SGSubhankar Ghosh

Subhankar Ghosh

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Subhankar Ghosh has a total of 35 co-authors. Here are the top co-authors by number of shared publications:

  • Madan Kumar Jha

    AgFE Department, Indian Institute of Technology Kharagpur, Kharagpur, India

    5 shared publications
  • Amulya Kumar Mohanty

    Indian Council of Agricultural Research (ICAR), Agricultural Technology Application Research Institute (ATARI), Umiam (Barapani)

    3 shared publications
  • Laishram Kanta Singh

    Indian Council of Agricultural Research-Krishi Vigyan Kendra (ICAR-KVK) Imphal West, Indian Council of Agricultural Research (ICAR) Research Complex for North Eastern Hill (NEH) Region, Imphal, Manipur, India

    3 shared publications
  • Sougrakpam Roma Devi

    Indian Council of Agricultural Research (ICAR), Krishi Vigyan Kendra-Churachandpur

    2 shared publications
  • Abhaya Kumar Pradhan

    Department of Soil and Water Conservation Engineering, College of Agricultural Engineering & Technology, OUAT, Bhubaneswar, India

    1 shared publication
  • Afrin Naz

    1 shared publication

Latest Publications

  • Open Access

    Machine learning-based prediction of soil aggregate stability for sustainable land management in the eastern Himalayan region of India

    by Suryakant Gupta, Prachi Yadav, Jagabandhu Panda, Kiran Bala Behura, Subhankar Ghosh, Laishram Kanta Singh, Sougrakpam Roma Devi, Mainak Ghosh, Amulya Kumar Mohanty, Prasanta Kumar Pathak

    Introduction: Plot-scale rainfall-induced soil erosion data are scarce in the eastern Himalayas of India. Soil aggregate stability (SAS), commonly quantified by mean weight diameter (MWD), is a key indicator of erosion susceptibility. As laboratory determination of MWD is labor-intensive, machine learning (ML)-based pedotransfer functions (PTFs) using readily measurable soil properties offer an efficient alternative. Methods: In this study, 126 topsoil samples (0–15 cm depth) were collected from agricultural (n = 61) and forest (n = 65) lands in a sub-watershed of Sikkim, India, to develop and evaluate ML-based models for soil MWD prediction. Eleven soil physico-chemical parameters, i.e., ten ‘input’ (%Sand, %Silt, %Clay, %OC, BD, pH, Ca 2+ , Mg 2+ , CO 3 2− , and HCO 3 −), and one ‘target’ (MWD), estimated in the laboratory, were used in this study. Three supervised ML techniques, such as ANN, RF and SVM, along with empirical MLR, were used to predict soil MWD. Results and Discussion: The mean BD, %OC and MWD of agricultural land were 1.25 g/cm 3 , 3.37% and 1.09 mm, respectively, whereas those of forest land were 1.17 g/cm 3 , 4.23% and 1.11 mm, respectively. The sand fraction was found more in agricultural land (61.34%), while the clay proportion was found more in forest land (30.11%). All three ML models performed better than MLR in predicting soil MWD. The RF model performed best for the agricultural land (R 2 = 0.89, NSE = 0.89 for training; R 2 = 0.66, NSE = 0.66 for testing), while the ANN model was found to be superior for the forest land (R 2 = 0.92, NSE = 0.92 for training; R 2 = 0.91, NSE = 0.91 for testing). The SVM model predicted soil MWD more accurately for the pooled dataset (R 2 = 0.98, NSE = 0.98 for training; R 2 = 0.72, NSE = 0.72 for testing). No single technique is superior for all the land uses; however, the SVM model is either best or second best in all the cases. Therefore, the developed SVM-based SAS model can be employed to accurately predict soil MWD across the eastern Himalayan region and facilitate regional soil erosion investigations.

  • Open Access

    Monthly soil temperature modelling across different depths using machine learning algorithms in northeast India

    by Abhaya Kumar Pradhan, Deepak Jhajharia, Kiran Bala Behura, Subhankar Ghosh, Muhammad Touseef, Laishram Kanta Singh, Sougrakpam Roma Devi, Mainak Ghosh, Amulya Kumar Mohanty, Ghanashyam Singh Yurembam, et al.

