Fig. 4 Redundancy analysis (RDA) showing the relationships between enzyme activity and enzyme stoichiometry (blue arrows) with soil physicochemical properties and soil stoichiometry (red arrows) in different land use patterns. (a): The relationships in alfalfa artificial grassland (AG). RDA analysis explained 90.80% of the total variation in AG, RDA axis 1 and axis 2 explained 88.26% and 2.44% of the variation respectively. TP and NH4+-N had significant effects on biological factors (p < 0.01), and the explanatory degrees of biological factors were 75.4% and 7.4% respectively. (b): The relationships in wheat-maize rotation field (WM). RDA analysis explained 64.50% of the total variation in WM, RDA axis 1 and axis 2 explained 34.72% and 29.49% of the variation respectively. TP and WSA (2-1 mm) had significant effects on biological factors (p < 0.01), and the explanatory degrees of biological factors were 27.1% and 18.3% respectively. (c): The relationships in native grassland (PC). RDA analysis explained 93.80% of the total variation in PC, RDA axis 1 and axis 2 explained 86.53% and 6.91% of the variation respectively. NO3--N had significant effects on biological factors (p < 0.01), and the explanatory degree of biological factors was 56.00%. WSA: soil water stable aggregate. pH: hydrogen ion concentration. EC: soil electrical conductivity. TC: total carbon. TN: total nitrogen. TP: total phosphorus. SOC: soil organic carbon. NH4+-N: ammonium nitrogen. NO3--N: nitrate nitrogen. AP: soil available phosphorus. βG: β-1,4-glucosidase. NAG: β-1,4-N-acetyl-glucosaminidase. LAP: L-leucine aminopeptidase. AP: acid phosphatase. VL: vector length. VA: vector angle. C: carbon. N: nitrogen. P: phosphorus.