FEEDING A HOT AND HUNGRY PLANET

Tim Searchinger

Princeton University

tsearchi@princeton.edu

Ag’s Contribution to World Greenhouse Gases ~ 30%

(IPCC 2007; Bellarby 2008 , in CO2 eq.)

• Nitrous oxide – fertilizer, manure

• Methane – cows, rice, manure, biomass burning

• Energy – farm machinery, fertilizer, irrigation pumps

• 33 million acres/yr gross deforestation

Non-Ag33.5 Gt

69%

Ag Land Expansion

8.5Gt17%

CO2 from Energy

Use1 Gt2% N2O and

NH46 Gt12%

2004

49 Gt Total

Land Use Change

Biofuels send price signal:

Undernourished People and Recent Changes

Source: FAO. 2009. The State of Food Insecurity In the World, 2008

- 16% of world malnourished- 1/3 of children in developing world stunted- 30 million babies born impaired due to lack of natal nutrition - 5 million children die annually from causes related to lack of nutrition

The Hunger ChallengeThe Hunger Challenge

Source: Keyzer 2006

Food for a Week, Darfur Refugees, Chad© 2007 PETER MENZEL PHOTOGRAPHY

Food for a Week, Germany© 2005 PETER MENZEL PHOTOGRAPHY

Tilman, Forecasting Global Agricultural Driven Environmental Consequences

IrelandUK

DenmarkSweden

Netherlands

Belgium

Germany

France

Rwanda

Zimbabwe

Namibia

Zambia

Egypt

R2 = 0.8788

0

0.5

1

1.5

2

2.5

3

0 2000 4000 6000 8000 10000

Yield (kg ha-1)

Cro

p W

ater

Pro

du

ctiv

ity

(kg

m-3

)

Asia

Europe

North America

South America

Africa

Oceania

Ireland

UK

DenmarkSweden

Eritrea

Chad

Sudan

Mali

R2 = 0.7859

0

0.5

1

1.5

2

2.5

3

0 2000 4000 6000 8000 10000 12000 14000

Yield (kg ha -1)

Cro

p W

ater

Pro

du

ctiv

ity

(kg

m-3

)

Asia

Europe

North America

South America

Africa

Oceania

Potential yield and yield gap at a global scale

GEPIC Outputs: global map of wheat yield

crop water productivity of wheat

Areas of Water Stress

Can we limit LUC for food alone

Optimism

• Big yield gaps• Much food waste• Biotechnology• Livestock intensification

potential

Pessismism

• Declining trend lines• Irrigation limits• Improved knowledge for

tropical forest agriculture• Population? • Climate change (high

temperature; declining snowpack)

IPCC SRES ScenariosPredicted Emissions from Land Use Change

Year B-1 B-2

2020 2.2 0

2050 -0.4 -0.2

Other Uses of Land

• More cropland and pasture for food –200-500 million hectares by 2050

• Terrestrial carbon sink – 9.5 GT CO2/yr

• Avoided deforestation potential – 9 Gt/year

• Afforestation mitigation potential – 4 Gt/year

• Restore peatlands – 1.3 gigatons/year

All from IPCC 2007 Mitigation Report, chapters 8 & 9

Agriculture Mitigation Potential at $100/t

Some Technical Options

Somewhat Known

• Crop & livestock yield gains

• Chickens, tilapia over beef

• Manure biogas

• Improved N use efficiency

• Livestock feeding

• Pasture improvement

• Some rice water management

Innovations Required

• Perennial feed crops & other new cropping systems

• Cover crops?

• Cow stomach bug changes

• More nitrogen crops

• Upland or alternative rice

• New nitrification inhibitors

• Biochar

Conventional approach But . . .