    Soil temperature is one of the most crucial factors for agricultural systems, which affects various hydrogeological processes in soil. Soil temperature is a frequently used variable in machine learning (ML) approaches for modelling those intricate processes. In regions with limited data, such as northeast India, soil temperature is a meteorological variable that is rarely available. The present study aims at soil temperature modelling in three different soil depths (5, 15 and 30 cm), using eight long-term meteorological variables, for example, minimum, mean and maximum air temperatures, rainfall, sunshine hours, minimum and maximum relative humidity, and number of rainy days over the Tochlai region of Jorhat, Assam (northeast India). To accomplish this goal, five suitable ML techniques, for example, Random Forest Regression (RFR), Support Vector Regression (SVR), Boosted Regression Trees (BRT), Classification And Regression Tree (CART), and Bayesian Neural Network (BNN) were employed. Results indicate that the RFR (R2 = 0.983; NSE = 0.983) and the BNN (R2 = 0.981; NSE = 0.981) outperformed other models in 5 cm soil temperature prediction. In the case of 15 cm soil temperature prediction, the BNN (R2 = 0.988; NSE = 0.988) and the RFR (R2 = 0.987; NSE = 0.987) performed better than other models. However, the BNN (R2 = 0.975; NSE = 0.975) and CART (R2 = 0.970; NSE = 0.970) accurately predicted 30 cm soil temperatures in the study region. Based on the analysis for determining the variable importance, air temperature (mean temperature, followed by minimum and maximum temperatures) was found to be the most influential variable for predicting soil temperature. The findings of the study will be beneficial in formulating future research directions, like soil temperature-based crop modelling in data-limited regions and beyond, especially in the absence of multi-depth soil temperature information.


  • Open Access

    Comparative Evaluation of Multi-Criteria Decision Analysis Techniques for Groundwater Potential Mapping in a Coastal Alluvial Basin of Eastern India

    by Afrin Naz, Subhankar Ghosh, Madan Kumar Jha

    Groundwater potential zoning (GWPZ) using multi-criteria decision analysis (MCDA) and data-driven techniques has emerged as a vital tool for sustainable groundwater management, particularly in data-scarce coastal regions where surface water availability is limited and aquifers are vulnerable to overexploitation and seawater intrusion. Coastal aquifer systems are inherently heterogeneous and dynamically influenced by geomorphology, lithology, land use, and climatic conditions, making integrated decision-support frameworks essential for identifying zones with varying groundwater potential. In this context, the present study evaluates and compares six widely used subjective, objective and probabilistic MCDA techniques—Analytic Hierarchy Process (AHP), Fuzzy-AHP (F-AHP), Frequency Ratio (FR), Weight of Evidence (WoE), Entropy, and Multi-Influencing Factor (MIF), for analyzing groundwater prospect in a complex coastal alluvial setting. The study area with an areal extent of 6468.60 km2, comprises the Haldi–Kansabati–Subarnarekha interfluve located along the Bay of Bengal in the southern coastal region of West Bengal, eastern India. The region is characterized by a ‘Tropical Wet-and-Dry’ climate under the Köppen–Geiger classification and exhibits pronounced hydrogeological heterogeneity. The geological framework is dominated by younger alluvium (quaternary sediments) in the floodplains and deltaic tracts, followed by older alluvium (tertiary sediments) inland, coastal alluvium along the shoreline, and lateritic formations in the landward uplands. These formations exert strong control on groundwater occurrence, storage, and movement. Multiple groundwater-conditioning factors such as: ‘runoff coefficient’, ‘land slope %’, ‘drainage density’, ‘geology’, and ‘proximity to surface water bodies’ representing topography, hydrology, geology, land surface characteristics, and recharge conditions were integrated within the ArcGIS pro v3.4.2 environment to generate GWPZ maps using each of the six MCDA techniques. The resulting groundwater potential maps were classified into three categories—‘high’, ‘moderate’, and ‘low’ potential zones—using consistent classification criteria across all methods to enable inter-model comparison. Model validation was performed using observed pumping well yield (discharge) data, which were independently categorized into three classes: ‘low’ (90 m³/hr). The predictive performance of each GWPZ model was quantitatively evaluated using Pearson’s correlation coefficient (r), and Receiver Operating Characteristic (ROC) curve-derived Area Under the Curve (AUC) statistics generated through the ArcSDM v5.00.22 toolbox in ArcGIS Pro v3.4.2. The validation results demonstrate notable variability in model performance. Among the six MCDA techniques, Fuzzy-AHP (F-AHP) exhibited the highest predictive accuracy with a correlation coefficient (r) of 0.912 and an AUC value of 0.853, followed by AHP (r=0.897; AUC=0.815), WoE (r=0.868; AUC=0.773), FR (r=0.809; AUC=0.724), MIF (r=0.778; AUC=0.702), and Entropy (r=0.727; AUC=0.683). The superior performance of the hybrid Fuzzy-AHP technique highlights its robustness in handling uncertainty and subjectivity while maintaining logical consistency. The F-AHP technique further sustains gradual transitions in factor importance when dealing with complex coastal hydrogeological systems. Overall, the study underscores the robustness of MCDA-based approaches, particularly F-AHP and AHP, for groundwater potential assessment in coastal alluvial environments and provides a comparative framework to support groundwater planning and management in similar vulnerable coastal regions.

