CSC. S. Sumathi

C. S. Sumathi

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


C. S. Sumathi has a total of 46 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Long-term nutrient management reshapes soil carbon fractions and biological fertility: Evidence from a 43-year semi-arid Vertisol experiment

    by S S Kaviya, S B Bhakiyathu, B K Sathiya, K Boomiraj, C S Sumathi, G Guru, V Sanjivkumar, M Manikandan, J V N S Prasad, V K Singh, et al.

    Improving soil carbon quality and biological fertility through long-term nutrient management is essential for sustaining productivity, reducing input dependence and enhancing climate resilience in rainfed Vertisols. This study evaluated the effects of nine treatments including control, inorganic fertilisation (50 % RDF, 100 % RDF, RDF + Zn), organic amendments [farmyard manure (FYM) (50 % NFYM, sole FYM), crop residues (CR) (50 % NCR)] and integrated nutrient management (INMFYM, INMCR) on soil carbon pools, oxidisable carbon fractions, carbon management indices (CMI) and enzyme activities in a permanent manurial experiment. Soil organic carbon (SOC) ranged from 5.66 (control) to 11.19 g kg-1 (INMFYM), both in the cotton phase. Inorganic treatments enhanced soil inorganic carbon (SIC), suggesting an inverse relationship with organic inputs. Sole FYM enriched the very labile fraction, whereas INM enhanced the labile and less labile fractions and non-labile carbon was highest under RDF + Zn. CMI ranged from 136.7–307.1 and was highest under sole FYM. The 50 % NFYM recorded higher permanganate oxidisable carbon (POXC) and microbial biomass carbon (MBC) than 100 % RDF despite lower SOC, highlighting the importance of substrate quality. Enzyme activities were consistently lower in the cotton phase. Structural equation modelling indicated that SOC was associated with dehydrogenase activity indirectly through MBC, which mediated 86.5 % of this effect, whereas POXC was the primary driver of β-glucosidase (BGA) and cellulase activities (CA). These findings demonstrate that biological fertility in dryland Vertisols is governed by active, biologically accessible carbon pools rather than total SOC and that sustained organic and integrated inputs are essential for enhancing carbon quality and microbial enzyme function.

  • Open Access

    Analysis of rainfall variability and spell dynamics in Coimbatore using copula function modelling

    by P Preethi, R Ravikumar, V Geethalakshmi, R Sivakumar, R Karthikeyan, C S Sumathi

    This study investigates the coupled dynamics of rainfall and evapotranspiration in Coimbatore, a major urban and agricultural hub in southern India, over 31 years (1994–2024). Aside from areas near the Noyyal River basin and episodic overflow from the Western Ghats, the region is predominantly dry, receiving approximately 665.5 mm of annual rainfall over about 45 rainy days and exhibiting a yearly average temperature of 25.4 °C. Advanced statistical techniques, including Gaussian, t, Frank, Roch-Alegre and BB5 copula-based models, are employed to analyse key hydroclimatic indicators: the conditional probability index (CPI), moisture availability index (MAI) and potential evapotranspiration (PET). These models facilitate forecasting of wet and dry spells and provide insights relevant to drought preparedness, sustainable water management and agricultural planning in monsoon-dependent regions. Rainfall patterns and drought risk are quantified using MATLAB, enabling discrimination between low- and high-rainfall trends that are critical for rainfed agriculture. Wet and dry spell dynamics are further evaluated using the Mann-Kendall and Spearman’s Rho non-parametric tests at the 5 % significance level, supporting the identification of atmospheric demand patterns and characteristic monsoon periods. The results reveal substantial variability in hydrological regimes, characterized by abrupt transitions between extreme wet and dry spells and highlight years of pronounced moisture deficiency alongside periods of excessive moisture. This comprehensive framework, which employs the Gaussian copula function, enhances understanding of climate variability and offers actionable strategies for managing hydroclimatic risks in vulnerable regions.

  • Open Access

    Weather-driven groundnut price forecasting and profitability assessment of cropping patterns in Tamil Nadu using boosting algorithms

    by Kalpana Muthuswamy, Shrishail Dolli, Kedar Khandeparkar, Chandre Gowda, Venkatesa Palanichamy Narasimma Bharathi, K. M. Shivakumar, C. S. Sumathi, Suresh Appavu, Balakrishnan Natarajan, Krupesh Sivakumar

