Agronomic phosphorus imbalances across the world's croplands

Abstract

Increased phosphorus (P) fertilizer use and livestock production has fundamentally altered the global P cycle. We calculated spatially explicit P balances for cropland soils at 0.5° resolution based on the principal agronomic P inputs and outputs associated with production of 123 crops globally for the year 2000. Although agronomic inputs of P fertilizer (14.2 Tg of P·y(-1)) and manure (9.6 Tg of P·y(-1)) collectively exceeded P removal by harvested crops (12.3 Tg of P·y(-1)) at the global scale, P deficits covered almost 30% of the global cropland area. There was massive variation in the magnitudes of these P imbalances across most regions, particularly Europe and South America. High P fertilizer application relative to crop P use resulted in a greater proportion of the intense P surpluses (>13 kg of P·ha(-1)·y(-1)) globally than manure P application. High P fertilizer application was also typically associated with areas of relatively low P-use efficiency. Although manure was an important driver of P surpluses in some locations with high livestock densities, P deficits were common in areas producing forage crops used as livestock feed. Resolving agronomic P imbalances may be possible with more efficient use of P fertilizers and more effective recycling of manure P. Such reforms are needed to increase global agricultural productivity while maintaining or improving freshwater quality.

Fig. 1.

Global map of agronomic P imbalances for the year 2000 expressed per unit of cropland area in each 0.5° grid cell. The P surpluses and deficits are each classified according to quartiles globally (0–25th, 25–50th, 50–75th, and 75–100th percentiles).

Fig. 2.

Distributions of P surpluses and deficits by quartiles shown as percent of total cropland area in each continent and as percent of global cropland area.

Fig. 3.

Cumulative distributions of global cropland P imbalances (surpluses or deficits, sorted from largest to smallest) in relation to cumulative global cropland area.

Fig. 4.

Agronomic drivers of P surpluses based on the magnitude of fertilizer or manure P applied relative to crop P use in different locations (A) and summarized according to percent of cropland area by continent and globally (B). Each category is mutually exclusive based on locations where either fertilizer alone or manure alone exceeded crop P use, where fertilizer and manure each individually exceeded crop P use, or where only the sum of fertilizer and manure exceeded crop P use.

Fig. 5.

Map of total agronomic PUE (kg of crop·kg of P·input⁻¹) for 123 crops in 2000 classified from low to high based on quartiles globally. Ovals indicate examples of contiguous areas with model P management (relatively balanced P situations, with PUE and crop production each above the global median; based on Fig. S5B).