Debates about the agronomic feasibility of feeding the world organically have often focused on potential nitrogen (N) limitations, overlooking the challenges raised by the availability of phosphorus (P) in soils. Here we address this missing piece by simulating, over a 100-year time span, a 100% organic world, in which mineral fertilizers are banned. We show that, in the short term, soil available P would barely limit organic crop production due to (i) existing soil P stocks, (ii) a replenishment of the available P pool by more stable pools, and (iii) a strong N limitation that reduces crop yields and consequently plant P demand. Conversely, in the long term (~100 years), the simulated 60% drop in cropland soil available P would result in a -12% reduction in cropland production globally, ranging from -4% in Eastern Europe to -29% in Latin America. This P limitation would be greater in places with high yields and a high proportion of legumes in crop rotations, highlighting the need to address N-P interactions simultaneously to design more sustainable food systems. The expansion of organic farming therefore runs the risk of slowly but surely depleting soil P unless measures are taken to recycle all P flows within the agri-food system.

Improved management of phosphorus (P) is essential for achieving a range of Sustainable Development Goals (SDGs), including maintaining food security, preserving water quality, and mitigating climate change. This requires an integration of comprehensive mechanistic understanding with accurate spatial data. In this interdisciplinary review, we combine insights from empirical P research, digital soil mapping, biogeochemical modeling, and environmental law to critically examine the current state, pinpoint challenges and propose novel pathways for desperately needed P maps. We first elucidate the relevance of spatial data on P for different SDGs. Subsequently, we summarize the current efforts in mapping P pools at regional to global scales, and discuss the challenges of mapping “available P” due to substantial local scale variability and poor correlation with predictors relative to other soil properties. The practical applicability of these recently published maps is tested by evaluating them with independent measurement data. Finally, we outline ways forward to enhance the accuracy and reliability of P maps, as a basis for science-informed management of P resources.

Soil phosphorus (P) directly impacts major sustainability outcomes, namely crop yields, water quality, and carbon sequestration. Optimally managing P to improve sustainability outcomes requires a mechanistic understanding of P availability and transfer, alongside high-resolution spatial data. However, it is unclear if current measurement techniques, models, and maps meet the demands for science-informed management. Here, we review recent advances in measuring P fluxes, quantifying P availability, and mapping soil P resources and discuss implications for sustainability outcomes. We find that the understanding of soil P availability has significantly improved but that agronomical applications and climate models are still largely based on outdated concepts. Also, we find that spatial data on soil P resources are highly uncertain, limiting the usefulness of current P maps. We highlight steps to improve existing tools and emphasize that these improvements need to go hand in hand with policy and technological development to successfully address P-related sustainable development goals.

Numerous drivers such as farming practices, erosion, land-use change, and soil biogeochemical background, determine the global spatial distribution of phosphorus (P) in agricultural soils. Here, we revised an approach published earlier (called here GPASOIL-v0), in which several global datasets describing these drivers were combined with a process model for soil P dynamics to reconstruct the past and current distribution of P in cropland and grassland soils. The objective of the present update, called GPASOIL-v1, is to incorporate recent advances in process understanding about soil inorganic P dynamics, in datasets to describe the different drivers, and in regional soil P measurements for benchmarking. We trace the impact of the update on the reconstructed soil P. After the update we estimate a global averaged inorganic labile P of 187 kgP ha−1 for cropland and 91 kgP ha−1 for grassland in 2018 for the top 0–0.3 m soil layer, but these values are sensitive to the mineralization rates chosen for the organic P pools. Uncertainty in the driver estimates lead to coefficients of variation of 0.22 and 0.54 for cropland and grassland, respectively. This work makes the methods for simulating the agricultural soil P maps more transparent and reproducible than previous estimates, and increases the confidence in the new estimates, while the evaluation against regional dataset still suggests rooms for further improvement.