    This research explores the integration of weather-based forecasting and profitability analysis to optimize groundnut-based cropping patterns in Tamil Nadu. Groundnut, a crucial oilseed crop, is significantly influenced by weather variability, which impacts its price and profitability. The study leverages advanced boosting algorithms, including Light Gradient Boost, XGBoost, HistGradientBoosting, and CatBoost, to forecast groundnut prices using a multivariate approach that incorporates weather parameters like Minimum and Maximum Temperature, Relative Humidity and Rainfall. Weather parameters were discretized using \(\:k\)-means clustering. Crop prices were decomposed using Seasonal-Trend decomposition based on LOESS and each component was forecasted separately. HistGradientBoosting consistently outperforms other models, achieving the lowest multivariate Mean Absolute Error (MAE) across most crops and districts, underscoring its capability to handle complex, high-dimensional data effectively. The results reveal a substantial improvement in forecasting accuracy with multivariate models compared to univariate ones, establishing the importance of integrating weather features. Groundnut-related cropping patterns were analyzed for profitability using forecasted prices, with patterns involving high-value crops, such as onion in Namakkal, achieving the highest benefit-cost ratio (BCR) of 2.18. Patterns involving black gram also consistently outperformed green gram in economic efficiency. The findings emphasize the need for region-specific, weather-informed cropping strategies to maximize returns for farmers while mitigating risks.


  • Open Access

    Insecticidal longevity and resistance trends in cauliflower pests: A survival analysis approach

    by M Naveenasarika, Kannan Balaji, S Vishnu Shankar, M Kannan, C S Sumathi, M Vijayabhama, Bakthavachalam Gajendra

    Survival analysis is widely used to evaluate the effectiveness of insecticide trials on insect survival, which can also be used to estimate the durability of insecticides. This study focuses on the major pests of cauliflower that are capable of causing significant damage. Depending on the pest species and their population density, these infestations can lead to yield losses of up to 100 %. Major pest species in cauliflower are generally assumed to develop resistance more rapidly than minor pests. However, few studies have systematically analyzed published resistance data to compare resistance development among different species. Using 412 records from the Arthropod Pesticide Resistance Database covering 16 species, this study applied survival analysis to estimate the number of generations required for resistance to emerge following insecticide introduction. The results revealed significant variation among species in resistance development rates. On average, resistance first appeared after 178 generations in tropical regions and 56.5 generations in temperate regions. Insecticide durability also varied by Mode of Action (MoA) and year of introduction. On average, insecticides remained effective for 184.6 generations in tropical regions and 54.73 generations in temperate regions. For Diamondback moth control, estimated longevity in tropical regions was 7 years for Diamides, 8.5 years for Spinosyns and 12.9 years for Milbemycins. In temperate regions, effectiveness was estimated at 7 years for Diamides, 9.75 years for Spinosyns and 18.75 years for Milbemycins. Unlike traditional methods that depend on periodic field surveys or lab tests, survival analysis uses time-based data, including censored information, to give more reliable and consistent estimates of how quickly resistance develops and how long insecticides remain effective across different pests and modes of action.

  • Open Access

    Predictive analysis, data visualization and artificial intelligence tools for agricultural research and communication

    by M Prahadeeswaran, S Senthilnathan, C S Sumathi, S D Sivakumar, D Suresh Kumar

    Data Science offers powerful tools that predict and promise to enhance crop yields, optimize the farm resource utilization. Artificial Intelligence (AI) can perform tasks that typically require human intelligence, which encompasses a wide range of applications in agriculture starting from seed selection to robotic harvest. Synergy between data science and AI will grow strong to drive technological advancements and shape the future of agriculture sector. This paper presents application of few data science and AI tools in agriculture. Trends, patterns and variations in cost, benefit and returns from crop like Paddy, Groundnut, Sugarcane, Cotton and Maize across different states from 2011 to 2020 were analysed through data visualization. An experimental attempt was also made through web scrapping to understand the United States consumer preference towards the brands (Spice Train, Tellicherry), quantity (11 and 14 oz) and price (15-25 USD package) of pepper. Predictive modeling had shown that the export of pepper may fluctuate over the years while export of basmati rice, cashew nuts and tea will gradually increase for the next few years. High-definition images of tomatoes of different shape, colour, size were used to train the algorithm and tested for identification and classification of tomatoes and images of FAW infested maize were trained and tested for deduction accuracy, which was 82.5 %. Prediction accuracy will increase when the algorithm is trained with large number of images. AI avatars are widely used in social communication for various purposes like short communications, storytelling and documentation, which will be also used effectively for agricultural research communication and learning purposes.

