Studying the impact of rainfall and LULC changes in the reservoir water level of the Hatia dam, Ranchi, India

Rainfall is a crucial factor that shapes vegetation, water quality, and hydrology of a region. Crop selection in agriculture is largely determined by temperature and precipitation patterns, which can vary greatly across regions. Understanding rainfall features, such as its amount, distribution, and seasonal variations, is essential for increasing agricultural productivity and adapting to changing climatic conditions (Gajbhiye et al. 2016; Meshram et al. 2017). India's economy, food security, energy security, and agriculture are all directly or indirectly dependent on timely access to a sufficient quantity of rainfall (Jain & Kumar 2012). The rainfall dynamics are crucial in understanding regional hydrological systems. It is equally important to consider the role of anthropogenic factors, such as changes in land use and land cover (LULC) that also significantly impact water resources.

To analyse changes in LULC in river basins, especially those in arid and semi-arid regions, large geospatial datasets may be processed using cloud computing platforms like Google Earth Engine (GEE). This allows for the making of well-informed decisions on sustainable water management (Souza et al. 2020; Niu et al. 2022). Changes in LULC have a direct influence on climate change, hydrological processes, and other environmental issues (Iban & Sachin 2022; Rautela et al. 2023). Several factors including population increase, socioeconomic development, deforestation, urbanization, agriculture, and decreasing water quality, have contributed to human-caused changes in LULC (Sertel et al. 2010; Napoli et al. 2017). Urbanization, agriculture, and soil nutrient fluxes affect the growth of native species and the composition of soil textures. Changes in land cover may also influence the way reservoir function. Changes in land cover is the main factor affecting surface runoff, and these changes also affect the groundwater recharge, evapotranspiration, infiltration, and other hydrological parameters (Ranjan & Singh 2022). It is important to investigate the spatial distribution, size, extent, and position of land cover changes to ascertain the extent to which these changes may impact hydrological processes locally, regionally, and globally (Sertel et al. 2010; Lee & Berbery 2012). As land cover changes and hydrological consequences are closely related, it is critical to evaluate the potential effects of fast urbanization, deforestation, and agricultural practices, especially in surrounding areas of sensitive ecosystems, on water availability and ecosystem health.

Sustainable urban development strategies are necessary because urbanization causes changes in LULC patterns, which may have negative ecological effects (Long et al. 2008). In the present context, LULC changes affect hydrological systems and are crucial for sustainable development, optimal use, and management of water resources (Ayivi & Jha 2018; Sam & Khoi 2022). Debnath et al. (2022) observed the impact of temperature-rainfall and LULC changes on the hydrological regime of a forested watershed in Northeast India from 1986 to 2016. Results showed significant decreases in water discharge, level, and hydraulic variables, with LULC changes being more responsible than rainfall variations (Shree & Kumar 2018). Using Mann–Kendall (MK) and Sen's slope approaches, the study examined trends in seasonal and annual rainfall in Ranchi district of Jharkhand from 1901 to 2014. The susceptibility of region to droughts and floods was shown by the results, which indicated large drops in annual, winter, and southwest monsoon rainfall and increases in pre- and post-monsoon rainfall. The Hatia dam, located on the river Subarnarekha in the upstream of Ranchi, (Jharkhand), India, is the main source of water supply for the Ranchi city, located in Jharkhand state of India. After formation of Jharkhand state and declaration of Ranchi city as its capital, there is rapid increase in urbanization and population growth. Only 2.25% of Jharkhand's total land area is urbanized with a population of 2,350,612 in Ranchi district, according to the Census of India (2001). As per the Census of India (2011), the population of Ranchi has increased to 19%. This surge in population expansion and thereby urban development has led to significant shifts in land-use patterns, particularly in regions that were previously dominated by tribal populations. These changes have not only altered the landscape but also exacerbated water management challenges.

Land-use patterns in tribal-dominated regions have changed due to rapid urbanization and population expansion, especially around the outer edges of major towns. The creation of advanced industry, especially in the vicinity of the state capital, has led to an increase in urbanization in Jharkhand (Kumar et al. 2011). People have been forced to relocate because of their land being taken for industrial purposes, making it difficult to provide for and rehabilitate them (Priyadarshi & Dutt 2000). The rapid increase in urbanization and population growth in the region has led to an increased demand for water, imposing pressure on existing water sources. The construction of the Ranchi ring road has affected the drainage and runoff capacity of the basin, with infrastructure development encroaching on the drainage and entry of water into the reservoir. Rapid urbanization in Ranchi over the last few decades has altered the landscape, disturbed water retention capacity, and depleted groundwater, leading to a water crisis and contamination.

