Analyzing water level variability in Odisha: insights from multi-year data and spatial analysis

Abstract

A comprehensive analysis of long-term water level trends is essential for freshwater sustainability. Given that Odisha heavily relies on agriculture, the monitoring and management of groundwater and its fluctuations are imperative for ensuring future sustainability in the state. Here, we analyzed the trend in Groundwater using water level data for a 30-year period (1990–2020) for the entire Odisha region. Moreover, to determine the long term variability, critical zones of future groundwater variability and controlling parameters of the water level change, we used spatio-temporal water level data of 746 locations. Water level rise of coastal districts during post-monsoon (POM), corresponds to the intensity of rainfall received, thus rising, however other districts of Odisha, showing decline in water level during the same season is due to shortage of rainfall, increase in population at a sudden, and over pumping due to industrial activities. Similarly, during pre-monsoon (PRM), water level shows an increasing trend in hard rock terrain of Odisha implying rabi crop irrigation, high density drainage network and lesser population density. Feature selection techniques were used in this study to know the parameters controlling most to this water level fluctuation in the entire Odisha state. Precipitation followed by landuse & landcover, lithology and population density are controlling the most for the long term water level change. Drainage, elevation, lithology and slope are positively related to the water level change while others are negatively related. It is also inferred that the districts like Mayurbhanj, Sundargarh, Keonjhar, Kandhamal, Boudh, Dhenkanal, Gajapati, Koraput and Kalahandi contain most of the high critical zone concerning future availability of groundwater while most of the coastal regions are safe.

Article Highlights

Water level trend analysis was performed taking 30 years of historical data.

Hardrock regions are prominent zones of declining in water level than the alluvium parts.

Precipitation followed by landuse & land cover, lithology and population density are controlling the most for the long-term water level change.

Hardrock regions contain most of the high critical zone concerning future availability of groundwater.

Most of the coastal regions are safe pertaining to future availability of groundwater.

Introduction

Groundwater, a vital component of the Earth's hydrological cycle, plays a pivotal role as a primary reservoir of freshwater, essential for human needs, agricultural productivity, industrial operations, and the maintenance of ecosystems [18]. However, in recent decades, many regions worldwide are experiencing a declining trend [9]. This phenomenon poses significant challenges to water resource management, sustainability, and ecosystem health, with far-reaching implications for both current and future generations [32]. The decline in water levels can be attributed to a number of factors, including rapid population growth, agricultural practices, saline water infiltration, accelerated urban development in arid and semi-arid regions already experiencing water scarcity, increased contamination from agricultural, industrial, and anthropogenic sources, and unusual weather [31, 36]. Groundwater contributes significantly the freshwater availability on the Earth, often utilized in economic activities such as agriculture, textiles, and other industries [1], Dandge et al. [8, 12].

In the year book of 2022–23 of CGWB, a decline in water level for the month of April 2022 with respect to average decadal (2012–2021) is seen in 30.54% of the observed wells (CGWB year book 2021–2022). Moreover, the decline in water level varies from a minimum of 8.33% of wells in Bhadrak district to a maximum of 52.63% of wells in Nabarangapur district. Besides, water level for the month of November 2022 with reference to decadal mean (2012–21) shows a rising trend in 58.03% wells. Besides, CGWB has suggested 4043 sq.km for artificial recharge in the state of Odisha by storing the water in the form of dam, nala bandha, pond, rainwater harvesting and canal system. These are suggested because of the declining water level in the subsurface. CGWB suggested these areas for artificial recharge because of 3 reasons: Areas showing post monsoon depth to water level more than 5 m, areas showing declining decadal water level trend (2010–2020) more than 10 cm/year and areas showing EC more than 2000 µS/cm at 25 °C.

As per the 2023 assessment, the Total Annual Ground Water Recharge of the entire state is 17.35 BCM out of which 11.7 BCM is from rainfall recharge and the rest 5.65 BCM is recharge from other sources (CGWB, 2022, Resource assessment). The stage of groundwater extraction is a numerical value in the form of percentage and it is calculated as the gross groundwater extraction for all uses to annual extractable groundwater resource. The overall stage of ground water extraction of the state in 2023 is estimated to be 46.33%, whereas it was 44.25% in 2022. Out of 314 assessment units (blocks) in the state, 299 are safe, nine are semi-critical and six are saline. With due course of time the number of safe blocks is going to decrease due to over exploitation of groundwater and lesser groundwater recharge.

