Net Position Report 2009

• Executive Summary
• 1 New Zealand’s Emissions Profile 1990–2007
• 2 Kyoto Protocol Compliance Equation
• 3 Projections of Emissions and Removals
• 4 Projections Uncertainty
• 5 Reconciliation Against Previous Projections
• References
• Appendix A: Agriculture Emission Projections provided by the Ministry of Agriculture and Forestry
• Appendix B: Projected Balance of Emissions for the Energy, Transport and Industrial Processes Sectors for the Kyoto Commitment Period, 2008–2012
• Appendix C: Land Use, Land-use Change and Forestry Sector provided by the Ministry of Agriculture and Forestry
• Appendix D: Projected Waste Sector Emissions for the First Commitment Period of the Kyoto Protocol

Appendix C: Land Use, Land-use Change and Forestry Sector provided by the Ministry of Agriculture and Forestry

Summary

This report provides projections of carbon dioxide (CO₂) removals and emissions from New Zealand’s Land Use, Land-use Change and Forestry (LULUCF) sector, limited to post-1989 afforestation, reforestation and deforestation activities in exotic planted forests as accounted for under the Kyoto Protocol. There is no information on post-1989 afforestation and deforestation of New Zealand’s indigenous forests and shrublands. Therefore, these activities have not been included in these projections.

The projections cover Commitment Period One (CP1: 2008–2012) of the Kyoto Protocol. The LULUCF projections are an input to the “Net Position Report 2009 – Projected balance of Kyoto Protocol units during the first commitment period” report produced by the Ministry for the Environment. This “net position” report brings together the projected quantity of greenhouse gas (GHG) emissions and removals from all sectors of the economy and is an input to estimating New Zealand’s forecast liability or credit under the Kyoto Protocol during CP1.

Net removals by the LULUCF sector (that is, removals by post-1989 planted forests minus emissions from deforestation) for the period 2008–2012 are projected to be between 46 and 108 Mt CO₂. Net removals for the most likely scenario are projected to be 85 Mt CO₂, which is about 18 Mt CO₂ higher than the 2008 most likely scenario projection of 67 Mt CO₂. This increase is mainly due to new information that indicates post-1989 planted forests are growing at faster rates than previously thought. This new growth rate information has come from a preliminary analysis of forest measurements taken from sample plots during 2007 and 2008, as part of a national forest inventory under the Land Use and Carbon Analysis System (LUCAS) programme.

The large uncertainty range is mainly due to gaps in information and scientific knowledge. There is also uncertainty about the level of deforestation that will take place between 2009 and 2012 and the emissions forecast to come from this deforestation. Uncertainty has been incorporated into the LULUCF projections through the use of scenarios that represent maximum, most likely and minimum emissions (termed “upper emissions”, “mostlikely” and “lower emissions”; Table C1). Monte Carlo simulation was used to analyse uncertainty in the removals calculation. Measurement error, and uncertainty about the accuracy of growth models, are expected to be further reduced as information from the LUCAS programme is gathered between now and 2012. However, until the LUCAS plots are re-measured at the end of the commitment period, the forests have been mapped, and the final development of methodologies to estimate biomass are in place, the LULUCF projections are likely to remain the least certain of all sectors in the net position report.

Table C1: LULUCF projected CO₂ removals and emissions (in million tonnes) during CP1 (2008–2012). See Table C5 for further detail.

Contributing factor	Upper emissions scenario	Most likely scenario	Lower emissions scenario
Mean CO2 removals estimated through Monte Carlo simulation	70.2	92.3	115.4
Less deforestation emissions	–24.2	–7.3	–7.3
Net removals (total removals less deforestation emissions)	46.0	85.0	108.1

Projected deforestation areas in CP1 are based on a Deforestation Intentions Survey undertaken between December 2008 and February 2009. The lower emissions and most likely scenario are based on deforestation forecast under the Emissions Trading Scheme (ETS) policy scenario in the survey. The upper emissions scenario is based on forecast deforestation under a No ETS policy scenario.

Introduction

Under the terms of the Kyoto Protocol to the United Nations Framework Convention on Climate Change (UNFCCC), New Zealand has agreed to take responsibility for its greenhouse gas emissions in Commitment Period One (CP1: 2008–2012).

As forests grow they remove carbon dioxide (CO₂) from the atmosphere (removals). Under the Kyoto Protocol, parties must account for CO₂ emissions and removals by forests established on non-forested land after 31 December 1989 (post-1989 planted forests). Net removals can be used to offset greenhouse gas emissions from other sectors.

This report provides projections of CO₂ removals and emissions from New Zealand’s Land Use, Land-use Change and Forestry (LULUCF) sector, presently limited under the Kyoto Protocol to post-1989 afforestation, reforestation and deforestation. These projections only cover Kyoto-compliant planted forests.

The six key factors used to estimate these projections are:

• the estimated area of post-1989 planted forests (Kyoto forest area)

• forecast afforestation rates

• forecast deforestation rates of planted forests

• post-1989 planted forest growth rates based on a preliminary analysis of Land Use and Carbon Analysis System (LUCAS) forest inventory measurements

• the proportion of exotic forest area planted since 1 January 1990 that may be “ineligible Kyoto forests” (over-planted onto land which was already defined as forest as at 1 January 1990)

• the potential loss of soil carbon following afforestation of grassland.

Assumptions around the likelihood of these factors in the future provide the range of values for the upper, most likely, and lower emissions scenarios.

Forestry trends and drivers affecting forecasts

Forecasts are greatly influenced by recent historic and prevailing conditions. This section briefly summarises the economic and policy environment the New Zealand forest sector has been operating in.

From 2004 until mid-2008, the New Zealand forestry sector faced a high exchange rate, increasing costs (particularly shipping costs), increasing international competition and changing international markets – all of which impacted negatively on forest-growing profitability. More recently, international demand for forestry products has fallen sharply, with lumber exports badly affected by the global economic situation. The domestic forest products market is also forecast to slow further during 2009.