The process of urbanization has significant implications for changes in demographic characteristics and the physical transformation of landscapes. However, when urbanization occurs in an unplanned, unsystematic, and rapid manner, it can lead to profound impacts on various environmental components, particularly on land and water resources (Patra et al. 2012). Urbanization also leads to a reduction in infiltration, which in turn affects the recharge of aquifers and alters streamflow regimes. This can result in an increase in surface runoff (Saddique et al. 2020). By changing the quantity and timing of precipitation, the way water penetrates and percolates through soils, and the way water is released into streams and rivers, changes in land use can have a direct effect on the hydrological cycle. Additionally, utilization of land changes in vegetation cover, soil qualities, and land management techniques are examples of how physical, chemical, and biological aspects of the landscape can change and have an indirect impact on the hydrologic response (Gabiri et al. 2020). Further affecting water resources and ecosystem health, these changes may have an impact on the way water travels through the environment and the mechanisms that control it (Zhang et al. 2017). Land-use change can have profound and multifaceted impacts on the hydrological cycle, with consequences for water resources and the broader environment. Hence, the main objectives are (i) to study the trends of historical data of rainfall and temperature in the catchment of the Hatia dam and its impacts on the water spread area of the reservoir and (ii) to study the LULC changes and its impact on the reservoir surface area in the dam. The findings of the study can be used for conservation efforts and water resource management decisions by accounting for the hydrological impacts of LULC changes in the catchment of the Hatia dam.

The meteorological characteristics are essential datasets for hydrological process modelling. These characteristics were received in gridded form from the Indian Meteorological Department (IMD) Pune from 1991 to 2023. The datasets include rainfall, maximum temperature (T_max), and minimum temperature (T_min). Table 1 presents the data used and its sources. The Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) Digital Elevation Model (DEM) with a resolution of 30 m was utilized for the preparation of stream network. This resolution is appropriate for precisely capturing important topographical features such as elevation variations, drainage patterns, and watershed boundaries, which are essential for hydrological modelling and LULC analysis. ASTER DEM data are freely available and offers global coverage, making it accessible for consistent use across different regions, including the study area in Ranchi, India. Also, it is compatible with major GIS platforms, such as ArcGIS and GEE, which were used for LULC and hydrological analyses in this study. The 30 m resolution is sufficient for watershed delineation, stream network analysis, and slope calculations, which are critical components of hydrological studies. In the present study, GEE was utilized to generate the LULC data for the basin by using images from Landsat 5 and Landsat 8.

The time series analysis of gridded temperature and rainfall data throughout several seasons, such as annual, pre-monsoon, post-monsoon, winter, and monsoon, provides a thorough spatial description of the distribution and variability of precipitation in the research region. These graphical depictions help to provide a comprehensive knowledge of seasonal precipitation dynamics and their effects on hydrological systems by observing the various patterns, trends, and intensities of rainfall throughout various time periods. These graphs are vital for examining the interactions between seasonal precipitation variation and their impact on hydrological systems because they provide a visual representation of the geographical subtleties of rainfall throughout different seasons.

The area-averaged gauge rainfall was interpolated using the Thiessen polygon technique. This method involves creating polygons around each rain gauge, with the area of each polygon representing the region of influence for that gauge. In the study area, only one of the four gridded points was found to be usable and available very close to the study area. Inverse distance weighting (IDW) interpolation technique was used for interpolation of rainfall data in the present study.

Trend analysis of rainfall is a method used to detect changes in temperature and rainfall patterns over time. It involves analysis of long-term temperature and rainfall data to identify patterns, trends, and changes in rainfall amounts, frequency, and distribution. This analysis helps to forecast future rainfall patterns and understand the impacts of climate change on hydrological systems.

This is widely used in non-parametric method for detecting monotonic trends in hydrological and meteorological time series data (Mann 1945; Kendall 1975; Patra et al. 2012; Himayoun & Roshni 2019; Sa'adi et al. 2019; Deoli & Kumar 2020) analyzing time series data that deviates from a normal distribution, like hydrological data, is appropriate for the MK test.

The hypothesis that the MK test would examine was that there is no trend in the time series (H₀), and that there is a monotonic trend (H₁) that can be either upward or downward.

Sen's slope estimator test is used to get a non-parametric trend slope, which indicates how considerable a trend is. The overall trend slope is determined by first calculating the slope for each pair of points throughout time, and then taking the median of all the slopes (Sen 1968). If the calculated value of β is positive, it represents an ‘upward trend’ while a negative value depicted ‘downward trend’ (Xu et al. 2007; Karpouzos et al. 2010). Sen's slope has an advantage over parametric approaches in that it is not affected by extreme values or outliers, which might affect the results. Because of this, it is the best option for evaluating environmental data that could see abrupt changes because of severe weather or natural variability.