Taking all these into account, this water resource is insufficient due it’s over and unplanned use for various purposes as mentioned earlier, basically in developing countries [16, 26]. Therefore, it is vital to formulate a robust monitor and utilization strategy for sustainable and fair, specifically in arid and semi-arid environments. The increasing population (in developing countries) drives up the demand of food, grains, infrastructure, and industrial setup consequently amplifying the demand of water and often rely on groundwater if efficient and reliable public water utilities are lacking [34]. This excessive extraction of groundwater results in rapid depletion of the water table in the region. The major concern in climate change is it leads to change in global temperature and precipitation forming flood, sea level change in the coastal area (IPCC, 1997; Woldeamlak et al., 2007). Hence, there is a critical need of studies focused on predicting future climate change and its effects on fluctuations in groundwater levels. The factors that control the groundwater trend or groundwater fluctuation are surface water, rainfall frequency, lithology, human influence, plants, soil, aquifer type atmosphere and manmade recharge structures [11, 35]. There are several natural sources that contribute to recharge the groundwater,with precipitation, infiltration of open water and atmospheric moisture being a few primary sources [21], Wohling et al. [37]. Industrialization and mining activities also play a very significant role in influencing the groundwater level fluctuation, despite groundwater reportedly having a limited role in industrial water supply [2]. Among the most urgent groundwater problems at the moment are eutrophication of natural waterways and microchemical pollution by synthetic organic compounds, detergents, phenols, thermal pollutants, hydrocarbons, and nitrates [15, 19].

A geographic information system (GIS) is considered as a very powerful tool for evaluating the drainage system of an area, geological anomalies, surficial anomalies and non-spatial information of a particular region and extracting information regarding the status of groundwater [33, 35]. This study is conducted in the Eastern Indian region, Odisha, where the temperature is hot and humid with an average annual rainfall around 1502 mm. Mahanadi river is the primary source for groundwater recharge in the region (CGWB, 2020). The poor socioeconomic conditions and increasing human influence in the region are causing further decline. Thereby, sustainable and efficient water management requires monitoring and long-term assessment of resources. Here, we collected 30 years of water level data for comprehensive assessment of the underlying trend since 1990 to 2020. Further the study aims to derive the parameters influence for the variation in groundwater level in the region.

Study area

The study area is located in Eastern India, with a geographical area coverage of around 1,55,707 sq. km, bounded between North latitude of 17^°49’ to 22^°34’ and East longitude of 81^°24’ to 87^°29’ (Fig. 1). The State Odisha is considered as an agricultural state and most part of the economy is contributed by farming activities. For a successful agriculture and to develop the economic status of the state, the availability of water is crucial. The annual average rainfall received in the state is about 1502 mm/year (CGWB 2020), mainly contributed by southwest monsoon (SWM; 86%). The state is underlain by diverse rock types ranging in age from Precambrian to Cenozoic in which 80% is contributed by Precambrian formations. Most of the wells in coastal aquifers are located in alluvium. Topographically, the state has been divided into five different units like coastal plains, northern uplands, erosional plains of the river Mahanadi and other river valleys, south-western hilly region and plateaus (CGWB 2020). Hydrologically, the state has been divided into three different formations: 1. Consolidated formation of Precambrian age 2. Semi-consolidated Gondwana and Tertiary formation 3. Unconsolidated Quaternary formation. (CGWB 2020). For un-consolidated formation, groundwater occurs under unconfined conditions in the shallow zone and under semi-confined to confined conditions in the deeper zone (CGWB 2020).

Fig. 1 [Images not available. See PDF.]

Study area shows the Odisha state of India; here water level data location points since last 30 years (1990–2020) plotted in various shapes for different years over lithology of Odisha

There are eleven principal rivers traversing the entire state that can be grouped under eight major river basins within the state, whereas the Indravati, Kolab, Machkund sub-basins in the south forms part of Godavari River basin. Most of the major river’s flow lies in easterly and southeasterly direction with gentle slope. The climate experienced in the state is sub-tropical with the maximum summer temperature around 45 ℃, however the average temperature varies between 18 to 22 ℃. The average annual rainfall of Odisha is 1451 mm with 74 annual rainy days.

Methodology

Data acquirement and analysis

Water level data

Water level data from 1990 to 2020 is taken from Central groundwater board (CGWB), South Eastern Region (SER), Bhubaneswar. The PRM (April) and POM (November) water level data was considered excluding the water level data for the month of January and August. A total of 30,000 numbers of data were taken for each PRM and POM, respectively. Approximately 1000 water level data each year is considered for PRE and POM since 1990.

Climate data

Precipitation (0.25° grid size) and temperature (1° grid size) data is acquired from Indian meteorological departments (IMD) in grid format for the period of 1990–2022 [24]. For higher resolution temperature and precipitation data of 0.25°, we took Aphrodite data. To know the variability between two datasets, we took these 2 types of data. Moreover, for better resolution we took the Aphrodite data.

Population density

Population density data is downloaded from human.org website (https://data.humdata.org/dataset/worldpop-population-density-for-india).

DEM derived thematic layers

Elevation, slope, aspect, curvature and surface texture are the attributes of DEM data which are mostly controlling groundwater recharge in a certain location [23]. We obtained SRTM DEM from USGS earth explorers. Besides hydrologic parameters like drainage, basin size and basin types are also helping the phreatic aquifer to get their recharge water.

Geological parameters

Lithology, Lineament and Fault are the crucial parameters that control groundwater recharge in a hardrock terrain. These datasets are taken from the geological society of India (https://gsi.gov.in/webcenter/portal/OCBIS/pageMAPS/pageMapsSeries).

