Impact of rainfall variability and land use/land cover dynamics on runoff in the Madamsilli dam catchment, India

Introduction

Land use and land cover (LULC) changes caused by anthropogenic activities have become one of the most significant drivers of hydrological alteration in watershed systems. Rapid urbanization, agricultural expansion, deforestation, and mining activities modify the natural landscape, thereby influencing infiltration, evapotranspiration, groundwater recharge, and surface runoff processes. These alterations ultimately affect water availability, sediment transport, and water quality within river basins. Several studies have demonstrated that land use transitions significantly influence watershed hydrology and runoff generation (Ololade et al., 2008; Duncan et al., 2009; Sewnet, 2016; Pande et al., 2021) [6,4,12,8] . Rainfall variability is another critical factor affecting hydrological processes. In India, the South-West monsoon contributes nearly 75-80% of the annual rainfall and plays a crucial role in sustaining agriculture and water resources. However, climate change has altered rainfall patterns, leading to increased frequency of extreme events such as droughts and floods. Rising temperatures, erratic precipitation distribution, and shifts in seasonal rainfall trends have significantly impacted rainfed agricultural systems (Prakasam, 2010) [10] . Therefore, understanding long-term rainfall variability and its relationship with land use changes is essential for sustainable watershed management. Remote sensing and Geographic Information System (GIS) techniques have emerged as powerful tools for monitoring spatio-temporal changes in land use and hydrological parameters.

Multi-temporal satellite imagery enables accurate mapping of land cover dynamics and supports hydrological modelling applications (Zubair, 2006; Bedini et al., 2008; Agaton et al., 2016) [17,3,1] . These tools facilitate the assessment of watershed characteristics, including slope, drainage, vegetation cover, and impervious surfaces, which are critical inputs for runoff estimation models. Among various hydrological models, the Soil Conservation Service Curve Number (SCS-CN) method is widely used for estimating direct surface runoff from rainfall events. The method integrates land use, soil type, and antecedent moisture conditions to compute runoff depth. Due to its simplicity and reliability, the SCS-CN approach has been successfully applied in several Indian watersheds (Patel et al., 2017; Sutradhar, 2018; Satheeshkumar et al., 2017) [9,13,11] . Integration of GIS with the SCS-CN method enhances spatial runoff estimation accuracy and enables watershed-scale analysis (Malekani et al., 2014) [5] . Trend analysis of rainfall and runoff is equally important for evaluating climate variability impacts. Non-parametric statistical tests such as the Mann-Kendall (MK) test are widely applied to detect monotonic trends in hydrological time series without assuming normal data distribution (Vivekanandan, 2007; Malik & Kumar, 2020; Tiwari & Pandey, 2019) [16,6,15] . These methods are particularly useful in identifying significant increasing or decreasing rainfall trends, which can inform water resource planning and reservoir management strategies. Dams and reservoirs play a vital role in regional economic development by supporting irrigation, hydropower generation, and drinking water supply. However, land use transformation in dam catchments can significantly alter runoff patterns and sediment yield. Studies have shown that deforestation and urban expansion increase impervious surfaces, thereby enhancing runoff and reducing groundwater recharge (Bagwan & Gavali, 2021; Tewabe & Fentahun, 2020) [2,14] . Conversely, vegetation cover promotes infiltration and reduces peak discharge. The Madamsilli Dam catchment, located in Dhamtari district of Chhattisgarh, is an important watershed that supports irrigation, hydroelectric power generation, and domestic water supply. The findings of this study will contribute to improved watershed management, reservoir sustainability planning, and climateresilient water resource strategies in central India.

Study Area

The Madamsilli Dam, is an earth-fill embankment dam located on the Silari River, a tributary of the Mahanadi in central eastern India also known as the Babu Chhotelal Shrivastav Dam. It was constructed between 1914 and 1923 under the supervision of the British Raj governor, Madamsilli, after whom it was named. Situated in Chhattisgarh's Dhamtari District, the dam is approximately 95 kilometers from Raipur and is considered a notable architectural marvel in the region. Its primary function is to provide irrigation water to agricultural areas. The Madamsilli Dam serves various purposes, including irrigation for the agricultural regions of Dhamtari district, hydroelectric power generation, and fulfilling the drinking water needs of the area. It has played a crucial role in the economic development of the region by boosting agricultural production, providing employment opportunities, and generating hydroelectricity. However, the damhas faced challenges such as silting and environmental damage, which pose risks to sustainable development. To tackle these issues, the government has implemented measures such as reservoir desilting.

