Experience of France for reporting under the Kyoto Protocol Etienne MATHIAS (CITEPA) Nicolas STACH (IFN) – Michel-Paul MOREL (SSP) Technical workshop on LULUCF reporting issues under the Kyoto Protocol 13-14 November 2008 13-14 November 2008 JRC - LULUCF

Plan 1. Global scheme of French LULUCF inventory 2. French reporting under the Kyoto protocol 3. Specific issues related to the Kyoto protocol 3.1. Land use monitoring 3.1.1. Annual survey – Mainland 3.1.2. Data treatment 3.1.3. Remote sensing analysis – French Guiana 3.2. Reporting by region - improvements/difficulties 4. Conclusion 13-14 November 2008 JRC - LULUCF

1. Global scheme of the LULUCF inventory France applies the 2003 Guidelines for LULUCF since the submission of April 2006 for the UNFCCC. The discussions on the implementation of the reporting under the Kyoto protocol have begun in the same time. The inventory is based on land use change matrixes which represent all the possible land use changes. For each of this land the net carbon balance is estimated for each carbon pool. Forestland Grassland Cropland Settlements Wetlands Other lands Aboveground biomass Underground biomass Litter Dead wood Soil 13-14 November 2008 JRC - LULUCF

1. Global scheme of the LULUCF inventory On forestland the carbon balance for living biomass is estimated thanks to the method of fluxes. The net balance is estimated by difference between the growth and the harvests. Net carbon balance = carbon growth – carbon harvests For the dead organic matter and the soils, a tier 1 default method is used; emissions and removals are reported when there are land use changes. ∆Carbon = Carbon before conversion - Carbon after conversion A A=>B B=>A Dead organic matter 1year 20years Soil B 13-14 November 2008 JRC - LULUCF

2. French reporting under the Kyoto Protocol France chose to submit annually the estimate of the figures for the articles 3.3 and 3.4 under the Kyoto Protocol. France elected the only Forest management option under the article 3.4. The cap for France for this option is 0.88tC/year (3.2 tCO2/year) for the commitment period 2008-2012. A trial on the year 2006 within the Kyoto format was submitted to the UNFCCC for the 15th of April 2008. 13-14 November 2008 JRC - LULUCF

3. Specific issue related to the Kyoto Protocol The reporting under the UNFCCC doesn’t perfectly correspond to what is demanded for the Kyoto Protocol especially on two points according to France: Areas have to be monitored from 1990 to the end of the commitment period The reporting must be done for smaller regions than the entire territory. 13-14 November 2008 JRC - LULUCF

3.1. Land use change survey Many ways have been imagined to be able to report under the Kyoto Protocol with : accuracy (0.5 ha of deforested forestland must be detected), completeness (all the different land uses, forestlands, croplands, grasslands… must be considered) representativeness (the entire territory must be covered). France decided to use a national survey based on fixed and systematic sample plots which is named TERUTI. Since 2005, it is named TERUTI-LUCAS because of the harmonization with the other European countries within the framework of LUCAS. In French Guiana, this survey was not available a remote sensing based study has then been implemented. 13-14 November 2008 JRC - LULUCF

3.1.1. TERUTI concept An annual area frame sampling survey A two-stage systematic sampling design PSUs (Primary Sample Unit) of 324 ha SSUs (Secondary Sample Unit) of 9 m2 A double observation : for each SSU, the surveyor has to determine, on the spot, land cover and land use (2 nomenclatures). land cover : wetland, broadleaved forest, wheat, roads … land use : agriculture, forestry, residential, leisure … A location based on aerial photographs field documents used since 1962 13-14 November 2008 JRC - LULUCF

Altogether, more than 15,500 PSUs et 555,000 SSUs visited every year 3.1.1. TERUTI sample A grid of 4,700 elements covers the whole territory 4 to 8 PSUs of 324 ha in each grid element 36 SSUs observed in each PSU Altogether, more than 15,500 PSUs et 555,000 SSUs visited every year 13-14 November 2008 JRC - LULUCF

