) is formulated by Eq. 1.

$$ S = \mathop \sum \limits_{i = 1}^{n - 1} \mathop \sum \limits_{j = i + 1}^{n} {\text{sgn}} \left( {Y_{j} - Y_{i} } \right) $$
(1)

where \(Y_{j} \;{\text{and}}\;Y_{i}\) are the sequence values of j, n is the range of time series and \( {\text{sgn}} \left( {Y_{j} - Y_{i} } \right)\) is the sign function as followed Eq. 2.

$$ {\text{sgn}} \left( {Y_{j} - Y_{i} } \right) = \left\{ {\begin{array}{*{20}l} { + \;1,} \hfill & { {\text{if}}\quad \left( {Y_{j} - Y_{i} } \right) > 0} \hfill \\ {0, } \hfill & {{\text{if}}\quad \left( {Y_{j} - Y_{i} } \right) = 0} \hfill \\ { - \;1,} \hfill & { {\text{if}}\quad \left( {Y_{j} - Y_{i} } \right) < 0} \hfill \\ \end{array} } \right. $$
(2)

Mann (1945) and Kendall (1975) had reported that statistics S is near to normally distributed when n ≥ 8, with mean and variance of statistics S as followed Eqs. 3 and 4.

$$ F\left( {\text{S}} \right) = 0 $$
(3)
$$ V\left( S \right) = \frac{{n\left( {n - 1} \right)\left( {2n + 5} \right) - \mathop \sum \nolimits_{i = 1}^{l} k_{i} \left( {i - 1} \right)\left( {2i + 5} \right)}}{18} $$
(4)

where\({ }k_{i} \) is the number of data in the tied group and l is the number of the tied value. Null hypothesis (Ho) is tested using statistics Zo as followed Eq. 5.

$$ {\text{Zo}} = \left\{ {\begin{array}{*{20}l} {\frac{S - 1}{{\sqrt {V\left( S \right)} }};} \hfill & {S > 0} \hfill \\ {0;} \hfill & {S = 0} \hfill \\ {\frac{S + 1}{{\sqrt {V\left( S \right)} }};} \hfill & {S > 0} \hfill \\ \end{array} } \right. $$
(5)

Zo demarks the direction of a trend, if the value of Zo is found positive, it indicates an increasing trend and negative value indicate a decreasing trend.

3.3 Sen’s slope estimation

Sen’s slope estimation (Sen, 1968) is a nonparametric test for estimating the magnitude of trend (Ashraf et al., 2021; Kumar et al., 2017). Sen’s slope estimator method uses a linear model and is expressed as Eq. 6.

$$ \begin{aligned} & M_{i} = \frac{{p_{j} - p_{i} }}{j - i} \\ & {\text{ for}} i = 1,2, \ldots N \\ \end{aligned} $$
(6)

where \(p_{i} \) and \(p_{j}\) are the data values at time j and i (j > k). The median value of these N value of \(M_{i}\) is presented by Sen’s slope estimator which is given as Eq. 7.

$$ Ri = \left\{ {\begin{array}{*{20}l} {M_{{\frac{N + 1}{2}}} } \hfill & {N\; {\text{is}}\;{\text{ odd}} } \hfill \\ {\frac{1}{2}\left( {M_{\frac{N}{2}} + M_{{\frac{N + 2}{2}}} } \right)} \hfill & {N\;{\text{ is}}\;{\text{ even}} } \hfill \\ \end{array} } \right. $$
(7)

A negative value of \(Ri\) gives a falling trend and positive value of \(Ri\) gives a rising trend in a time series.