The Madamsilli Dam is an important and versatile structure with a surface area of 110 square kilometers and 1,170 million cubic meters of reservoir capacity. During its construction, the government relocated and rehabilitated people residing in the reservoir area, providing compensation and rehabilitation packages to ensure a smooth transition for the affected individuals.

Despite the challenges faced, the Madamsilli Dam has made substantial contributions to the region's economic growth by increasing agricultural productivity, generating hydroelectric power, and supplying potable water to the local population. The government has taken proactive steps, including desilting the reservoir, to address issues like silting and environmental damage and ensure the dam's sustainable development.

Methodology Data Collection and Preparation

The present study integrates remote sensing, GIS, hydrological modelling, and statistical trend analysis to assess the impact of rainfall variability and land use/land cover (LULC) changes on runoff generation in the catchment of Madamsilli Dam. The following datasets were used:

All spatial datasets were projected to a common coordinate system and processed using ArcGIS software.

Land Use/Land Cover Classification Supervised classification using the Maximum Likelihood Algorithm was performed to classify the satellite images into four major classes:

Runoff Estimation Using SCS-CN Method

Surface runoff was estimated using the Soil Conservation Service Curve Number (SCS-CN) method, which integrates rainfall, land use, soil type, and antecedent moisture condition. For AMC III:

AMC II was considered as the standard condition for annual runoff estimation.

Trend Analysis Using Mann-Kendall Test

Rainfall trend analysis was conducted using the non-parametric Mann-Kendall (MK) test, which does not require normally distributed data. The test statistic S is defined as:

he standardized test statistic Z is calculated as:

The test was applied to: The MK test evaluates the null hypothesis (H₀) that the data are independent and randomly ordered against the alternative hypothesis (H₁) that a monotonic increasing or decreasing trend exists over time. A key assumption of the test is that the observations are serially independent.

Mann-Kendall Test Statistic

The Mann-Kendall test statistic (S) is defined as:

• and are sequential data values at time i and j, • n is the total number of observations. A positive value of S indicates an increasing trend, while a negative value indicates a decreasing trend.

Mean and Variance of S

Under the assumption that the data are independent and identically distributed, the mean of S is:

The variance of S is calculated as:

• n = number of observations • m = number of tied groups

•

= number of data points in the k-th tied group

Standardized Test Statistic (Z)

The standardized test statistic Z is computed as:

The corresponding p-value is then obtained from the standard normal distribution.

Interpretation of Results

At a 95% confidence level (α = 0.05):

96 → The trend is statistically significant.

• If |Z| ≤ 1.96 → The trend is not statistically significant.

A positive Z value greater than 1.96 indicates a significant increasing trend, whereas a negative Z value less than -1.96 indicates a significant decreasing trend.

Since Z approximately follows a standard normal distribution for large sample sizes (n > 10), the MK test is suitable for longterm rainfall datasets such as the one used in this study.

Antecedent Moisture Condition (AMC)

Antecedent Moisture Condition (AMC) represents the wetness condition of the watershed prior to a rainfall event. It reflects the soil moisture status at the beginning of rainfall and plays a critical role in determining the magnitude of surface runoff. Higher antecedent soil moisture generally results in greater runoff generation due to reduced infiltration capacity. AMC is classified into three categories based on the total rainfall received during the previous five days:

• AMC I -Dry soil condition, characterized by low soil moisture and high infiltration capacity.

• AMC II -Average soil moisture condition, representing normal field conditions. • AMC III -Wet or saturated soil condition, resulting from significant rainfall during the preceding five days.

In the SCS-CN method, Curve Number (CN) values are typically provided for AMC II conditions. However, equivalent CN values for AMC I and AMC III can be computed using the following empirical relationships:

Calculation of Curve Number for AMC I Calculation of Curve Number for AMC III

Where: The definition of Barren area as used here denotes agricultural land and open spaces.

The absence of a significant monotonic trend suggests that rainfall variability in the region is characterized by short-term fluctuations rather than systematic long-term increase or decrease. However, when combined with land use transformation, even statistically insignificant rainfall variability may produce noticeable changes in runoff response.

