3.B - Manure Management

In 2022, NH₃ emissions from category 3.B (manure management) were 42.4 % from total agricultural emissions, which is equal to ~ 199.2 kt NH₃. Within those emissions 51.2 % originate from cattle manure (~ 101.9 kt), 31.9 % from pig manure (ca. 63.5 kt), and 12.9 % from poultry manure (~ 25.7 kt). Calculations take into account the impact of anaerobic digestion of manure on the emissions.

NO_x emissions from category 3.B (manure management) contribute only 1.3 % (~ 1.3 kt) to the total agricultural NO_x emissions. They are calculated proportionally to N₂O emissions, see Vos et al. (2024) ¹⁾.

NMVOC emissions from category 3.B (manure management) contributed 96.9 % (278.2 kt) from total agricultural NMVOC emissions (287.1 kt).

In 2022, manure management contributed, respectively, 64.4 % (38.0 kt), 36.6 % (12.1 kt) and 68.6 % (3.6 kt) to the total agricultural TSP, PM₁₀ and PM_2.5 emissions (TSP: 59.1 kt, PM₁₀: 33.1 kt, PM_2.5: 5.3 kt, respectively).

The Federal Statistical Agency and the Statistical Agencies of the federal states carry out surveys in order to collect, along with other data, the head counts of animals. The results of these surveys are used for emission calculations, for details see Vos et al, 2024, Chapter 2.3.

The animal population figures used in the inventory are presented in Table 1. Buffaloes are included in the cattle population figures, mules and asses are included in the horse population figures (IE), see Vos et al. (2024), Chapter 2.3. In the first years after the German reunification in 1990 animal livestock decreased markedly. The head counts for cattle continued to decrease significantly until 2006/2007, followed by a more or less stable period until 2014. Since 2015 a slight decrease occurred. In 2022, dairy cattle numbers are 60.0% of 1990 numbers, while the total population of other cattle is at 54.7 % of 1990. Swine numbers decreased until 1995 and then increased slightly. Since 2014 a new decrease occurred which became significant between 2020 and 2022 (total pig numbers were reduced by around 18 %). In 2022 swine numbers are 66.8 % of 1990 numbers. The 2022 numbers of horses, sheep and goats are, respectively, at 92.0 %, 55.3 % and 181.1 % of 1990.

Figures for broilers and turkeys are showing a massive increase since 1990. Since the year 2013, there have been only minor changes of total poultry numbers. In total, 2022 poultry population figures are at 154.3 % of 1990. A detailed description of the animal numbers used can be found in the National Inventory Report 2024 2), Chapter 5.1.3.2.3. Emissions of deer, rabbits, ostrich and fur-bearing animals are reported since submission 2024, see Vos et al. (2024)¹⁾. The underlying animal numbers of these categories were estimated in different ways because there are no surveys which collect those animal numbers, for details see Vos et al, 2024, Chapter 2.3. However, the impact of those animal categories on the total emissions is small.

Table 1: Population of animals, in [1,000 individuals]

Emission calculations in accordance with a Tier 2 or Tier 3 method require data on animal performance (animal weight, weight gain, milk yield, milk protein content, milk fat content, numbers of births, numbers of eggs and weights of eggs) and on the relevant feeding details (phase feeding, feed components, protein and energy content, digestibility and feed efficiency). To subdivide officially recorded total numbers of turkeys into roosters and hens, the respective population percentages need to be known. Details on data requirements for the modelling of emissions from livestock husbandry in the German inventory can be found in Vos et al. (2024), Chapter 2 ²⁾.

Most of the data regarding feed and performance is not available from official statistics and was obtained from literature, from publications by agricultural associations, from regulations for agricultural consulting in Germany and from expert judgments.