3.1.1 TERUTI sample The SSU represents a surface of 100 ha 13-14 November 2008 JRC - LULUCF

3.1.1. LULUCF nomenclature for France 13-14 November 2008 JRC - LULUCF

3.1.1 From TERUTI to LULUCF 13-14 November 2008 JRC - LULUCF

3.1.1. LUCF matrix from 2003 to 2004 13-14 November 2008 JRC - LULUCF

3.1.1. Teruti-Lucas PSUs 13-14 November 2008 JRC - LULUCF

3.1.2. Creation of matrixes to implement the inventory For the UNFCCC inventory different matrixes are needed for each year of inventory to separate the land keeping the same land use and the land changing of land use: 1 Land use change matrix for a period of 20 years 1 Land use change matrix for a period of 1 year With the Kyoto reporting another matrix is needed to separate the areas under the articles 3.3 and the 3.4: 1 Land use change matrix for the period between 1990 and the inventory submission. 13-14 November 2008 JRC - LULUCF

3.1.2. Creation of matrixes to implement the inventory All those matrixes are very difficult to implement for two main reasons: The 1 year matrixes are used especially to detect deforested areas but according to the survey it is difficult to estimate if a change is temporary or permanent. For example when a forested area is cut for wood production it is not possible to know if there will be a land use change or if the forest will come back. The choice made by France is to consider that if the area remains more than five years without forest it is counted as deforested. A special treatment has then been applied to estimate this deforestation but it means that you can estimate the deforested areas only after five years. There are different time series of the TERUTI database (1982-1990; 1992-2004; 2005-…) the differences are some definitions of categories, a different number of sample plots and different places of plots. Consequently, it is not possible to have the monitoring of one plot from 1970 to today. Treatments are then needed to build the 20 year matrixes To make consistent the different time series of TERUTI. To estimates all the changes during the period. 13-14 November 2008 JRC - LULUCF

3.1.3. Case of the French Guiana TERUTI doesn’t cover the entire territory, and especially not the department of French Guiana which is essentially covered by rainforest. Moreover there is currently no forest inventory in this area. 2 specific studies have then been implemented within the framework of the Kyoto Protocol 1 study to estimate the area deforested 1 study to estimate the biomass released with this deforestation. 13-14 November 2008 JRC - LULUCF

3.1.3 Context : French Guiana French overseas “Department” ~84 000 km² Cayenne Brazil Surinam Oyapock Maroni Managed forest National Park Petit Saut Human activity concentrated in a narrow coastal strip and along main rivers French Guiana forest ~80 000 km² of tropical forest. ~ 95% of the total territory ~ 1/3 of total French forest (including mainland) Deforestation causes in French Guiana Traditional shifting cultivation: Small individual areas (0.3 to 1 ha) Extension of modern agriculture Gold mining: Small individual areas - Legal or illegal – About 12 000 ha since 1990 - Along streams and rivers Petit Saut dam: Hydroelectricity – reservoir about 30 000 ha, Dam filling in 1994 Limited deforestation concerning small parcels : a challenge for monitoring 13-14 November 2008 JRC - LULUCF

3.1.3. French Guiana and Kyoto protocol Specific constraints in French Guiana France is the only Annex 1 country with tropical forests No preexisting systematic and exhaustive land use survey Very recent and partial Land use inventory from the Ministry of agriculture (TERUTI LUCAS) French NFI permanent inventory of forest resources is limited to mainland France Impossibility to carry out a systematic ground based survey No exhaustive Aerial photography mission Important cloud cover 13-14 November 2008 JRC - LULUCF

3.1.3. Satellite Data : LANDSAT 1990 7 Images from Global Landcover Facility (GLCF – Uni. Maryland) + 1 image USGS All Images around 1990 From 1986 to 1992 62% of the territory covered by 1990 images Mean acquisition date October 1989 Landsat images georeferenced again using SPOT orthos as references 24 to 49 tie points between each Landsat scene and the corresponding SPOT images Bilinear model between Landsat Coordinates and SPOT Coordinates Geometric accuracy from 19 to 27 meters 13-14 November 2008 JRC - LULUCF