3.4 Standard groundwater level index (SGLI)

As ground water level is a state variable and not a flux like recharge, rainfall and stream flow, the deficit volume calculated with the threshold level approach can identify groundwater droughts or scarcities better compared to other approaches. Although the fixed threshold provides quite acceptable results, the cumulative deficit is preferred as the major droughts can be identified more clearly. The best results can be obtained for a fixed threshold level and the cumulative deficit. To understand the periodic variance in drought conditions, “Standard groundwater level index (SGLI)” is an effective statistical method (Halder et al., 2020). Here, we used annual groundwater depth data from 2006 to 2018 from 65 well stations in the study area to understand the temporal and spatial variation of drought intensity. The dynamic range and categories of drought intensity with respect to SGLI have been presented in Table 1. SGLI has been formulated by using Eq. 8 as given by Halder et al., (2020).

$$ x = \frac{{d - \overline{d}}}{\sigma } $$
(8)

where d is the groundwater depth of respective year, \(\overline{d}\) is the mean groundwater depth from 2006 to 2018 and \(\sigma\) is the standard deviation.

Table 1 Dynamic range and categories of drought intensity with respect to SGLI
Full size table

3.5 Buishand U statistic

The Buishand’s U test was used to measure the change point of ground water depth. This statistical test is a parametric test widely used for monthly or yearly timeseries data to reveal the breaking point. The nullifying condition statement is in the timeseries data there is homogeneity and alternate hypothesis is there was change in the data. If the computed p-value is lower than the significance level alpha = 0.05 it reviled there is change in the data. This equation was developed for single change point detection (Buishand, 1984). The U test is formulated as Eq. 9.

$$ U = \frac{{\sum\nolimits_{k = 1}^{n - 1} {(S_{K} /D_{X} )^{2} } }}{n(n + 1)} $$
(9)

where Sk is the cumulative of deducted value from the mean, and D2x is the standard deviation presented in Eqs. 10 and 11.

$$ S_{k} = \sum\limits_{i = 1}^{k} {(X_{i} - \overline{X})^{{}} } $$
(10)
$$ D_{x}^{2} = 1/n\sum\limits_{i = 1}^{n} {(X_{i} - \overline{X})^{2} } $$
(11)

3.6 Depth variation and recharge probability

Depth variation and recharge probability was calculated using growth rate formula computed as Eq. 12.

$$ {\text{Df}}\;\& \;{\text{Pr}} = \left\{ {\left( {\sum {{\text{Dt}}_{{{\text{pos}}}} } - \sum {{\text{Dt}}_{{{\text{pre}}}} } } \right)/\sum {{\text{Dt}}_{{{\text{pos}}}} } } \right\}*100 $$
(12)

where Dtpre is total depth of water in pre-monsoon season, Dtpos is total depth of water in post-monsoon season. Df and Pr is depth variation and recharge probability.

3.7 Hierarchical cluster analysis

Cluster analysis is a tool for categorising phenomena based on their dissimilarity or similarity. The semi-arid region's groundwater level is constantly changing. Cluster analysis was attempted for the sampled wells of the study area based on the increasing or decreasing magnitude (Sen’s slope) for the period 2006 to 2018. Sen’s slope was chosen since it provides a future trend analysis of water level, which is the major controlling factor of drought. Though there are several clustering techniques, for the present study, the most popular technique of hierarchical clustering (Bloomfield et al., 2015; Subbarao et al., 2013) was chosen. Ward’s linkage has been applied to identify similar wells according to their changing magnitude (Murtagh & Legendre, 2014; Ward, 1963). Similarity measurements between the wells have been done by using the square Euclidean distance method (Shiau & Lin, 2015). The whole statistical process has been done with the help of the hclust package in the statistical RStudio environment.

