View all publications

Annex 3: Detailed methodological information for other sectors

A3.1 The agriculture sector

A3.1.1 Uncertainty of animal population data

Details of the surveys and census are included to provide an understanding of the livestock statistics process and uncertainty values. The information documented is from Statistics New Zealand. Full details of the surveys are available from Statistics New Zealand’s website. See for information about surveys and census before 2006 and for information about the 2007 census.

Agricultural production surveys

The target population for the 2007 Agricultural Production Census was all units that were engaged in agricultural production activity (including livestock, cropping, horticulture and forestry) or that owned land that was intended for agricultural activity during the year ended 30 June 2007. The response rate was 87 per cent. These businesses represent 89 per cent of the total estimated value of agricultural output. Statistics New Zealand imputes using a random “hot deck” procedure for values for farmers and growers who did not return a completed questionnaire. The imputation levels for the 2007 Agricultural Production Census are given in Table A3.1.1.

The 2007 Agricultural Production Census was not subject to sampling error. Sampling error arises from selecting a sample of businesses and weighting the results, rather than taking a complete enumeration, and is not applicable when there is a census. Non-sampling error arises from biases in the patterns of response and non-response, inaccuracies in reporting by respondents, and errors in the recording and classification of data. Statistics NZ adopts procedures to detect and minimise these types of errors, but they may still occur and are not easy to quantify.

Table A3.1.1 Imputation levels for the 2007 Agricultural Production Census

Statistic Percentage of total estimate imputed
Ewe hoggets put to ram 11
Breeding ewes, 2 tooth and over 12
Total number of sheep 12
Lamb born to ewe hoggets 11
Lambs born to ewes 11
Beef cows and heifers (in calf) 2 years and over 11
Beef cows and heifers (in calf) 1–2 years 11
Total number of beef cattle 11
Calves born alive to beef heifers/cows 11
Dairy cows and heifers, in milk or calf 16
Total number of dairy cattle 15
Calves born alive to dairy heifers/cows 11
Female deer mated 10
Total number of deer 10
Fawns born on farm and alive at 4 months 10
Area of potatoes harvested 9
Area of wheat harvested 11
Area of barley harvested 12


A3.1.2 Key parameters and emission factors used in the agricultural sector

Table A3.1.2.1 Parameter values for agriculture emissions of nitrous oxide

Parameter (Fraction) Fraction of the parameter Source Parameter value
FracBURN Crop residue burned in fields Ministry of Agriculture and Forestry (expert opinion) 0.3
FracBURNL Legume crop residue burned in fields Ministry of Agriculture and Forestry (expert opinion) 0
FracFUEL Livestock nitrogen excretion in excrements burned for fuel Practice does not occur in New Zealand 0
FracGASF Total synthetic fertiliser emitted as NOx or NH3 IPCC (1996) Reference Manual Table 4.19 0.1
FracGASM Total nitrogen emitted as NOx or NH3 IPCC (1996) Table 4.19 0.2
FracGRAZ Livestock nitrogen excreted and deposited onto soil during grazing See Table 6.3.1 Livestock specific
FracLEACH Nitrogen input to soils that is lost through leaching and run-off Thomas et al (2002) 0.07
FracNCRBF Nitrogen in N-fixing crops IPCC (1996) Reference Manual Table 4.19 0.03
FracNCR0 Nitrogen in non-N-fixing crops IPCC (1996) Reference Manual Table 4.19 0.015
FracR Crop residue removed from the field as crop IPCC (1996) Reference Manual Table 4.19 0.45


Table A3.1.2.2 Emission factor for agriculture emissions of N2O

Emission factor Emissions Source  
EF1 Direct emissions from nitrogen input to soil Kelliher and de Klein (2006) 0.01 kg N2O-N/kg N
EF2 Direct emissions from organic soil mineralisation due to cultivation IPCC (2000) Table 4.17 8 kg N2O-N/ha-yr
EF3AL Direct emissions from waste in the anaerobic lagoons AWMS IPCC (2000) Table 4.12 0.001 kg N2O-N/kg N excreted
EF3SSD Direct emissions from waste in the solid waste and drylot AWMS IPCC (2000) Table 4.12 0.02 kg N2O-N/kg N excreted
EF3PRP Direct emissions from waste in the pasture range and paddock AWMS Carran et al (1995); Muller et al 1995; de Klein et al (2003) 0.01 kg N2O-N/kg N excreted
EF3OTHER Direct emissions from waste in other AWMSs IPCC (2000) Table 4. 13 0.005 kg N2O-N/kg N excreted
EF4 Indirect emissions from volatising nitrogen IPCC (2000) Table 4.18 0.01 kg N2O-N/kg NHx-N
EF5 Indirect emissions from leaching nitrogen IPCC (2000) Table 4.18 0.025 kg N2O-N/kg N leached & runoff