For 1991, 1995 and 1999, frequency distributions of feeding strategies, husbandry systems (shares of pasturing/stabling; shares of various housing methods), storage types as well as techniques of farm manure spreading were obtained with the help of the RAUMIS agricultural sector model (Regionalisiertes Agrar- und UmweltInformationsystem für Deutschland/ Regionalised agricultural and environmental information system for Germany). RAUMIS has been developed and is operated by the Institute of Rural Studies of the Thünen Institute (Federal Research Institute for Rural Areas, Forestry and Fisheries). For an introduction to RAUMIS see Weingarten (1995)²⁾; a detailed description is provided in Henrichsmeyer et al. (1996)³⁾.

RAUMIS did not model complete time series but only selected years. RAUMIS data for the years 1991, 1995, and 1999 are used in the inventory for years 1990 – 1993, 1994 – 1997, and 1998 – 1999, respectively. For the year 2009, respective data are used that were derived from the 2010 official agricultural census and the simultaneous survey of agricultural production methods (Landwirtschaftliche Zählung 2010, Statistisches Bundesamt/ Federal Statistical Office) as well as the 2011 survey on manure application practices (Erhebung über Wirtschaftsdüngerausbringung, Statistisches Bundesamt/ Federal Statistical Office).

For the year 2015, data on techniques of farm manure spreading from the 2016 official agricultural census (Agrarstrukturerhebung 2016, Statistisches Bundesamt / Federal Statistical Office) are used. The gaps between the latest RAUMIS model data (1999) and the first official data (2009) were closed by linear interpolation on district level. For the year 2019 data from the 2020 official agricultural census (Landwirtschaftszählung 2020, LW20) are used for housing systems, storage systems and manure spreading systems. For 2010 to 2018 the housing and storage systems data was linearly interpolated between the censuses of 2010 and 2020. The data on manure spreading techniques was linearly interpolated between the census data from 2009 and 2015, and for 2016 to 2018 between the censuses conducted in 2016 and 2020. In addition, it was taken into account that, as of 2012, slurry spread on bare soil has to be incorporated within four hours. For a description of the RAUMIS data, the data from official surveys and additional data from other sources see Vos et al. (2024), Chapter 2.5. Time series of frequency distributions of housing systems, storage systems and application techniques as well as the corresponding emission factors are provided in NID 2024, Chapter 17.3.1.

In order to determine NH₃ and NO_x emissions from manure management of a specific animal category, the individual N excretion rate must be known as well as, for NH₃, the TAN content of the N excretions. Default excretion rates are provided by IPCC Guidelines and default TAN contents can be found in the EMEP Guidebook, 2019⁴⁾. However, the German agricultural emission inventory uses N mass balances to calculate the N excretions and the TAN contents of almost all reported animal categories. N mass balance calculations (see below) consider N intake with feed, N retention due to growth, N contained in milk and eggs, and N in offspring. Table 2 presents national means of N excretions and TAN contents. For methodological details and mass balance input data see Vos et al. (2024), Chapter 4.2 as well as Chapter 4.1.2.

Table 2: National means of N excretions and TAN contents

The calculation of the emissions of NH₃, N₂O, NO_x and N₂ from German animal husbandry is based on the so-called N mass flow approach (e. g. Dämmgen and Hutchings, 2008 ⁵⁾). This approach differentiates between N excreted with faeces (organic nitrogen Norg, i. e. undigested feed N) and urine (total ammoniacal nitrogen TAN, i. e. fraction of feed N metabolized). The N flow within the manure management system is treated as depicted in the figure below. This method reconciles the requirements of both the Atmospheric Emission Inventory Guidebook for NH₃ emissions (EMEP, 2019), and the IPCC guidelines for greenhouse gas emissions (IPCC (2006)⁶⁾. Reidy et al. (2008),⁷⁾, showed for several European countries (Germany, the Netherlands, Switzerland, United Kingdom) that their N-flow based inventory models yielded, in spite of national peculiarities, comparable results as long as standardised data sets for the input variables were used.

Not explicitly shown in the N mass flow scheme is air scrubbing in housing and anaerobic digestion of manure. These issues are separately described further below. Note that emissions from grazing and application are reported in sector 3.D.

