Country inventory: Austria

Sadie Shelton

Overview of Austria’s current Tier 2 approach

Livestock typesTier used for enteric fermentation (CH4)Year adopted*Tier used for manure management (CH4)Year adopted*
Dairy cattleT22003T22003
Non-dairy cattleT22003T22003
SheepT1-T1-
Pigs--T22003
OtherT1-T1-

*Year refers to the year of NIR submission

Livestock categorization method:

Dairy CattleNon-dairy CattleSwine
1 Category8 categories defined by:
Age, physiological status, production system (organic non-organic)
3 categories: young & fattening pigs >20 kg; breeding sows > 50 kg; piglets <20 kg.

Enteric fermentation

Approach used: Intake-based estimate of gross energy

Why was this approach adopted? Before 2003, livestock emissions were estimated using the CORINAIR system for GHG inventories. Because this model is not consistent with IPCC GPG requirement to use a higher Tier approach for key sources, Austria developed a Tier 2 approach in NIR 2003. Available research on nitrogen-flows in livestock systems was used for key sources (i.e. cattle).

Description of approach: Austria’s initial Tier 2 approach was based on research on nitrogen flows in livestock production systems that had been conducted as part of Austria’s compliance with the European Commission’s Nitrates Directive, which limits nitrogen application rates on agricultural land. An EC methodology was applied to estimate the N content of manure based on dietary N intake, N content of livestock products, and gaseous N losses. DMI was estimated on the basis of prior research that used 20-year feeding experiment data to predict feed intake on the basis of nutritional (forage quality and composition, concentrate level) and animal factors (milk yield, live weight, stage of lactation, breed). In the initial version of the N-flow model, crude protein was the main nutritional content of the ration considered. Crude protein content in different diets required to achieve different levels of milk yield enabled estimation of DMI of those diets, and DMI is then converted to GE. The national GHG inventory uses data from statistics agencies on milk yield and live weight to estimate GE. GE is then converted to methane emissions using the IPCC equation (EF=GE*Ym/55.65).

Implementation of the approach: For dairy cattle, GE is estimated from annual statistical data on milk yield. The EF thus changes with fluctuation between years in average milk yield, which is assumed to reflect change in the underlying diet.

Table A: Relationship between energy intake and milk yield for dairy cattle in Austria

Milk yield3500400045005000
GE (MJ GE day-1)214.96227.63240.22252.75

Source: Austria NIR 2017

For non-dairy cattle, diet varies depending on whether they are in organic or non-organic production systems. Typical diets in organic and non-organic systems were characterised for different classes of non-dairy cattle. Expert opinion suggests that typical diets did not change over time, thus GE per animal remains constant in the time series. However, the proportion of cattle in organic and non-organic systems does change. Annual activity data on numbers of cattle of different classes in each production system are used. Thus, the implied emission factor changes year to year, depending on the structure of the cattle population in different production systems.

Table B: Typical diets and gross energy of non-dairy cattle in conventional and organic production systems in Austria

ConventionalSuckling cowsCattle <1 yearCattle 1-2 yearsCattle >2 years
Live weight600 kg210 kg530 kg600 kg
Diet50% green feed
20% hay
30% grass silage
15% green feed
20% hay
30% grass silage
35% maize silage
20% green feed
15% hay
30% grass silage
35% maize silage
40% green feed
20% hay
30% grass silage
10% maize silage
GEI (MJ GE day-1)191.5684.36166.96163.44
OrganicSuckling cowsCattle <1 yearCattle 1-2 yearsCattle >2 years
Live weight600 kg190 kg480 kg580 kg
Diet50% green feed
20% hay
30% grass silage
355% green feed
20% hay
45% grass silage
40% green feed
15% hay
45% grass silage
40% green feed
15% hay
45% grass silage
GEI (MJ GE day-1)191.5672.06151.14159.93

Source: Austria NIR 2017

Inventory improvements:

 ImprovementYear*
Activity data-
Livestock characterization-
Emission factorsRe-estimation of milk yield GE relationship2007
Revision of GE estimates for non-dairy cattle2010
Adoption of IPCC 2006 GL Ym default value2015
Uncertainty estimationReplaced UNCAD literature value with value based on review of statistical data2016

*Year refers to the year of NIR submission

Re-estimation of milk yield GE relationship (2007): 2005 and 2006 inventory reviews suggested improving the relationship between GE and milk yield. The main improvement in the inventory method was a re-estimation of the milk yield-GE relationship for dairy cattle. This was based on research publication (Gruber & Putsch, 2006) and included in the 2007 NIR. The research reviewed actual feed rations based on expert opinion from farm advisors, and forage quality based on field studies in representative grassland and dairy farm areas. The re-estimation led to higher EFs because the revised model considered more indicators of forage composition and quality than the original model, which considered protein only.

Revision of GE estimates for non-dairy cattle: In NIR 2010, new studies on suckler calf growth suggested higher growth than previously assumed and thus higher milk yields to support calf growth. This resulted in changes in the estimated GE per animal in non-dairy cattle systems.

Adoption of IPCC 2006 GL Ym value: Prior to NIR 2015, the IPCC 1996 Ym value of 0.60 was used. In 2014, work focused on revising the agricultural model according to the IPCC 2006 GL, which was reviewed by external Austrian agricultural experts.

Manure management (Methane)

Approach used: IPCC approach (T2 for cattle and swine), T1 for other livestock.

Description of approach: The Austrian Tier 2 approach uses the IPCC Tier 2 model for manure management.