Table A3.1.2.3 Emission factor for Tier 1 enteric fermentation livestock and manure management

Emission factor Emissions Source  
EFGOATS Enteric fermentation – Goats NZ specific 9 kg/head/yr
EFHORSES Enteric fermentation – Horses IPCC (1996) Table 4.3 18 kg/head/yr
EFSWINE Enteric fermentation – Swine IPCC (1996) Table 4.3 1.5 kg/head/yr
MMGOATS Manure management – Goats IPCC (1996) Table 4.5 0.18 kg/head/yr
MMHORSES Manure management – Horses IPCC (1996) Table 4.5 2.1 kg/head/yr
MMSWINE Manure management – Swine IPCC (1996) Table 4.6 20 kg/head/yr
MMPOULTRY Manure management – Poultry IPCC (1996) Table 4.5 0.117 kg/head/yr

A3.2 Supplementary information for the LULUCF sector: the Land Use and Carbon Analysis System (LUCAS)

A3.2.1 Background

The aim of the Land Use and Carbon Analysis System (LUCAS) programme is to develop a robust and comprehensive data gathering, data management, analysis and reporting system that is consistent with IPCC good practice guidance for Land Use, Land-use Change and Forestry (GPG-LULUCF) and designed to:

  • be appropriate for UNFCCC LULUCF sector reporting

  • enable reporting under Article 3.3 of the Kyoto Protocol for the first commitment period

  • support and underpin New Zealand climate change policy development through to 2012 and beyond.

A3.2.2 Approach

Data collection

Data collection is separated into three components; forest, soil and land-use mapping. Data collection methodologies have been designed to provide unbiased carbon estimates at the national scale. The methods are supported by relevant scientific research. Analysis of the data will provide nationally applicable values for carbon stock and stock change for each of the five Kyoto carbon pools. The data collection methodologies and the data analysis approach are being independently reviewed to provide transparency and published in the international scientific literature to ensure the activities within the LUCAS programme are widely understood.

Natural forests

Natural forest permanent inventory plots have been established on a systematic eight-kilometre grid across New Zealand. Collection of the data from these plots occurred over a five-year period and was completed in early 2007. The subsequent results will determine the carbon-status of these managed forests. Currently, quality assurance and checking is on schedule to be completed by July 2009. Required re-measurement work is underway.

Planted forests

A planted forest carbon inventory and New Zealand-specific parameters are being developed for New Zealand to provide an unbiased estimate with known uncertainty for reporting and accounting purposes. Carbon stock estimates will be derived from plot measurements for the four biomass carbon pools. Most – around 90 per cent – of the planted forests species in New Zealand are exotic, primarily Pinus radiata.

Forests planted after the 31 December 1989

the 31 December 1989

Change in carbon stocks in forests planted after 31 December 1989 (called “post-1989 planted forests”) will be determined by measurement of trees within permanent inventory plots established on a four-kilometre grid across New Zealand. A combination of field measurements and airborne LiDAR (Light Detection and Ranging) are being used. The field measurement programme was completed at the end of 2008. LiDAR data capture occurred during February–June 2008. The LiDAR data has yet to be calibrated against the field measurements. For forest plots that are inaccessible, LiDAR data will be processed to provide the total amount of carbon per plot. The total carbon per plot will be separated into the four biomass carbon pools using a specific carbon allocation model. The measurement process will be repeated at the end of the first commitment period, based on the same set of plots. Where new forest planting covers a point on the four-kilometre grid, new plots will be established. The 2008 LiDAR mapping is already completed, while the quality assurance and checking is due to be completed by April, 2009.

Forests planted before 1 January 1990

The LUCAS programme will also establish the change in carbon stocks in New Zealand’s forests planted before 1 January 1990 – these are called pre-1990 planted forests. The post-1989 planted forest approach will be applied, but it is expected that the measurement plots will be on a coarser grid network (on an eight-kilometre grid intersects). It is planned that plot measurements will only be made in the middle of the commitment period, and will be used to forward- and back-cast carbon values to cover the five-year commitment period. Data analysis is scheduled for 2010.


