Petr Havlík & Michael Obersteiner+ >30 collaborators

International Institute for Applied Systems Analysis (IIASA), AustriaInternational Livestock Research Institute (ILRI), Kenya

University of Natural Resources and Applied Life Sciences, Vienna (BOKU), AustriaUniversity of Hamburg, Sustainability and Global Change (FNU), Germany

Soil Science and Conservation Research Institute, Bratislava, Slovakia

Global Perspectives on Agriculture and Forest Mitigation with Emphasis on

Induced Land Use Change

Forestry and Agriculture GHG Modeling Forum, September 27, 2011, West Virginia

2

In low stabilization scenarios LULUCF becomes the single most important GHG emitter by mid-century

For efficient mitigation policy design global and comprehensive tools needed

Global - Design of globally consistent national baselines- Accounting for potential international leakage effects

Comprehensive- Capture co-benefits and leakages across sectors

linked through land

3

Outline

I. GLOBIOM Presentation

II. LULUCF Assessments

III. For further discussion…

IV. Conclusions

4

I. GLOBIOM Presentation

5

Global Biosphere Management ModelBasic resolution: 28 regions

6

Supply functionsimplicit:production system 1 (grass based) productivity 1 + constant cost 1

production system 2 (mixed) productivity 2 + constant cost 2

Demand functions explicit: linearized non-linear functions

Partial equilibrium model (endogenous prices)Agriculture: major agricultural crops and livestock products

Forestry: traditional forests for sawnwood, and pulp and paper production

Bioenergy: conventional crops and dedicated forest plantations

Recursively dynamic (10 year periods)Maximization of the social welfare (PS + CS)

eqqpp /1)ˆ/(*ˆ

International trade

Spatial equilibrium modelTrade flows between individual regions (BACI database, CEPII)

Homogeneous goods assumption - Within a region imported and domestically produced goods are valued equally

no mutual trade - Differences in prices between regions are due to external trade costs

Trade costsTrade barriers (MacMap database, ITC/CEPII)

+ Transport cost (Hummels, 2001)+ Calibration

8

Output: Production Q - land use (change) - water use- GHG, - other environment (nutrient cycle, biodiversity,…)

Consumption QPricesTrade flows

Main exogenous drivers: PopulationGDPTechnological changeBio-energy demand (POLES team)Diets (FAO, 2006)

9

Wood Processing

Bioenergy Processing

Livestock Production

Unmanaged Forest

Managed Forest

Short RotationTree Plantations

Cropland

Grassland

Other Natural Vegetation

Energy products:Ethanol (1st gen.)Biodiesel (1st gen.)Ethanol (2nd gen)MethanolHeat …

Forest products:SawnwoodWoodpulp

Livestock:Cattle meat & milkSheep & Goat meat & milkPork meatPoultry meat & egg

Crops:BarleyCornCotton …

Supply chains

10

LandSimulation Units (SimU) = HRU & PX30 & Country zone

Source: Skalský et al. (2008)

Country HRU*PX30

PX5

SimU delineation relatedstatistics on LC classes and

Cropland management systems

reference for geo-coded data on crop management;

input statistical data for LC/LU economic optimization;

LC&LUstat> 200 000 SimU

11

EPIC

Rain, Snow, Chemicals

Subsurface FlowSurface

Flow

Below Root Zone

Evaporation and

Transpiration

• Weather• Hydrology• Erosion• Carbon sequestration• Crop growth• Crop rotations• Fertilization• Tillage• Irrigation• Drainage• Pesticide• Grazing• Manure

Processes

Major outputs:Crop yields, Environmental effects (e.g. soil carbon, )

20 crops (>75% of harvested area)4 management systems: High input, Low input, Irrigated, Subsistence

Cropland - EPIC

Relative Difference in Means (2050/2100) in Wheat Yields[Data: Tyndall, Afi Scenario, simulation model: EPIC]

13

Forests – G4MStep 1: Downscaling FAO country level information on

above ground carbon in forests (FRA 2005) to 30 min grid

Source: Kindermann et al. (2008)