Data sorting and cleaning analysis

Water level Data from CGWB SER Bhubaneswar (National Hydrograph Network Station) are obtained for 2 seasons PRM and POM (Fig. 2). CGWB monitors water level in four different seasons: January, April, August and November in 1300 different wells distributed in the state (Fig. 2). However, data is disorganised and have following issues:

1. Data is not continuously monitored in a single well,

2. Number of wells increases with time.

Fig. 2 [Images not available. See PDF.]

Work flow for the water level analysis and critical zone identification concerning to future water availability

To overcome this problem, we took the well which has been monitored for more than 20 years, while the rest 10 years were assigned to its close years (Fig. 2). Linear trend analysis is performed to get PRE and PMO trends of water level separately. Finally, hotspot analysis is performed to identify the area where the trend is significantly positive or negative (Fig. 2).

Spatial auto-correlation with Getis-OrdGi*

Spatial autocorrelation of water level trend is performed using Getis-OrdGi* that calculates the cluster of higher or lower values of water level trend based on the z-scores and p-values [13, 14]. For being statistically significant hotspot or cold spot, a water level trends higher or lower values should be surrounded by higher or lower value features [13, 14]. Here, we took the inverse distance option for interpolation analysis on the spatial auto-correlation result. A simple form of the Gi* statistics is [22] calculated by using Eq. 2:

1

$${\text{G}}_{\text{i}}^{\ast }=\frac{{\sum }_{\text{j}=1}^{\text{n}}{\text{W}}_{\text{ij}}{\text{X}}_{\text{j}}}{{\sum }_{\text{j}=1}^{\text{n}}{\text{x}}_{\text{j}}}$$where Gi* is the spatial dependency of the water level trend of a well over ‘n’ number of wells. The term, X_j, characterizes the magnitude of the water level trend at event j over all n number of wells. The w_ij is called weight value between well i and j represents their spatial interrelationship. The distribution of the Gi* statistics are normal when normality is observed in the underlying five distributions of the water level change. The standardized Gi* is essentially a Z-value and can be associated with statistical significance is calculated by using the Eq. (2):

2

$${\text{Z(G}}_{\text{i}}^{\ast })=\frac{{\sum }_{\text{j}=1}^{\text{n}}{\text{W}}_{\text{ij}}{\text{X}}_{\text{j}}-\overline{\text{X}}{\sum }_{\text{j}=1}^{\text{n}}{\text{W}}^2\text{ij}}{\text{S}\sqrt{\frac{\text{n}{\sum }_{\text{j}=1}^{\text{n}}{\text{W}}_{\text{ij}}^2-{\sum }_{\text{j}=1}^{\text{n}}{\text{wij}}^2}{\text{n}-1}}}$$

The standardized Gi* is essentially a Z score and therefore, can be attached to the statistical significance. A random distribution of the observed water level trend happens when we got a Gi* value of zero [13, 14]. Besides, the clusters of high and low valued water level trend occur with high absolute positive and negative Gi* statistics values [22]. The negative Gi*, however, indicates a tendency of clusters of water level change with short incident durations. In summary, if the calculated index values are more significant than a threshold associated with statistical significance, the location of a cluster is identified as a hotspot while the opposite is for cold spots [22].

Result

Water level trend analysis

Hotspot analysis and IDW (inverse distance weightage) interpolation methods are used for spatial distribution of water level trend in Odisha (Fig. 7a, b). The PRM water level showed a declining trend, from 5.3 m in 1990 to 6.3 mbgl in 2020 (Fig. 3a). Moreover, the POM water level was 1.9mbgl in 1990, whereas 3.1mbgl in 2020 (Fig. 3b). Thus, there is a gradual decrease of water level trend with time (1990 to 2020) for POM (Fig. 3b). PRM water level trend is due to long term change in Nala or irrigation system concerning Rabi crop [28] however the POM trend is due to various causes like less monthly average precipitation than extreme one day precipitation 30.

Fig. 3 [Images not available. See PDF.]

The scatter plot shows water level along Y-axis versus year along X-axis and their trend is also shown with R²: Here in fig. (a) pre-monsoon water level trend of Odisha, (b) post-monsoon water level trend of Odisha (c), post monsoon water level trend in hard and soft rock terrain of Odisha and (d) pre-monsoon water level trend in hard and soft rock terrain of Odisha is shown

Water level trend in hard/soft rock terrain of the state

Water level for the hard/soft rock region follows the same trend as that of the water level of the whole trend of Odisha (Fig. 3a, b) for both PRE and POM period. The pre-monsoon water level in hard rock was 5.7 mbgl in 1990 however it increased up to 6.1 mbgl in 2020 (Fig. 3c). Moreover, the Post Monsoon water level shows very less deviation, i.e., the level was at 2 mbgl in 1990 and 3 mbgl in 2020 (Fig. 3d). Slight decline in water level can be seen from the soft rocks for both the PRE & POM time. Thus, PRE and POM water levels showed a minimal declining trend in hard rock terrain. In general, the slight decline trend of hard rock terrain in pre-monsoon may be due to increasing Rabi crop irrigation [27]. In sedimentary terrain, it might be due to reduced outflow from Hirakud dam to the Mahanadi River [10].