Better returns from alternative land uses, and the greater separation of forest ownership and forest land ownership, have led to the conversion of forest land to other land uses. The area of deforestation accelerated in anticipation that government climate policy would require forest landowners to pay for deforestation emissions from the start of 2008. A survey of forest owners undertaken between December 2008 and February 2009 indicates that intentions for future land-use changes between 2008 and 2020 is forecast to be in the range of 29,000 to 90,000 hectares.

The results of these changes and the perceptions about forestry’s future profitability have resulted in:

• a major decline in the rate of afforestation: from an annual average of 38,000 hectares over the last 30 years, to around 2,000 hectares per years in 2007 and 2008. These are the lowest levels of afforestation recorded since 1945

• forest land being converted to other land uses, particularly dairy farming. It was estimated that approximately 20,000 hectares were deforested in the year ended December 2007, before the Emission Trading Scheme (ETS) legislation was enacted (Manley, 2009). Estimated deforestation for the year ended 2008 under current ETS policy was approximately 3,000 hectares.

Modelling methodology

This report provides scenario-based forecasts (projections) of CO₂ removals and emissions for the LULUCF sector for the period 2008 to 2012. The projections are based on information available as at February 2009 and only cover planted forests.

These forecasts are derived from data and assumptions provided by the Ministry of Agriculture and Forestry (MAF) and the Ministry for the Environment (MfE). The modelling was undertaken by Scion (formerly the New Zealand Forest Research Institute). The underpinning science incorporated in the forest carbon models used in these projections, together with scientific assumptions, come from work carried out by New Zealand’s Crown Research Institutes, predominantly Scion and Landcare Research.

Scientific uncertainty, information gaps and the range of possible future outcomes (such as future afforestation and deforestation rates) are reflected in a scenario-based analysis. The scenarios represent the circumstances expected to result in the maximum, most likely and minimum emissions (termed the “upper emissions”, “mostlikely” and “lower emissions” scenarios). The scenarios include the likely ranges of the major contributing factors that influence planted forest LULUCF sector removals and emissions, based on the current economic conditions, policy settings, land-use statistics, and scientific knowledge. More detailed information on these factors are contained in the section on model assumptions.

The projected post-1989 planted forest removals were calculated using LUCAS field inventory data collected from 273 sample sites. At each sample site, four sample plots were measured. The sample sites were located on a 4*4km grid laid across New Zealand, with sample sites established where the grid intersected with post-1989 planted forest. Because there are still a number of outstanding measurement issues, the analysis must be regarded as preliminary at this stage.

The removals calculation methodology used in these projections is based on the design intended to be used by the LUCAS Calculation and Reporting Application. This largely replaces the previous approach used to project CP1 CO₂ removals, which was based on data from the National Exotic Forest Description and models developed for UNFCCC reporting in the early 1990s.

Carbon stocks were estimated from the plot data using an empirical forest growth model – the 300 Index model (Kimberley et al, 2005) and the carbon allocation model C_change (Beets et al, 1999). The 300 Index model uses the LUCAS inventory data to estimate stem volume growth from establishment to a future harvest age. The C_change model uses the 300 Index generated stem volumes along with forest management information to estimate forest carbon stocks.

A model that links the 300 Index and C_change has been developed in Microsoft Excel; this model is called “Forest Carbon Predictor” (Version 2.1).

The change in carbon stock (tonnes C/ha) from 1 January 2008 to 31 December 2012 has been predicted for each LUCAS plot using the Forest Carbon Predictor. From this the average change in carbon stock per hectare was calculated and multiplied by the estimated total post-1989 planted forest area based on national afforestation statistics collected by MAF. This gives the total forecast change in carbon stock over the commitment period, which is then converted to CO₂-equivalents.

The scenarios modelled included uncertainty around total afforested area, and the adjustment of forest areas to deduct ineligible forest areas planted onto existing forest land (ie, shrubland that met New Zealand’s forest definition). In the latter case, the over-planted proportion was removed from the calculations.

The spreadsheet simulation model described in previous net position reports was still used as a cross check on the Forest Carbon Predictor model forecasts and also to provide estimates of projected removals from post-2007 afforestation, soil carbon, and deforestation.

Model assumptions

Kyoto forest area

Kyoto forest areas have been estimated from national afforestation statistics collected by MAF. These statistics are based on a combination of:

• an annual survey of the number of planting stock sold by forest nurseries (Eyre, 1995). From this survey national estimates of total planting, restocking and new planting are calculated. This survey has been in operation since 1992 and the methodology used was reviewed in 2003 (Manley et al, 2003)

• the National Exotic Forest Description (NEFD) database. The NEFD data is maintained through an annual census of major forest growers and a biennial survey of forest owners with 40 hectares of forest or more. Since the mid-1990s, much of the afforestation that has occurred has been by smaller-scale forest owners, many of them new entrants to forest growing. Obtaining complete statistics from these small-scale owners (using postal survey methods) has been problematic

• Statistics New Zealand estimates of afforestation from the Agricultural Production Survey

• a small reduction in area to allow for known deforestation of post-1989 planted forest area.

In developing regional wood availability forecasts which are based on NEFD forest areas, the small-scale owners’ areas were reduced by 15 per cent. This was done because small-scale owners’ areas are often reported on a gross area basis rather than the actual net stocked area basis. For that reason, the uncertainty range used for the post-1989 afforestation area (much of which is owned by small-scale owners) was set to ±15 per cent.

Future afforestation of exotic planted forests

The average new planting rate over the last 30 years has been 38,000 hectares per year. In the period 1992 to 1998, new planting rates were high, averaging 69,000 hectares per year. Since then, new planting has declined, to around 2,000 ha in 2008.

Figure C1: New forest planting, 1920–2008

Source: National Exotic Forest Description (MAF, 2008)

Text description for figure C1

Table C3 shows the afforestation rates used in the 2009 net position projections.

Table C3: Future plantation afforestation (hectares)

Calendar year	Upper emissions scenario	Most likely scenario	Lower emissions scenario
2008	0	2,000	2,700
2009	0	2,000	4,000
2010	0	2,000	5,200
2011	0	2,000	6,500
2012	0	2,000	17,900
Average (2008–2012)	0	2,000	7,300

The most likely scenario assumes annual afforestation of 2,000 hectares per year during CP1, based on current afforestation levels.