3.1.3. SEAS-Guyane Multiple partner Project leaded by IRD Technologic platform for acquisition and processing of High resolution satellite data (SPOT et ENVISAT) 2005: Installation of a new SPOT/ ENVISAT Ground Receiving Station in Cayenne The Antenna The receiving area Dramatic Increase of SPOT image acquisition capacities over the region 13-14 November 2008 JRC - LULUCF

3.1.3. Annex SPOT Receiving station Network 13-14 November 2008 JRC - LULUCF

3.1.3. Partenaires Inventaire Forestier National IGN Espace Porteur du projet / Photo-interprétation / Articulation avec inventaire Kyoto / Appui méthodologique et technique au traitement d’image IGN Espace Spatiotriangulation, géoréférencement IRD Unité Espace Acquisition et Traitement des images Transfert local (résultats et applications) Cemagref UMR TETIS Appui scientifique (traitement d’images et protocole de Kyoto) ONF Direction régionale Guyane Expertise forestière locale / Aide à la stratification 13-14 November 2008 JRC - LULUCF

3.1.3. Satellite Data : SPOT 2006 Several thousands SPOT scenes acquired at the Cayenne Ground station during 2006 SPOT 5 SPOT 4 SPOT 2 Cloud cover < 75 % Cloud cover < 25 % Cloud cover < 10 % Selection of 2 to 7 images per frame (depending on the cloud coverage) 65 SPOT 2 20m multispectral 63 SPOT 4 20m multispectral 47 SPOT 5 10m multispectral Mean acquisition date : September 2006 11/03/08 Assessment of forest cover changes in French Guiana using satellite imagery

3.1.3. Definition of the strata Sources : ONF levés terrain lors de missions aéroportées exploitation légale et clandestine Sources : ONF DIREN Source: ARUAG Source: DAF 2005 Source : Agence Régionale d’Urbanisme et d’Aménagement de la Guyane (ARUAG) 2003 Source: ONF Source: ONF Source: ONF ONF GIS analysis on existing information Roads Agriculture Human settlements Cities Urban areas Gold mining Maroni villages Camopi & 3 Sauts Villages 2km Buffer → Strata R definition 13-14 November 2008 JRC - LULUCF

Sample points distance 3.1.3. Sampling design Localization of the 3 strata Systematic/random grid Square grid - 932 m between points Random origin and inclination Sample definition Strata R : All Grid points Strata P : All Grid points Strata N : sub-sample 1 point out of 9 Strata Total Area Sampling type Sample size Sample points distance N 7 021 597 Normal 973 8388 P 212 641 Reinforced 2453 932 R 1 162 273 13360 Permanent sampling : same sample in 1990 and 2006 Good estimation of Land Use areas AND Land Use Changes 11/03/08 Assessment of forest cover changes in French Guiana using satellite imagery

3.1.3. Photo interpretation Shifting Cultivation Gold mining Landsat 5 TM 1990 - GLCF SPOT 5 2006 – © CNES SPOT 5 2006 – © CNES Landsat 5 TM 1990 - GLCF Petit Saut Reservoir Agriculture extension SPOT 4 2006 – © CNES Landsat 5 TM 1990 - GLCF Landsat 5 TM 1990 - GLCF SPOT 4 2006 – © CNES 13-14 November 2008 JRC - LULUCF

