C
yit carbon stock of land converted fromland-use x to land-use y on soil-type i at time t years after conversion (Gg C ha-1)
t
years since land-use change to land-use y And this carbon stock was filled in the first formula to calculate the mineral soil emissions involved in another land-use change.This results in net sources of 20.7 (2008), and net sink of 59.5 (2009), 159.4 (2010), 257.4 (2011) and 353.9 (2012) kton CO2 per year for deforestation and a net sink of 27.9 (2008), 29.4 (2009), 27.14 (2010), 24.9 (2011) and 22.7 (2012) kton CO2 per year for reforestation/afforestation. The net sink for deforestation after 2008 is the result of the conversion of a relatively large area of forest to grassland between 1 January 2009 and 1 January 2013. Method of estimating carbon stock change in ARD land in organic soils
The area of organic soils under forests is very small: 11,539 ha (4% of the total peat area), based on the land-use map of 2004. The area of AR land on organic soils was 49 ha in 2012 (0.1% of total AR area) and of deforested land on organic soils was 2,905 ha (5.4% of deforested area) in 2012. The majority of this change (79% for AR and 60% for Deforestation in 2012) was a conversion between Kyoto Forest and agricultural land (Cropland or Grassland). Drainage of organic soils to sustain forestry is not part of the land management nor is it actively done. However, organic soils under forests are indirectly also affected by drainage from the nearby cultivated and drained agricultural land.
Based on the land use-maps of 1990 and 2004, the locations of deforestation and reforestation/afforestation were determined (Kramer et al., 2009) and overlaid with the subsidence map of peat areas. The emissions from organic soils were then calculated using the subsidence rate, the bulk density of the peat, the organic matter fraction and the carbon fraction in organic matter (see Kuikman et al., 2005). For organic soils under
deforestation, the assumption that emissions are equal to the emissions of cultivated organic soils is realistic. For reforestation/afforestation, this assumption is rather conservative, as active drainage in forests is not common practice. For this reason and since no data are available on emissions from peat soils under forest or on the water management of forests, we have assumed that emissions remain equal to the emissions on cultivated organic soils before reforestation/afforestation.
The result of the overlay of the subsidence map of peat soils with the locations of reforestation/afforestation and deforestation (land-use changes from 1990 to 2004) results in area (ha) and emissions (kton CO2). The average CO2 emission from organic soils under reforestation/ afforestation is 23.7 ton CO2 per ha per year and under deforestation 23.9 ton CO2 per ha per year.
Method of estimating nitrous oxide emissions associa- ted with disturbance of soils when deforested areas are converted to Cropland
Nitrous oxide emissions associated with the disturbance of soils when deforested areas are converted to Cropland are calculated using equations 3.3.14 and 3.3.15 of the Good
Practice Guidance for LULUCF (IPCC, 2003) for each aggregated soil type (see mineral soils above). The default EF1 of 0.0125 kg N2O-N/kg N was used. For three
aggregated soil types, average C:N ratios, based on measurements, were available and used. For all other aggregated soil types, we used the default C:N ratio of 15 (GPG p. 3.94, IPCC, 2003). For aggregated soil types where conversion to Cropland led to a net gain of carbon, the nitrous oxide emission was set to zero.
Method of estimating carbon stock change in ARD land due to liming
The liming of forests in the Netherlands might occur occasionally, but no statistics are available. All liming based on quantities of product sold is attributed to agricultural land (Cropland, Grassland), which is the main sector where liming occurs. Liming is therefore reported only for deforested land that is converted to either of these categories. The total amount of liming is reported in sector 5G of the Convention and described in section 7.11. There is no information on how much of the total amount of lime is applied to Cropland and Grassland that are reported under deforestation (as opposed to other Cropland and Grassland). A mean per ha lime application was calculated on the basis of the total amount of lime applied and the total area under Grassland and Cropland. This was multiplied by the total area of Grassland and Cropland reported under Article 3.3 deforestation to calculate the amount of CO2 emission due to liming.
Due to changes in the implementation of the new land-use matrix 2008-2012, the area of Grassland and Cropland reported under Article 3.3 deforestation changed. The emissions from lime application were recalculated to reflect this change.
Statistics on lime application lag behind by one year. The
2012 emissions from lime application were therefore estimated using the 2011 quantities of lime applied, resulting in an emission of 0.89 Gg CO2.
