A Multivariate Adaptive Regression Splines Based Study of Soil Parameters and Their Impact on Onion Yield in Bhavnagar District

Notice

This is an unedited manuscript accepted for publication and provided as an Article in Press for early access at the author’s request. The article will undergo copyediting, typesetting, and galley proof review before final publication. Please be aware that errors may be identified during production that could affect the content. All legal disclaimers of the journal apply.

Year : 2026 | Volume : 15 | 01 | Page :
    By

    Arjun Nayak,

  • K.S. Tailor,

  1. Research scholar, Department of Statistics, M. K. Bhavnagar University, Bhavnagar- 364002, Gujarat, India
  2. Assistant Professor, Department of Statistics, M. K. Bhavnagar University, Bhavnagar- 364002, Gujarat, India

Abstract

Bhavnagar district is one of the prominent onion-growing areas in the Saurashtra region of Gujarat, encompassing key talukas such as Mahuva, Talaja, Ghogha, Jesar, and Palitana. Onion cultivation in the district is carried out across three distinct seasons: rabi, kharif, and late kharif with harvesting periods extending from April to May for the rabi crop and from October to March for the kharif and late kharif crops. The productivity of onion crops is strongly influenced by soil characteristics, particularly pH balance and nutrient availability. This study examines the effects of soil pH and NPK (Nitrogen, Phosphorus, and Potassium) levels on onion yield in the Bhavnagar district of Gujarat, a major onion-producing region. A multivariate adaptive regression splines (MARS) modeling approach was employed to capture nonlinear relationships and complex interactions between soil parameters and crop yield. The MARS model effectively identified key spline functions representing critical thresholds in pH and nutrient levels that significantly impact yield. The findings provide valuable insights for site-specific soil management and precision agriculture practices, enabling farmers to optimize fertilizer application and soil conditioning strategies. This data-driven approach supports sustainable agricultural planning, improved resource-use efficiency, and enhanced crop productivity, offering a practical decision-support framework for policymakers, agronomists, and extension services involved in onion cultivation in semi-arid regions.

Keywords: Statistical Analysis, Soil Parameters, Onion crop, Regression, MARS model

How to cite this article:
Arjun Nayak, K.S. Tailor. A Multivariate Adaptive Regression Splines Based Study of Soil Parameters and Their Impact on Onion Yield in Bhavnagar District. Research & Reviews : Journal of Agricultural Science and Technology. 2026; 15(01):-.
How to cite this URL:
Arjun Nayak, K.S. Tailor. A Multivariate Adaptive Regression Splines Based Study of Soil Parameters and Their Impact on Onion Yield in Bhavnagar District. Research & Reviews : Journal of Agricultural Science and Technology. 2026; 15(01):-. Available from: https://journals.stmjournals.com/rrjoast/article=2026/view=242187


References

  1. 1. Singh GP, Meena ML, Pankaj TR. Effect of different levels of nitrogen, phosphorus and potassium on growth and bulb yield of onion. Pharma Innov J. 2021;10(10):1504–1507.
    2. Kabir MR. Effect of different sources of nitrogen on growth and yield of tomato (Solanum lycopersicum L.) [doctoral dissertation]. Dhaka: Sher-e-Bangla Agricultural University; 2021.
    3. Garai S, Paul RK, Yeasin M, Paul AK. CEEMDAN-based hybrid machine learning models for time series forecasting using MARS algorithm and PSO-optimization. Neural Process Lett. 2024;56(2):92.
    4. Mirani A, Memon MS, Chohan R, Wagan AA, Qabulio M. Machine learning in agriculture: A review. LUME. 2021;10:5.
    5. Hastie T, Tibshirani R, Friedman J. The elements of statistical learning: data mining, inference and prediction. New York: Springer-Verlag; 2001.
    6. Celik S, Akın M, Aliyev P, Eyduran SP, Eyduran E. A hybrid approach of combining factor analysis scores with MARS predictive model for regression problems. In: Proceedings of the 1st International Erciyes Agriculture, Animal & Food Science Conference (AGANFOS); 2019 Apr 24–27; Kayseri, Turkey.
    7. Friedman JH. Multivariate adaptive regression splines. Ann Stat. 1991;19(1):1–67.
    8. Sephton P. Forecasting recessions: can we do better on MARS™? Fed Reserve Bank St Louis Rev. 2001;Mar/Apr:39–49.
    9. Kornacki J, Ćwik J. Statistical learning systems. Warsaw: WNT; 2005.
    10. Willmott C, Matsuura K. Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance. Clim Res. 2005;30:79–82.
    11. Liddle AD, Davies AH. Pelvic congestion syndrome: chronic pelvic pain caused by ovarian and internal iliac varices. Phlebology. 2007;22(3):100–104.
    12. Takma C, Atil H, Aksakal V. Comparison of multiple linear regression and artificial neural network models goodness of fit to lactation milk yields. Kafkas Univ Vet Fak Derg. 2012;18:941–944.
    13. Eyduran E, Akin M, Eyduran SP. Application of multivariate adaptive regression splines in agricultural sciences through R software. Ankara: Nobel Bilimsel Eserler; 2019. ISBN: 978-605-2149-81-82.
    14. Milborrow S. Derived from mda:mars by T. Hastie and R. Tibshirani. earth: Multivariate adaptive regression splines. R package; 2011.
    15. Milborrow S. Notes on the earth package. 2018. Available from: http://www.milbo.org/doc/earth-notes.pdf
    16. R Core Team. R: A language and environment for statistical computing. Vienna: R Foundation for Statistical Computing; 2014. Available from: http://www.R-project.org
Ahead of Print Subscription Original Research
Volume 15
01
Received 06/01/2026
Accepted 24/01/2026
Published 30/04/2026
Publication Time 114 Days
1

Login


My IP

PlumX Metrics