A comparative analysis of soil health index calculation methods and minimum dataset construction approaches: a case study from a 40-year tillage experiment

Accurate assessment of the health status of arable soils is crucial for the sustainable development of agriculture. This study aimed to compare the pros and cons of various procedures involved in soil health assessment, providing a foundation for the development of a scientific soil health evaluation system. This study was based on a long-term, 40-year experimental platform involving different tillage practices in Mollisol farmland, a total of 27 indicators, encompassing soil physical, chemical and microbial properties, were measured in 0–20- and 20–40-cm soil layers to constitute the total dataset. Two minimum dataset (MDS) construction methods, were employed to construct MDS using three indicator selection approaches. In addition, along with two scoring methods and two soil health index (SHI) calculations, were employed to evaluate soil health status under different tillage practices in agricultural fields. Results showed that compared with no-tillage (NT) and conventional tillage (CT), moldboard ploughing tillage (MP) and sub-soiling tillage (ST) improved SHI in the 0–40-cm soil layer by 4.81 % to 11.82 %, respectively. All MDS construction methods and indicator selection approaches identified soil organic carbon, dissolved organic carbon and soil compaction as the most important indicators for evaluating the impact of different tillage practices on SHI. A comparative analysis of different soil health assessment methods revealed that the SHI calculated using the network analysis (NA), soil ecosystem multifunctionality grouping MDS (EMDS) and non-linear (NL) methods exhibited higher average L fit R2 values of 0.63, 0.65 and 0.75, respectively. In addition, SHI calculated using NA, NL and area method (Area) demonstrated greater sensitivity index, with variation ranges of 1.30–2.92, 1.20–2.92 and 1.24–2.92, respectively. Therefore, for MDS construction, indicator selection, indicator scoring and SHI calculation in soil health assessment, it is recommended to employ NA, EMDS, NL and Area. These approaches are more effective for evaluating the impact of agricultural management practices on soil health.