PurposeKarst regions in Southwest China are confronted with severe ecological challenges due to rocky desertification and intensive agricultural practices, which exacerbate soil degradation and threaten the stability of the ecosystem. Therefore, it is necessary to evaluate quality and degradation of the soils in this region under different farmland-use types.MethodsThis study evaluated soil quality under different types of farmland use in a karst peak-cluster lowland area, including dragon fruit field (HF), cornfield (CF), sugarcane field (SF), and paddy field (PF), with forestland and grassland being selected as control. A total of 23 indicators, including physicochemical and microbial properties, were used for soil quality evaluation in this study. Soil quality indices (SQI) were constructed using factor analysis (FA) and network analysis (NA) to identify a minimum data set (MDS).ResultsThe MDS indicators through the FA method were silt, clay, total nitrogen (TN), total phosphorus (TP), fungal abundance-based coverage estimator (ACE) index (FA), bacterial ACE index (BA), and fungal Simpson index (FSP). Soil quality indicators identified through the NA approach were sand, bacterial Simpson index (BSP), and fungal observed species (FOP). Soil quality in karst region is significantly affected by agricultural management practices. HF exhibited a higher SQI, while SF had a lower SQI. This difference may be attributed to the conservation tillage practices implemented in HF, which improved soil quality, whereas the cultivation of the nutrient-intensive crop, sugarcane, in SF contributed to exacerbated soil degradation. The SQI derived from NA exhibited higher max/min ratios (1.49) compared to that from FA (1.29). Therefore, NA proved to be a simpler and more stable tool than FA for computing the SQI, as it reduced data redundancy and more effectively differentiated changes in land use.ConclusionSustainable land management practices like conservation tillage are critical for improving soil quality and mitigating degradation in karst regions. Network analysis demonstrates strong potential as an effective approach for soil quality assessment, offering valuable insights for optimizing agricultural practices and conserving fragile karst ecosystems.