Precipitation and temperature change rate pattern

Average annual precipitation in monsoon time showed a rising trend for Odisha while it is not significant in other seasons (Fig. 4a, b). Temperature change rate shows a higher increase for the central and western part of the Odisha region during the last 30 years (Fig. 5B). Odisha observed a rising trend in long term temperature; however, precipitation showed contrasting response i.e. increasing trend in coastal and northern part of Odisha and declining trend in southern part of Odisha (Fig. 5A).

Fig. 4 [Images not available. See PDF.]

Trend line for precipitation is shown for monsoon (a) and non-monsoon (b) period, respectively

Fig. 5 [Images not available. See PDF.]

Precipitation and temperature change rate map of Odisha is shown. Here, light blue to light red and deep blue to deep red is assigned for temperature and precipitation change rate, respectively

Long term (1990–2022) precipitation trends showed a decline trend in the coastal region (Kendrapara, Jagatsinghpur, Bhadrak, Cuttack, and Balasore) and south western part (Nabarangpur, Kandhamal, Kalahandi, Nuapada and Bolangir) of Odisha in all months other than monsoon (Fig. 6A, B). While South, south eastern, west and north western parts (Sundargarh, Mayurbhanj, Sambalpur, Bargarh, Koraput, Gajapati and Nayagarh) showed a rising trend. Besides, in monsoon time coastals regions (Balasore, Bhadrak, Puri) and most parts of central and western Odisha showed decline in precipitation, while parts of Mayurbhanj, Malkangiri, Koraput, Sundargarh, Ganjam, Jharsuguda and Gajapati showed a rising trend.

Fig. 6 [Images not available. See PDF.]

Long-term precipitation trend is shown for other season (a) and monsoon season (b) along with post (a) and pre monsoon (b) water level trend is shown. Here rising and decline in water level trend can be seen in violet star and black triangle respectively

The decline in precipitation during monsoon time is resonating with post-monsoon water level decline in parts of Mayurbhanj, Nabarangpur, Sundargarh and Deogarh. In some parts of Mayurbhanj, Sundargarh, Ganjam and Koraput, the rise in post monsoon water level matches well with the monsoon time precipitation. Similarly, decreasing trend in pre-monsoon water level in places like: Balasore, Mayurbhanj, Nabarangpur, Kandhamal and Angul. Besides, the coastal region water level trend is mostly controlled by alluvium and surface runoff [7]. Besides,the higher population density in coastal regions is controlling the water level trend in that region.

Discussion

Long term water level trend

Groundwater levels are typically measured by monitoring wells, which are drilled into the ground and equipped with instruments to measure water levels [4, 25]. The measurements are usually taken at regular intervals, such as daily or monthly, and are used to create a groundwater level monitoring network. The groundwater level variations can have significant impacts on the environment and human activities. For example, excessive drop in water level can lead to land subsidence, reduced water availability for agriculture, industrial, household needs and other uses [3, 20]. On the other hand, if the groundwater level rises too high, it can cause flooding, waterlogging and damage to infrastructure [29]. The water logging problems are obvious in coastal regions of Kendrapara, Jagatsinghpur, Puri districts where the alluvium is the major rock type.

Similarly, in hardrock terrain, most of the wells showed a declining trend except a few that showed a rising trend. However, most of the wells in coastal regions show a rising trend in both PRE and POM time. The high to medium rate of declines in water level is observed in the hard rock region during pre-monsoon and post-monsoon, respectively. During field visits in pre monsoon, most of the wells in hardrock regions (Boudh and Kandhamal) showed dryness which created problems for local people to have water for daily use. Moreover, for the coastal region a moderate rate of rising trend is noted (Fig. 7b). The average water level during pre-monsoon goes to 9 m in hardrock region.

Fig. 7 [Images not available. See PDF.]

Spatial distribution of pre (a) and post-monsoon (b) water level trend since last 30 years for Odisha; here blue color shows declining trend, while red color shows rising trend and the yellow color shows the trend is not significant (p > 0.005). The ddistribution of water level trend is shown for hard (c) and soft rock region (d), here blue color shows rising while brown color shows declining trend

Generally, aquifers get replenished during monsoon and depleted during pre-monsoon time leaving less groundwater available for the extraction. This phenomenon is majorly observed in hard rock terrain since there is no open water resource available for the recharge of aquifers. Further the degree of weathering and topography in the hard rock region also plays as an influential factor in governing the yield of wells. There are several factors which influence the groundwater level variation in aquifers for both hard rock and sedimentary terrain (Coastal area) like (1) Rainfall variation. (2) Groundwater withdrawal during the rainy season for irrigating kharif (rain-fed) crops. (3) Increase in the withdrawals due to development. (4) Extraction from the deeper confined aquifers through bore wells. (5) The groundwater in hard rock in shallow aquifers is dynamic and hence joins the surface water as base flow. We found groundwater level is increasing during post-monsoon time in sedimentary terrain (Coastal area) which may be attributed to the high amount of rainfall received. Whereas the alluvium in coastal regions serves as a very good source of recharge, it has the more capacity of holding water [5]. Other than that, in coastal areas, nearby open water source also plays an important role in recharging the aquifer so there is a steady increase of water level noted in this area.

