Towards optimal use of phosphorus fertiliser.

Because phosphorus (P) is one of the most limiting nutrients in agricultural systems, P fertilisation is essential to feed the world. However, declining P reserves demand far more effective use of this crucial resource. Here, we use meta-analysis to synthesize yield responses to P fertilisation in grasslands, the most common type of agricultural land, to identify under which conditions P fertilisation is most effective. Yield responses to P fertilisation were 40-100% higher in (a) tropical vs temperate regions; (b) grass/legume mixtures vs grass monocultures; and (c) soil pH of 5-6 vs other pHs. The agronomic efficiency of P fertilisation decreased for greater P application rates. Moreover, soils with low P availability reacted disproportionately strong to fertilisation. Hence, low fertiliser application rates to P-deficient soils result in stronger absolute yield benefits than high rates applied to soils with a higher P status. Overall, our results suggest that optimising P fertiliser use is key to sustainable intensification of agricultural systems.

Long-term nitrogen loading alleviates phosphorus limitation in terrestrial ecosystems.

Increased human-derived nitrogen (N) deposition to terrestrial ecosystems has resulted in widespread phosphorus (P) limitation of net primary productivity. However, it remains unclear if and how N-induced P limitation varies over time. Soil extracellular phosphatases catalyze the hydrolysis of P from soil organic matter, an important adaptive mechanism for ecosystems to cope with N-induced P limitation. Here we show, using a meta-analysis of 140 studies and 668 observations worldwide, that N stimulation of soil phosphatase activity diminishes over time. Whereas short-term N loading (≤5 years) significantly increased soil phosphatase activity by 28%, long-term N loading had no significant effect. Nitrogen loading did not affect soil available P and total P content in either short- or long-term studies. Together, these results suggest that N-induced P limitation in ecosystems is alleviated in the long-term through the initial stimulation of soil phosphatase activity, thereby securing P supply to support plant growth. Our results suggest that increases in terrestrial carbon uptake due to ongoing anthropogenic N loading may be greater than previously thought.

Vertical gradients of delta15N and delta18O in soil atmospheric N2O--temporal dynamics in a sandy soil.

The greenhouse gas nitrous oxide (N(2)O) can be both formed and consumed by microbial processes in the soil. As these processes fractionate strongly in favour of (14)N and (16)O, delta(15)N and delta(18)O gradients of N(2)O in the soil profile may elucidate patterns of N(2)O formation, consumption or emission to the atmosphere. We present the first in situ data of such gradients over time for a mesic typic Haplaquod seeded with potatoes (Solanum tuberosum L.). On two adjacent fields in 2002 and 2003, topsoil N(2)O fluxes were measured and the soil atmosphere was regularly sampled for N(2)O concentrations, delta(15)N and delta(18)O signatures of N(2)O at depths of 18, 48 and 90 cm during approximately 400 days. During the entire sampling period, the N(2)O concentrations were the highest and the delta(15)N signatures the lowest in the subsoil (48 or 90 cm depth) as compared with the topsoil, indicating production of N(2)O in the subsoil. For delta(15)N, differences greater than 30 per thousand between topsoil and subsoil on the same date were regularly observed. The highest N(2)O concentration of 100385 microL m(-3) at 90 cm depth on 1 July 2003, was preceded by the lowest delta(15)N value of -43.5 per thousand one week earlier. This was followed by a 150-day general decrease of N(2)O concentrations at 90 cm depth to 1723 microL m(-3) and a simultaneous enrichment of delta(15)N to +7.1 per thousand, mostly without a significant topsoil flux. There was a negative logarithmic relationship between N(2)O concentration at 90 cm depth and its delta(15)N signature. This relationship indicated a delta(15)N signature of -40 to -45 per thousand during the production of N(2)O in the subsoil, and a subsequent enrichment during the consumption of N(2)O. We conclude that the isotopic signature of the N(2)O topsoil flux is the result of various processes of consumption and production at different depths in the soil profile. It is therefore not a reliable estimator for the overall delta(15)N signature of N(2)O in the soil atmosphere, nor for indirect losses of N(2)O to the environment. Therefore, these findings will pose a further challenge to ongoing efforts to draw up a global isotopic budget for N(2)O.