GHG emission due to biomass burning in units of land subject to Article 3.3 ARD
Greenhouse gas emissions (CO2, CH4 and N2O) related to controlled biomass burning in areas that are afforested or reforested (AR) does not occur, as no slash burning, etc., is allowed; they are therefore reported as not occurring (NO). No recent statistics on wildfires are available (only 1980–1992, see Wijdeven et al., 2006). Greenhouse gas emissions (CO2, CH4 and N2O) from wild fires on ARD land are therefore estimated using the Tier 1 method. Average annual area AR land burned was estimated from the historical series of total forest area burned between 1980 and 1992 (on average 37.8 ha, approximately 0.1% of the total area of forest land; Wijdeven et al., 2006) scaled to the proportion of AR to total forest area (approximately 11%–16%; see Table 11.5) and average annual carbon stock in living biomass, litter and dead wood. These estimates are reported in Table 5(KP-II)5 and are subject to
recalculation compared with the NIR 2013, due to the new land-use change matrix 2009-2012, resulting in a change in the proportion of AR to total forest area and the availability of new average annual carbon stock in living biomass data from the NBI6.
Average annual area D land burned was estimated from the same historical series of area burned between 1980 and 1992 (difference between total area and area of forest fire, on average 210 ha; Wijdeven et al., 2006) scaled to the proportion of FAD converted to grassland to total area Grassland (approximately 1.4%–1.9%; see Table 11.5) and average annual carbon stock in living biomass (6.7 t ha-1) in Grassland.
Table 11.5 Estimates area and GHG emissions from wildfires on AR land and D land. Fraction of total area gives the proportion AR to total forest area for AR land and the proportion of area of FAD converted to Grassland to total grassland area for Deforestation.
Year fraction of total
area
area burned (ha) CO2 (Gg) CH4 (Gg) N2O (Gg)
AR 2008 0.112 4.24 0.809 0.004 0.00003 2009 0.130 4.92 0.954 0.004 0.00003 2010 0.139 5.24 1.034 0.005 0.00003 2011 0.147 5.56 1.115 0.005 0.00004 2012 0.156 5.87 1.200 0.006 0.00004 D 2008 0.014 2.99 0.049 0.00023 0.000002 2009 0.015 3.16 0.052 0.00024 0.000002 2010 0.016 3.40 0.056 0.00026 0.000002 2011 0.017 3.64 0.060 0.00028 0.000002 2012 0.019 3.89 0.064 0.00030 0.000002
The estimated GHG emissions for wildfires have a high level of uncertainty due to the uncertain areas of wildfires and the large year-to-year variation in area burned over the period 1980–1992, which was used to estimate an average area.
Forest fires are estimated only for AR land because, after deforestation, all biomass is assumed to have been removed already.
In the Netherlands, wild fires seldom lead to total loss of forest cover and therefore do not lead to Deforestation.
11.3.1.2 Justification for omitting any carbon pool or GHG emissions/removals from activities under Article 3.3 and elected activities under Article 3.4
Carbon stock change due to changes in dead wood and litter in units of land subject to Article 3.3 AR
The national forest inventory provides an estimate for the average amount of litter (in plots on sandy soils only) and the amount of dead wood (all plots) for plots in
permanent forests. The data provide the age of the trees and assume that the plots are no older than the trees. However, it is possible that several cycles of forest have been grown and harvested on the same spot. The age of the plot does not take into account this history or any effect it may have on litter accumulation from previous forests in the same location. Therefore, age does not necessarily represent the time since reforestation/ afforestation. This is reflected in a very weak relation between tree age and carbon in litter (Figure 11.2) and a large variation in dead wood, even for plots with young trees (Figure 11.1).
Apart from Forest, no land use has a similar carbon stock in litter (in Dutch Grassland, management prevents the built-up of a significant litter layer). The conversion of non-forest to forest, therefore, always involves a build-up of carbon in litter. But because good data are lacking to quantify this sink, we report the accumulation of carbon in litter for reforestation/afforestation conservatively as zero. Similarly, no other land use has carbon in dead wood. The conversion of non-forest to forest, therefore, involves a build-up of carbon in dead wood. But as it is unlikely that much dead wood will accumulate in very young forests (having regeneration years in 1990 or later), the accumulation of carbon in dead wood in reforested/ afforested plots is most likely a very tiny sink that is too uncertain to quantify reliably. We therefore report this carbon sink conservatively as zero.
N2O emissions due to nitrogen fertilization in units of
land subject to article 3.3 AR
Forest fertilization does not occur in the Netherlands. Therefore, fertilization in re/afforested areas is reported as NO.
11.3.1.3 Information on whether or not indirect and natural GHG emissions and removals have been factored out
For all article 3.3 AR activities, forests were created only after 1990 and the factoring-out of effects on age structure of practices and activities before 1990 is not relevant. For article 3.3 D activities, the increase in mean carbon stock since 1990 may be an effect of changes in management as well as a change in age structure resulting from activities and practices before 1990. However, it is not known which factor contributes to what extent. There has been no Figure 11.1 Volume of dead wood (standing and lying) in Dutch NFI plots in relation to tree age.