The increase in groundwater level could be attributed to excessive rainfall during the post-monsoon season in the coastal region. However, groundwater levels also have dropped significantly in hard rock terrain as mentioned, which may be due to over-pumping, lithology and less rainfall. The groundwater level data demonstrates that the aquifer is generally very sensitive to monsoonal precipitation. Besides, the occurrence of groundwater in hard rock is confined to the shallow weathered zone [17]. The saturation in the zone gets depleted due to extraction and also due to out flow of groundwater which ultimately leaks out with the streams and drainages as a base flow. The system is dynamic and has hydraulic gradients parallel to surface contours. This indicates that the groundwater, whether extracted or not, does not remain static and gets depleted and de-saturated and reaches its minimum saturation during summer months. This situation compels the optimum utilization of groundwater in hard rock by its extraction during kharif and rabi cropping seasons. If not extracted and utilized, it flows out to the river and streams.

Based on the POM depth to water level of the year 2018 and long term (2009–2018) ground water level trends, it has been estimated that approximately 45,592 sq. kms area is feasible for artificial recharge (CGWB, 2021 annual report). Coastal regions like Boudh (448 sq.km) and Balasore (409 sq.km) have these kinds of problems. Moreover, hardrock regions like Koraput (974 sq.km), Rayagada (751 sq.km), Sundargarh (536 sq.km), Mayurbhanja (204 sq.km), Gajapati (150). and Kandhamal (127 sq.km). Recharge is envisaged to be carried out through four sources, viz., uncommitted surplus run off, surplus canal water, surface run off from agricultural land in large agricultural farms and roof top rainwater harvesting (RTRWH). The uncommitted surplus run off is to be used by proposed check dam and recharge shaft, while surplus canal water is envisaged to be recharged through injection wells and run off from the large agricultural Land through the farm ponds in respective farms and provision of RTRWH in urban area for rainwater from roof top.

Regional scale controlling factor of water level/Feature selection techniques

All the controlling parameters of water level were classified into 5 groups: climatic (temperature, precipitation, humidity), topographic (elevation, slope, aspect, texture, terrain rugged texture), hydrologic (drainage, basin size, basin type), Geologic (lithology, lineament, fault) and human intervention (population, construction, factory, industry). We used multiple regression techniques which is a feature selection technique to know the parameter controlling for water level change in the long-term basis (Table 1). Here, slope, elevation, aspect, drainage distance, lithology, land use and land cover, geomorphology, precipitation change, population density and lineament density were considered for the analysis as independent variables, while water level change was taken as dependent parameters. Moreover, 860 controlling samples are considered for this analysis. Precipitation is given higher weightage towards water level change in Odisha region followed by land use and land cover, lithology, population density, elevation and slope. Drainage, elevation, lithology and slope are positively related to the water level change while others are negatively related.

Table 1. Coefficient of controlling parameters for water level trend

Parameters	Coefficients	SE	t Stat	P-value
Intercept	1.86083	0.494097	3.766114	0.000177
Slope	0.00618	0.020435	0.3023	0.762497
Elevation	0.002	0.000643	3.104207	0.001971
Aspect	− 0.00042	0.000936	− 0.4449	0.656507
Preciptation	− 0.22643	0.079958	− 2.83193	0.004735
Lithology	0.05486	0.042806	1.281682	0.200301
Population density	− 0.00009	0.00007	− 1.22908	0.219377
Landuse and landcover	− 0.11044	0.061414	− 1.79828	0.072484
Geomorphology	− 0.02911	0.05683	− 0.51221	0.608637
Lineaments distance	− 0.00038	0.000149	− 2.56564	0.010468
Drainge_Distance	0.00002	0.000161	0.126978	0.898988

Critical zone demarcation concerning groundwater availability

All the controlling factors of future availability of groundwater can be divided in 3 major classes: climate parameter (temperature and precipitation and humidity), water level parameter (fluctuation, trend average, etc.) and human related parameter (infrastructure, agriculture, etc.). Three parameters for future water level availability zone demarcation in Odisha using overlay analysis has been carried out (Table 2). Here, water level fluctuation, average water level and water level trend are used for the critical zone identification in Odisha in ArcGIS 10.3. Here the negative water level trend, high average water level and high fluctuation is taken as critical for the future availability of groundwater. We give a higher weightage to the long-term water level trend followed by decadal water level fluctuation and decadal average water level for the estimation of future water availability.

Table 2. Classification of different thematic maps and their weightage

	Classification	Weightage	Critical level
Decadal water level fluctuation	< 3	1	Less critical
	3–5	2	Semi critical
	> 5	3	Critical
Decadal average water level	< 3	1	Less critical
	3–5	2	Semi critical
	> 5	3	Critical
Long term water level trend (30 years)	< − 1	3	Critical
	− 1–1	2	Semi critical
	> 1	1	Less critical

Here, most of the coastal regions are showing less critical towards future availability of groundwater except some patches in Balasore district (Fig. 8). As the coastal region contains alluvium with dendritic pattern of major rivers of this state allowing groundwater recharge easily in monsoon time [6], thus there will be no issue regarding near future availability of groundwater. Besides, Mayurbhanj, Sundargarh, Keonjhar, Kandhamal, Boudh, Dhenkanal, Gajapati, Koraput and Kalahandi contain most of the higher critical zones concerning future availability of ground water. These regions contain mostly hard rocks with high relief opposing groundwater to get recharge in monsoon time. Recharge structures like nala bandha, dam and barrages should be constructed to overcome this problem in hard rock terrain (CGWB, 2020).