Soil carbon changes very slowly in response to land-use changes. A New Zealand-specific soil carbon model is being used. Soil data for input to the model either exists from previous soil mapping programmes across New Zealand, has been collected under the natural forests work stream, or has been collected through a number of related research programmes and a specific soil sampling programme. Collation of this data and initial analysis was completed at the end of 2008. This data will allow estimates of soil carbon change where land-use changes occur. The key drivers of change are: soil type, climate, topography, and the nature of land-use change. The soils data will be analysed to identify gaps in its coverage across the country. Where significant gaps exist in important land-use areas, further samples and analysis will be performed. The quality assurance/quality checking of soil carbon changes is due to be completed by July 2009. The remaining soil deliverables are being prioritised.

Land-use mapping

The LUCAS system has been designed to achieve the following objectives in relation to land-use mapping:

  • determine changes in land use between 1990 and the start of the first commitment period by providing a New Zealand-wide map of land use at 1990 and at 2008

  • determine changes in land use through the first commitment period by providing a New Zealand-wide map of land use at 2012

  • determine where forests have been harvested, and where deforestation has occurred.

Land use will be mapped as per the UNFCCC categories, namely forest land, cropland, grassland, wetlands, settlements and other land.

Mapping of land use at 1990 is 99 per cent completed and is scheduled to be concluded in April 2009. The 1990 mapping uses a combination of Landsat 4 and Landsat 5, as well as some SPOT-2, satellite imagery and aerial photography. To assist in the interpretation of the 1990 imagery, land use at 2000–2001 is being determined first, and used to track land use back to 1990. The 2000–2001 mapping is using Landsat ETM+ imagery and aerial photography.

Mapping of land use at 2008 is due to be delivered by August 2009. Currently, four of 16 New Zealand regions have been mapped. Mapping of land use at 2008 will use cloud-free, SPOT-5 satellite imagery. This imagery has being captured over the 2006/07 and 2007/08 summers (November to March).

The LUCAS programme will be using medium, spatial resolution (250 m), MODIS satellite imagery to identify the location and timing of forest harvesting. The intention is to create cloud-free MODIS images of New Zealand on an annual basis. As the MODIS is not high enough to map deforestation at one hectare, the actual area of harvesting and deforestation will be determined from high resolution satellite systems or aerial photography. MODIS data will also be used to update forest changes in the 2008 land-use map, particularly those changes that occurred immediately prior to 1 January 2008.

The forest area estimates will be available for use by the New Zealand’s Emissions Trading Scheme (ETS). The ETS is the price-based mechanism established to reduce net greenhouse gas emissions below business-as-usual levels, and to comply with international obligations, including our Kyoto Protocol obligations. The forestry sector was the first sector to enter the ETS – effective 1 January 2008. It is expected that LUCAS-derived, forest area estimates will be used to verify applications to the Forestry ETS and to establish whether deforestation is being satisfactorily reported by forest-land owners.

Database system

A component of the LUCAS programme is the development of a database to store and manipulate all data. The modular database system will achieve the following objectives:

  • provide a transparent system for data and calculations

  • provide a repository for the storing, versioning and validation of plot measurements

  • store static land-use layers and LUCAS-derived polygon layers to determine land use and land-use change area nationally

  • calculate carbon stocks per hectare for land uses and soils based on the plot and spatial data collected

  • produce the common reporting format reports for the LULUCF sector and reporting under Article 3.3 of the Kyoto Protocol.

The calculation and reporting application is under development.

A3.2.3 Statistical design and uncertainty

The statistical methods and assumptions will be independently reviewed to ensure they are consistent with best practice statistical design. Opportunities for ongoing improvement of data collection systems will be sought, while considering the cost-effectiveness of alternatives.

Uncertainty in estimated carbon values will be determined as the data collection, land-use mapping and analysis approaches are developed. The uncertainty information will be used to prioritise future improvements in methodologies and data collection.

A3.2.4 Quality Assurance

A quality assurance framework has being developed for the LUCAS project. This is being implemented across all LUCAS activities. It is consistent with New Zealand’s greenhouse gas inventory QAQC plan and with IPCC good practice guidance (IPCC, 2000 and 2003).