Quantitative relationships between soil test phosphorus (STP) methods are needed to guide P management especially in manured soils with high P. Our objectives were: (i) to compare amounts of P extracted by different methods; (ii) to develop and verify regression equations to convert results among methods; and (iii) to establish environmental P thresholds for different methods, in manured and non-manured soils of Manitoba. We analyzed 214 surface soil samples (0–15 cm), of which 51 had previous manure application. Agronomic STP methods were Olsen (O-P), Mehlich-3 (M3-P), Kelowna-1 (original; K1-P), Kelowna-2 (modified; K2-P), Kelowna-3 (modified; K3-P), Bray-1 (B1-P) and Miller and Axley (MA-P), while environmental STP methods were water extractable (W-P), Ca Cl2 extractable (Ca-P) and iron oxide impregnated filter paper (FeO-P) methods. The different methods extracted different amounts of P, but were linearly correlated. For an O-P range of 0–30 mg kg-1, relationships between O-P and other STP were similar for manured and nonmanured soils, but the relationships diverged at higher O-P levels, indicating that one STP cannot be reliably converted to another using a single equation for manured and non-manured soils at environmentally critical P levels (0–100 mg kg-1 O-P). Suggested environmental soil P threshold ranges, in mg P kg-1, were 88–118 for O-P, 138–184 for K1-P, 108–143 for K2-P, 103–137 for K3-P, 96–128 for B1-P, 84–111 for MA-P, 15–20 for W-P, 5–8 for Ca-P and 85–111 for FeO-P. Key words: Phosphorus, soil test phosphorus, manured soils, non-manured soils, environmental threshold
