Anthropogenic phosphorus flows in Denmark : Quantification and critical analysis

Abstract

Phosphorus (P) is an essential plant nutrient mined from the earth’s crust as phosphate rock. It cannot be substituted, making it a crucial resource for food production. For the EU, future phosphate scarcity is a potential geopolitical and strategic threat. An increasing worldwide phosphate demand is coupled with dependence on imports from a limited number of suppliers outside the EU-28, so that the EU updated its list of critical raw materials in 2014 to include phosphate rock. As a plant nutrient, P is not destroyed by human use, but dissipated into the environment, where it is a pollutant contributing to eutrophication of water bodies and soils. The anthropogenic P is open on the global scale, with global shipments of animal feed, fertiliser, and food; and on the local scale, through the inefficient use of fertiliser or animal manure by application in excess of plant P demand, and losses in waste and wastewater treatment due to insufficient recycling. The focus of this PhD project was on the resource aspect, as opposed to the pollution aspect, of P in Denmark. The overall goal was to quantify and evaluate the country’s anthropogenic P flows, i.e. those flows caused or significantly influenced by human action, based on a comprehensive material flow analysis (MFA). MFA is a method widely applied to establish resource budgets within a spatial – such as a country – and temporal system boundary, establish a material balance, and handle data uncertainties and data conflicts. When looking at P from a resource efficiency perspective, the most important flows to consider are those linked to agriculture, as a consumer and producer of large P flows, and waste/wastewater management, as the key processes for treating the resulting P-containing wastes. Country-wide average values regarding these processes hold limited informative value. Moreover, it became clear at the outset of the study that there were distinct differences between the P flows across regions of the country, especially between the east, with the largest urban agglomeration, and the northwest. Apart from population and industrial density, a contrast also exists in agricultural practice, with animal husbandry concentrated in the west and northwest, and the east being dominated by crop production. For the agriculture and waste management processes, the MFA was divided into 3 “typical” regions between the northwest (North Jutland), the east (Zealand and the capital region), and a middle part with more mixed characteristics (Mid-Jutland and Southern Denmark); the regional subdivisions formed a part of a complete country-scale MFA. As is typical for a European country, the Danish P budget showed a strong dependence on P imports in fertilisers and animal feed; with food products being the dominant export of P. The regional contrasts in agricultural P budgets were pronounced as expected, with a slight P deficit in the east and the largest per-hectare surplus, due to large amounts of manure, in the northwest. Manure was shown to hold the most salient potential for P recovery, yet stays quite local and adds to the surplus in the country’s northwest, posing a environmental problem. In the waste management system, two streams were identified to hold significant potential for P recovery. Sewage sludge, while already applied to land on a considerable scale, still holds potentially recoverable P not yet utilised today; and vegetable and animal kitchen and food waste from consumption currently not collected separately, with residues being lost to P recovery. These amounts are furthermore located in the east, with a slight P deficit in agricultural soils, suggesting themselves for substituting some fertiliser imports in the future. The total P quantities in these streams amounted to approximately 35% of concurrent mineral P imports. Since MFA for regional resource budgets is often the groundwork for further analysis, the robustness and comparability of MFA studies’ outcomes when using them as sources of information is important. To this end, the MFA for Denmark was compared to a recent and methodically similar P MFA for Austria, and the effects of the structure of the data material and an MFA practitioner’s modelling choices on the outcomes identified and measured. It was demonstrated that the data available do, in fact, influence model layout. Moreover, the approach to assess uncertainty is subject to a certain degree of arbitrariness, and reflects the modeller’s belief in the quality of the data material. This, however, leads to incomparability of data quality between MFAs, as the comparison showed, since data uncertainties can be only evaluated against those in the same model. Lastly, data conflicts are normal in country-scale MFA; the extent of the necessary reconciliation of conflicting data provides a useful proxy measure for the quality of an MFA. Metadata matter; this comparison showed the quantitative effects of those aspects of MFAs not resulting from the real-world systems studied. The results thus gave a quantitative basis to requiring a transparent system definition and data characterisation in regional MFA beyond the case of P in Denmark. A third part of this PhD project consisted of exploring the potentials for increased P recovery efficiency in the Danish anthroposphere based on the results of the initial MFA. An aspect of a secondary resource (recovered P fertiliser) is its ability to fulfil the functions of the resource substituted (mineral P fertiliser). For this purpose, the MFA system and values obtained in this study were adapted to reflect the typical values for availability of P from various material flows to crops, and to allow for transport of less bulky secondary-P material flows. An optimal distribution of recovered P flows (from sewage sludge and composted organic household waste) was then determined by formal optimization via linear programming. The outcome showed a gradual decline of both mineral P inputs, and net additions to soil P stocks, stabilising at a distinctly lower level than evident from the static MFA, due to P applied gradually becoming available for plants over time, showing a significantly higher (82%) potential for substituting mineral P imports than evident from the initial, static MFA (35%). While the potential improvements in closing the P cycle could be shown, this can, however, not be expected to change the reliance on imported P on one or another form.