The upper emissions scenario assumes no further afforestation occurs after 2007. This assumption is the same as last year’s projections (worst-case scenario).

The lower emissions scenario assumes average afforestation of 7,300 hectares per year between 2008 and 2012. These rates are based on the projected afforestation rates estimated by Cairns et al (2008) at a carbon price of $25 per tonne (CO₂ equivalent). Afforestation areas include 1,500 per year afforested through the Afforestation Grant Scheme (AGS). This scheme is currently scheduled to run until 2012. The balance of the area is attributable to the ETS.

Afforestation rates may increase once the forestry schemes (Forestry ETS, Permanent Forest Sinks Initiative (PFSI) and AGS) are fully implemented. There are currently only a small number of participants that have registered in these schemes. However, it is expected that when national climate change policy is fully defined and international carbon market trading is more established, afforestation rates would increase in time. The lower emissions scenario takes into account this assumption of increased future afforestation rates, compared with current trends (as at February 2009).

As previously noted, the impact of future afforestation on the amount of CO₂ removed in CP1 is very limited. However, these forests will remove increasing amounts of CO₂ as the forests mature, resulting in larger removals in future commitment periods.

Future deforestation

Since 2004, a clear trend has emerged of not replanting all forest after harvesting and in a number of cases even immature forest has been converted to pasture. These land-use changes have been driven by changing commodity prices between forest products and those from competing land uses, particularly dairy farming. New Zealand has traditionally had dynamic land-use change that is responsive to price signals, so these changes in land use are not unusual. However, prior to 2002 almost all forest was replanted after harvest.

It has been estimated that approximately 20,000 hectares of plantation forests were deforested in the year ended December 2007 (Manley, 2009). The latest Deforestation Intentions Survey forecast deforestation under three scenarios (Manley, 2009):

1. ETS policy (with deforestation liabilities accruing to the forest owner)
2. amended ETS policy (offset planting required⁷; no deforestation liabilities for forest owners)
3. no ETS (no deforestation liabilities for forest owners).

The survey results indicated that deforestation between 2008 and 2012 would be approximately 13,000, 27,000 and 34,000 hectares for each scenario respectively. In previous surveys, it was assumed that all deforestation was of pre-1990 planted forest. In this year’s deforestation intentions survey, forest owners also provided new information on the areas of immature and mature trees that are intended to be deforested. This has allowed a more refined forecast of deforestation emissions. For all the deforestation scenarios, it was assumed that 6,000 hectares of immature post-1989 planted forest would be deforested, with the remainder being pre-1990 planted forest.

All deforestation of pre-1990 planted forest is assumed to be mature radiata pine (28-year-old), releasing approximately 800 tonnes of CO₂ per hectare. Deforestation from post-1989 planted forest was assumed to release approximately 280 tonnes of CO₂ per hectare, assuming an average age of 12 years⁸.

Deforestation estimates do not include indigenous forest or shrubland that meets New Zealand’s adopted Kyoto forest thresholds, as there are currently insufficient national statistics available on the area cleared of either indigenous forest or shrubland (that meets the forest definition). A Landcare Research report provided estimates for indigenous forest and scrub area cleared between 1989/90 and 1996/97, using visual interpretation of ground cover from satellite images (Stephens et al, 2001). Although a complete coverage for New Zealand was not achieved because of insufficient cloud-free images, it was estimated that around 0.03 per cent of the total area of indigenous forest and 0.05 per cent of the total area of scrub were cleared between 1990 and 1996.

However, it is considered that under current legislation no significant deforestation of indigenous forest is likely. Until improved national mapping of forest area and change is available through the LUCAS programme, the actual level of indigenous forest and shrubland clearance remains unknown.

Growth rates

Forest growth rates used in this report were based on a preliminary analysis of the data collected from LUCAS sample plots established in post-1989 planted forests. These new growth rates replace the NEFD-based yield table used in previous net position reports. The preliminary results from the LUCAS plots indicate that post-1989 planted forests have a higher biomass per unit area compared with the NEFD data. This difference seems to be a result of post-1989 planted forests owned by small-scale foresters having received less intensive forest management and so these have higher stockings than those managed by large-scale forest owners. In addition, much of the post-1989 planted forest is established on former farm sites, which are likely to be more fertile than traditional forestry sites.

The estimates of the removals of Kyoto forests for the most likely scenario were calculated by projecting the carbon stock gain during the commitment period for each plot. Each plot was modelled individually using standardised forest management decision rules (Paul et al, 2009).

The lower emissions scenario assumes no silviculture occurs in post-1989 planted forests during CP1 (therefore more forest biomass and removals).

The upper emissions scenario was defined by a 10 per cent reduction in the average carbon stock increase. This represents the lower end of the sampling error for the most likely estimate (6%), with additional allowance made for modelling error and greater losses due to wind damage or disease than are assumed by the growth model.

Ineligible planting

Initial research has suggested that a proportion of the post-1989 exotic planting may have occurred on land that already met New Zealand’s forest definition due to the presence of indigenous shrubland species that had already reached the Kyoto forest thresholds adopted by New Zealand. Under carbon accounting rules, such land does not qualify as Kyoto forest, as the land was already deemed to be forest land on 31 December 1990.

The estimated proportion of “ineligible” post-1989 planted forests used in the 2006, 2007, and 2008 LULUCF projections were 8 per cent (lower emissions), 16 per cent (most likely), and 21 per cent (upper emissions). The proportions for the most likely and upper emissions scenarios have been updated to 12 and 16 per cent respectively, based on a preliminary analysis of newly developed datasets of land cover at 1990 (Kirschbaum et al, 2009). These figures represent the best estimates currently available.

