Bhagwan Pandit Vaibhav, Anjaiah Theerthala, Ravali Chitteti, Chary Darshanoju Srinivasa, Zamani Abu Taha, Ullah Sajid, Rebouh Nazih Y, Tariq Aqil
Department of Soil Science and Agricultural Chemistry, School of Agriculture, SR University, Warangal, 506371, Telangana, India.
Institute of Soil Health Management, Professor Jayashankar Telangana State Agricultural University, Hyderabad, 500030, Telangana, India.
Sci Rep. 2025 Jul 1;15(1):20991. doi: 10.1038/s41598-025-07283-0.
Expansive soil spatial variability plays a key role in the over- and under-application of fertilizers, contributing to environmental pollution. Assess soil variability and delineate it into management zones to adopt site-specific nutrient management for balanced fertilization and sustainable agriculture. To assess spatial variability by geostatistical methods and delineate and evaluate nutrient management zones for site-specific nutrient management and variable rate fertilizer application using fuzzy c-means clustering. Overall, 200 soil samples (0-15 cm depth) with geographical coordinates were collected with a grid size of 14.2 m × 14.2 m from a 4-ha maize cultivated 4-ha of Mahagoan village of Bhainsa Mandal, Nirmal district, Telangana, India. The collected samples were tested with different reagents to determine the soil reaction and available nutrient status. Soil spatial variability was assessed by the geostatistical method, and delineation of nutrient management zones was carried out by integrating principal component analysis and fuzzy c-means clustering. Geostatistical analysis revealed spherical (pH, electrical conductivity, organic carbon, available sulfur, and available Zn) and Gaussian (available nitrogen, available PO, available KO, available Fe, available Zn, and available Cu) as the best-fit semivariogram model with strong spatial dependence. Five management zones were delineated by principal component analysis and fuzzy c-means clustering based on fuzzy performance index (FPI) and normalized classification entropy (NCE) indices. Variable rates of fertilizer recommendations in different management zones were calculated using a soil test crop response equation. The results show the highest grain yield and fertilizer saving in MZ, followed by MZ, MZ, MZ, and MZ, compared to farmer fertilizer practices. The study aims to delineate the management zone to reduce fertilizer application, ensure balanced fertilizer application, minimize environmental pollution, and increase crop grain yield and profitability.
膨胀土的空间变异性在肥料的过量施用和施用不足方面起着关键作用，从而导致环境污染。评估土壤变异性并将其划分为管理区，以采用因地制宜的养分管理实现平衡施肥和可持续农业。通过地统计学方法评估空间变异性，并使用模糊 c 均值聚类法划分和评估养分管理区，以实现因地制宜的养分管理和变量施肥。总体而言，在印度特伦甘纳邦尼尔马尔区拜恩萨曼达尔马哈戈安村一块4公顷的玉米种植地上，以14.2米×14.2米的网格大小采集了200个具有地理坐标的土壤样本（深度0至15厘米）。对采集的样本使用不同试剂进行检测，以确定土壤反应和有效养分状况。通过地统计学方法评估土壤空间变异性，并通过主成分分析和模糊 c 均值聚类法进行养分管理区的划分。地统计学分析表明，球形模型（pH值、电导率、有机碳、有效硫和有效锌）和高斯模型（有效氮、有效磷、有效钾、有效铁、有效锌和有效铜）是具有强空间依赖性的最佳拟合半方差图模型。基于模糊性能指标（FPI）和归一化分类熵（NCE）指数，通过主成分分析和模糊 c 均值聚类法划分出五个管理区。使用土壤测试作物响应方程计算不同管理区的变量施肥建议率。结果表明，与农民的施肥方式相比，MZ区的谷物产量最高且肥料节省，其次是MZ区、MZ区、MZ区和MZ区。该研究旨在划分管理区，以减少肥料施用量，确保平衡施肥，最大限度减少环境污染，并提高作物谷物产量和盈利能力。