  • Open Access

    Mapping wheat potential: a GIS-based agro-land suitability analysis in Gujarat’s Bhal Region, Western India

    by Ranjan Kumar, Subhankar Ghosh, Gautam R. Patel, Laishram Kanta Singh, Amulya Kumar Mohanty, Arun Kumar Singha

    Introduction: Agro-land suitability evaluation is vital to determine the optimal use of agricultural land for specific crops, ensuring sustainable cultivation and maximizing yields.Methods: This study focuses on assessing the suitability of land for wheat (Triticum L) cultivation in the Bhal region of Gujarat state, western India, a prominent agricultural area where fluctuating production levels have been attributed to sub-optimal crop selection. With wheat being a staple crop essential in meeting the demands of a growing population, systematic evaluation of the region’s land suitability is both timely and necessary. The analysis integrates a comprehensive range of parameters (17 in total), including land use/land cover, slope, soil texture, soil depth, maximum water-holding capacity, pH, Electrical Conductivity, Soil Organic Carbon, and the availability of essential macro-nutrients (N, P, K and S) and micro-nutrients (B, Cu, Fe, Mn and Zn). Using an advanced Multi-Criteria Decision Analysis (MCDA) methodology such as the Analytic Hierarchy Process (AHP) in a Geographic Information System (GIS) framework, a detailed land suitability map was developed for wheat cultivation.Results and discussion: The soils in the Bhal region are found to be deficient of Nitrogen (N), Manganese (Mn), Iron (Fe) and Zinc (Zn), which necessitates fertilizer application. The highest and lowest normalized weights were obtained for soil texture (0.291) and Zinc concentration (0.017), respectively. The findings revealed that 24.80% (1121.1 km2) of total area is categorized as ‘highly suitable’, 31.37% (1418.2 km2) as ‘moderately suitable’, 12.40% (560.6 km2) as ‘least suitable’, and remaining 31.43% (1420.9 km2) as ‘not suitable’ for wheat cultivation. Ahmedabad, Kheda and Surendranagar districts are ‘highly’ to ‘moderately’ suitable, while Anand district is ‘moderately’ suitable and Bhavnagar district is ‘least’ suitable for wheat cultivation.Concluding remarks and significance: These results underscore the critical importance of soil fertility and crop-specific land assessment in guiding agricultural strategies. The present study provides valuable insights into agricultural land-use planning and crop management, offering a data-driven approach to enhance wheat productivity in the Bhal region. By addressing the challenges of crop selection and resource optimization, these findings contribute to ensuring food security and promoting sustainable agricultural practices in the region.


  • Open Access

    Evaluation of Index-Based Methods for Analyzing Seawater Intrusion Vulnerability in a Coastal Alluvial Aquifer of Eastern India