Most phosphorus (P) in soils is unavailable for direct biological uptake, as it is locked within primary or secondary mineral particles, adsorbed to mineral surfaces, or immobilized inside of organic material. Deciphering the composition of different P forms in soil is critical for understanding P bioavailability and its underlying dynamics. However, widely used global estimates of different soil P forms are based on a dataset containing few measurements in which many regions or soil types are unrepresented. This poses a major source of uncertainty in assessments that rely on these estimates to quantify soil P constraints on biological activity controlling global food production and terrestrial carbon balance. To address this issue, we consolidated a database of six major soil P “forms” containing 1857 entries from globally distributed (semi-)natural soils and 11 related environmental variables. These six different forms of P (labile inorganic P (Pi), labile organic P (Po), moderately labile Pi, moderately labile Po, primary mineral P, and occluded P) were measured using a sequential P fractionation method. As they do not represent precise forms of specific discrete P compounds in the soil but rather resemble operational pools, we will now refer to them as P pools. In order to quantify the relative importance of 11 soil-forming variables in predicting soil P pool concentrations and then make further predictions at the global scale, we trained random forest regression models for each of the P pools and captured observed variation with R2 higher than 60 %. We identified total soil P concentration as the most important predictor of all soil P pool concentrations, except for primary mineral P concentration, which is primarily controlled by soil pH and only secondarily by total soil P concentration. When expressed in relative values (proportion of total P), the model showed that soil pH is generally the most important predictor for proportions of all soil P pools, alongside the prominent influences of soil organic carbon, total P concentration, soil depth, and biome. These results suggest that, while concentration values of P pools logically strongly depend on soil total P concentration, the relative values of the different pools are modulated by other soil properties and the environmental context. Using the trained random forest models, we predicted soil P pools' distributions in natural systems at a resolution of 0.5∘×0.5∘. Our global maps of different P pools in soils as well as the pools' underlying drivers can inform assessments of the role of natural P availability for ecosystem productivity, climate change mitigation, and the functioning of the Earth system.

The use of mineral phosphorus (P) fertilizers, often referred to as anthropogenic phosphorus, has dramatically altered the global phosphorus cycle and increased soil phosphorus fertility and crop yields. Quantifying agriculture’s reliance on anthropogenic phosphorus requires estimates of its contribution to agricultural soil fertility. Here we present a model of soil phosphorus dynamics simulating phosphorus availability in agricultural soils for individual countries from 1950 to 2017. Distinguishing between anthropogenic and natural phosphorus pools and accounting for farming practices, agricultural trade and crop–livestock recycling, we estimate that the global anthropogenic contribution to available phosphorus in agricultural soils was 47 ± 8% in 2017. Country-level anthropogenic phosphorus signatures vary according to cumulative fertilizer use and phosphorus availability in soil inherited pre-1950, with negligible influence of the trade of feed and food products. Despite different historical trajectories, we find that Western Europe, North America and Asia are similarly reliant on anthropogenic phosphorus, with nearly 60% of the total available phosphorus of anthropic origin in 2017. Conversely, anthropogenic phosphorus inputs in Africa remained low over the study period, contributing only around 30% of available phosphorus. The unequal reliance of agricultural soil fertility and food production systems on anthropogenic phosphorus resources highlights the need for a fairer management of the world’s remaining phosphate rock resources. About half of the current available phosphorus in agricultural soil globally is derived from anthropogenic sources, according to country-scale simulations of phosphorus dynamics between 1950 and 2017.