Table C4: Percentage of existing forest (shrubland) ineligible under the Kyoto Protocol

	Upper emissions scenario	Most likely scenario	Lower emissions scenario
Percentage of post-1989 forest planted into shrublands that could already have met New Zealand’s Kyoto forest definition	16%	12%	8%

Confirmed estimates of ineligible post-1989 exotic forest planting will not be available until the LUCAS land-use mapping for the 1990 and 2008 years have been completed and undergone quality assurance. This will provide more definitive data.

Changes in soil carbon

Soil carbon values used in this report are based on the New Zealand Soil Carbon Monitoring System (Soil CMS) model and the soils dataset that will be used in LUCAS.

The Soil CMS model was developed for New Zealand conditions to meet Intergovernmental Panel on Climate Change (IPCC) reporting requirements. This model estimates soil carbon stocks and the forecast change in stock with land-use change (the stock change factor). The Soil CMS model has been determined to be appropriate for meeting soil carbon reporting requirements by an International Review Panel (Ministry for the Environment, 1999) and has been reported in a number of peer-reviewed international scientific publications (eg, Scott et al, 2002, Tate et al, 2005). With the Soil CMS model, LUCAS uses the Historic Soils dataset which has been extracted from the National Soils Database and five other smaller soils datasets. Future refinements are planned including additional data collection to fill gaps in the current dataset and model refinements to reduce uncertainty.

Initial calculations from the Soil CMS model and Historic Soils dataset predicted a soil carbon loss of 18.4 t C/ha for afforestation. This is assumed to occur over the IPCC default transition period of 20 years. This estimate is assumed to be the upper emissions scenario for this year’s projections, as it was in the 2008 projections. A review of national and international studies, and process-based modelling, by Kirschbaum et al (2009) – and the expert judgment of researchers and officials – indicated that this initial predicted carbon loss associated with afforestation may be overstated.

Attempts have been made recently to recalibrate the Soil CMS model in a way that better accounts for the broad differences in soil profiles between typical grassland and forest sites, by weighting apparently spatially auto-correlated grassland data, and by rejecting grassland sites that are a long way from forest sites. Preliminary analyses based on these approaches indicate mean soil carbon losses of between 8 to 13 t C/ha with afforestation. Further refinement of the spatial auto-correlation approach is underway. Based on expert judgement considering all evidence currently available, a soil carbon loss of 11 t C/ha with afforestation was used in the most likely scenario.

The lower emissions scenario assumes no soil carbon change following afforestation, as in the 2008 projections.

Projection results

Table C5 provides a breakdown of the major contributing factors on which the removals and emissions projections are based. Net removals from the LULUCF sector for the period 2008 to 2012 are projected to be between 46 and 108 Mt CO₂. Net removals for the most likely scenario are projected to be 85 Mt CO₂ (compared to 67 Mt CO₂ in the previous year’s projection).

Table C5: LULUCF projected carbon removals and emissions (Mt CO₂-e) during CP1: comparison of the 2008 “most likely” projection with the three 2009 scenarios

Contributing factor	2009 projections			2008 projection (most likely scenario)
	Upper emissions scenario	Most likely scenario	Lower emissions scenario
Removals based on afforestation only
Post-1989 planted forest CO2 removals (based on existing 664,000 ha)	109.7	109.7	109.7	95.5
Future afforestation (2008 to 2012): 0; 2,000; 7,300 ha/year	0.0	0.1	0.2	0.2
Adjustment factors (assumptions see text)
Area of Kyoto forest planted between 1990 and 2007 ± 15%	-16.5	0.0	16.5	0.0
Kyoto forest growth rates	-11.0	0.0	15.2	0.0
Ineligible afforestation	-17.5	-13.2	-8.8	–14.6
Soil carbon change with afforestation	-11.1	-6.6	0.0	–2.9
Mean removals estimated through Monte Carlo simulation	70.2	92.3	115.4	84.1
Emissions from deforestation1,2,3	-24.2	-7.3	-7.3	-16.94
Removals less deforestation emissions	46.0	85.0	108.1	67.2

Notes:

1. The deforestation rates were based on the latest Deforestation Intentions survey results. The most likely and lower emissions scenarios have estimated deforestation emissions of -7.3 Mt CO₂. This is based on the “Current ETS policy” scenario with 13,000 hectares of deforestation in CP1. The upper emissions scenario is based on intended deforestation without an ETS and results in 34,000 hectares in CP1 (-24.2 Mt CO₂).

2. It has been assumed that all forest carbon is instantly emitted upon the deforestation activity taking place.

3. All scenarios include the deforestation of 6000 hectares of post-1989 planted forest, with emissions estimated at approximately 280 t CO₂/ha (assuming an average age of 12 years).

4. The 2008 projections assumed all deforestation was pre-1990 planted forests and resulted in emissions of approximately 800 t CO₂/ha (28-year-old trees). The most likely scenario did not assume all carbon was instantly emitted. Instead, it is assumed that harvesting residues left on site decayed over a 10-year period.

Data limitations

There are acknowledged limitations in the data used in the LULUCF sector projections due to information gaps and scientific uncertainty. MfE commenced the implementation of LUCAS in 2005. LUCAS is being designed to provide more robust inventory data specifically for Kyoto carbon accounting purposes. This is a long-term and large-scale project that will not be fully operational until 2011. LUCAS uses a network of permanent plots across New Zealand’s planted and natural forest. This permanent plot network along with national forest mapping has been designed to provide unbiased national estimates of carbon stocks and carbon stock change for New Zealand’s forests.

Preliminary analysis of LUCAS sample plots in post-1989 planted forests was used in this report. LUCAS mapping products that will allow the estimation of post-1989 planted forest areas, and land-use changes, are not currently complete. Until this information is available, other existing planted forest information such as the NEFD and the Land Cover Databases (LCDBs) will continue to be used for projecting CO₂ removals, even though these data sources were not designed for forest carbon accounting purposes and have known limitations.

The NEFD describes pre-1990 planted forests well (with ownership dominated by large-scale forest growers). NEFD information on plantation forests established by a large number of smaller-scale forest owners since 1992 is of poorer quality. Information on carbon stock changes in New Zealand’s 6.5 million hectares of indigenous forest and 2.6 million hectares of shrubland remains scant (Ministry for the Environment, 2004).

