Data on mineral fertiliser sales are generally available on national scale, deriving from two sources: reports of the Member States and from the association “Fertiliser Europe”. According to Eurostat (2018), due to the inclusion of intermediary goods and non-agricultural use, these statistics may overestimate the use of mineral fertilisers. Due to differences in reference periods, data sources and methodology, both sources cannot be directly compared.
Estimates on mineral fertilisation at farm level, however, can be obtained by interlinking data on crop production (e. g. agricultural preconditions, crop pattern, qualitative and quantitative yield).
Mineral fertilisation on farm or local level may be obtained by interrogating the farmers. In some Member States, the national transformation of Nitrates Directive obliges farmers to set up a precise, farmwide fertiliser planning, at least for nitrogen (for both, organic and mineral fertilisers). The purchased fertiser amounts are registered by the farm’s accountancy and the plant nutrient concentration has to be declared according to EU or national rules (EU 2003/2003).
Consequently, as far as mineral fertilisers are concerned, data on fertiliser use may be the most readily usable nitrogen input data (providing the data are available), which may be used as indicator, at least in regions with intensive arable crop production and marginal animal production (e.g. Paris Basin, for instance Case study 4, La Voulzy).
In regions with a pronounced animal production and a large amount of manure which has to be utilised on agricultural land as fertiliser, mineral fertiliser usage data do not possess indicator fuction: Osterburg and Techen (2012) combined data from a survey carried out between 2008 and 2010 among different types of farms in three German Federal States on mineral and organic fertilisation. They could show that independently from the amount of organic manure applied, the mineral fertiliser use varied considerably within one farm type. In Figure 5.2, farms were classified according to the nitrogen im- and export on field level. Exported nitrogen mirrors the yield and therefore the fertilising demand. The vertical axis shows a standard deviation of 30-50 kg/ha from the average applied nitrogen. This signifies, that the additional mineral fertiliser dose varies broadly, with no relation to the level of organic fertilisation with manure, ranging from 0 to 200 kg/ha. The variation in the mineral fertiliser dose is so high that it cannot be explained by different frame conditions but rather is due to differences in individual farm management (Osterburg, 2016).
According to this figure, farmers may use approximately the same amount of mineral N, whether they have zero or 200 kg N as organic fertilisers from keeping farm animals at their disposal.
According to Eurostat (2018), more detailed data on mineral fertiliser use in agriculture would be useful for several environmental EU policies. Collecting data in a geo-referenced sample survey would allow the use of these data for modelling environmental impacts at finer spatial scales.
Mineral fertilisation/ha could be used as indicator to explain nitrogen leaching in arable land areas. However, as explained, in animal breeding areas or mixed farming area it is necessary to use a combination of the two indicators mineral fertilisation by hectare and organic fertilisation by hectare.