14

Forests – G4MStep 2: Forest growth functions estimated from yield tables

Major outputs:Mean annual incrementTree sizeSawn wood suitabilityHarvesting cost

15

Livestock Production System ApproachLivestock

16

Livestock Production System Parameters

Livestock

Input parameters

Cut&Carry

Grains

Stover

Grazing

Occasional

Bovines

Sheep & Goat

Pigs

Poultry

Output parameters

Bovine Milk & Meat

Shoat Milk & Meat

Pig Meat

Poultry Meat & Eggs

CH4Manure

17

Herrero, Havlik et al (PNAS forthcoming)

Non-CO2 intensity of milk production

18

II. LULUCF Assessments

19

Model cluster approach

20

G4M - Spatially explicit results

21

Recent applied projects (Highlights)DG Climate Action: EU LULUCF Reference Level for Forest

Management accounting Baseline runs for the construction of country specific

Reference Levels Accounting of emissions from FM will compare development

of emissions from forestry against RL- Reviewed by UNFCCC

DG Climate Action: EU Roadmap for moving to a low-carbon economy in 2050

- Contribution to the impact assessment

DECC (UK, Depatment of energy and climate change), DEA (Dannish Energy Agency)

– Global Forestry Emissions Projections and Abatement Costs– Feeding MACCs for forestry activities into GLOCAF model

World Bank: Congo BasinWWF Living Forest ReportPackard Foundation: USA climate policies international leakage

DO NOTHING scenario – Projected forest area

DO NOTHING scenario – Projected tropical deforestation

TARGET

Zero Net Deforestationand Forest

Degradation by 2020(ZNDD)

REDD policy scenario

Diet Shift Bioenergy Plus Pro-Nature Pro-Nature Plus

Alternative futures scenarios

Diet Shift Bioenergy Plus Pro-Nature Pro-Nature Plus

Diet Shift Bioenergy Plus Pro-Nature Pro-Nature Plus

Diet Shift Bioenergy Plus Pro-Nature Pro-Nature Plus

Kapos et al. (2008)

Total land cover change (2010-2050)

Agricultural commodity prices compared to DO NOTHING

Agricultural input usecompared to DO NOTHING

33

A Roadmap for moving to a competitive low carbon economy in 2050

34

GLOBAL – GHG emissions from agriculture and gross deforestation

Global baseline - globally no additional climate action is undertaken up to 2050. The EU implements the climate and energy package but nothing additional is undertaken.

Global Action - global action that leads to a reduction of global emissions of 50% by 2050 compared to 1990

35

EU27 – Alternative LULUCF emission pathways

36

Packard: International leakage effects of US biofuels policies

Total LUC Fertilizer use Livestock Sub-TOTAL Fossil fuel

replacement TOTAL

BioenScen_50 -32 1769 1478 3182 -266 2916BioenScen_75 -30 1792 1446 3178 -396 2781BioenScen_100 -24 1830 1440 3223 -531 2692BioenScen_125 8 1889 1432 3337 -668 2669BioenScen_150 12 1956 1432 3412 -810 2602

US cumulative GHG emissions from agriculture and LUC over 2000-2050 [MtCO2eq]

preliminary results

37

Total LUC Fertilizer use Livestock Sub-TOTAL Fossil fuel

replacement TOTAL

BioenScen_50 11556 16785 20278 48619 -1329 47290BioenScen_75 11666 16832 20315 48814 -1476 47338BioenScen_100 11745 16903 20322 48970 -1628 47342BioenScen_125 11859 16973 20336 49167 -1781 47386BioenScen_150 11985 17077 20360 49422 -1941 47482

Packard: International leakage effects of US biofuels policies

World cumulative GHG emissions from agriculture and LUC over 2000-2050 [MtCO2eq]

preliminary results

38

III. For further discussion…

W. Africa 1966 – pastoral system 2004 – crop-livestock system

Courtesy of B. Gerard

What is the potential contribution of LPS changeto food security, land sparing and GHG reduction?

40

Total abatement calorie cost (TACC) curves for different policy options by 2030

Herrero, Havlik et al (PNAS forthcoming)

41

IV. CONCLUSIONS

42

Bottom-up modeling of global agriculture and forestry sectors becoming feasible through integration of economic and bio-physical models

REDD still appears as the low hanging fruit

Sustainable intensification can provide benefits in terms of food security, reduced LUC, and GHG emissions

Not only intensification but also reduction of yield volatility can act as land sparing measure in view of extreme weather events

Large uncertainties in very basic datasets need to be properly handled…

43

Thank you!

havlikpt@iiasa.ac.atwww.globiom.org