Changes in soil phosphorus, perennials versus annuals
Publication: Agriculture, Ecosystems and Environment 184 (2014) 168–181
To achieve robust phosphorus sustainability in agriculture, it will be necessary to reduce unintentional P losses through erosion and leaching, improve P availability and synchrony, and cycle the P that is intentionally exported from the farm in harvests out of the “waste” stream and back to the farm. This study suggests that the development of perennial grain crop agroecosystems shows encouraging promise for advancing the first two of these three conditions.
We compared inorganic and organic P fractions to a soil depth of 92 cm in two long-term Classical Experiments at Rothamsted Research in the U.K. The predominant soil-forming factor that differentiated the sites was vegetation type. The Broadbalk plots feature annual wheat and have been in continuous production since 1843, while the Park Grass plots feature perennial grassland vegetation that has been hayed every year since 1856. To evaluate the long-term effects of annual versus perennial vegetation on soil P forms, we carried out Hedley P fractionations and microbial biomass-P fumigation-extraction analyses on soils from fertilized and unfertilized treatments of both experiments. In both P-fertilized and unfertilized soils we found an inverse relationship between pool sizes of actively cycling Po (0.5 M bicarbonate + 0.1 M NaOH fractions) and recalcitrant Pi (hot conc. HCl + final digest fractions) with Po dominant in the perennial hay meadow and recalcitrant Pi dominant in the annual wheat. Microbial biomass-P in the surface horizons of fertilized and unfertilized perennial hay meadow was an order of magnitude greater than in annual wheat. To investigate how P fractions changed through time we conducted Hedley P fractionations on archived soils sampled from Broadbalk wheat in 1893, and Park Grass hay meadow in 1876. Since 1893, unfertilized Broadbalk soils experienced almost no change in P fractions in the surface 23 cm, but substantial depletion in labile and recalcitrant Pi and Po in deeper strata. The Park Grass perennial vegetation showed greater depletion of surface soil fractions over time. When fertilized for over 100 years, almost all P fractions in the surface 23 cm were enriched in both crop types, but below 70 cm, only the active Po pool in Park Grass showed a substantial increase under fertilization. Even when fertilized, low available or occluded Pi fractions in both annual and perennial systems were substantially depleted below 70 cm. Our findings suggest that herbaceous perennials maintain a greater proportion of native or fertilizer-P in relatively available organic forms compared to annual wheat. By reducing the fraction of P held in recalcitrant forms, P fertilizer requirements could be reduced.