General scheme of N flows in animal husbandry

m: mass from which emissions may occur. Narrow broken arrows: TAN (total ammoniacal nitrogen); narrow continuous arrows: organic N. The horizontal arrows denote the process of immobilisation in systems with bedding occurring in the house, and the process of mineralisation during storage, which occurs in any case. Broad arrows denote N-emissions assigned to manure management (E_yard NH₃ emissions from yards; E_house NH₃ emissions from house; E_storage NH₃, N₂O, NO_x and N₂ emissions from storage; E_applic NH₃ emissions during and after spreading; E_graz NH₃, N₂O, NO_x and N₂ emissions during and after grazing; E_soil N₂O, NO_x and N₂ emissions from soil resulting from manure input).

The model allows tracing of the pathways of the two N fractions after excretion. The various locations where excretion may take place are considered. The partial mass flows through the livestock systems are represented. During storage Norg can be transformed into TAN and vice versa. Both, the way and the magnitude of such transformations may be influenced by manure treatment processes like, e. g., anaerobic digestion where a considerable fraction of Norg is mineralized to TAN. For details see Vos et al. (2024), Chapter 4.2. Wherever NH₃ is emitted, its formation is related to the amount of the TAN present. N₂O emissions are related to the total amount of N available (Norg + TAN). NO_x emissions (i. e. NO emissions) are calculated proportionally to the N₂O emissions, see section 'Emission factors'. Note that the N₂O, NO_x and N2 emissions from the various storage systems include the respective emissions from the related housing systems.

For pig and poultry production the inventory model considers the effect of air scrubbing. Data on frequencies of air scrubbing facilities and the removal efficiency are provided by KTBL (Kuratorium für Technik und Bauwesen in der Landwirtschaft / Association for Technology and Structures in Agriculture) supplemented by data from the 2020 agricultural census. The average removal efficiency of NH₃ is 80 % for swine and 70 % for poultry, while for TSP and PM₁₀ the rates are set to 90 % and for PM_2.5 to 70 % for both animal categories. For swine two types of air scrubbers are distinguished: systems of “first class” that remove both NH₃ and particles, and ”second class” systems that remove only particles reliably and have a NH₃ removal efficiency of 20%.

According to KTBL, 7.5 % of all pig places were equipped with ”first class” systems in 2022, another 12.5 % were equipped with “second class” systems. For poultry 0.85 % of all laying hen places and 2.2 % of all broiler places were equipped with air scrubbers that remove both NH₃ and particles. The amounts of NH₃-N removed by air scrubbing are completely added to the pools of total N and TAN for landspreading. For details see Vos et al. (2024), Chapter 4.2.2.

According to IPCC (2006), anaerobic digestion of manure is treated like a particular storage type. In the German Inventory it comprises three sub-compartments (pre-storage, fermenter and storage of digestates). For details see Vos et al. (2024), Chapters 2.6 and 4.2.5. The resulting digestates are considered as liquid. Two different types of digestates storage systems are considered: gastight storage and open tank. For open tanks formation of a natural crust because of co-fermentation with energy crops is taken into account. Furthermore, the modelling of anaerobic digestion and spreading of the digestates takes into account that the amount of TAN in the digestates is higher than in untreated slurry and that the frequencies of spreading techniques differ from those for untreated slurry.

NH₃ and NO emissions occur from pre-storage of solid manure, from non-gastight storage of digestates and from application of digestates (NH₃ emissions and NO emissions from application of digested manure are reported in 3.Da.2.a). There are no such emissions from pre-storage of slurry, from the fermenter and from gastight storage of digestates. Note that NH₃ and NO emissions calculated with respect to the digestion of animal manures do not comprise the contributions by co-digested energy crops. The latter are dealt with separately in 3.D.a.2.c and 3.I.

Application of the N mass flow approach requires detailed emission factors for NH₃, N₂O, NO_x and N₂ describing the emissions from the various housing and storage systems.