Overall, the results indicate that while rainfall trends remain statistically stable, land use changes-particularly urban expansion and forest reduction-are likely playing a progressively important role in modifying runoff generation and hydrological behavior within the catchment of Madamsilli Dam. The definition of Barren area as used here denotes agricultural land and open spaces. Overall, the study confirms that rainfall variability combined with significant land use and land cover changes has influenced the hydrological response of the watershed. The increase in built-up areas and reduction in forest cover may enhance surface runoff, reduce infiltration, and alter watershed hydrodynamics. These findings emphasize the need for sustainable land management, watershed planning, and integrated water resource management strategies to ensure long-term hydrological stability and ecological balance in the catchment.

Acknowledgement

AcknowledgmentThe authors gratefully acknowledge the support provided by India & State Data Centre, Chhattisgarh Council of Science & Technology and Water Resources Department, Chhattisgarh.

References

1. Agaton M, Setiawan Y, Effendi H. Land use/land cover change detection in an urban watershed: A case study of Upper Citarum Watershed, West Java Province, Indonesia. Procedia Environmental Sciences. 2016;33:654-660.
2. Bagwan WA, Gavali RS. Dam-triggered land use land cover change detection and comparison (transition matrix method) of Urmodi River watershed of Maharashtra, India: A remote sensing and GIS approach. Geology, Ecology, and Landscapes. 2021.
3. Bedini E, Naço P, et al. Automated change detection from remote sensing data: A case study at the Pali Cape-Erzeni River mouth coastal sector. In: BALWOIS Conference Proceedings; 2008.
4. Duncan EE, Kuma JS, et al. Open pit mining and land use changes: An example from Bogosu-Prestea area, southwest Ghana. Electronic Journal of Information Systems in Developing Countries. 2009;36(3):1-10.
5. Malekani L, Khaleghi S, Mahmoodi M. Application of GIS in modeling Zilberchai Basin runoff. International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. 2014;40(2):181-186.
6. Malik A, Kumar A. Spatio-temporal trend analysis of rainfall using parametric and non-parametric tests: Case study in Uttarakhand, India. Theoretical and Applied Climatology. 2020;140:183-207.
7. Ololade O, Annegarn HJ, Limpitlaw D, Kneen MA. Landuse/cover mapping and change detection in the Rustenburg mining region using Landsat images. In: IEEE International Geoscience and Remote Sensing Symposium Proceedings; 2008.
8. Pande CB, Moharir KN, Khadri SFR. Assessment of landuse and land-cover changes in Pangari watershed area (MS), India, based on remote sensing and GIS techniques. Applied Water Science. 2021;11(6):96-96.
9. Patel J, Singh NP, Prakash I, Mehmood K. Surface runoff estimation using SCS-CN method-A case study on Bhadar Watershed, Gujarat, India. International Journal of Innovative Research. 2017;3(5):1213-1218.
10. Prakasam C. Land use and land cover change detection through remote sensing approach: A case study of Kodaikanal taluk, Tamil Nadu. International Journal of Geomatics and Geosciences. 2010;1(2):150-158.
11. Satheeshkumar S, Venkateswaran S, Kannan R. Rainfallrunoff estimation using SCS-CN and GIS approach in the Pappiredipatti watershed of the Vaniyar sub basin, South India. Modeling Earth Systems and Environment. 2017;3:18.
12. Sewnet A. Land use/cover change at Infraz watershed by using GIS and remote sensing techniques, northwestern Ethiopia. International Journal of River Basin Management. 2016;14(2):133-142.
13. Sutradhar H. Surface runoff estimation using SCS-CN method in Siddheswari River basin, Eastern India. Journal of Geography, Environment and Earth Science International. 2018;17(2):1-9.
14. Tewabe D, Fentahun T. Assessing land use and land cover change detection using remote sensing in the Lake Tana Basin, Northwest Ethiopia. Cogent Environmental Science. 2020;6(1):1778998.
15. Tiwari H, Pandey BK. Non-parametric characterization of long-term rainfall time series. Meteorology and Atmospheric Physics. 2019;131:627-637.
16. Vivekanandan N. Analysis of trend in rainfall using nonparametric statistical methods. In: AIP Conference Proceedings. 2007;923(1):101-113.
17. Zubair AO. Change detection in land use and land cover using remote sensing data and GIS: A case study of Ilorin and its environs in Kwara State. Ibadan: University of Ibadan; 2006.