Fig. 8 [Images not available. See PDF.]

Critical map of Odisha concerning to future availability of groundwater is shown; Here critical region is shown in red color while less critical is shown in green color. Different locations is marked in the main figure and their pre and post water level trend is shown in the figure

Details field verification

Detailed verification and hydrograph is studied in some locations of Odisha (Figs. 9, 10, 11). Field photographs, Google earth view and hydrograph are shown in Boriguma and Miriguda region of Koraput district which contain hardrock and these regions show a rising trend in water level (Fig. 9). In Boriguma trend, 2 dugwell data can be seen i.e., one is from 1990 to 2008 data which shows a decline trend while the other one (After 2010) gives a rising trend. Both these locations gave flat trend lines in post monsoon water level meaning a good recharging condition of these areas. Likewise, in Miriguda 2 dugwells data shows a contrasting trend; one of them (1990–2010) shows a rising while another (2010–2022) shows a declining trend in water level.

Fig. 9 [Images not available. See PDF.]

Field photo od dugwells along with elevation profile and pre and post monsoon water level trend is shown for Miriguda (A, B and E) and Boriguma (C, D and F)

Fig. 10 [Images not available. See PDF.]

Google Earth view of well location and their pre and post monsoon water level trend for Hirakud (left) and Boudth town (right side)

Fig. 11 [Images not available. See PDF.]

Elevation profile from Google Earth along with water level trend is shown for different dug wells in Odisha (a, b) Daspallah, (c, d) Kendrapara, (e, f) Khajuripada and (g, h) Bhubaneswar town

Most of the hydrographs in Odisha are showing a decline in water table while a dug well near to Hirakud shows a steep rising trend due to the presence of a bigger reservoir nearby (Fig. 10). Besides, a high pre-monsoon rising trend in water level can be seen in Boudh town while a lesser rising trend is seen in post monsoon. This location is near to the Mahanadi River which is the largest river system in Odisha but due presence of hard rock the recharge is lesser even in flood time.

Kendrapara and Khajuripada regions show a rising trend in pre and post monsoon water level while Daspalla and Bhubaneswar show a steep decline trend (Fig. 11). The decline in water level in Daspalla is due to hard rock. The steep decline trend in pre as well as post monsoon time in Bhubaneswar caused by high population density and large infrastructure. This region contains khondalite, laterite and sandstone formation. Besides, Kendrapara is an alluvium deposit with flood plains distributed in most parts of the district showing a stable and rising trend in water level. Besides, the rising trend in post-monsoon time is attributed to the good recharge condition of an aquifer. In Kajuripada, the dugwell hydrograph clearly shows 2 different well’s data; one was measured till 2005 which shows a decline trend in water level and another was measured after 2015 which also showed a decline trend in pre-monsoon time. This trend is happening due to Hardrock and hilly terrain.

Conclusions

• Water level trend shows a steep decline for post-monsoon time for the whole state. Besides, this trend is steeper for the hardrock region than the alluvium part of the state. This might be due to the inconsistency in precipitation in most parts of Odisha and Runoff has sufficiently decreased in the Mahanadi River after the construction of a dam in 2015 in Chhattisgarh, which controls the water level in the coastal region.

• Pre-monsoon water level showed a declining trend in coastal regions (Balesore, Khurda, Bhadrak, Cuttack, Puri and Jajpur), which is not correspondingly matched with precipitation change rate. This trend is controlled mostly by population density, construction and industry. Moreover, water level trend shows a rising trend in Ganjam, Kalahandi, Nuapada, Jagatsinghpur, Kendrapara, Nayagarh, Phulbani and parts of Sambalpur, Baragarh, Keonjhar with 95% significant level for the same season. This might be due to the Rabi crop (Summer crop) irrigation, higher major drainage network and lesser population density.

• The declining trend of water level in both the seasons in the southern parts of Odisha and Balasore, Sambalpur and Angul can be explained by the decreasing trend in precipitation in those locations and lesser major drainage networks. Moreover, in post-monsoon the coastal region got sufficient recharge from precipitation and inland flow from drainage network, thus except Balasore, all are showing rising or no change in water level trend.

• Feature selection techniques show that precipitation has higher weightage towards long term water level change in Odisha followed by land use and land cover, lithology. Besides, precipitation, population density and landuse & landcover are negatively related to the water level change while lithology, elevation and slope are negatively related.

• Mayurbhanj, Sundargarh, Keonjhar, Kandhamal, Boudh, Dhenkanal, Gajapati, Koraput and Kalahandi contain most of the zone with high critical zone concerning future availability of groundwater, while most of the coastal region is in safe zone except Balasore. Western and southern parts of Odisha contain mostly hard rocks with high relief opposing groundwater recharge. Recharge structures like Nala bandha, dam and barrages should be constructed to overcome this problem in hard rock terrain.