4 Results and discussions

4.1 Pre-monsoon groundwater depth

Using the 13-year average (2006–2018) of the sampled 65 stations covering the study region, an average water level depth map for the pre-monsoon season was prepared. The analysis of pre-monsoon groundwater depth is shown in Fig. 2a and Table 2. During the study period, the average pre-monsoon water level depth ranged between 7.72 and 61.38 M (bgl). The average shallowest groundwater depth was found at Chapar (Table 3). This denotes the prevalence of comparatively high water tables in this area. However, the deepest groundwater depth was found towards the western part of the study area in Sadasar in Sardarshahar tehsil with a 13-year average of 61.38 M (bgl). This proves the prevalence of a comparatively low water table in this area. Based on ground water depth, the region has been divided into 4 zones—low (0–20 Mbgl), very low (20–40 Mbgl), critical (40–60 bgl) and extremely severe (> 60 Mbgl). The critical depth zone had the highest areal extent (Table 2) in 2006, followed by the areal extent of the very low depth zone of 20–40 M (bgl), while the situation reverses itself in 2018. On an average, the lowest depth of groundwater table was found in the southern and some pockets of the western part of the study area (> 40 Mbgl), while the northern part had low to very low water table depth at around 25 M (bgl). Satisfactory water table depth was absent in none of the places in the study area. This is according to the relief and landform of the study area, which has an elongated shape with low relief. A comparative analysis of the areal extent of the various ground water zones between the base year and terminating year of sampling (Table 2) revealed a considerable increase of 607.73 Km2 of area in the extremely severe zone of > 60 M (bgl) and a reduction of 290.38 Km2 in the low depth zone.

Fig. 2
figure 2

a Pre-monsoon groundwater depth of the study area, b post-monsoon groundwater depth of the study area

Full size image
Table 2 Area under various pre-monsoon ground water depth zone
Full size table
Table 3 Average groundwater fluctuation (2006–2018)
Full size table

4.2 Post-monsoon groundwater depth

Akin to pre-monsoon analysis of groundwater depth aspect, the post-monsoonal investigation was also accomplished. Using the 13-years average (2006–2018) of the sampled stations, average water level depth map for post-monsoon season was prepared (Fig. 2b). Very little variation in the spatial distribution of the various ground water depth zones can be viewed here. The areal extent of extremely severe zone seems to have been recharged owing to monsoonal rains which is the reason of its shrinkage in area. However strangely enough the 13 years average exhibits shrinkage of the low ground water depth zone also which is a matter of concern since it is evident that the areas surrounding this zone has not been much benefitted by the rains. It is evident that haulage in these parts overrides recharge. Further, low rainfall, lower permeability may also be additional reasons of negative recharge. Average groundwater depth varied from 7.82 M (bgl) {Sujangarh 1} to 57.79 M(bgl) {Sadasar} (Table 3). The very low depth zone of 20–40 M (bgl) had the highest areal extent throughout the study period as is evident from Table 4.

Table 4 Area under various post-monsoon ground water depth zone
Full size table

On an average, the lowest ground water depth was obtained in some pockets of the western part of the study area in Dungargarh and Sardarshahar tehsils, as is evident from Fig. 2b. Similar groundwater characteristics were also observed during the pre-monsoon season, which denotes the prevalence of the lowest water table in this region. A comparative analysis of the areal extent of the various ground water zones between the base year and terminating year of sampling (Table 4) revealed a colossal increase of 810.04 Km2 of area in the extremely severe zone of > 60 M (bgl) which is a matter of concern for both habitation and natural vegetation.

4.3 Groundwater fluctuation

To understand the groundwater scenario of the study area, the average (2006–2018) groundwater fluctuation was determined by processing the pre-and post-monsoon groundwater depth layers under the GIS environment. A positive fluctuation is obtained when the depth of the pre-monsoon (in Mbgl) is greater than the post-monsoonal depth, indicating monsoonal recharge, while a negative fluctuation is just the opposite. A negative fluctuation occurs when pre-monsoonal level value (in mbgl) is lower than post-monsoonal level value (in mbgl). This occurs when the groundwater level falls further in the post-monsoonal season (due to over drafting/and no monsoon). This indicates negative recharge or no recharge in ground water or ground water depletion. The same has been represented in Fig. 3. An overall low to negative fluctuation is observed throughout the study area during the study period.