Sen's slope estimator and the MK tests were selected for this study as they are appropriate for assessing skewed and non-normally distributed environmental data, such as rainfall and temperature records.

The term LULC describes how natural components and human activity are categorized on a landscape during a certain period. The LULC maps are essential for tracking changes in land-use patterns over time, which supports environmental management and conservation efforts. Developing infrastructure, planning urban expansion, and ensuring sustainable urban growth all depend on LULC maps. Using scientific and statistical methods, GEE is used to create LULC classifications, which facilitate the study and mapping of different land cover types within a given region during a certain period. The description of all the classes of LULC is shown in Table 2.

The validation techniques for accuracy and classification were combined into a single script in GEE. The Random Forest (RF) classifier was used to generate a confusion matrix using the same training and validation data. These techniques can be categorized into two approaches: supervised classification methods like RF, or those based on pixels, objects, or a combination of both, which also utilize supervised classification techniques like RF (Gislason et al. 2006; Tatsumi et al. 2015; Wang et al. 2015). The RF method is highly effective for non-parametric machine learning strategy. For this investigation, RF was used as it yielded as almost precise quality findings for land use (Qu et al. 2022). When compared to other well-known remote sensing technologies, its usage in mapping land use and land cover offers several advantages, including greater accuracy (Zeferino et al. 2020).

For identifying LULC, classify the study area using pixels and pixel-based RF machine learning approach that was applied in GEE cloud computing. The high-resolution photos from Google Earth were manually processed to collect the training and validation samples for this study. In the GEE, 80% of the sample data is used for model training and validation, while 20% is reserved for testing. A confusion matrix is generated based on the results, and the model accuracy is evaluated. For every LULC map, the model accuracy must be more than 90%. Until the required accuracy standards are reached, and the procedure is repeated if the accuracy is less than 90%. By using a continuous improving method, the model performance is ensured to meet the necessary requirements. From the six bands that are available, which are B1, B2, B3, B4, B5, and B7, a subset of three bands is frequently used. The selected bands, which are commonly used for analysis, are B2, B3, and B4. A total of approximately 350 sample data points are typically collected, which are then divided evenly among all the classification categories. This ensures that each class has a representative number of samples to facilitate accurate model training and validation.

The trend analysis of maximum and minimum temperatures using MK and Sen's slope estimators revealed significant findings. Table 5 demonstrates an increasing trend in annual maximum average temperature, as well as in winter, monsoon, and post-monsoon maximum temperatures, with pre-monsoon temperature also showing a rising trend. This analysis suggests a correlation between the rising maximum temperatures, particularly in recent years, and a decreasing trend in rainfall. Studies on climate change and its impacts on biodiversity and wetlands support the idea that annual and seasonal temperatures are indeed on the rise. The escalation in maximum temperature is associated with reduced rainfall, leading to diminished drainage, likely due to altered atmospheric conditions affecting precipitation patterns and subsequent drainage systems. Furthermore, Table 6 indicates a significant increasing trend in annual minimum average and monsoon temperatures, while winter, pre-monsoon, and post-monsoon temperatures exhibit an insignificant increasing trend. The notable increase in annual minimum average and monsoon temperatures aligns with the overall warming of the climate. In contrast, the insignificant trends in winter, pre-monsoon, and post-monsoon temperatures suggest that these seasons are less impacted by climate change or that other factors, such as local weather patterns, play a more dominant role during these periods. This nuanced pattern of temperature changes across different seasons underscores the diverse impacts of climate change on temperature trends throughout the year.

The MK test and Sen's slope estimator were applied to identify a pattern in the time series data. The performed test was executed at the 5% level of significance, and the results of the MK test and Z-statistics showed both positive and negative trends for the seasonal and annual rainfall time series. Monsoon rainfall showed a statistically significant negative trend (P < 0.05), while winter and post-monsoon rainfall showed a statistically insignificant negative trend (P < 0.05).

