Characterizing the fine-scale spatial distribution of soil phosphorus for efficient phosphorus management in an...

Closed depressions in post-glacial landscapes can accumulate phosphorus (P) due to repeated flooding and become hotspots for P loss when underlain by subsurface (tile) drainage. Soil P mapping is routinely based on the interpolation of samples from a 1-ha grid, which may miss closed depressions and underestimate soil P levels leading to overfertilization and nutrient loss. Our objective was to improve the characterization of the spatial distribution of soil P at the sub-field scale by accounting for depressions and assess their importance for fertilizer prescriptions and tile P loss. We evaluated the effectiveness of stratified sampling that included closed depressions within a 1-ha grid and nonstationary interpolation (external drift kriging) that leverages information about depression depth to estimate the distribution of soil P (0-16 cm) under a corn (Zea mays L.) and soybean (Glycine max (L.) Merr) rotation in Douglas County, IL. Our novel approach produced an improved soil P map, which resulted in a 47% increase in land area that does not require P fertilizer (a reduction of 7.14% or ca. 4 metric tons of P). Additionally, soil P estimated from the improved map was a stronger predictor of the flow-weighted mean concentration of dissolved reactive P during the non-growing season than soil P estimated from ground-based sampling and other interpolation approaches. These results demonstrate that improved characterization of the spatial distribution of soil P through stratified sampling and interpolation with depression depth can better match soil P with crop P requirements, protecting water quality and conserving a finite resource.