Fig. 3
figure 3

Groundwater fluctuation map (2006–2018)

Full size image

4.3.1 Low ground water zone

Pre-and post-monsoon fluctuation studies prove that the study area is devoid of any abundant groundwater zone. Much of the study area falls under the low groundwater zone, as is evident from Fig. 3 and Table 3, with fluctuations ranging between 7 and 15 M(bgl). Dungargarh and Sardarshar tehsils have pockets of low fluctuation zone.

4.3.2 Very low groundwater/potential groundwater scarce zone

This is the dominant zone in the study area with most of the area falling under this category. The rate of fluctuation is very low at 1–7 M(bgl).

4.3.3 Negative fluctuation zone/groundwater scarcity zone

This is the critical zone or water scarcity zone where negative fluctuation or depleting ground water was noticed. In this zone post-monsoonal depth was deeper than pre-monsoonal depth indicating negative recharge/fluctuation. Pockets of scarcity zone of − 6 M(bgl) to 1 M(bgl) was found mainly in the northern part of the study area in Sujangarh, Chapar and Bidasar.

4.4 Pre-monsoon season trend analysis

Descriptive results of the statistical tests along with trend statistics using the MK test of the surveyed stations are shown in Table 5. Sen’s slope estimator was used to find out the rate of groundwater decline in m/year (Fig. 4a). It is evident from the analysis of the statistical tests that there is a marked declining trend in the groundwater levels throughout the study area. The trend of decline was highest in Sardarshahar tehsil. According to Sen's slope estimate, the declining trend in the entire studied region ranged from 0.01 m/year (Churu tehsil) to 0.85 m/year (Ratangarh tehsil), implying a loss of 1–85 m/decade. On an average, parts of Sardarshahar, Ratangarh, and some parts of Rajgarh had the highest rate of ground water decline, as is observed from Fig. 4a, with a total area of 497.39 km2.

Table 5 Mann–Kendall Z statistics and Sen’s slope estimate for pre- and post-monsoon seasons (2006–2018)
Full size table
Fig. 4
figure 4

a Sen’s slope map (pre-monsoon) of the study area, b Sen’s slope map (post-monsoon) of the study area

Full size image

4.5 Post-monsoon season trend analysis

Descriptive results of the statistical tests along with trend statistics using the MK test of the surveyed stations are shown in Table 5. It is evident from the given statistics that there is a similar marked decline in groundwater level even in the post-monsoonal season throughout the study area since declining trends at 95% significance level are observed throughout the study area. Analysis of average trend statistics shows the highest declining trend in Rajgarh tehsil. The rate of decline according to Sen’s slope estimator varied between 0.04 (Churu tehsil) and 1.95 m/year (Dungargarh tehsil), indicating a decline of 4–195 m/decade in the entire study area. The region of highest rate of decline of above 0.84 m/year is observed in the south western part of the study area in Dungargarh tehsil, as is observed from Fig. 4b, with a total area of 369.14 km2.

The statistical analysis of ground water seasonal trend proves the alarming situation in the study area since ground water is declining at a faster rate in the post-monsoonal season than in the pre-monsoonal season instead of getting recharged in the study area. This is a serious threat to the natural vegetation and human endeavours in the study area.

4.6 Change point of ground water depth

The groundwater depth change point status of 65 stations was evaluated using the Buishand test. This statistical test shows the groundwater depth status of pre- and post-monsoon seasons according to their data homogeneity or whether there was a change in depth during the study period (2006–2018). During the pre-monsoon season, 66.15% of stations experienced a change in groundwater depth, while the remaining 33.84% remained unchanged. During the post-monsoon period, 50.76% of stations accounted for changes in ground water depth. During the pre-monsoon season, the p value of 0.05 in parts of the study area (detail in Table 6) indicates that there was no significant change in the depth of groundwater in these stations, while during the post-monsoon season, no significant change in groundwater depth was accounted for at stations like Bhalautibba, LudiKhuda, etc. (detail in Table 6). There was a significant change in groundwater depth in the remaining stations, having a significant alpha p value > 0.05 (detail in Table 6).

