Cluster Analysis for Sustainable Soybean Production in India

Abstract

Clustering spatial growth and dynamics is crucial for understanding output and concentration patterns for optimizing resources and promoting sustainable development. India, with Madhya Pradesh (MP) as the leading soybean state, exhibits significant growth and instability patterns. Thus, we explored the spatiotemporal patterns of soybean growth and instability since 1964. Nationally, the pooled growth of soybean is positive, accompanied by an unstable area and production, along with medium yield stability. However, despite stable and positive decadal growth, yield growth has recently declined to negative. In MP as well, these variables grew positively and outpaced national growth; however, they have declined to negative in the last decade. Stability has increased to a stable state but is currently at a medium level. The clustering of MP districts identified 5, 5 and 4 optimum clusters using the \'Elbow\' method, with high precision in the growth of these variables. The clustered maps highlighted the Malwa Plateau, Jhabua Hills, Nimar Valley, Vindhiyan Plateau, Gird Region, Northern Hills, and Harda and Betul districts as potential high-growth regions, whereas the Kymore Plateau stands out as a low-growth region. Supervised clustering revealed a more stable soybean area in western MP over the last decade, whereas production stability generally remained low throughout the state. Therefore, developing climate-resilient technologies, improving extension services and formulating clusterwise policies should be prioritized to boost sustainable soybean production in India.

Introduction

Sustainable agriculture requires not only increased production but also reduced instability caused by biotic and abiotic stresses, which threaten food security—particularly critical in India, where 65% of the population depends on agriculture. Soybean cultivation in India, especially in Madhya Pradesh (MP), has expanded rapidly over the past five decades due to low cultivation costs. MP, known as the 'Soya State,' contributes over 42% of India's soybean production, primarily under rainfed conditions.

Despite substantial area and production growth (area increased nearly 400 times since 1970), soybean yields in India and MP remain low (~1 to 1.3 tons/ha), significantly below the global average (~2.75 tons/ha) and potential yields (over 3 tons/ha achievable with improved technology). Yield instability persists due to climatic variability, poor management practices, and low adoption of advanced technologies.

This study analyzes district-level growth and instability in soybean production in MP using compound annual growth rates (CAGR), instability indices (notably the Cuddy-Della-Valle Index), and K-means clustering to identify spatial patterns. Results indicate continuous area and production growth but fluctuating and unstable yields, with notable setbacks linked to adverse weather and events like COVID-19. The study underscores the need for targeted, cluster-specific policies to optimize resource use, improve yield stability, and promote sustainable soybean development in MP and India.

Conclusion

This study explored soybean growth and dynamics in India from 1964-2023. Nationally, soybean area and production have increased at CAGRs of 11.15% and 12.27%, respectively, since 1970. However, yields stagnated around 1 - 1.3 tons per hectare. Decadal growth is positive in all periods except for yield in Period V; with initially greater growth, it has gradually declined to its lowest value in the last decade. The APY became highly stable over the last two decades, with the area becoming stable faster than production and productivity. However, in the pooled period, area and production are highly unstable, whereas yield has medium stability due to greater fluctuations. In Madhya Pradesh, soybean area, production and productivity are increasing at rates of 14.29%, 16%, and 1.5% CAGR, respectively, outperforming national growth; however, productivity has plateaued at approximately 1 ton per hectare. Except for the PC and last decade periods, growth in area and production was significantly positive but steadily decreasing, turning negative in the last decade, whereas yield growth was negative only during the first and last decade periods. The stability of all three variables gradually increased and became highly stable in period IV but rose again. The area is becoming stable more rapidly than production and yield. The pooled (VI) period shows a highly unstable area and production but medium stability for yield. Unsupervised clustering of Madhya Pradesh districts identified optimal clusters for area, production, and yield growth using the K-Means algorithm with high precision. The spatiotemporal clustered map highlighted the Malwa Plateau, Nimar Valley, Jhabua Hills, Vindhiyan Plateau, Northern Hills, Gird Region, and Harda and Betul districts as potential high-growth regions and the Kymore Plateau as a low-growth region. The yield growth trend reversed, with higher yields in western MP during the pooled period than during the last decade. Supervised clustering of instability indicated that the soybean area is more stable in western MP during period A, whereas production generally has low stability throughout the state. However, during period B, both area and production were unstable and volatile, with no clear pattern of yield stability in either period. These findings, along with an understanding of the spatiotemporal growth and dynamics of soybean, lead to the formulation of cluster-specific policies for rapidly adopting new technologies and extension activities in low-contributing regions with the aim of improving productivity, ensuring food security, and providing sustainable solutions for reducing yield gaps, such as developing climate-resilient technologies, enhancing extension services, and expanding soybean cultivation in other seasons to boost productivity.