    by Subhankar Ghosh, Madan Kumar Jha, Vimlendra Mani Pandey

    The present study aims to explore seawater intrusion vulnerability in a coastal alluvial aquifer of West Bengal state, eastern India, using two GIS-based indexing techniques, viz., GALDIT and DRASTIC. The study region, with an area of 6358.70 km2, is underlain by two main aquifer systems (‘leaky confined’ and ‘confined’ aquifers). Daily rainfall data of 2020–2021, Pre-Monsoon (PRM) and Post-Monsoon (POM) seasons’ groundwater-level and groundwater-quality (EC, Clˉ and HCO3ˉ) data of 2021 for leaky confined aquifer, and lithology logs data were used. The GALDIT method incorporates six hydrogeochemical parameters as inputs, viz., groundwater occurrence (G), aquifer hydraulic conductivity (A), groundwater elevation (L), distance from the seashore (D), impact of existing seawater intrusion status (I), and aquifer thickness (T). Conversely, inputs to the DRASTIC method are: depth to groundwater level (D), net recharge (R), aquifer media (A), soil media (S), topography (T), impact of vadose zone (I), and aquifer hydraulic conductivity (C). All these thematic layers and their features were assigned weights and ratings, respectively, following original GALDIT and DRASTIC methodology. Seasonal seawater intrusion vulnerability maps were prepared using weighted overlay analysis in ArcGIS environment. Based on GALDIT Vulnerability Indices (GVI), the study area was delineated into three vulnerability zones, viz., ‘low’ (GVI=2.5–5.0), ‘moderate’ (GVI=5.0–7.5), and ‘high’ (GVI>7.5). Similarly, the whole area was categorized into three vulnerability zones depending on DRASTIC Vulnerability Indices (DVI), viz., ‘low’ (DVI=18–61), ‘moderate’ (DVI=61–104), and ‘high’ (DVI=104–146). Results of the GALDIT method indicated 19–31% of the total area under ‘low’, 66–78% under ‘moderate’ and 3–4% under ‘high’ vulnerability classes in different seasons. Outcomes of the DRASTIC method revealed 21–78% area under ‘moderate’ and 22–79% under ‘high’ vulnerability classes. Finally, results of GALDIT and DRASTIC methods were validated with measured Electrical Conductivity (EC) concentrations. As per drinking and irrigation suitability, seasonal EC maps were categorized into three classes, viz., ‘acceptable/low hazardous’ (EC3000 μS/cm). The GALDIT method predicted 50–64% less area as ‘low’, 47–61% higher area as ‘moderate’, and 3–3.5% more area as ‘high’ vulnerable zones compared to the corresponding EC classes. Conversely, the DRASTIC technique estimated 81–84% less area as ‘low’, 2–61% higher area as ‘moderate’, and 22–79% more area as ‘high’ vulnerable zones. Moreover, moderate correlations were found between GVI and EC in both PRM (r=0.518) and POM (r=0.589) seasons, whereas poor correlations were found among DVI and EC in both PRM (r=0.442) and POM (r=0.118) seasons. Additionally, Receiver Operating Characteristic (ROC) curves revealed high Area Under the Curve (AUC) values for the GALDIT method in both PRM (AUC=0.872) and POM (AUC=0.891) seasons, whereas lower AUC values were obtained for the DRASTIC method in both PRM (AUC=0.810) and POM (AUC=0.573) seasons. Therefore, these results suggest that the GALDIT method delineated seawater intrusion vulnerable zones much better than the DRASTIC method. The outcomes of this research will aid in identifying priority zones (moderate-to-high vulnerable) to implement efficient groundwater-quality management programs.

  • Evaluating trends in groundwater quality of coastal alluvial aquifers of Eastern India for sustainable groundwater management

