Livestock density does not give detailed information on nutrient (especially nitrogen and phosphorus) excretion. Livestock unit (LU) is a derived value, using animal counts from animal censuses. According to Regulation EC 1165/2008 concerning livestock and meat statistics, animal censuses categorise bovine, pigs, sheep and goats far more detailed (Table 5.1; EC, 2008). Information on animal categories and nutrient excretions (= the total amount of N and P excreted by livestock per year as urine and faeces) can be combined to obtain a more substancial estimation of the nutrient concentration in livestock excretion. This approach is suitable on national, regional and farm level.
Velthof et al. (2015) could show in a review on methods to determine N-excretion factors for different animal categories, that methods differ significantly between policy reports and between countries. Part of these differences may be related to different animal production methods, size or weight of the animals and feed composition, but partly also to differences in the aggregation of livestock categories and estimation procedures. Additionally, methodologies and data use often are not well described. Consequently, there is a need for harmonisation.
Hou et al. (2016) propose a method for the estimation of annual feed use and N excretion per animal category for all countries of the EU-27, based on the energy and protein requirements of the animals and statistics of feed use and composition, animal number and productivity (Hou et al., 2016).
Using an input-output-model, in Germany, nutrient excretion for different animal categories and production varieties are calculated (DLG, 2014). These data are also part of the current German Fertilising Ordinance (DüV, 2017). In a simplified approach, the total volume of excretion produced by livestock is calculated by multiplying the number of animals with a rough estimation on the manure production per animal and the nutrient concentration of manure per t. Estimates by this approach on manure production though differ, in France for cattle 8.4 t per year (CA Bretagne, 2014), with a concentration of 5.9 kg of N per t, what equals 50 kg N per year; in Germany for young cattle 8 t per year, but for dairy cows around 20 t per year. N-excretion are calculated from 45 to 57 per year for a young cattle and from 100 to 153 kg N per year for a dairy cow, depending on breeding intensity and feeding regime (DLG, 2014, DüV, 2017).
Table 5.1: Categories of livestock statistics (EC, 2008)
Bovine animals bovine animals aged not over 1 year |
calves and young cattle for slaughter | ||
other | male | ||
female | |||
Bovine animals aged over 1 year but under 2 years (except females that have calved) | male | ||
female (heifers; animals that have not yet calved) | animals for slaughter | ||
other | |||
Bovine animals of 2 years and over | male | ||
female | heifers | heifers for slaughter | |
other | |||
cows (bovine animals that have calved, including those that are under 2 years old) | dairy cows | ||
other | |||
Buffaloes | female breeding buffaloes | ||
other buffaloes | |||
Pigs | piglets with a live weight of less than 20 kg | ||
pigs with a live weight of 20 kg or more but less than 50 kg | |||
fattening pigs, including cull boars and cull sows with a live weight | of 50 kg or more but less than 80 kg | ||
of 80 kg or more but less than 110 kg | |||
110 kg or more breeding | |||
pigs with a live weight of 50 kg and over | boars | ||
covered sows, of which | sows covered for the first time | ||
other sows | |||
gilts not yet covered | |||
Sheep | ewes and ewe lambs put to the ram | milk ewes and milk ewe lambs put to the ram | |
other ewes and ewe lambs put to the ram | |||
other sheep | |||
Goats | goats which have already kidded and goats which have been mated | ||
goats which have already kidded | |||
goats mated for the first time | |||
other goats |
Therefore, depending on the calculation method, the livestock excretion is a rather precise indicator, but not easy to calculate. Default values for each type of livestock are available to perform estimations (DLG, 2014), as well as calculation methods (Hou et al., 2016).