Uncertainty analysis

A Monte Carlo analysis was carried out using @Risk software (Palisade Corporation), as in the 2008 projections. The ranges for afforestation factors in Table C5 were represented by triangular probability distributions, with the upper emission values set to the 97.5th percentile of the distribution and the lower emissions level set to the 2.5th percentile (except for future afforestation where the low value – associated with zero hectares of afforestation – was set as the distribution minimum). The uncertainty analysis used 10,000 iterations to derive the 95th percentile range for CO₂ removals of, which range from 70 to 115 Mt CO₂. Deforestation emissions were then deducted to give an uncertainty range of about 46 to 108 Mt CO₂ (Table C5).

Review of past projections

Since 2005, greenhouse gas projections have been subject to a number of reviews, the most comprehensive being two AEA Technology (United Kingdom) reviews (2005 and 2007). These reviews identified a number of improvements for producing future projections, most of which have been incorporated in the current report. The overall finding of the review of the 2005 projections was that “the methodologies employed to project emissions and sinks across the different sectors are generally sound and reasonable in their approach”. AEA Technology noted the uncertainties are inherent in all countries’ approaches to projecting future greenhouse gas emissions, and that it is “not uncommon” for projections to change on re-analysis. The reviewers recognised that many of their recommendations built upon improvements already in train. AEA Technology’s key conclusions for the LULUCF sector review were:

• methodologies and input assumptions are reasonable and the resulting removal and emission projections are of a good standard

• a single document should be produced for any future projection estimates that provides a detailed basis and sources for all calculations

• four key issues will require further consideration to minimise uncertainty in future projections:

  1. reasons and drivers for the downward trend in new forest planting
  2. the areas of post-1989 forest planting at a national scale into existing shrublands that meet the Kyoto Protocol definition of forest
  3. estimation of areas deforested and drivers for this process
  4. time patterns of loss of carbon soil after afforestation

1. the New Zealand Carbon Accounting System (now called Land Use and Carbon Analysis System) will provide valuable data in assessing removals and emissions for land use, land-use change and forestry.

Of the four key issues above, issues 1 and 3 have been addressed. For issue 1, a report examining the financial returns from forestry and its relationship to forestry planting rates has been published (Horgan, 2007). This report is available on MAF’s website. In respect to issue 3, deforestation intentions surveys have been undertaken yearly since 2005 (Manley, 2005, 2006, 2008 and 2009) examining major forest owners’ deforestation intentions and determining where deforestation is taking place and why. The survey results have been incorporated in the present projections. The 2006 and 2007 deforestation intention survey reports are available on the MAF website.

Issues 2 and 4 are expected to be informed by data and analysis undertaken within the LUCAS programme, though obtaining data for item 4 is very costly since changes are small and highly spatially variable. For further details on LUCAS see

http://www.mfe.govt.nz/issues/climate/lucas/.

References

Baisden WT, Beets P, Carran RA, Clark H, Ford-Robertson JB, Francis GS, Maclaren P, Marshall H, Manning MR, Newton PCD, Saggar S, Tate KR, 2001. An Assessment of the Significance to New Zealand of Article 3.4 Activities under the Kyoto Protocol. Unpublished contract report prepared for the Ministry of Agriculture and Forestry.

Beets PN, Robertson KA, Ford-Robertson JB, Gordon J, Maclaren JP, 1999. Description and Validation of C–Change: A model for simulating carbon content in managed Pinus radiata stands. New Zealand Journal of Forestry Science 29(3): pp. 409-427.

Cairns I, Livesey C and Twaddle D, 2008. Forestry Mitigation Response and Cost Curves. Internal report. Ministry for the Environment.

Eyre J, 1995. Predicting and Measuring New Planting from Nursery Surveys. New Zealand Journal of Forestry, August 1995, pp 45–46.

FRI (Forest Research Institute), 1995. IFS/FOLPI Database and Utilities, Interactive Forest Simulator Manual. FRI Software Series No 19.

García O, 1981. IFS, an Interactive Forest Simulator for Long Range Planning. New Zealand Journal of Forestry Science 11: 8–22.

Horgan G, 2007. Financial Returns and Forestry Planting Rates. Ministry of Agriculture and Forestry report.

Kimberley MO, West GG, Dean MG, Knowles LR, 2005. The 300 Index – a volume productivity index for radiata pine. New Zealand Journal of Forestry 50(2) 13–18.

Kirschbaum M, Trotter C, Wakelin S, Baisden T, Curtin D, Dymond J, Ghani A, Jones H, Deurer M, Arnold G, Beets P, Davis M, Hedley C, Peltzer D, Ross C, Schipper L, Sutherland A, Wang H, Beare M, Clothier B, Mason N and Ward M, 2009. Carbon stocks and changes in New Zealand’s soils and forests, and implications of post-2012 accounting options for land-based emissions offsets and mitigation opportunities. Unpublished contract report prepared for the Ministry of Agriculture and Forestry.

Manley B, Somerville O, Turbitt M and Lane P, 2003. Review of new forest planting estimates. New Zealand Journal of Forestry 43 (3).

Manley B, 2005. 2005 Deforestation Survey. Canterprise, University of Canterbury.

Manley B, 2006. 2006 Deforestation Survey. Canterprise, University of Canterbury.

Manley B, 2008. 2007 Deforestation Survey. Canterprise, University of Canterbury.

Manley B, 2009. 2008 Deforestation Survey. Canterprise, University of Canterbury.

Ministry for the Environment, 1999. International Review of New Zealand’s Carbon Monitoring Project, ME 340, Wellington.

Ministry for the Environment, 2004. New Zealand Land Cover Database version 2.

Ministry for the Environment, 2008. Net Position Report 2008: Projected balance of Kyoto Protocol units during the first commitment period. 15 May 2008.

Ministry of Agriculture and Forestry, 2008. A National Exotic Forest Description as at 1 April 2007.