Implementation of the approach:

  • Activity data are taken from national statistics.
  • N excretion rates for the different types of cattle are derived from the model used to estimate GE for enteric fermentation (see above). For non-dairy cattle, VS excretion rates are converted using country specific research on GE intake, digestibility and ash content. For swine, there is no data on performance, and VS excretion rates of swine were kept constant for the whole time series.
  • Values for Bo and MCF initially used IPCC default values, but these were later updated using national research.
  • The fraction of manure handled in different management systems initially used data from an academic study. These were later updated using a new study, and a combination of extrapolation and expert opinion were used to recalculate the time series for each type of MS.

Inventory improvements:

 ImprovementYear*
Activity data-
Manure management systemsNew data on distribution of manure in different management systems2010
Including biogas in management systems2013
Emission factorsRe-estimation of milk yield N excretion relationship2007
Country-specific values for MCF2010
Estimation and re-estimation of biogas MCF2013
Uncertainty estimationReplaced UNCAD literature value with value based on review of statistical data2016

*Year refers to the year of NIR submission 

Re-estimation of N excretion rates: The research used to re-estimate GE values for enteric fermentation in NIR 2007 (Gruber & Putsch, 2006) was also used to re-estimate N and VS excretion values for different types of cattle. A time series for VS was generated based on the times series for milk yield and the distribution of livestock between production systems.

Improvements in manure management system (MMS) data: Austria’s initial inventories noted the lack of national statistics on MMS. NIRs 2003-2009 used data from an academic publication reporting a survey conducted in 1989-1992. Due to lack of alternative data, this data was applied to the whole reporting period 1990-2001. Inventory review reports in 2006 and 2008 noted that the distribution of housing and storage systems has undergone major changes. In 2008, the inventory agency commissioned a review of the estimation method, and a nationally representative survey of MMS conducted in 2005 by a national research project was identified (Amon et al. 2007). To use the survey data on MMS in the NIR 2010, a plausible time series using the earlier survey and new survey data was created using expert opinion for years prior to 2005, and using linear extrapolation for years after 2005. The survey also provided improved information on the timing of storage, which could be used together with measurements of emission factors (see below) to improve emission estimates.

Country-specific values for MCF for liquid systems: The agriculture and education ministries had funded a 3-year measurement campaign on emissions from manure stores. Results were published in peer reviewed publications (1), and were used for MCF values for liquid manure systems in NIR 2010.

Adding biogas storage to the MMS and MCF data: Inventory review in 2013 recommended to include consideration of biogas as a management method. This was done in NIR 2015 using data from different sources for different years. Initially, methane losses were not considered. A centralized expert review recommended to consider this, and the MCF for biogas storage was revised in NIR 2016.

Uncertainty management

Uncertainty of activity data: Prior to NIR 2016, UNCAD was estimated on the basis of a literature value. In 2016, livestock statistics were reviewed. Uncertainties were derived by analysing official Austrian livestock numbers published in June and December each year. Comparing these two data sets the standard deviation was calculated. As a conservative approach the doubled standard deviation was taken, leading to uncertainties for dairy cattle of 2%, for non-dairy cattle of 1%, and for swine of 4%.

Uncertainty of emission factors: In the 2003 inventory, uncertainties for enteric fermentation were estimated using Monte Carlo simulation. Assuming a normal probability distribution, the calculated standard deviation is 4%. This indicates there is a 95 % probability that CH4 emissions are between +/- 2 standard deviations, i.e. between 153 Gg and 178 Gg in the year 1990 and between 138 Gg and 162 Gg in the year 2001.

The Monte Carlo uncertainty method used has the advantage, compared to the default propagation method, that it produces better results if the uncertainty is in a higher range (Winiwarter & Orthofer, 2000). Uncertainties that were taken into account for calculations of the total uncertainty include:

  • Gross Energy Intake (GE): +/- 20% (estimated by expert judgement of Dr. Amon)
  • Methane Conversion Factor (Ym) cattle: +/- 8.3% (IPCC Guidelines, 1997)
  • Livestock: (Source: Statistic Austria; sample survey –) statistical accuracy 95%
  • Share of organic farming: +/- 10% (estimated by expert judgement)
  • EF for Sheep, Swine, Horses, Goats (IPCC default values): +/- 30% (IPCC Guidelines, 1997)
  • The emission factors for the “Tier 2” method are determined by the uncertainty of the gross energy intake (GE) and the CH4 conversion rates (Ym). The uncertainty was estimated to be to be about +/- 20% (Amon et al. 2002).

(1)  AMON et al. 2002, 2006, 2007


Resources

Austria National Inventory Reports 2003, 2010, 2016, 2017.

Amon B, Hörtenhuber S. 2010. Revision of Austria’s National Greenhouse Gas Inventory, Sector Agriculture. Final Report. Division of Agricultural Engineering (DAE) of the Department for Sustainable Agricultural Systems of the University of Natural Resources and Applied Life Sciences (BOKU), study on behalf of Umweltbundesamt GmbH. Wien. (unpublished)

Amon B, Hopfner-Sixt K, Amon T. 2002. Emission Inventory for the Agricultural Sector in Austria – Manure Management, Institute of Agricultural, Environmental and Energy Engineering (BOKU – University of Agriculture, Vienna), July 2002.

Amon B, Kryvoruchkp V, Amon T. 2006. Influence of different levels of covering on greenhouse gas and ammonia emissions from slurry stores. International Congress Series (ICS) No 1293 “2nd International Conference on Greenhouse Gases and Animal Agriculture.”

Amon B, Fröhlich M, Weissensteiner R, et al. 2007. Tierhaltung und Wirtschaftsdüngermanagement in Österreich. Studie im Auftrag des Bundesministeriums für Landund Forstwirtschaft, Umwelt- und Wasserwirtschaft, Wien.

Gruber L, Putsch E. 2006. Calculation of nitrogen excretion of dairy cows in Austria. Die Bodenkultur, Vol. 57, Heft 1–4, Vienna.