Agricultural productivity has dramatically increased in the last sixty years, with undeniable benefits for global food security. Yet, our agricultural production systems have been built on the use of non-renewable resources, thereby altering their sustainability. Agriculture depends on fossil fuel energies – mainly to produce nitrogen fertilizers - but also on another non-renewable resource: phosphate rocks. Here we propose to quantify the reliance of our global food production on the use of fertilizer and additives derived from phosphate rocks, referred to as anthropogenic phosphorus (P). To do so, we simulated the evolution of the soil available P for 132 countries during the 1950-2017 period, with a distinction between both anthropogenic vs. natural soil P stocks. Natural P refers to P that is not derived from mined phosphate rocks. We also explicitly simulated the international trade of feed and food products, given that these fluxes participate in the transfer of anthropogenic P between countries. Finally, for each country, we calculated the P anthropogenic signature of their food production by dividing the anthropogenic P content of agricultural products by their total P content. Our results show that in 2017, the global P anthropogenic signature of food production was ~37%, with large variations across world regions. North America displayed the largest anthropogenic signature (63% ±9% in 2017), followed by Western Europe (55% ±10%), Asia (47% ±7%), Eastern Europe (35% ±10%), South America (33% ±6%), and Africa (20% ±5%). Also, the temporal evolutions of the P anthropogenic signatures reflect the dynamics of agricultural intensification observed in the different world regions. Overall, trade had a negligible effect on the P anthropogenic signature of food production, even when it contributed significantly to increase the soil P fertility of some countries (e.g. The Netherlands). Our estimates of soil P anthropogenic signatures were associated with large uncertainties, raising questions about the best way to estimate soil P legacy and about the data availability to calibrate the models. Eventually, our results highlight the large dependence of global food production to the non-renewable resources that are phosphate rocks. They suggest the urgent need to engage the transition of our food production systems toward more sustainable, input-free and circular agriculture.

Abstract The global phosphorus (P) cycle has been dramatically altered by human activities through the use of mineral P fertilizers, often referred to anthropogenic P. The application of mineral P fertilizers on agricultural soils has driven the planet beyond its safe operating space but the extent to which the global P cycle relies on anthropogenic P has never been quantified. To fill this gap, we developed a model that simulates, at the country scale, the evolution of agricultural soil available P by distinguishing anthropogenic vs. natural P pools, and by accounting for farming practices, crop-livestock recycling loop, and agricultural trade, over the 1950-2017 period. At the global scale we found that the anthropogenic signature of soil available P was 45% ± 8% in 2017. The national anthropogenic signatures varied according to the cumulative mineral P fertilizer use as well as to the soil P availability in 1950. Despite having different historical trajectories, Western Europe, North America and Asia displayed similar reliance on anthropogenic P, close to 60% in 2017. Conversely, African soil anthropogenic signature remained around 30%. Trade did not modify the simulated signatures. Overall, our results unravel the strong reliance of our soil fertility and food production systems on anthropogenic P resources.

Abstract. How global gridded crop models (GGCMs) differ in their simulation of potential yield and reasons for those differences have never been assessed. The GGCM Intercomparison (GGCMI) offers a good framework for this assessment. Here, we built an emulator (called SMM for simple mechanistic model) of GGCMs based on generic and simplified formalism. The SMM equations describe crop phenology by a sum of growing degree days, canopy radiation absorption by the Beer–Lambert law, and its conversion into aboveground biomass by a radiation use efficiency (RUE). We fitted the parameters of this emulator against gridded aboveground maize biomass at the end of the growing season simulated by eight different GGCMs in a given year (2000). Our assumption is that the simple set of equations of SMM, after calibration, could reproduce the response of most GGCMs so that differences between GGCMs can be attributed to the parameters related to processes captured by the emulator. Despite huge differences between GGCMs, we show that if we fit both a parameter describing the thermal requirement for leaf emergence by adjusting its value to each grid-point in space, as done by GGCM modellers following the GGCMI protocol, and a GGCM-dependent globally uniform RUE, then the simple set of equations of the SMM emulator is sufficient to reproduce the spatial distribution of the original aboveground biomass simulated by most GGCMs. The grain filling is simulated in SMM by considering a fixed-in-time fraction of net primary productivity allocated to the grains (frac) once a threshold in leaves number (nthresh) is reached. Once calibrated, these two parameters allow for the capture of the relationship between potential yield and final aboveground biomass of each GGCM. It is particularly important as the divergence among GGCMs is larger for yield than for aboveground biomass. Thus, we showed that the divergence between GGCMs can be summarized by the differences in a few parameters. Our simple but mechanistic model could also be an interesting tool to test new developments in order to improve the simulation of potential yield at the global scale.