The detailed NH₃ emission factors are, in general, related to the amount of TAN available at the various stages of the N flow chain. The emission factors for laying hens, broilers, pullets, ducks and turkeys are related to N. Most NH₃ emission factors are country-specific but some are taken from EMEP (2019). No specific NH₃ emission factors are known for the application of digested manure. However, due to co-fermentation with energy crops, the viscosity of digested manure resembles that of untreated cattle slurry. Hence, the emission factors for untreated cattle slurry are adopted for the application of digested manure. For the detailed emission factors of livestock husbandry see Vos et al. (2024), Chapter 4.3.

Table 3 provides, by animal category, the implied NH₃ emission factors for manure management (housing and storage). The overall German NH₃ IEF for manure application is reported in section 3.D.a.2.a.

The the detailed emission factors for N₂O, NO_x and N₂ relate to the amount of N available which is N excreted plus, in case of solid manure systems, N input with bedding material. The N₂O emission factors are taken from IPCC (2006). The emission factors for NO_x and N₂ are approximated as being proportional to the N₂O emission factors, i.e. the NO-N and N₂ emission factors are, respectively, one-tenth and three times the value of the N₂O-N emission factor, see Vos et al. (2024), chapter 4.2.4. This proportionality is also applied to anaerobic digestion of manure, where N₂O emissions occur from pre-storage of solid manure and non-gastight storage of digestates with the emission factors being those used for normal storage of solid manure and the storage of untreated slurry with natural crust provided by IPCC (2006). Note that the inventory model calculates NO rather than NOx. The conversion of NO emissions into NO_x emissions is achieved by multiplying the NO emissions with the NO₂/ NO molar weight ratio of 46/30. This relationship also holds for NO and NO_x emission factors.

Table 3 shows the implied emission factors of NH₃ and NO_x for the various animal categories. These emission factors normalize emissions from an animal category as the ratio of the total emission to the respective number of animals.

Table 3: IEF for NH₃ & NO_x from manure management, in [kg per animal place]

Dairy cattle, other cattle and swine are key sources of NH₃ emissions from manure management. The time series of the total NH₃ emissions from all three categories are predominantly driven by the development of the animal numbers, see Table 1. However, the effect of decreasing animal numbers is partly compensated by the continuously increasing animal performance. This leads to increasing N excretions per animal, see Table 2, which, in principle, is reflected by increasing implied emission factors, see Table 3. Increasing dairy cattle emissions since 2010 are also due to a sharp decline of tied housing systems, which have a lower NH₃ emission factor than loose housing systems. For swine the IEF is decreasing over time due to lower raw protein contents in feed and the use of air scrubbing systems that, to a high degree, remove NH₃ from the housings.

For NO_x there are no key categories.

All timeseries of the emission inventory have completely been recalculated. Tables REC-1 and REC-2 compare the recalculated time series for NH₃ and NO_x from 3B with the respective data of last year’s submission. The total emissions of NH₃ and NO_x are slightly higher than those of submission 2023.

One reason for this is that the animal categories deer, rabbits, ostrich, and fur animals are reported for the first time in submission 2024 (recalculation No. 1). Big differences in poultry emissions before 2013 are due to recalculation No. 6 (correction of poultry numbers). The update of activity data for anaerobic digestion (recalculation No. 13) generally leads to higher emissions from cattle and lower emissions from pigs and poultry. Further details on recalculations are described in Vos et al. (2024), Chapter 1.3.

Table 4: REC-1: Revised NH₃ emissions, in kilotonnes

Table 5: REC-2: Revised NO_x emissions, in kilotonnes

No improvements are planned at present.

In 2022, NMVOC emissions from manure management amount to 278.2 kt which is 96.9 % of total NMVOC emissions from the agricultural sector. 84.8 % originate from cattle, 15.2 % from other animals.