0 50 100 150 200 250 years 0 50 100 150 200 250 m 3 www.prtr.n l
factoring-out of indirect GHG emissions and removals due to the effects of elevated carbon dioxide concentrations or nitrogen deposition. To our knowledge, there is no internationally agreed methodology to factor out the effects of these that could be applied to our data.
This increase in mean carbon stock results in higher carbon emissions due to deforestation. Thus, not factoring out the effect of age structure dynamics since 1990 results in a more conservative estimate of emissions due to article 3.3 D activities.
11.3.1.4 Changes in data and methods since the previous submission (recalculations)
1. A new land-use map for 1-1-2013 is available, allowing the calculation of a the land-use change matrix over the period 2009-2012. Until the NIR 2013 the rate of land-use change was extrapolated from the period 2004-2008. This resulted in changes in the ARD data for 2009, 2010 and 2011.
2. Over the period 2012-2013 the 6th Dutch Forest Inventory (NBI6) was carried out. Based on this new forest carbon stock data are available . Because the methodology was the same as the previous forest inventory in 2000 (MFV), the actual carbon stock changes in living biomass between 2000 and 2013 could be determined. Previously changes in living biomass since 2000 were calculated using a simple forest growth model. Consequently emission factors involving living forest biomass were recalculated. Also the emissions from forest fires were updated, using the new estimates of carbon stocks in living biomass on forest land.
3. CO2, N2O and CH4 emissions from wild fires on D land were estimated and included in this NIR for the years 2008-2012
4. Emissions from liming for 2011 were updated. In the previous NIR fertilizer data were not available for 2011 and therefore 2011 emissions were set equal to 2010 emissions. These fertilizer data have become available and have been used to calculate 2011 emissions. These recalculations correspond with part of the recalculations described in par. 7.4 for the submission under the Convention.
11.3.1.5 Uncertainty estimates
The Tier 1 analysis in Annex 7, Table A7.3 provides estimates of uncertainties of LULUCF categories. the Netherlands uses a Tier 1 analysis for the uncertainty assessment of the LULUCF sector. The analysis combines uncertainty estimates of the forest statistics, land use and land use change data (topographical data) and the method used to calculate the yearly growth in carbon increase and removals (Olivier et al., 2009). The uncertainty analysis is performed for Forests according to the Kyoto definition (par. 7.2.5) and is based on the same data and calculations as used for KP article 3.3 categories.
Thus, the uncertainty for total net emissions from units of land under article 3.3 afforestation/reforestation is estimated at 63%, equal to the uncertainty in Land converted to forest land. Similarly, the uncertainty for total net emissions from units of land under article 3.3
deforestation is estimated at 66%, equal to the uncertainty in Land converted to grassland (which includes for the sake of the uncertainty analysis all Forest land converted to any other type of land use; see Olivier et al., 2009). As a result of recent improvements in both maps and calculations (compare NIR 2009), it is likely that the current estimate is an overestimate of the actual uncertainty.
Figure 11.2 Thickness of litter layer (LFH) in Dutch NFI plots in relation to tree age. LFH measurements were conducted only in plots on sandy soils. 0 50 100 150 200 years 0 20 40 60 80 cm www.prtr.n l
11.3.1.6 Information on other methodological issues
There is no additional information on other methodological issues.
11.3.1.7 The year of the onset of an activity, if after 2008
The forestry activities under article 3, paragraph 3 are reported from the beginning of the commitment period.
11.4 Article 3.3
11.4.1 Information that demonstrates that
activities under Article 3.3 began on or
after 1 January 1990 and before 31
December 2012 and are directly
human-induced
Land use and land-use change is mapped using regularly updated land-use maps covering the whole land area of the Netherlands. Land use maps with dates 1 January 1990,2004, 2009 and 2013 have been used to track changes in land-use on units of land. All ARD activities between 1 January 1990 (map 1 January 1990) and 31 December 2012 (map 1 January 2013) are taken into account.
In the Netherlands, forests are protected by the Forest Law (1961), which stipulates that ‘The owner of ground on which a forest stand, other than through pruning, has been harvested or otherwise destroyed, is obliged to replant the forest stand within a period of three years after the harvest or destruction of the stand’. A system of permits is applied for deforestation, and compensation forests need to be planted at other locations. This has in the past created problems for (local) nature agencies that wanted to restore the more highly valued heather and peat areas in the Netherlands and, as a result, will not allow forest regeneration on areas where it is not intended.
With the historic and current scarcity of land in the Netherlands (which has the highest population density of any country in Europe), any land use is the result of deliberate human decisions.