Author contributions

LKM-defining problem, methodology, original writing and editing, BRP- editing and writing, AD- methodology, modifying original draft, SS- Data providing, SB-data providing.

Funding

No external funding was used.

Data availability

All the data will be available on request by mailing to the corresponding author.

Declarations

Competing interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

1. Abascal, E; Gómez-Coma, L; Ortiz, I; Ortiz, A. Global diagnosis of nitrate pollution in groundwater and review of removal technologies. Sci Total Environ; 2022; 810, 152233. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2021.152233]

2. Abiye, T; Masindi, K; Mengistu, H; Demlie, M. Understanding the groundwater-level fluctuations for better management of groundwater resource: a case in the Johannesburg region. Groundw Sustain Dev; 2018; 7, pp. 1-7. [DOI: https://dx.doi.org/10.1016/j.gsd.2018.02.004]

3. Ashwini, K; Vera, RK; Sriharsha, S; Chourasiya, S; Singh, A. Delineation of groundwater potential zone for sustainable water resources management using remote sensing-GIS and analytic hierarchy approach in the state of Jharkhand; 2023; India, Groundwater for Sustainable Development: 100908.

4. Attwa, M; Zamzam, S. An integrated approach of GIS and geoelectrical techniques for wastewater leakage investigations: Active constraint balancing and genetic algorithms application. J Appl Geophys; 2020; 175, [DOI: https://dx.doi.org/10.1016/j.jappgeo.2020.103992]

5. Bhanja SN, Mukherjee A, Rodell M (2018) Groundwater storage variations in India. Groundwater of South Asia, 49–59

6. Chakraborty, S; Maity, PK; Das, S. Investigation, simulation, identification and prediction of groundwater levels in coastal areas of Purba Midnapur, India, using MODFLOW. Environ Dev Sustainab; 2020; 22, pp. 3805-3837. [DOI: https://dx.doi.org/10.1007/s10668-019-00344-1]

7. Choudhury A, Sarkar S, Das PK, Pati GC (2009) Coastal aquifers of Orissa–a simplistic overview. In: Conference: proceedings of the on groundwater for emergency situation-UNESCO–KIIT, pp: 1–15

8. Dandge, KP; Patil, SS. Spatial distribution of ground water quality index using remote sensing and GIS techniques. Appl Water Sci; 2022; 12, pp. 1-18. [DOI: https://dx.doi.org/10.1007/s13201-021-01546-7]

9. Dangar, S; Asoka, A; Mishra, V. Causes and implications of groundwater depletion in India: a review. J Hydrol; 2021; 596, [DOI: https://dx.doi.org/10.1016/j.jhydrol.2021.126103]

10. Dhar, A; Sahoo, S; Mandal, U; Dey, S; Bishi, N; Kar, A. Hydro-environmental assessment of a regional ground water aquifer: Hirakud command area (India). Environ Earth Sci; 2015; 73, pp. 4165-4178. [DOI: https://dx.doi.org/10.1007/s12665-014-3703-x]

11. Fenta, MC; Anteneh, ZL; Szanyi, J; Walker, D. Hydrogeological framework of the volcanic aquifers and groundwater quality in Dangila Town and the surrounding area Northwest Ethiopia. Groundw Sustain Dev; 2020; 11, [DOI: https://dx.doi.org/10.1016/j.gsd.2020.100408]

12. Galkate RV, Yadav S, Pandey RP, Negm AM, Yadava RN. An overview: water resource management aspects in India. Water Qual Assess Manag India, 2022; 29–55

13. Getis, A. Cliff, ad and ord, jk 1973: spatial autocorrelation. London: Pion: Progress Hum Geogr; 1995; 19, 2 pp. 245-249. [DOI: https://dx.doi.org/10.1177/030913259501900205]

14. Getis, A; Ord, JK. The analysis of spatial association by use of distance statistics. Geogr Anal; 1992; 24, 3 pp. 189-206. [DOI: https://dx.doi.org/10.1111/j.1538-4632.1992.tb00261.x]

15. Harter T. Groundwater quality and groundwater pollution (2003)

16. Jamei, M; Karbasi, M; Malik, A; Abualigah, L; Islam, ARMT; Yaseen, ZM. Computational assessment of groundwater salinity distribution within coastal multi-aquifers of Bangladesh. Sci Rep; 2022; 12, 1 11165. [DOI: https://dx.doi.org/10.1038/s41598-022-15104-x]

17. Kar, AK; Hembram, R; Mohanty, H. Study of morphological changes in deltaic river of odisha using GIS. Water Manag Water Gov Hydrol Model; 2021; 96, pp. 35-46.