The Tier 2 methodology provided by EMEP (2019)-3B-28 was used to assess the NMVOC emissions from manure management for dairy cattle and other cattle. For all other animals the Tier 1 methodology (EMEP (2019)-3B-17) was used. The use of the Tier 2 methodology deliversyields NMVOC emissions which formally could be reported in the sectors 3.D.a.2.a (application of manure to soils) and 3.D.a.3 (grazing emissions). However, to be congruent with the NMVOC emissions for other animal categories, Germany reports these emissions in the NMVOC emissions reported from manure management (3.B). For the NFR codes 3.D.a.2.a and 3.D.a.3 the key note IE is used for NMVOC emissions.

Animal numbers serve as activity data, see Table 1.

For the Tier 2 methodology applied to dairy cattle and other cattle the following data was used:

• gross feed intake in MJ per year, country specific data from the annual reporting of greenhouse gas emissions, see NIR 2024, Chapter 5.1.3.3,
• proportion x_house of the year the animals spend in the livestock building: country specific data, being equal to 1 – x_graz with x_graz the proportion of the year spent on pasture, see NIR 2024, Chapter 19.3.2,
• FRAC_silage: 1 as proposed by EMEP (2019)-3B-29, since silage feeding for cattle is considered dominant in Germany
• FRAC_{silage store}: 0.25 as proposed by EMEP (2019)-3B-30 for European conditions
• EF_{NMVOC, silage_feeding}, EF_{NMVOC, house}, EF_{NMVOC, graz} are taken from EMEP (2019)-3B-32, table 3.11 as 0.0002002, 0.0000353 and 0.0000069 kg NMVOC/MJ feed intake, respectively,
• EF_{NH₃,storage}, EF_{NH₃,building} and EF_{NH₃,application} are taken from the NH₃ reporting (see above and 3.D).

For all other animal categories the Tier 1 emission factors for NMVOC were used as provided in EMEP (2019)-3B-18, Table 3.4. For horses the emission factors for feeding with silage was chosen, for all other animals the emission factors for feeding without silage. Due to missing country-specific emission factors or emission factors that do not correspond to the inventory’s animal categories, the emission factors provided in EMEP (2019)-3B-18, Table 3.4, were used to define specific emission factors for weaners, boars, lambs, ponies/light horses and pullets, ostriches, and deer see Vos et al. (2024), Chapter 4.3.3. The implied emission factors given in Table 4 relate the overall NMVOC emissions to the number of animals in each animal category. The IEFs for dairy cattle and other cattle are much higher than the EMEP Tier 1 EF, which are 17.937 kg NMVOC for dairy cattle and 8.902 kg NMVOC for other cattle. The only possible explanation for those huge differences is that the EMEP Tier 2 and Tier 1 methods are not consistent.

The IEFs for the other categories provided in Table 4 correspond to the EMEP Tier 1 emission factors, except for horses, sheep and swine. These categories comprise subcategories with different emission factors so that their overall IEFs in Table 4 represent subpopulation-weighted national mean values. Note that other poultry in Germany includes not only geese and ducks but also pullets. For pullets no default EF is given in the EMEP guidebook (EMEP, 2019), hence the EF of broilers has been adopted (because of similar housing). This assumption significantly lowers the overall IEF of other poultry in Table 4 the IEFs are listed separately for each poultry category). The IEF of the sheep category is significantly lower than the EMEP Tier 1 emission factor, because for lambs the EF is assumed to be 40% lower compared to an adult sheep in accordance with the difference in N excretion between lambs and adult sheep.

Table 6: IEF for NMVOC from manure management, in [kg NMVOC per animal place]

Dairy cattle and other cattle are key sources of NMVOC emissions from manure management. The total NMVOC emissions from both animal categories strongly correlate with the animal numbers given in Table 1 (dairy cattle: R² = 0.87; other cattle: R² = 0.99).

All timeseries of the emission inventory have completely been recalculated. Table REC-3 compares the recalculated time series of the NMVOC emissions from 3.B with the respective data of last year’s submission. The recalculated total emissions are slightly higher for other animals. This is mostly due to recalculations No. 6 (correction of poultry numbers before 2013) and (to a lesser extent) No. 1(new animal categories), see main page of the agricultural sector). Minor changes for dairy cattle and other cattle emissions are due to changes of NH3 emissions which have impact on the Tier 2 methodology which is applied for cattle NMVOC emissions (especially through recalculation No. 13 (anaerobic digestion)). Further details on recalculations are described in Vos et al. (2024), Chapter 1.3.