Table 6 Change point status of groundwater
Full size table

4.7 Depth fluctuation and recharge probability of groundwater

Irregularity in Ground water depth fluctuation has been noticed in the study area. The fluctuation of water table was calculated by growth rate formula by using the total depth of ground water. The detailed result has been given in Table 7. The high positive values designate negative growth of ground water table owing to less recharge of ground water while negative value is signifying positive growth of ground water table, i.e., the negative value represents ground water recharge during the study period. Based on 13 years ground water depth monitoring, 46.15% of the stations recorded positive value indicating dwindling ground water availability which could be on account of excess use of ground water and less ground water recharge at the stations Bhaleri (19.91), Melusar (18.16), Sadasar (8.51), Bhalautibba (6.85), Rampura (6.15). The stations carried much water recharge potentiality recorded at the stations 59, 2, 18, 3, 32, 7,5 0, (− 39.24, − 29.80, − 22.01, − 16.69, − 16.69, − 11.98, − 9.60 respectively) that is indicating ground water availability for drinking and irrigation purpose along with having positive imprints of recharge potentiality (Table 7). The value zero or near to zero indicates no change in water table or ground water recharge over the study period (Tidiyasar, Togawas, Kitasar, Abadsar).

Table 7 Depth variation and recharge status (calculated based on total depth of 13 years depth data)
Full size table

4.8 Hierarchical cluster analysis

For the study period, cluster analysis was used to categorise the wells based on their pre-monsoon and post-monsoon groundwater level magnitude trends. A dendrogram has been prepared to show the cluster (Fig. 5). For choosing the similarity, the ward method with a 0.99 agglomerative coefficient has been applied, which has the highest coefficient value among average, single, and complete methods. Cluster analysis resulted in four groups having 17, 20, 10 and 18 wells, respectively. The Box plot (Fig. 6) shows the average Sen’s slope value that indicates the fluctuation between the pre- and post-monsoon depth trend. It shows that cluster 4 has the highest fluctuation, followed by 2, 3, and 1. The average increasing or decreasing magnitude is represented in Table 8. It is observed that variation of the Sen’s slope in clusters has happened owing to land use and land cover changes (Fig. 7). Clusters 1 and 3 have an upward tendency, while Clusters 2 and 4 have a downward trend. The north-east portion of Sardarshahar and Dungargarh, erstwhile a mixed area with agricultural crop land and agricultural fallow land, has turned into a sandy area. This has resulted in the downward tendency.

Fig. 5
figure 5

Dendrogram showing the clusters based on Sen’s slope

Full size image
Fig. 6
figure 6

Average Sen’s slope of each cluster

Full size image
Table 8 Average trend magnitude of each Cluster
Full size table
Fig. 7
figure 7

a: LULC, 2006; b LULC, 2018

Full size image

4.9 Standard groundwater depth level index (SGLI)

SGLI has been done to examine the spatial and temporal variation of drought for the study area during the study period. Both the pre- and post-monsoonal period 7 years were suffered extreme drought of 12 years. In pre-monsoon 2006, 2009, 2012, 2014, 2015, 2017, and 2018 were drought event years and for post-monsoon 2009, 2011, 2012, 2013, 2014, 2007, and 2018 were drought event years. It was found that 2006, 2009, 2012, 2014, 2017 and 2018 (both pre-monsoon and post-monsoon season) had drought period (Table 9 and Fig. 8). As winter season is dry in nature in monsoon climate, drought areas and drought years have been more significant during the pre-monsoon season. A spatial and temporal analysis in Fig. 9 shows that the drought area has increased during the pre-monsoon season (from 2006 to 2018), whereas it has decreased during the post-monsoon season (Table 10). In general, it is observed that the study area is prone to drought with incidences of drought more 2011 onwards (Fig. 8).

