Responses of soil enzyme activities to increasing salinity: A quantitative meta-analysis

Soil salinization is a critical degradation process threatening global agricultural sustainability and ecosystem functioning. Soil extracellular enzymes, as key agents of microbial metabolism, play a central role in driving soil biogeochemical cycles; however, a large-scale and cross-regional synthesis understanding of their response patterns to increasing salinity and the underlying drivers remains limited. This study synthesized 1379 observations from 47 publications using meta-analysis to quantitatively assess the overall response of soil enzyme activities to increasing salinity and to explore the moderating effects of enzyme functional group, ecosystem type, plant growth form, and climatic conditions. The results revealed that increased salinity significantly suppressed overall soil enzyme activity by 9.1 %. However, responses varied dramatically among functional enzymes: P-cycling (-26.2 %) and S-cycling (-26.3 %) enzymes were the most sensitive, whereas no significant changes were detected at the group level for C- and N-cycling enzymes as well as oxidases, although individual enzymes within these groups exhibited contrasting responses. Ecosystem type was a critical moderator, with forest ecosystems exhibiting unique resistance and a significant overall stimulation of enzyme activity (+69.3 %), while grasslands (-49.3 %) and wetlands (-23.3 %) were more vulnerable. Woody plants alleviated the suppression of, or even stimulated, C- and N-cycling enzyme activities, but exacerbated the inhibition of P-, S-cycling enzymes and oxidases compared to the widespread sensitivity observed under herbaceous plants. Increases in mean annual precipitation and mean annual temperature significantly intensified the suppressive effects of salinity on most enzyme activities. Furthermore, sensitivity analysis revealed that the magnitude of enzymatic responses depended critically on the salinity metric used (ECe vs. soil salt content), a pattern intrinsically linked to the ecosystem type predominantly associated with each metric. Our findings underscore the context-dependent nature of salinity effects, highlight the importance of methodological consistency in cross-study comparisons, and provide a critical theoretical basis for the ecological function assessment and adaptive management of salt-affected soils.