Paul T, Andersen C, Kimberley M and Goulding C, 2009. Analysis of the 2007-2008 Planted Forest Carbon Monitoring System Inventory Data of Post-1989 Forests. Scion. Unpublished contract report prepared for the Ministry for the Environment.

Scott, NA, Tate KR, Giltrap DJ, Tattershall Smith C, Wilde RH, Newsome PFJ and Davis MR, 2002. Monitoring land-use change effects on soil carbon in New Zealand: quantifying baseline soil carbon stocks. Environmental Pollution, 116, S167–S186.

Stephens P, Heke H, Sutherland A, Sheperd S and Pinkney T, 2001. Estimating Clearance of Indigenous Forest and Scrubland between 1898/90 and 1996/97. Landcare Research. Unpublished contract report prepared for the Ministry of Agriculture and Forestry.

Tate, KR, Wilde RH, Giltrap DJ, Baisden WT, Saggar S, Trustrum NA, Scott NA and Barton JP, 2005. Soil organic carbon stocks and flows in New Zealand: system development, measurement and modelling, Canadian Journal of Soil Science, 85, 481–491.

Wakelin S, 2008. Carbon Inventory of New Zealand’s Planted Forests – Calculations revised in October 2008 for New Zealand’s 2007 Greenhouse Gas Inventory. Scion. Unpublished contract report prepared for the Ministry of Agriculture and Forestry.

Annex 1: Description of the LULUCF net position models

Background

Article 3.3 of the Kyoto Protocol allows the net changes in greenhouse gas emissions (by sources) and removals (by sinks) resulting from afforestation, reforestation and deforestation since 1 January 1990 to be used to meet the commitments of Annex B parties. Emissions and removals are to be measured as verifiable changes in carbon stocks in each commitment period. The carbon stocks to be accounted for are above-ground biomass, below-ground biomass, dead wood, litter, and soil organic carbon pools.

The Ministry for the Environment is implementing the Land Use and Carbon Analysis System (LUCAS) so that New Zealand can meet its international obligations for reporting changes in the LULUCF sector under the Kyoto Protocol. The LUCAS programme will report emissions and removals of greenhouse gases across the carbon pools for land-use changes since 1 January 1990. Key land-use changes include afforestation and deforestation since 1990. The basis for estimating stock changes in the carbon pools will be representative sampling of the Kyoto forest estate taken at the start and the end of CP1⁹ and the mapping of land-use change from 1990 to 2012.

Summary of changes in the modelling approach used for the 2009 projections

During 2007 and 2008 forest inventory plots have been measured across the post-1989 planted forest estate. This is the first time this national forest inventory data has been available. In order to use this new data, a revised approach is used to forecast emissions and removals in this 2009 net position report.

Previous approach: Simulation

Up until this year, projected removals were calculated using a spreadsheet simulation model of the post-1989 planted forest estate. This previous approach used a carbon yield table derived from the National Exotic Forest Description (NEFD) yield tables. This national carbon yield table provides carbon stock estimates by age on a per hectare basis for the four forest biomass pools. All forest areas planted in the same year were modelled as a single forest area for that planting year. The model tracked these planted areas through time and generated annual estimates of carbon stock by multiplying the area at a given age by the carbon yields per hectare for that age (see Table C6). This approach is the same as that employed by routinely-used forest estate planning simulators, such as the Interactive Forest Simulator (IFS) (FRI, 1995, García 1981).

Table C6: Calculating annual carbon stocks from forest areas using a national carbon yield table*

Age	Yield (t CO2)	1990		1991		1992
		Area (ha)	Stock (t CO2)	Area (ha)	Stock (t CO2)	Area (ha)	Stock (t CO2)
0	0.0	x 15,400	0	x 15,800	0	x 50,200	0
1	9.2			x 15,400	= 141,680	x 15,800	= 145,360
2	13.9					x 15,400	= 214,060
3	27.9
..
..
Total area (ha)		15,400		31,200		81,400
Total stock (t CO2)		0		141,680		359,420
Stock increase (t CO2)		0		141,680		217,740

* Cohorts of planted areas “move down one row in the table” each successive year.

For each net position report, the national average carbon yield table created for the most recent UNFCCC national planted forest carbon inventory has been used (Wakelin, 2008). In previous net position reports, this national average carbon yield table was used to calculate both removals from existing planted forest and future afforestation, as well as emissions from all deforestation.

Revised approach

The modelling approach described above was still used to calculate removals associated with future afforestation. The only difference was that a specific post-1989 yield table was derived from the LUCAS plot data for this purpose. This yield table was also used to model deforestation of post-1989 planted forests. The latest NEFD-based national average carbon yield table used in the 2007 UNFCCC planted forest inventory (Wakelin, 2008) was only used to model deforestation occurring in the pre-1990 planted forest.

Removals associated with the existing post-1989 planted forest as at January 2008 were not estimated using the previous simulation approach. Instead, these removals were calculated directly using the LUCAS plot data, total forest area and the LUCAS methodology described in more detail in the next section (LUCAS method).

There was no change in the way soil carbon changes were modelled.

Table C7 summarises the methods used for each contributing factor in the projections.

Table C7: Summary of the 2009 and previous modelling approaches and source of yield tables

Contributing factor	Previous approach		Revised approach
	Methodology	Yield table	Methodology	Yield table
Post-1989 planted forests CO2 removals	Simulation	NEFD-based national average	LUCAS method	Not required
Future afforestation removals	Simulation	NEFD-based national average	Simulation	Derived from LUCAS plots
Area of Kyoto forest planted between 1990 and 2007 ± 15%	Simulation	NEFD-based national average	LUCAS method	Not required
Kyoto forest growth rate – lower emissions	Simulation	300 Index model	LUCAS method with no future thinning	Not required
Kyoto forest growth rate – upper emissions	Simulation	NEFD-based national average minus 10%	LUCAS method minus 10%	Not required
Ineligible afforestation	Simulation	NEFD-based national average	LUCAS method	Not required
Soil carbon change with afforestation	Simulation	Soil carbon estimates	Simulation	Soil carbon estimates
Emissions from deforestation: pre-1990 planted forests	Simulation	NEFD-based national average	Simulation	NEFD-based national average
Emissions from deforestation: post-1989 planted forests	Not modelled11	Not required	Simulation	Derived from LUCAS plots

 

2009 modelling based on LUCAS data and methodology

Post-1989 planted forest CO₂ removals

The projected post-1989 planted forest removals were calculated using LUCAS field inventory data collected from 273 sample sites. At each sample site up to 4 sample plots were measured. The sample sites were located on a 4*4km grid laid across New Zealand with sample sites established where the grid intersected with post-1989 planted forest. There are still a number of outstanding measurement issues that need to be resolved before the analysis of the forest inventory measurements can be finalised. At this stage the analysis must therefore be regarded as preliminary.