Single and combined fertilization additions are a common tool to assess the interactions between nutrients in a given ecosystem. While such experiments can allow systems to be defined into categories of nutrient interactions, for example, simultaneous co‐limitation or single resource response, this categorization may itself be sensitive to way nutrient interactions are mathematically formulated. To this end, we developed a theoretical analysis of nitrogen (N) and phosphorus (P) fertilization experiments based on the computation of ratios between plant demand and soil supply for each nutrient to explore two mathematical interaction formalisms: Liebig's law of minimum (LM) and the multiple limitation hypothesis (MH). We defined, for each interaction formalism, what conditions (in terms of supply and demand in N and P) are required to make the ecosystem in each category of nutrient interaction. Notably, we showed that synergistic co‐limitation could occur even using LM formalism under certain conditions. We then applied our framework to global maps of soil nutrient supply and of crop nutrient demand to achieve the potential yield. This was done to examine how the choice of interaction formalism influenced the occurrence of nutrient interaction categories. MH predicts true co‐limitation for ∼40% of the global maize area where LM predicts other categories of nutrient interaction, particularly single resource P limitation (whose the exact occurrence is, however, sensitive to the amount of P applied in the fertilization experiments). Our study identified areas where real fertilization experiments are required to choose between LM or MH to best represent nutrient interaction in croplands.

Abstract. Soils represent the largest phosphorus (P) reserves on land and determining the amount is a critical first step for identifying sites where ecosystem functioning is potentially limited by P availability. However, global patterns and predictors of soil total P concentration remain poorly understood. To address this knowledge gap, we constructed a database of the total P concentration of 5,275 distributed globally natural soils. We quantified the relative importance of 13 soil-forming variables in predicting soil total P concentration and then made further predictions at the global scale using a random forest approach. Soil total P concentration varied significantly among parent material types, soil orders, biomes, and continents, and ranged widely from 1.4 to 9,630.0 (median 430.0 and mean 570.0) mg kg−1 across the globe. About two-thirds (65 %) of the global variation was accounted for by the 13 variables that we selected, among which soil organic carbon concentration, parent material, mean annual temperature, and soil sand content were the most important. While global predictions of soil total P concentration increased significantly with latitude, they varied largely among regions with similar latitudes due to regional differences in parent material, topography, and/or climate conditions. Global soil P stocks (excluding Antarctica) were estimated to be 26.8 ± 3.1 (mean ± standard deviation) Pg and 62.2 ± 8.9 Pg (1 Pg = 1 × 1015 g) in the topsoil (0–30 cm) and subsoil (30–100 cm), respectively. Our global map of soil total P concentration as well as the underlying drivers of soil total P concentration can be used to constraint Earth system models that represent the P cycle and to inform quantification of global soil P availability. Raw datasets and global maps generated in this study are available at https://doi.org/10.6084/m9.figshare.14583375 (He et al., 2021).

Soil represents the largest phosphorus (P) stock in terrestrial ecosystems. Determining the amount of soil P is a critical first step in identifying sites where ecosystem functioning is potentially limited by soil P availability. However, global patterns and predictors of soil total P concentration remain poorly understood. To address this knowledge gap, we constructed a database of total P concentration of 5275 globally distributed (semi-)natural soils from 761 published studies. We quantified the relative importance of 13 soil-forming variables in predicting soil total P concentration and then made further predictions at the global scale using a random forest approach. Soil total P concentration varied significantly among parent material types, soil orders, biomes, and continents and ranged widely from 1.4 to 9630.0 (median 430.0 and mean 570.0) mg kg−1 across the globe. About two-thirds (65 %) of the global variation was accounted for by the 13 variables that we selected, among which soil organic carbon concentration, parent material, mean annual temperature, and soil sand content were the most important ones. While predicted soil total P concentrations increased significantly with latitude, they varied largely among regions with similar latitudes due to regional differences in parent material, topography, and/or climate conditions. Soil P stocks (excluding Antarctica) were estimated to be 26.8 ± 3.1 (mean ± standard deviation) Pg and 62.2 ± 8.9 Pg (1 Pg = 1 × 1015 g) in the topsoil (0–30 cm) and subsoil (30–100 cm), respectively. Our global map of soil total P concentration as well as the underlying drivers of soil total P concentration can be used to constraint Earth system models that represent the P cycle and to inform quantification of global soil P availability. Raw datasets and global maps generated in this study are available at https://doi.org/10.6084/m9.figshare.14583375 (He et al., 2021).