Table 9 Temporal drought events from 2006 to 2018
Full size table
Fig. 8
figure 8

Temporal changes of SGLI (2006–2018)

Full size image
Fig. 9
figure 9

Spatial distribution of SGLI showing wet and drought, a pre-Monsoon, 2006, b post-monsoon, 2006, c pre-Monsoon, 2018, d post-monsoon, 2018

Full size image
Table 10 Areal extension of pre-monsoon and post-monsoon groundwater depth (2006–2018)
Full size table

5 Conclusion

With an ever increasing world population, natural resources have been facing critical pressure. The problem is more acute in the semi-arid regions of the world, where the basic natural resource of water is facing a perilous fate. The present research work on groundwater depth vulnerability, trend analysis, drought condition and recharge probability has been done in a semi-arid region of India—Churu district and Dungargarh tehsil of Rajasthan. The study period has been taken from 2006 to 2018. Owing to dwindling groundwater resources and semi-arid conditions, the study area has been forced to opt for dry farming, with large swaths of land slowly turning into barren land. Various statistical techniques like Mann–Kendall, Sen’s slope, SGLI, Cluster analysis, Depth variation and recharge probability, and Buishand U Statistics have been carried out to understand the objectives of the research. The Mann–Kendall analysis shows the dwindling nature of groundwater. Drought area increased by 5.90% during the pre-monsoon period from 2006 to 2018, while post-monsoon extreme drought area increased by 3.50%. According to the analysis, the study area mainly falls under the groundwater scarce zone with pockets having negative fluctuation or withdrawal exceeding recharge. Analysis of Sen’s slope estimator reveals that the rate of groundwater decline is greater in the post monsoon (1.95 m/year) than in the pre-monsoon season (0.85 m/year), indicating groundwater is not able to recharge post rainfall. The reasons may be insufficient rainfall and semi-arid conditions along with anthropogenic reasons. The change point of ground water status shows that 66.15% of the stations experienced a significant change in water depth during the pre-monsoon season and 50.75% of the stations experienced a significant change in water depth during the post-monsoon season. A relatively lower number of stations encountered depth change after the rainy season (post-monsoon), indicating a lesser groundwater recharge post-rainfall. The depth fluctuation and ground water recharge probability indicate that ground water availability was decreasing at 46.15% of the stations. Thus, from change point and depth fluctuation status, it was understood that the ground water table is continuously decreasing in the study area. The statistical analysis of SGLI was undertaken to assess the nature of groundwater drought. The SGLI results show an increase in drought period and an increase in the area of drought area since 2011. The increase is more apparent during the pre-monsoon season. The pre-monsoon period is extremely drought prone. Proper mitigation steps can be implemented for reduce drought intensity. Connection of surface water distribution is one of them. The govt. of India has taken the project “Accelerated Irrigation Benefit Programme (AIBP)”, “Har khet ko Pani (HKKP)”, “Per drop more crop” were working as beneficial project on it. Advance irrigation techniques namely Sprinkler and drip irrigation system also be beneficial for reducing drought intensity. For long term drought management water budget preparation, afforestation and integrated basin planning are helpful for reduce drought intensity.

Thus, if the present trend continues, it may result in the conversion of the semi-arid region into an arid or dry region. Low rainfall accompanied with high evapo-transpiration is the reason for drought like condition in the study area. Further increasing population pressure has resulted to unabated ground water withdrawal for domestic and irrigation purpose. This restricts groundwater recharge. Necessary steps are thus required towards reducing drought intensity and ensuring water sustainability. Both short-term and long-term management techniques will be required to mitigate drought. The following should be incorporated into management strategies: I. Climate prediction; Implementation of Common Monitoring System; II. Creating new water sources (by building dams, reservoirs, wells, and canals, controlling flooding and capturing water otherwise lost to the sea, using non-conventional water sources like treated wastewater, desalinated brackish and saline water, water transfers, harvesting rainwater, developing salt tolerant crops that can be irrigated with saline water.