3.1.3. Results: Land uses 1990 & 2006 Land use 1990 / 2006 13-14 November 2008 JRC - LULUCF

42 050 ha (standard error 4%) 2 486 ha/year 3.1.3. Results Analysis forest changes in French Guyana 1990-2006 Total forest / non forest conversion: 94 061 ha – standard error 12 % 3 181 ha 16 947 ha 35 064 ha 17 403 ha 10 379 ha 11 087 ha 5 000 ha 10 000 ha 15 000 ha 20 000 ha 25 000 ha 30 000 ha 35 000 ha 40 000 ha Usual deforestation Punctual event 1994 std. err. 4% Natural changes St. err. 67% 42 050 ha (standard error 4%) 2 486 ha/year Forest 1990 / Gold mining 2006 Forest 1990 / Other settlements 2006 Forest 1990 / Croplands 2006 Forest 1990 / Other Land uses 2006 Forest 1990 / P. Saut Reservoir 2006 Mangrove 1990 / sea 2006 0 ha 13-14 November 2008 JRC - LULUCF

3.1.3. Emission factors ONF / CIRAD/ CNRS Dendrometric expertise Dispositifs Biomasse épigée totale des arbres de plus de 10cm > Volume sur écorce, densité, facteurs d’expansion > Terre ferme 350 T/ha ± 25 > Marécages 290 T/ha ± 30 • Autres compartiments - étude bibliographique : > Biomasse épigée vivante autre que les arbres > 10 cm : 15 à 30 T/ha ; > Biomasse épigée morte (litière et bois mort) : 20 à 40 T/ha ; > Matière organique du sol : 75 à 100 TC/ha 13-14 November 2008 JRC - LULUCF

3.2. Reporting by regions In the framework of the Kyoto protocol, France has to report by smaller geographical units than the entire country. The administrative level of the regions has been chosen for this reporting which means that 26 results are reported for the LULUCF sector for the 26 different regions (with oversea territories). It has been an opportunity to take more specific parameters Many parameters and data don’t exist with this accuracy and must be estimated It represents a lot of work. 13-14 November 2008 JRC - LULUCF

3.2. An opportunity for improvements One of the base of the LULUCF inventory is the cross between the land use data (TERUTI) and the Forest inventory data (IFN), the results of land use survey can perfectly be given by region the result for => France is the sum of the regional data, of course the accuracy is lower when we just look at the regional level, but it doesn’t change anything for the total. The IFN was able to provide data of growth and carbon stocks by super regions (5 super regions in France). => This changes a lot the results compared to the previous submission to the UNFCCC and should improve the relevancy of the result. The data for soils were also improved with updated data by region for each type of land use. => This changes the results too. 13-14 November 2008 JRC - LULUCF

3.2. Estimation of unknown regional parameters The harvests are well known by department (smaller geographical units than regions) except for fuel wood, the national consumption of wood fuel is estimated thanks to the energy data to not forget non commercial consumption. But this data is not available by region each year. Moreover it is very difficult to estimate the transport of fuel wood between regions. => For the last test submission we considered that there is no transport of fuel wood between regions but inconsistencies can appear and corrections could be necessary. 13-14 November 2008 JRC - LULUCF

4. Conclusion on the submission within the Kyoto format This regional reporting demands a lot of work but has been an opportunity to look for more accurate data. It is difficult to estimate the relevancy at the level of the region but French local actors feel really more concerned with this level of result. The monitoring of the surfaces remains a very difficult issue especially to have time and geographical consistency (especially because the Kyoto Protocol put the stress on afforested and deforested areas which are very small areas compared to the lands without change) No big trouble was met with the filling of the new reporting tables. 13-14 November 2008 JRC - LULUCF

4. Conclusion on the submission within the Kyoto format France was very involved to estimate the balance of the article 3.3, and the first estimate for the year 2006 gave a net sink of 23.000 tons of carbon which means clearly nothing according to the level of uncertainty (impossible to know whether the future balance will be positive or negative). If the trends and the methodology were completely stable we could assume that the balance would be positive for the commitment period but it is currently not the case. It is very difficult to communicate with these different estimates under the Convention and the Kyoto Protocol, for example because the afforested areas in the Kyoto Protocol (since 1990) are not the same than in the Convention reporting (for 20 years). The future French submissions for the UNFCCC and the Kyoto Protocol should be completely harmonized and consistent but there is no way to check this consistency. 13-14 November 2008 JRC - LULUCF