References

[1] Ayele A, Worku M, Bekele Y. Trend, instability and decomposition analysis of coffee production in Ethiopia (1993-2019). Heliyon. 2021; 20:7(9):e08022. doi:10.1016/j.heliyon.2021.e08022. PMID:34589632; PMCID: PMC8461354. [2] Chattopadhyay M, Mitra SK. Comparative decision models for anticipating shortage of food grain production in India. Theoretical and Applied Climatology. 2018; 131: 523–530. https://doi.org/10.1007/s00704-016-1961-0 [3] Patel RM, Sharma AN, Sharma P. GIS based pest-weather model to predict the incidence of Girdle beetle (Oberiopsis brevis) in Soybean crop. Journal of Agrometeorology. 2021; 23(2): 183-188. [4] Sharma P, Patel RM, Dupare BU. Profitability of Soybean vis-a-vis Major Kharif Crops in Madhya Pradesh. Soybean Research. 2015; 13(2): 65-77. [5] Patel RM, Sharma P, Bunkar K, Ramesh A, Khandekar N. Decadal cropping pattern and expansion dynamics of Soybean in Madhya Pradesh, India. Soybean Research. 2023; 21(1): 110-123 (2023). [6] Patel RM, Sharma P, Sharma AN. Weather Based Prediction Models for Forewarning Tobacco Caterpillar, Spodoptera Litura (Fabricius) Larval Population in soybean. Journal of the Indian Society of Agril. Statistics. 2020; 74(2): 129–1 [7] Sharma P, Dupare BU, Patel RM. Soybean Supply Chain Management in India: A Review. Soy. Res.. 2023; 21(2):1-16 [8] FAO-AMIS. https://app.amis-outlook.org/#/market-database/custom-query [9] Nataraj V, Kumar S, Kumawat G, Shivakumar M, Rajput LS, Ratnaparkhe MB, Ramteke R, Gupta S, Satpute GK, Rajesh V, Kamble V, Chandra S. Charcoal Rot Resistance in Soybean: Current Understanding and Future Perspectives. In: Disease Resistance in Crop Plants. 2019; [Wani S. (eds)] Springer, Cham, pp: 241-259. [10] Tiwari A, Choudhary VK, Shrey R. Analysis on growth rate and decomposition analysis of major cereals grown in kharif season in northern hills of Chhattisgarh. International Journal of Research in Agronomy. 2024; 7(5S):187-193. DOI:10.33545/2618060X.2024.v7.i5Sc.767 [11] Mall RK, Sharma NK, Singh N. Impacts of Climate Change on Agriculture Sector in Madhya Pradesh. 2016. https://doi.org/10.13140/RG.2.1.3010.0247 [12] Dam A, Chatterjee S, Sarkar D, Kumar P. Trends in Production of Pulses and Oilseeds Over Last Two Decades in Eastern India. Economic Affairs. 2023; 68(04): 2017-2023. [13] Mishra S, Balai HK, Malviya PK. Rabi Pulse Crops Cultivation in Madhya Pradesh (India): Growth and Decomposition Analysis. Economic Affairs. 2023; 68(01): 391-395. [14] Rana MJ, Islam S, Kamruzzaman M. Growth and instability in area, production and productivity of major spices in Bangladesh. Journal of Agriculture and Food Research. 2021; 6: 100216 [15] Wang Z, Zhang E, Chen G. Spatiotemporal Variation and Influencing Factors of Grain Yield in Major Grain-Producing Counties: A Comparative Study of Two Provinces from China. Land. 2023; 12:1810. https://doi.org/10.3390/land1209181 [16] Coppock JD. International Economic Instability. McGraw-Hill, New York. 1962; 523–525 [17] Joshi P, Gautam P, Wagle P. Growth and instability analysis of major crops in Nepal. Journal of Agriculture and Food Research. 2021; 6: 10023 [18] Cuddy JDA, Della-Valle PA. Measuring the Instability of Time Series Data. Oxford Bulletin of Economics and Statistics. 