  • Open Access

    Analyzing customer behavior and retention strategies in the edible vegetable oil e-commerce industry

    by A Apsana, K Divya, M Malarkodi, M Kavitha, Cs Sumathi

    The research article explores consumer behaviour on e-commerce platform and proposes strategies to enhance customer retention within the edible vegetable oil industry. The growing adoption of e-commerce presents a crucial opportunity to understand purchasing patterns and consumer preferences compared to traditional retail models. The study is based on data collected from 200 respondents, providing valuable insights into customer behaviour regarding edible oils like soybean oil, palm oil, and sunflower oil, which dominate both the Indian and global markets. The research examines factors influencing customer loyalty, including brand awareness, online promotions, pricing, and product quality. The findings are expected to offer significant insights for companies seeking to optimize operations in the edible oil industry while refining e-commerce strategies. Additionally, the study aims to support the development of long-term customer relationships in a highly competitive market.


  • Open Access

    Assessing the carbon footprint of marginal and smallholders farming systems: A typology driven approach

    by N Arivukkumar, Pm Shanmugam, B Kannan, Cs Sumathi, M Prahadeeshwaran, Sp Sangeetha, G Raghavi

    Agriculture is a significant contributor to greenhouse gas (GHG) emissions, with practices such as fertilizer application, soil tillage, and livestock management releasing carbon dioxide (CO2), nitrous oxide (N2O), and methane (CH4). To meet the demands of a growing population, it is essential to identify agricultural practices that boost food production while simultaneously reducing greenhouse gas emissions, contributing to climate change mitigation and adaptation. This study aimed to quantify GHG emissions from various farm typologies of marginal and smallholder households in the western regions of Tamil Nadu, India, using the Cool Farm Tool (CFT). Data was collected from 250 households in Coimbatore, Tiruppur, and Erode districts during 2022-2023 and farm typologies were identified through multivariate analysis, revealing four farm types: (i) cereal crop-dominated marginal farms, (ii) livestock-dominated marginal farms, (iii) cash crop-dominated marginal farms, and (iv) plantation crop-dominated small farms. The results show that cash crop and plantation crop dominated farms had the highest emissions, particularly from crop residue burning, fertilizer production and fertilizer application. In contrast, cereal crop-dominated farms recorded lower emissions. Livestock dominated farm type exhibited higher GHG emission from enteric fermentation and manure management due to higher number of livestock. The results of this study highlight the importance of developing farm-specific mitigation strategies to reduce emissions. The study also underscores the value of using tools such as CFT for comprehensive GHG quantification to ensure sustainable agricultural practices.


  • Open Access

    Adaptive routing in agricultural supply chains: Harnessing Q-learning for optimal decision-making in dynamic environments

    by Mantaw Suliyaa Chow, M Prahadeeswaran, V Karthick, Cs Sumathi, Sg Patil

    In this study, the authors try to emphasize how Q-learning, a model-free reinforcement learning (RL) technique can be used for optimizing routing in a grid-based environment. This study aims to assess the efficacy of Q-learning in enhancing routing for agricultural supply chains, investigate its flexibility in dynamic environments, and compare its performance across several real-world scenarios. In this specific case of the banana chain, an agent is moving through various entities in the system - from local growers to small traders and warehouses. It models the routing problem as a Markov Decision Process (MDP) and the goal is to optimize cumulative reward. Several possible cases are simulated, e.g. the finding of an optimal route for a given visit sequence that optimizes charging time and non-drivable paths left over when unexpected blockages occur to avoid energy wear penalties as well as how to best save costs; These results demonstrate the adaptability and durability of Q-learning in dynamic environments to obtain near-optimal solutions across diverse settings. Indeed, the present study adds to a growing body of research on the application of RL in logistics and supply chain management, highlighting its potential to enhance decision-making in complex and variable environments. The findings suggest that Q-learning can effectively balance multiple objectives, such as minimizing distance, reducing costs, and avoiding high-wear areas, making it a valuable tool for optimizing routing in real-world supply chains. Future work will explore broader applications and other RL algorithms in similar contexts.


  • Stability of tracking wheel mobile robot with teleoperation fuzzy neural network control system

    by C. S. Sumathi, R. Ravi Kumar, V. Anandhi

    The stability of the Tracking Wheel Mobile Robot with Teleoperation System and Path Following Method is discussed in this study. The path is to be tracked by the host computer which is the master robot. The response from the robot is captured on camera. As the slave robot approaches the target position, the camera captures the response robot’s position and as well as moving trajectory. The host computer receives all of the images, enabling mobile robot deviation recoveries. The slave robot can use teleoperation to follow the sensor based on the decisions made by the master robot. The Lyapunov function in the Fuzzy Neural Network (FNN) control structure assures the system’s stability and satisfactory performance. It supports a mobile robot’s ability to adhere to a reference trajectory without deviating from it. Finally, the outcome of the simulation demonstrates that our controller is capable of tracking different environmental conditions and maintaining stability.