Table 7 shows the statistics of rainfall including the maximum, mean, minimum, and standard deviation values. This provides a comprehensive overview of the rainfall statistics for that time frame, with the highest rainfall occurring at 2163.1 mm annually and the lowest being 98.9 in winter maximum rainfall. Table 8 shows the result of Kendall's tau, p-value, Sen's slope, and trend at 5% level of confidence. In this result, it can be clearly seen that annual and monsoon rainfall is decreasing with a significant trend. In the pre-monsoon season, there is an insignificant increasing trend, while winter and post-monsoon seasons exhibit an insignificant decreasing trend. Negative Sen's slope values indicate a decreasing trend, while positive values signify an increasing trend. The Z-statistics for annual rainfall revealed the largest negative value, while those for winter rainfall revealed the lowest negative value. This result is consistent with the findings of (Chandniha et al. (2017). The authors have observed that over the region of Ranchi district for the period of 1901–2014, there were positive Z-statistics values for pre-monsoon and post-monsoon rainfall time series and negative Z-statistics values for winter, monsoon, annual and post-monsoon rainfall time series. Sharma & Singh (2017) show that pre-monsoon and post-monsoon rainfall trends in Jharkhand were increasing, but trends in the winter, monsoon, and annual rainfall series were found to be decreasing for a period of 102 years. This pattern reflects varying temperature trends with rainfall pattern along different season.

These changes highlight the significant impact of urbanization and infrastructure development on the landscape, with a notable increase in built-up areas and a corresponding decrease in agricultural lands. The relatively stable forest cover suggests that conservation efforts may have helped maintain the existing forest areas during this period. Table 9 shows the changes in Land Use/Land Cover (LULC) classes and their corresponding areas during the post-monsoon season, along with the annual rainfall data. Table 10 shows the changes observed in the reservoir surface area during pre-monsoon and post-monsoon rainfall events. The data indicates that when rainfall is less, the drainage water that flows into the reservoir is also less. However, following the construction of the Ranchi Ring Road, the drainage into the reservoir is expected to be less than the previous runoff. This reduction in drainage can be attributed to the altered hydrological dynamics in the catchment area due to the construction of the ring road. The road infrastructure may have disrupted the natural flow of water, leading to reduced runoff into the reservoir. Tables 11 and 12 present the changes in the reservoir over five-year intervals during the pre-monsoon and post-monsoon periods, respectively from 1993 to 2023. In the pre-monsoon period, there is an increase in the percentage change in waterbody, built-up area, and forest cover by 3.74, 8.37, and 0.21%, respectively. Furthermore, there is a decrease in the percentage change in barren land and agricultural area by 1.81 and 10.53%, respectively. During the post-monsoon season, similar trends are observed, with a slight variation in the percentage changes. The waterbody, built-up area, and forest cover increased by 2.76, 8.53, and 0.25%, respectively. Barren land and agricultural area decreased by 1.68 and 9.88%, respectively. These changes highlight the significant impact of urbanization and infrastructure development on the landscape, with a notable increase in built-up areas and a corresponding decrease in agricultural lands.

The Hatia dam located in the Subarnarekha river is mainly built for the purpose of water supply to the Ranchi city. Due to the construction of the Ranchi ring road in the catch area of the Hatia reservoir, significant decrease in the surface area has been noticed. As a result, studies on the impacts of LULC changes on the water body are very essential in this study area. Also, the effect of rainfall and temperature pattern on the water body has been carried out.

The analysis of rainfall data from 1991 to 2023 shows a significant decrease in annual, monsoon, post-monsoon, and winter rainfall and an increase during pre-monsoon season in the Hatia dam catchment. This has led to reduced inflow and drying up of the reservoir resulting a decrease in water supply to Ranchi town. The LULC analysis using GEE reveals a substantial increase in built-up areas from 2.71 to 11.08% during the pre-monsoon period and 2.74 to 11.27% during the post-monsoon period. The decrease in Hatia dam's water surface area from 126.31 to 37.44 ha shows that urbanization may lead to severe water shortages, posing risks to both human populations and local ecosystems. Climate projections suggest that rainfall patterns will continue to shift, exacerbating the already stressed hydrological systems in the region. The results highlight the urgent requirement for sustainable approaches to water management to tackle the combined consequences of increasing urbanization and climatic variability. Failure to address these issues could result in further depletion of water supplies, increased flooding risks, and long-term damage to the environment. The Ranchi ring road development has significantly affected water drainage into the reservoir, highlighting the significance of integrating environmental assessments into urban planning. To ensure the sustainable management of water resources in rapidly urbanizing region like Ranchi, the following policy recommendations should be considered such as integrated urban planning, sustainable drainage systems, water conservation and efficiency programs, climate-resilient infrastructure, and monitoring and regulation. Thus, urgent policy interventions are needed to ensure that water resources are managed sustainably and that urban growth is aligned with climate adaptation strategies.

The author would like to thankfully acknowledge the India Meteorological Department, Pune.

S.K. contributed to literature review, conceptualization, and data curation. V.P.K. contributed to methodology, result analysis, conclusion, and writing the original manuscript. V.S. contributed to conceptualization AND supervision. R.T. contributed to reviewing and editing the original manuscript.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

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

The authors declare there is no conflict.