Large-scale spatial variability of eight soil chemical properties within paddy fields

• Published in: Catena (Giessen) 2020-05, Vol.188, p.104350, Article 104350
• Main Authors: Duan, Liangxia, Li, Zhenwei, Xie, Hongxia, Li, Zhiming, Zhang, Liang, Zhou, Qing
• Format: Article
• Language: English
• Subjects: Geostatistics; Interpolation; Rice fields; Soil chemical properties; Spatial variation
• ISSN: 0341-8162; 1872-6887
• DOI: 10.1016/j.catena.2019.104350

•8890 sites were sampled to identify spatial variability of eight soil properties.•High levels of SOC, TN, and AN were found in paddy fields at a large scale.•Distribution maps of eight soil chemical properties were produced.•Ordinary kriging method performed better than the inverse distance weighting method. Rice is one of the most important crops in the world. It provides food for 40% of the world’s population and 60% of China’s population. Soil nutrients have a significant impact on both agriculture and the environment, particularly with regards to soil fertility, soil quality, and rice production. However, minimal research has been conducted to address spatial patterns of soil nutrients at a large scale within paddy fields. This information is crucial for improving not only soil nutrient management, but also rice yields. Soil surface samples (0–20 cm of plough depth, totalling 8890) were collected from paddy fields to determine the spatial variability of eight soil chemical properties (soil organic carbon (SOC), total nitrogen (TN), available N (AN), total phosphorus (TP), available P (AP), total potassium (TK), rapidly available K (RAK), and slowly available K (SAK)). Inverse distance weighting (IDW) and ordinary kriging methods were applied to produce a continuous soil nutrient surface. Results indicated that soil chemical properties vary substantially. Paddy fields were characterized by high mean concentrations of SOC (19.7 g kg−1), TN (1.91 g kg−1), and AN (164.7 mg kg−1), implying that no additional C and N fertilizer was needed in regions with high SOC and N. SAK and AP demonstrated moderate (36.9%) and weak (96.2%) spatial dependence, respectively, while strong spatial dependence (10.1–13.5%) was observed across the remaining six soil chemical properties. Regional distribution maps of soil chemical properties were produced and ordinary kriging methods interpolated more accurately than the IDW method.