• Welcome
• Introductions
• Plan for the week
• Overview of project
• Where we are now with the EU
• Project Operation
– Project Management Team
– Programme Advisory Committee
– Matched funding
– Annual Project Meeting – COP-15, Copenhagen?
– Logo
– Website
– Communication issues within the project?
– Risk management
• Workplans

• What scales should we focus on?
• Linkage to the negotiation process?
• What do we mean by anthropogenic emissions – e.g. natural peatlands?
• Who are the relevant actors that we need to identify?
• What policy information is useful for modelling?
• How do we take into account leakage, permanence, additionality within the project?
• How do disseminate results wider than just the case study countries?

• C stock estimation – uncertainty
• Mitigation costs – marginal abatement cost curves
• Good knowledge from Indonesia, need to find out more about other study countries
• Vietnam – GHG, forestry inventory, allometric equations, monitoring forest change – RS data

• Quantifying GHG emissions from forest cover change
• Most focus so far on AGB – will improve CO2 emissions estimations from soil, particularly sub-soil (>30cm depth) – all field sites
• Improve methodologies for non-CO2 gases – N2O emissions depends on intensity of subsequent land use (fertiliser use) – focus on Jambi
• Literature review
• Relatively simple models
• Need: detailed land use maps, soils maps if possible – site selection; Jambi - land use history (burning, etc.)
• Forest fire – major CO2 loss
• Deep peat soils (>16m) – subsidence, relative importance of
CH4/CO2/N2O

• REDD hasn’t come out of the blue
• Forest policy architecture – how do REDD policies relate to others?
• What policy options have worked in the past?
• Provide policy options/scenarios for modelling work
• Local level impacts?
• Integrate global/local levels?
• Identification of relevant actors, and their networks
• Peter Minang will be ICRAF representative in this WP.

• Fairness vs. efficiency
• Relationship to COP15?
•
Hotspots
• Discussion tools – needs to be a combination of fairness and efficiency
• Forest use rights for local community
• Improvements in governance in landscape may be more important than REDD benefits
• REDD issues may need presenting in other contexts – different
‘languages’
• Value chains
• REDD ‘projects’ not feasible – need to be higher scales
(additionality, etc.) – what about international leakage?
• Certainty at different scales varies
• How are national level agreements translated into sub-national agreements?

Robin Matthews
Climate Change Theme Leader
Macaulay Institute
Aberdeen AB15 8QH
United Kingdom
REDD-ALERT Kick-off workshop, May 25, 2009

EU-FP7 Project
REDD-ALERT
Reducing Emissions from Deforestation and Degradation through Alternative Landuses in Rainforests of the Tropics
• Macaulay Land Use Research
Institute, United Kingdom
• Université Catholique de Louvain,
Belgium
•
Vrije Universiteit Amsterdam,
Netherlands
•
Georg August University of
Göttingen, Germany
• World Agroforestry Centre, Kenya
• Centre for International Forestry
Research, Indonesia
• International Institute of Tropical
Agriculture, Nigeria
• Centro Internacional de
Agricultura Tropical, Columbia
•
Indonesian Soils Research
Institute, Indonesia
•
Research Centre for Forest
Ecology and Environment,
Vietnam
•
Institut de Recherche Agricole pour le Développement,
Cameroon
•
Instituto Nacional de
Investigacion y Extension
Agraria, Peru

Overall goal: To contribute to the development and evaluation of mechanisms and the institutions needed at multiple levels for influencing stakeholder behaviour to slow tropical deforestation rates and hence reduce GHG emissions

• Documenting the diversity in social, cultural, economic and ecological drivers of forest transition and conservation, and the consequences, in the contexts of selected case study areas in Indonesia, Vietnam,
Cameroon, and Peru as representative of different stages of forest transition in Southeast Asia, Africa and South America.
• Quantifying rates of forest conversion and change in forest carbon stocks using improved methods.
• Improving accounting (methods, default values) of the consequences of land use change for GHG emissions in tropical forest margins including peatlands.
• Identifying and assessing viable policy options addressing the drivers of deforestation and their consistency with policy approaches on avoided deforestation currently being discussed in UNFCCC and other relevant international processes.
• Analysing scenarios in selected case study areas of the local impacts of potential international climate change policies on GHG emission reductions, land use and livelihoods.
• Developing new negotiation support tools and using these with stakeholders at international, national and local scales to explore a basket of options for incorporating REDD into post-2012 climate agreements.