1978; 40(1): 79–85 [19] Rani R, Sah U, Kumar H, Bishnoi R, Udgata AR. Multilevel growth assessment of mungbean production in Madhya Pradesh. Journal of Food Legumes. 2024; 37(2): 190-195. DOI: 10.59797/jfl.v37.i2.194 [20] Ikuemonisan ES, Mafimisebi TE, Ajibefun I, Adenegan K. Cassava production in Nigeria: trends, instability and decomposition analysis (1970–2018). Heliyon. 2020; 6: e05089 [21] Dhungana SM, Regmi PP, Dhakal SC, Devkota N R. Growth instability index, and decomposition effect on production of large cardamom in Nepal. Agriculture & Food Security. 2024; 13:5. https://doi.org/10.1186/s40066-023-00456-y [22] Zhang Z, Lu C. Clustering Analysis of Soybean Production to Understand its Spatiotemporal Dynamics in the North China Plain. Sustainability. 2020; 12: 6178. doi:10.3390/su1215617 [23] Sinaga KP, Yang MS. Unsupervised K-Means Clustering Algorithm. IEEE Access. 2020; 8: 80716-80727. doi:10.1109/ACCESS.2020.2988796 [24] Li D, Liang J, Wang X, Wu S, Xie X, Lu J. Rice Yield Simulation and Planting Suitability Environment Pattern Recognition at a Fine Scale. ISPRS International Journal of Geo-Information. 2021; 10: 612. https://doi.org/10.3390/ijgi10090612 [25] Mamun MAAl, Nihad SAI, Sarkar MAR, Aziz MA, Qayum MA, Ahmed R, Rahman NMF, Hossain MI, Kabir MS. Growth and trend analysis of area, production and yield of rice: A scenario of rice security in Bangladesh. PLoS ONE. 2021; 16(12): e0261128. https://doi.org/10.1371/journal.pone.026112 [26] Ye S, Song C, Shen S, Gao P, Cheng C, Cheng F, Wan C, Zhu D. Spatial pattern of arable land-use intensity in China. Land Use Policy. 2020; 99: 104845. doi:10.1016/j.landusepol.2020.10484 [27] Abirami B, Selvi RG, Kannan B, Vanitha G, Thavaprakaash N. Application of K-means Clustering Algorithm in Rice Production of Tamil Nadu, India. International Journal of Environment and Climate Change. 2022; 12(11): 1348-1355. [28] Ji D, Tian J, Zhang J, Zeng J, Namaiti A. Identification and Spatiotemporal Evolution Analysis of the Urban–Rural Fringe in Polycentric Cities Based on K-Means Clustering and Multi-Source Data: A Case Study of Chengdu City. Land. 2024; 13: 1727. https://doi.org/10.3390/land13111727 [29] Dupare BU, Sharma P, Billore SD, Verma SK. Impact of Climate Change on Soybean Cultivation in Malwa and Nimar Region of Madhya Pradesh: Farmers’ Perspective. Soybean Research. 2020; 18: 87–97 [30] Prashnani M, Dupare B, Vadrevu KP, Justice C. Toward food security: Exploring the spatiotemporal dynamics of soybean in India. PLoS ONE. 2024; 19(5):e0292005. https://doi.org/10.1371/journal.pone.0292005 [31] Shrivastava A, Pandey SK, Kobayashi S. Influence of Excessive Moisture Stress on Genetic Parameters in Soybean. Soybean Research. 2014; (Special Issue -II): 33–43 [32] Ellanki RE, Reddy SB. Centre for Economic and Social Studies H. Economics and technology of soybean cultivation in Central India. 2016; Volume 43 of CESS monograph [33] Mamun MAAl, Nihad SAI, Sarker MR, Sarkar MAR, Hossain MI, Kabir MS. Spatiotemporal mapping of rice acreage and productivity growth in Bangladesh. PLoS ONE. 2024; 19(3): e0300648. https://doi.org/10.1371/journal.pone.0300648

Copyright

Copyright © 2025 Ram Manohar Patel, Kamal Bunkar, Chhaya Arya. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.