The calculation methodology used in these projections is based on the design intended to be used by the LUCAS Calculation and Reporting Application. This largely replaces the previous approach used to project post-1989 planted forests CO₂ removals, which was based on data from the National Exotic Forest Description and models developed for UNFCCC reporting in the early 1990s.

Carbon stocks were estimated from the plot data using an empirical forest growth model – the 300 Index Model (Kimberley et al., 2005) and the carbon allocation model C_Change (Beets et al, 1999). The 300 Index model uses the LUCAS inventory data to estimate stem volume growth from establishment to a future harvest age. The C_Change model uses the 300 Index generated stem volumes along with forest management information to estimate forest carbon stocks. The 300 Index and C_Change have been linked in Microsoft Excel; this model is called “Forest Carbon Predictor” (Version 2.1).

Using the Forest Carbon Predictor the change in carbon stock (tonnes C/ha) from 1 January 2008 to 31 December 2012 has been predicted for each LUCAS plot. From this the average change in carbon stock per hectare was calculated, and multiplied by the estimated total post-1989 planted forest area based on national afforestation statistics collected by MAF. This gives the total forecast change in carbon stock over the commitment period, which is then converted to CO₂-equivalents. Carbon stocks include all living biomass in the crop (planted) trees, dead wood and litter, but exclude soil organic carbon and carbon in the forest understorey.

The scenarios modelled included uncertainty around total afforested area, and the adjustment of forest areas to deduct ineligible forest planted onto existing forest land (ie, shrubland that met New Zealand’s forest definition). In the latter case, the over-planted proportion was removed from the calculations.

The scenarios which used this LUCAS method are indicated in Table C7.

2009 modelling using the previous spreadsheet simulation model

The spreadsheet simulation models described in previous net position reports was still used as a cross check on the Forest Carbon Predictor model forecasts and also to provide estimates of projected removals from post-2007 afforestation, soil carbon, and deforestation.

Post-2007 afforestation calculations

Future afforestation is modelled using the spreadsheet simulation model used in previous net position reports. This model multiplies the area of forest planted each year (post-2007) by the appropriate value in the carbon yield table. This is done for each year in the simulation (1990 to 2012). The yield table used was derived by averaging projections made for each LUCAS inventory plot (described in previous section). This was felt to better reflect post-1989 planted forest growth rates than the NEFD-based yield table used previously.

Each year, carbon stocks are summed across planting cohorts to give the total annual carbon stock. Projected removal units are calculated as the stock change during the commitment period, defined as the stock as at 31 December 2012 minus the stock as at 31 December 2007.

Soil carbon calculations

Soil organic carbon stocks are estimated in a similar way, except that the separate national soil carbon “yield table” reflects changes in soil organic carbon resulting from afforestation of pasture. Both the magnitude of the change in carbon per hectare and the gradual rate of soil carbon change over time (rather than instantly at the time of afforestation) are modelled. Each hectare of land afforested is assumed to start with a high soil carbon stock representing steady-state soil organic carbon under pasture, which then declines before stabilising at a steady-state planted forest level.

Projected removal units are calculated as the soil carbon stock change during the commitment period.

Deforestation calculation for pre-1990 and post-1989 planted forests

Projected emissions from deforestation are determined by estimating the difference between pre- and post-deforestation carbon stocks on land deforested post-1989. Pre-deforestation stocks are calculated by multiplying the area assumed to be deforested by the per hectare carbon stock value in the relevant national carbon yield table at the nominal deforestation age.

For deforestation of post-1989 planted forests the national carbon yield table based on the LUCAS inventory plots was used, assuming an average deforestation age of 12.

For deforestation of pre-1990 planted forest the NEFD-based national carbon yield table developed for the 2007 National Greenhouse Gas Inventory was used, with a deforestation age of 28.

The IPCC methodology allows for an instantaneous loss of carbon at the time of harvest, or a gradual change over time (eg through decay of residues). The spreadsheet model allows either option to be applied, but instantaneous emissions were assumed for this report.

Projected emissions are calculated as the sum of annual emissions over the commitment period. Emissions and removals associated with post-deforestation land uses (ie, agriculture and lifestyle blocks/settlements) have not been modelled.

Limitations of the analysis

Calculations using the LUCAS plots assume that the measured plots provide an unbiased and statistically representative sample of post-1989 planted forests. Of the 400 planned inventory sites 273 have had field measurements taken. The field data will be supplemented with estimates based on remote sensing data (LiDAR) also collected during 2008. Analysis and quality assurance of the LiDAR data has not yet been completed. Once this analysis has been completed there is likely to be changes to the key estimators that underlie this report.

Of the field plots measured, 18 have been abandoned and 22 have been excluded from the calculations due to unresolved issues with the data. There are also some instances of discrepancies between stand history information provided by forest owners and field observations which still need to be resolved. Once all the data issues are resolved and the analysis finalised, it will undergo a formal independent review. Until this has taken place, the analysis needs to be regarded as preliminary and the results are subject to change.

Non-crop tree vegetation carbon has been excluded from the projections. In previous reports, an allowance was made for emissions from the clearance of scrub vegetation for afforestation, and subsequent CO₂ removals from the growth of understorey vegetation. The simple understorey growth model resulted in about 2 per cent of CP1 removals being attributable to understorey growth, and emissions from scrub clearance were of a similar magnitude. However, data to support these assumptions were weak, and neither has been included in these projections. The assumption used in this report is that there is no net change in non-crop vegetation biomass over CP1 as a result of afforestation or deforestation.