18. Kumar, P; Tiwari, P; Biswas, A; Acharya, T. Geophysical investigation for seawater intrusion in the high-quality coastal aquifers of India: a review. Environ Sci Pollut Res; 2023; 30, 4 pp. 9127-9163. [DOI: https://dx.doi.org/10.1007/s11356-022-24233-9]

19. Kurwadkar, S. Groundwater pollution and vulnerability assessment. Water Environ Res; 2017; 89, 10 pp. 1561-1577. [DOI: https://dx.doi.org/10.2175/106143017X15023776270584]

20. Kumar M, Singh DK, Sarangi A, Mani I, Khanna M. Regional groundwater sustainability assessment through enhancing conveyance and application efficiencies of irrigation water. 2022.

21. Lerner DN. Groundwater recharge. In: Geochemical processes, weathering and groundwater recharge in catchments, pp. 109–150. CRC Press (2020).

22. Manepalli UR, Bham GH, Kandada S. Evaluation of hotspots identification using kernel density estimation (K) and Getis-Ord (Gi*) on I-630. In: 3rd international conference on road safety and simulation (pp. 14–16) (2011).

23. Mohanty, LK; Maiti, S. Glacial lake formation probability mapping in the Himalayan glacier: a probabilistic approach. J Earth Syst Sci; 2022; 131, 1 54. [DOI: https://dx.doi.org/10.1007/s12040-021-01772-2]

24. Mohanty, L; Maiti, S; Dixit, A. Spatio-temporal assessment of regional scale evolution and distribution of glacial lakes in Himalaya. Front Earth Sci; 2023; 10, 1038777. [DOI: https://dx.doi.org/10.3389/feart.2022.1038777]

25. Mohamed Hanipha M, Zahir Hussain A. Study of groundwater quality at Dindigul town, Tamilnadu, India. Int J Environ Sci. 2013;2(1):68–73.

26. Nepal M, Bharadwaj B. Making urban waste management and drainage sustainable in Nepal. Clim Change Commun Resilience 2022:325–38

27. Pal, S; Chowdhury, P; Talukdar, S; Sarda, R. Modelling rabi crop health in flood plain region of India using time-series Landsat data. Geocarto Int; 2022; 37, 13 pp. 3761-3790. [DOI: https://dx.doi.org/10.1080/10106049.2020.1869328]

28. Panda, A; Sahu, N; Behera, S; Sayama, T; Sahu, L; Avtar, R; Yamada, M. Impact of climate variability on crop yield in Kalahandi, Bolangir, and Koraput districts of Odisha, India. Climate; 2019; 7, 11 126. [DOI: https://dx.doi.org/10.3390/cli7110126]

29. Pandey, AC; Singh, SK; Nathawat, MS; Saha, D. Assessment of surface and subsurface waterlogging, water level fluctuations, and lithological variations for evaluating groundwater resources in Ganga Plains. Int J Digital Earth; 2013; 6, 3 pp. 276-296. [DOI: https://dx.doi.org/10.1080/17538947.2011.624644]

30. Prabhakar, AK; Singh, KK; Lohani, AK; Chandniha, SK. Assessment of regional-level long-term gridded rainfall variability over the Odisha State of India. Appl Water Sci; 2019; 9, pp. 1-15. [DOI: https://dx.doi.org/10.1007/s13201-019-0975-z]

31. RamyaPriya, R; Elango, L. Geogenic and anthropogenic impacts on the water quality of cauvery river. Riverine Systems: Understanding the Hydrological, Hydrosocial and Hydro-heritage Dynamics; 2022; Cham, Springer International Publishing: pp. 149-170. [DOI: https://dx.doi.org/10.1007/978-3-030-87067-6_9]

32. Saha D, Ray RK. Groundwater resources of India: potential, challenges and management. Groundw Dev Manag Issues Challenges in South Asia. 2019:19–42.

33. Shahid, S; Nath, S; Roy, J. Groundwater potential modelling in a soft rock area using a GIS. Int J Remote Sens; 2000; 21, 9 pp. 1919-1924. [DOI: https://dx.doi.org/10.1080/014311600209823]

34. Sharma, V; Tripathi, AK; Mittal, H. Technological revolutions in smart farming: current trends, challenges & future directions. Comput Electron Agric; 2022; 201, 107217. [DOI: https://dx.doi.org/10.1016/j.compag.2022.107217]

35. Shirke, KD; Kadam, AK; Pawar, NJ. Temporal variations in hydro-geochemistry and potential health risk assessment of groundwater from lithological diversity of semi-arid region, Western Gujarat India. Appl Water Sci; 2020; 10, 6 pp. 1-20. [DOI: https://dx.doi.org/10.1007/s13201-020-01235-x]

36. Sunitha, V; Reddy, YS; Suvarna, B; Reddy, BM. Human health risk assessment (HHRA) of fluoride and nitrate using pollution index of groundwater (PIG) in and around hard rock terrain of Cuddapah, AP South India. Environ Chem Ecotoxicol; 2022; 4, pp. 113-123. [DOI: https://dx.doi.org/10.1016/j.enceco.2021.12.002]

37. Wöhling, T; Wilson, S; Wadsworth, V; Davidson, P. Detecting the cause of change using uncertain data: Natural and anthropogenic factors contributing to declining groundwater levels and flows of the Wairau Plain aquifer, New Zealand. J Hydrol Reg Stud; 2020; 31, [DOI: https://dx.doi.org/10.1016/j.ejrh.2020.100715]