Phosphorus (P) as a key element in DNA, RNA as well as ATP and phospholipids is essential for the growth, functioning and reproduction of all life on earth. However, if fertilization with animal wastes or human excreta is not available or not organized, P fertilizers stem from nonrenewable geological P deposits, which are an increasingly limited resource. The potential threats of a global P limitation due to “peak phosphorus” have been discussed intensively in the recent past including the socio economic as well as political consequences which will be dramatic. While a deficit in available soil P leads to a loss of agricultural yield, an excess of total P in soils triggers aquatic eutrophication, loss in biodiversity and wildlife habitat in surrounding water bodies in other regions of the world. We calculated global soil P balances considering input from atmosphere and plant management (as sum of manure and residue input minus plant uptake) versus depletion due to soil erosion in coupling P fluxes from (Ringeval et al., 2017) with soil erosion rates from (Borrelli et al., 2017). The world’s soils are currently being depleted in P in spite of high chemical fertilizer input. Considering the current high chemical fertilizer inputs most continents result in slightly positive P balances (e.g. net P input to soils). Exception are Africa with very low chemical fertilizer input of 1.7 kg ha-1yr-1 paired with high losses due to soil erosion of 2 kg ha-1yr-1 and Europe (the latter is the average for the geographic Europe including eastern European countries with very low chemical fertilizer input). Results indicate negative balances globally as well as for all continents (depletion between 4 and 19 kg P ha-1yr-1 ) if input of chemical fertilizers is neglected. Parallel to the distribution pattern and dynamics of global soil erosion by water (Borrelli et al., 2017), P losses from soils due to water erosion are most dramatic in countries and regions with intensive agriculture and/or extreme climates (e.g., high frequencies of heavy rain storm or droughts followed by significant rain events). References Borrelli, P., Robinson, D.A., Fleischer, L.R., Lugato, E., Ballabio, C., Alewell, C., Meusburger, K., Modugno, S., Schütt, B., Ferro, V., Bagarello, V., Oost, K.V., Montanarella, L. and Panagos, P., 2017. An assessment of the global impact of 21st century land use change on soil erosion. Nature Communications, 8(1): 2013. Ringeval, B., Augusto, L., Monod, H., van Apeldoorn, D., Bouwman, L., Yang, X., Achat, D.L., Chini, L.P., Van Oost, K., Guenet, B., Wang, R., Decharme, B., Nesme, T. and Pellerin, S., 2017. Phosphorus in agricultural soils: drivers of its distribution at the global scale. Global Change Biology

Soil phosphorus (P) loss from agricultural systems will limit food and feed production in the future. Here, we combine spatially distributed global soil erosion estimates (only considering sheet and rill erosion by water) with spatially distributed global P content for cropland soils to assess global soil P loss. The world’s soils are currently being depleted in P in spite of high chemical fertilizer input. Africa (not being able to afford the high costs of chemical fertilizer) as well as South America (due to non-efficient organic P management) and Eastern Europe (for a combination of the two previous reasons) have the highest P depletion rates. In a future world, with an assumed absolute shortage of mineral P fertilizer, agricultural soils worldwide will be depleted by between 4–19 kg ha −1 yr −1 , with average losses of P due to erosion by water contributing over 50% of total P losses.