Land grows plants (carbon) anyway

* forest

* food

Only ADDITIONAL plant growth helps

Biofuels & Greenhouse Gases

Direct Effects of Diverting Crops to Biofuels – No Change in Emissions

Corn uptake

Car, gasoline

Corn uptake

Car, ethanol

Indirect Scenario 1 – Ethanol Leads to Less Crop Consumption for Food, which Reduces CO2

Crop uptake

Car, gasoline

Livestock & human

respiration

Crop uptake

Car, ethanol

Reduced livestock & human

respiration

Indirect Scenario 2 – Ethanol Leads to Yield Growth on Existing Farmland to Replace Diverted Crops, which Absorb More Atmospheric Carbon and Reduces CO2

Car, gasoline

Crop uptake

Increased yields, which absorb more carbon on the same land but may

involve increased emissions from inputs, such as nitrous oxide

Car, ethanol

N2O Formation Rate

Greenhouse gases (CO2 eq.)/km

2% 401

3% 603

4% 803

5% 1003

6% 1204

GHGs from Nitrous Oxide For Yield Gains

Through Fertilizer (assuming 7 extra additional lbs of fertilizer/bushel of corn)

Compare 316 g/km from land use

Implied Yield Growth by 2020Scenario 10.2% Transport

Fuel(Etech 4)

Crop Biofuels, adjusted for by products

Non Biofuel food demand

Biofuel and Non Biofuel

1996-2006 Trend

Cereals (corn,wheat 0.8% 1.8% 2.6% 1.3%Oilseeds (soy, rape) 0.9% 2.2% 3.2% 1.5%Sugar (cane) 5.0% 0.6% 5.5% 0.8%

Palm 3.0% 3.9% 6.9% 1.9%

Viewed as Opportunity CostCarbon Benefit• 3 t/ha/yr – corn ethanol – GREET (incorporating by-product

credit)• 8.6 t/ha/yr – cellulosic ethanol – GREET (swtichgrass at 18

t/ha/yr, 359 l/t)Carbon Cost• Fallow land - forest regeneration temperate, 7.5-12 t/ha/yr• Forest regeneration tropical – 20+ t/ha/yr• Existing forest (lose 600-1000 tons) 15-35 t/ha/yr (over 30

years)• Existing grassland/savannah (lose 75-300 tons), 2.5-10 t/ha/yr

(over 30 years) plus lost forage

Bioenergy

How? • Large % of potential arable land – IPCC 2007 chapter

8 (based on Smith 2007) 1.3billion hectares and/or

• All forest growth in excess of harvest (Smeets 2008)and/or

• All “abandoned” cropland (Hoodwijk (2004) and/or

• Hundreds of millions of hectares of “grazing” or “other” land – savannah (Fischer 2001; Smith 2007)

Recounts existing forest, forest re-growth, net terrestrial carbon sink, land counted for grazing

Exemption for Bioenergy

Under a Cap & Trade System

• IPCC 2000 Land Use Report (p. 355): Because “fossil fuel substitution is already ‘rewarded’” by excluding emissions from the combustion of bioenergy, “to avoid underreporting . . . any changes in biomass stocks on lands . . . resulting from the production of biofuels would need to be included in the accounts.”

The Big Issue

Natural Forest

Natural Forest (Melillo, Gurgel, et al. 2008)

Natural Forest (“Deforestation” Scenario)

Natural Forest

Wise et al., Science 324:1183 (2009)

National Academy of Sciences (May 2009)

• "If food crops or lands used for food production are diverted to produce biofuels rather than food, additional land will probably be cleared elsewhere in the world and drawn into food production.  The greenhouse gas emissions caused by such clearing of land, especially forests, will decrease or even negate the greenhouse-gas benefits of the resulting biofuels."  p. 79

• "Producers need to grow biofuel feedstocks on degraded agricultural land to avoid direct and indirect competition with the food supply and also need to minimize land-use practices that result in substantial net greenhouse-gas emissions."  p. 79

Other 2008/09 Studies – Similar Conclusions

• UK Renewable Fuels Agency (Gallagher Review)

• EU Joint Research Center

• World Bank• FAO• Netherlands

Environmental Assessment Agency

• OECD• European Economic

and Social Committee• Scientific Committee

on Problems of the Environment

• British Royal Society