The models used to project forest growth are underpinned by an ongoing research programme and are therefore subject to future change. Currently all forest is assumed to be radiata pine. While radiata pine makes up around 90 per cent of the national forest estate there is are issues with using the functions developed for radiata pine for areas planted in other species. The silvicultural treatment each plot receives in future has the potential to significantly change actual net removals from those projected in this report. In the same way, actual afforestation and deforestation rates may be different from the rates forecast in this report. Finally catastrophic events such as forest loss dues to storms, disease and fire may also impact the uptake and emissions of CO₂ by New Zealand’s forests in future.

Annex 2: Forest climate change policy

Introduction

Climate change policy development has been ongoing and changeable for most of this decade. Up until 2002, there was no national climate change policy and New Zealand had not ratified the Kyoto Protocol. The New Zealand Government signed the Kyoto Protocol in 2002 and following the ratification by Russia in 2004, the Kyoto Protocol came into effect in 2005. In October 2002 the Government announced its broad climate change framework with respect to forests. This involved the Government using forest sink credits earned from post-1989 afforestation activities to offset increased greenhouse gas emissions from other sectors, and capping its liabilities for deforestation of pre-1990 planted forests at 21 million tonnes CO₂.

In late 2006, the Government promulgated its Sustainable Land Management (SLM) proposals and entered into a round of consultation with stakeholders. In the SLM proposals, the Government signalled its continued intention to use post-1989 afforestation credits to offset increasing emissions from other sectors of the economy; and that it would encourage afforestation through:

• an afforestation grant scheme

• giving growers a choice between being part of the afforestation grant scheme or devolution of the sink credits (and their associated liabilities) for all post-2007 afforestation

• that both these proposals would work alongside the Permanent Forest Sink Initiative.

Deforestation of non-Kyoto forests would be discouraged by either:

• a flat charge on land-use change from forestry to another use, for the loss of stored carbon

• a tradeable permit regime where the Government allocates tradeable deforestation permits to forest land owners and those who deforest are liable for emissions above the level of permits they hold

• centrally determined deforestation levels where the Government passes legislation to prevent deforestation of land unless government approval has been granted (to ensure total deforestation remains within a government-established target)

• Resource Management Act controls on deforestation: a national environmental standard would require local authorities to prescribe limits for greenhouse gas emissions for the explicit purpose of controlling deforestation.

Development of the Emissions Trading Scheme

After considering feedback from the Sustainable Land Management consultation the Government announced that it would establish a New Zealand Emissions Trading Scheme (ETS). The Government proposed that the ETS would cover all sectors and all gases but the sectors would commence participation within the scheme in a staged manner. Legislation¹¹ to enact the ETS was passed and came into force on 26 September 2008.

Forestry became the first sector to enter the scheme with obligations and entitlements applying retrospectively from 1 January 2008. For owners of non-exempt pre-1990 planted forests this means that New Zealand units (equivalent to 1 tonne CO₂) need to be surrendered to cover any deforestation emissions since 1 January 2008. Owners of post-1989 planted forests are able to receive New Zealand units if they choose to opt into the scheme (noting that owners who received credits would also be liable for CO₂ emissions from harvesting and other events).

The legislation, as it currently stands, also provides for a free allocation of New Zealand Units to owners of eligible pre-1990 planted forest land as part compensation for the land-use restriction placed on this land by deforestation liabilities. A number of deforestation exemptions are also available, for example forest owners with less than 50 hectares of pre-1990 planted forest as at September 2007 can apply to be exempt, as are owners deforesting less than 2 hectares of pre-1990 planted forest during CP1.

2009 review of the ETS

In November 2008, the establishment of a Special Select Committee Review on the Emissions Trading Scheme was announced. The Select Committee is expected to report back by May 2009 in time for amending legislation to be passed in the latter half of 2009. However, until the Select Committee and legislative review processes are complete the existing legislation applies. So while, at the time of publishing, the treatment of both pre-1990 and post-1989 planted forest land remains uncertain, the net position report assumes the application of existing legislation.

The main outcomes of the legislation are:

• post-1989 planted forest owners who opt into the ETS will receive New Zealand Units and therefore the Crown will not be able to use these devolved units to offset increased emissions from other sectors

• the Crown will also devolve 21 million units (minus an allocation for exemptions) in CP1 to owners of pre-1990 planted forests. It is proposed that an additional 0.8-1 million units are allocated for the removal of “weed” trees

• non-exempt owners of pre-1990 planted forest who deforest will be required to surrender New Zealand or Kyoto Units to cover deforestation emissions.

The Permanent Forest Sinks Initiative

The Permanent Forest Sink Initiative (PFSI) provides an opportunity for landowners to establish permanent forest, and obtain tradable Kyoto-compliant emission units (Assigned Amount Units or AAUs) in proportion to the carbon sequestered by their forests.

Accounting for Article 3.4 forest management

New Zealand has opted not to account under Article 3.4 of the Kyoto Protocol, which covers additional LULUCF activities in the first commitment period, such as forest management.

Under Article 3.4, New Zealand would have to account for carbon stock changes in its indigenous forests as well as its pre-1990 planted forests. At present, carbon stock changes in New Zealand’s indigenous forests are not well quantified, although preliminary data suggest that the carbon stocks in these largely old-growth forests are likely to be at steady state (Kirschbaum et al., 2009).

An assessment in 2001 (Baisden et al, 2001) of the significance to New Zealand of Article 3.4 forest management activities, estimated their contribution to be between –92 and +11 Mt CO₂-e over CP1. If New Zealand accounted for forest management under Article 3.4, it would also have been subject to a cap restricting the maximum amount of carbon dioxide removals it could claim in CP1 to 3.7 Mt CO₂. The substantial costs of measuring New Zealand’s entire forest estate did not warrant securing a possible maximum of 3.7 Mt CO₂ over CP1.

 

---

---

 

previous next