Table 7: REC-3: Revised NMVOC emissions, in kilotonnes

No improvements are planned at present.

In 2022, TSP emissions from manure management amount to 64.4 % of total emissions from the agricultural sector. Of these emissions 24.7 % originate from cattle, 32.0 % from pigs, and 42.6 % from poultry.

36.6 % of total PM₁₀ emissions from the agricultural sector are caused by manure management, where 35.7 % originate from cattle, 14.6 % from pigs, and 48.7 % from poultry.

68.6 % of total PM_2.5 emissions from the agricultural sector are caused by manure management, where 78.0 % originate from cattle, 2.3 % from pigs, and 18.0 % from poultry.

EMEP (2013-3B-26) provided a Tier 2 methodology. In the 2019 Guidebook (EMEP, 2019), this methodology has been replaced by a Tier 1 methodology. However, EF for cattle derived with the EMEP 2013 Tier 2 methodology remained unchanged. Therefore, the EMEP 2013⁸⁾ methodology was kept for cattle. For swine the EMEP 2013 methodology was formally kept but the EMEP 2019 Tier 1 EF was used both for slurry and solid based manure management systems. The same was done with the EMEP 2016 EFs for laying hens (used for cages and perchery). In case the EMEP 2019 EFs are simply rounded EMEP 2013 EFs, the unrounded EMEP 2013 EFs were kept. For rabbits the EFs from The Netherlands’ inventory were adopted (Huis In’t Veld et al, 2011)⁹⁾, for ostriches the EFs of goats were used. The inventory considers air scrubber systems in swine and poultry husbandry. For animal places equipped with air scrubbing the emission factors are reduced according to the removal efficiency of the air scrubber systems (90 % for TSP and PM₁₀, 70 % for PM_2.5). For details see Vos et al. (2024), Chapter 4.2.2.

Animal numbers serve as activity data, see Table 1.

Tier 1 emission factors for TSP, PM₁₀ and PM_2.5 from livestock husbandry are provided in EMEP (2019-3B-19), Table 3.5 and 55, Table A1.7. For cattle the Tier 2 emission factors provided in EMEP (2013-3B-29), Table 3-11 were used, because they differentiate between slurry and solid manure systems and were also used to develop the EMEP 2019 Tier 1 emissions factors.

The implied emission factors given in Table 5 relate the overall TSP and PM emissions to the number of animals in each animal category. The Guidebook does not indicate whether EFs have considered the condensable component (with or without).

Table 8: IEF for TSP, PM₁₀ & PM_2.5 from manure management

Swine and laying hens are key sources of TSP emissions from manure management. The total TSP emissions from swine mainly follow the animal numbers given in Table 1 for the earlier years of the time series. However, due to increases in places equipped with air scrubbing and different emission factors of the different housing systems of the four swine subcategories (sows with piglets, weaners, fattening pigs, boars) and the varying population shares in those housing systems the R² of the linear regression is lower than 1 (0.67). For laying hens (R² = 0.98) and broilers (R² = 0.99), due to the low prevalence of air scrubbing systems TSP emissions almost perfectly correlate with the animal numbers provided in Table 1.

The following table shows the effects of recalculations on emissions of particulate matter. Visible differences occur especially in the years before 2013, these are due to the correction of the numbers of laying hens and broilers (recalculation No. 6). The addition of new animal categories to the inventory (recalculation No. 1) shows only a very small effect. See main page of the agricultural sector. Further details on recalculations are described in Vos et al. (2024), Chapter 1.3.

Table 9: REC-4: Revised particle emissions (TSP, PM₁₀ & PM_2.5), in kilotonnes

No improvements are planned at present.

Details will be described in chapter 1.7.