    by Subhankar Ghosh, Madan Kumar Jha

    Groundwater is a precious natural element which ensures global water, food, and environmental security in the twenty-first century. Systematic monitoring, sustainable utilization, preservation and remediation are critical aspects of efficient groundwater resource management. This study deals with the analysis of spatial variability and trend in groundwater chemistry as well as identification of possible contamination sources in a coastal alluvial basin of eastern India. Pre-monsoon season data of 14 groundwater-quality variables measured in ‘leaky confined’ and ‘confined’ aquifers were analyzed for ten years (2012–2021). Mann–Kendall (M–K) test with the Sen’s Slope Estimator, Spearman Rank Order Correlation (SROC) and Innovative Trend Analysis (ITA) tests were employed to assess decadal (2012–2021) trends. The analysis of the results indicated that the ‘critical’ water-quality parameters exceeding the acceptable limits for drinking are TDS, EC, TH, pH, Mg2+, Na+, K+, Fe2+, HCO3ˉ, Clˉ and NO3ˉ. Weak negative correlations between rainfall and groundwater elevation for both the aquifers reveal poor rainfall recharge into the aquifers. Therefore, a reduction in groundwater abstraction and augmentation of groundwater recharge is recommended. Trend analysis results indicated that the concentrations of TH, Mg2+ and Fe2+ exhibit significant increasing trends in the ‘leaky confined aquifer’. In contrast, significant rising trends in TH, Mg2+, Na+, Fe2+, HCO3ˉ and NO3ˉ concentrations are identified in the ‘confined aquifer’. Further, the SROC test could not detect the trends in groundwater quality in most blocks and for many parameters. On the other hand, the ITA test revealed significant trends in most of the parameters of the two aquifers in almost all the blocks. Trend magnitudes of the groundwater-quality parameters based on the Sen’s Slope Estimator and the ITA test vary from –63.7 to 58.65 mg/L/year for TDS, –14 to 39.07 mg/L/year for TH, –1.49 to 4.83 mg/L/year for Mg2+, –7.14 to 22.96 mg/L/year for Na+, –0.32 to 0.44 mg/L/year for Fe2+, –8.33 to 20.75 mg/L/year for HCO3ˉ, –26.52 to 31.01 mg/L/year for Clˉ and 1.29 to 3.76 mg/L/year for NO3ˉ over the study area. The results of M–K and ITA tests were found in agreement in all the blocks for both the aquifers. Groundwater contamination in both the aquifers can be attributed to weathering, geogenic processes, mineral dissolution, seawater intrusion, poor recharge pattern and injudicious anthropogenic activities. It is strongly recommended that concerned authorities urgently formulate efficient strategies for managing groundwater quality in the ‘leaky confined’ and ‘confined’ aquifers which serve as vital sources of drinking and irrigation water supplies in the study area.


  • Open Access

    Analysis of Trend in Groundwater-Quality Parameters: A Case Study

    by Subhankar Ghosh, Madan Kumar Jha

    In the 21st century, groundwater has a pivotal role in ensuring water, food, and environmental securities worldwide. Systematic observation, protection and restoration are essential for sustainable management of water resources. Regular monitoring is key to investigate temporal changes in groundwater quality, and statistical trend tests define whether these changes are significant or not. This study focuses on investigating trend in seasonal groundwater quality in an alluvial coastal basin of West Bengal, India. The seasonal groundwater-quality data (pH, TH, TDS, Fe2+ and HCO3ˉ) of pre-monsoon and post-monsoon seasons were collected for 2011–2018 period and analyzed using three non-parametric statistical trend detection tests, namely: (i) Original Mann-Kendall (M-K) test, (ii) Modified Mann-Kendall (mM-K) test, and (iii) Spearman Rank Order Correlation (SROC) test. The trend magnitudes were estimated by using the Sen’s slope estimation test. Statistical analyses revealed that seasonal concentrations of all five groundwater-quality parameters have large spatial (block-wise) variation within the study area. The results of trend analyses indicated that seasonal TH and TDS concentrations mainly have significant decreasing trends (α = 5% or 1%), whereas seasonal HCO3ˉ and Fe2+ concentrations mostly show significant increasing trends (α = 5% or 1%) in different blocks. However, seasonal pH concentrations exhibited no trend. The mM-K test was found to be over-sensitive in finding trends than M-K and SROC tests. The SROC test was found to be less sensitive in detecting trends than M-K and mM-K tests. Trend magnitudes of seasonal pH, TH, TDS, HCO3ˉ and Fe2+ concentrations varied from –0.03/year to 0.23/year, –57.44 mg/L/year to 25.88 mg/L/year, –172.98 mg/L/year to 92.58 mg/L/year, –15.81 mg/L/year to 27.88 mg/L/year, and –0.05 mg/L/year to 0.61 mg/L/year, respectively. Continuous and proper groundwater-quality monitoring is critically required in all aquifer systems. The outcomes of this study will aid policy-makers in appropriately monitoring and managing groundwater quality.


  • Assessment of Agricultural Relevance on Groundwater Indicator in a Command Area of Eastern India

    by Laishram Kanta Singh, Srikanta Sannigrahi, Ingudam Bhupenchandra, Anup Das, Subhankar Ghosh, Samiron Dutta, Ratnakar Swain, Ranjeet Kumar Jha, Mohan Lal