Ucayali, Peru Southern Cameroon Indonesia Vietnam

• Humid tropics: combination of:
– commercial wood extraction
– conversion to cropland
– livestock development
– extension of overland transport infrastructure
• Southeast Asia
– timber concessions, plantations (paper, palm oil)
– slash-and-burn agriculture
• Latin America
– road building followed by migrant settlers practising S&B
– pasture creation for cattle
• Africa (Congo Basin)
– smallholder agriculture
– commercial logging
– fuelwood
(Geist & Lambin, 2002)

Cameroon Benchmark sites
?
Indonesia, Peru
Undisturbed forests
Vietnam
?
Forest frontiers
Forest, agricultural mosaics time
?
Forest, agricultural mosaics, plantations

• C stocks = f(forest area, C density)
• What is a forest?
– Continuum forced into binary classification
– ‘Forest without trees’, ‘trees outside forest’
•
Institutional definition of a forest
– e.g. intent to replant
• Agroforestry, e.g. rubber agroforests
– Hierarchical legends: ‘forest’,
‘tree-based systems’, ‘openfield agriculture’, but possibly more to minimise uncertainty

• C densities – degradation more serious than deforestation?
– Degradation: 300 to 50 t C ha -1
 250 t C ha -1 lost
– Deforestation: 50 to 0 t C ha -1
 50 t C ha -1 lost
– C density data for hierarchical categories  update databases
• Need to incorporate soil changes: time-averaged C stock
• Limiting factor is time series data rather than spatial resolution
(from F. Achard, JRC)

• Tier 1 focus on changes in C stocks in above-ground biomass – plan to quantify root biomass also (to 3 m)
• Soil C – most studies only in top 30 cm – plan to quantify to 3 m
• Significant CO
2 and N
2
O released from soil following conversion of forest
– Measurement and modelling of GHG emissions from ‘hotspot’ land use change
– oil palm plantations, deforested peatlands
• Methodologies to extrapolate to larger spatial and temporal scales
–
Controlling variables: N availability, soil aeration status
–
Can proxy variables be used? e.g. C/N ratio of litter, δ 15 N signatures of litter & soil
– Large soil variability: infra-red spectral analyses of the soils as covariates in the model

• Previous attempts have not always been positive due to:
– ‘Politics of scale’, i.e. decisions/concerns at global level don’t translate to those at a local scale
– Lack of synergy at all levels between forestry and other policies
• Looking for win-win policies – beneficial for GHG emissions reduction as well as local communities
• Aim to synergise the interests of different actors at different levels into a coherent multi-level policy framework
• Analyse trade-offs between CC policies (e.g. biofuels, C sequestration) and the larger goal of sustainable forestry
• Develop a menu of viable policy options to tackle REDD under the international climate regime

• Household agents
– decision-making
– inter-household interactions
• imitation of successful strategies
• buying/selling commodities
• Patches on a landscape
– CENTURY soil organic matter model
–
Ritchie water balance model
–
DSSAT crop models
– Trees model
• Livestock model
Papers
• Matthews, R.B., 2002. Chapter 15 in Crop-Soil Simulation Models: Applications in Developing Countries (R.B. Matthews &
W. Stephens, Editors), CAB International, Wallingford, UK, pp. 209-230.
• Matthews, R.B. & Pilbeam, C.J., 2005. Paper presented at MODSIM05 conference, Dec 9-12, 2005, Melbourne, Australia.
• Matthews, R.B. & Pilbeam, C.J., 2005. Agriculture, Ecosystems & Environment 111(1-4):119-139.
• Matthews, R.B. et al., 2005. Paper presented at MODSIM05 conference, Dec 9-12, 2005, Melbourne, Australia.
• Matthews, R.B., 2006. Ecol. Modelling 194(4):329-343.

• PALM model: Combining agentbased and biophysical modelling
• All instruments can encourage land managers to choose low emission options
•
But
– taxes reduce farmer returns
– incentives require external money
•
Combination of taxes and incentives may be cost-neutral
• Feasibility of carbon trading in the land use sector

• Challenge is how national level targets will be translated to changes in the behaviour of the people who will be expected to reduce deforestation activities
• Salience (relevance)
• Legitimacy – findings by foreign research groups not necessarily accepted – reconfirmation by local scientists – e.g. CO
2 emissions from Indonesia’s peatlands
• Credibility
• Synthesise peoples’ perceptions, fears, ambitions & expectations
– Q-methodology and Analytical Hierarchy Process techniques
• Role-playing games to explore benefit-sharing mechanisms
• Negotiation Support System (NSS) approach rather than
Decision Support System (DSS)

SAMBA-GIS (Vietnam)
• narrative conceptual model
• agent-based spatial computational model (ABM)
• role-playing game
• multiscale GIS
• Stakeholders exposed to other viewpoints of same problem
• ‘Buy-in’ and trust in models by participants more important than numerical accuracy
- Van den Belt (2004)
(Castella et al., 2005)