University of California, Santa Cruz August 8, 2011
50% increase in population by 205070% increase in food demand by 205040% increase in energy demand by 2030
The challenge is not simply meeting increased demand, but doing so sustainably.
Emerging policy innovationsSynergies with poverty alleviationPerhaps better to ask How?
How much land is available?What are the life-cycle impacts?What is bioenergy precarious role in the climate-energy nexus?
*1) Abandoned agriculture areas from historical land use data (HYDE, SAGE)
2) Exclude agriculture-to-forest and agriculture-to-urban (MODIS12C1)
3) High estimate of potential yields from ecosystem model (CASA)
4) Regional bioenergy potential on abandoned agriculture lands.
*(Campbell et al., ES&T, 2008)
Algae bioenergy sustainability (Wiley, Campbell, McKuin, WER, 2011)Wastewater co-benefitsEfficient harvesting with electrocoagulation and electrofloculation(Trent, 2010)
Electrocoagulation / ElectroflocculationSurface charge analysis of algae
(Wiley, Campbell, McKuin, WER, 2011)
A global resource Abandoned AgricultureRegional opportunities Mountaintop MiningNo land use Offshore AlgaeNot commercially viable yetElectrochemical approach is emerging
(Campbell et al., Science, 2009)
(Campbell, Lobell, & Field, Science, 2009)*
(Campbell, Sloan, Snyder, et al., In Prep)
Converting Brazilian residue to electricity has greater GHG benefits than conversion to ethanolResidue-based ethanol has small impact on US energy security but electricity would have massive impact on Brazilian energy security(Campbell & Block, ES&T, 2010)
(Campbell et al., In Prep)
(Casillas and Kammen, Science, 2010)
(McKuin & Campbell, In Prep)
(Pacala and Socolow, Science, 2004)
Role in stratosphere (Crutzen, 1976)A novel tracer of carbon sequestration?
(Campbell et al., Science, 2008)
(Campbell et al., In Prep)
Rapid growth with or without sustainability basis.Resources available for a multi-disciplinary approach to bioenergy research and education.Many opportunities for engaging with industry, policy, and mass media.
NSF/CAREER (Envl Sustainability Program #0955141)DOE/Institute for Climatic Change (#050516Z30)Stanford/Carnegie: Chris Field, Joe Berry, David LobellIowa: Jerry Schnoor, Greg CarmichaelNASA: Stephanie Vay, Randy KawaWonderful Students! Andrew Mckuin, Brandi McKuin, Chi-Chung Tsao, Patrick Wiley, Xianyu Yang
Question: What are the life-cycle GHG emissions of ethanol (g CO2e MJ-1)?Objective: Team presentations in 30 min (~4 slides)Materials: http://faculty.ucmerced.edu/ecampbell3/ucsc/Approach: Modify a widely referenced LCA model (Farrell et al., Science, 2006) with updated informationTeam 1: Crutzen et al. (N2O)Team 2: Plevin et al. (Feedstock location)Team 3: Searchinger et al. (Indirect land-use)Team 4: Fargione et al. (Direct land-use)
209.631.9312 | firstname.lastname@example.org
(Tilman , 2009)
(Raupach et al., PNAS, 2007)
(Raupach et al., PNAS, 2007)
MORTALITYHOSPITAL ADMISSIONMORTALITYHOSPITAL ADMISSION(Campbell, et al., In Preparation)
(Campbell et al., In Preparation)
(Fox & Campbell, ES&T, 2010)
Why this is the right place to look? Can we find win-win solutions?What is missing is the natural uptake by plants**Here we present a new spatially explicit estimate of the global area of abandoned agriculture and the plant production on these lands using historical land use data, satellite-derived land cover, and global ecosystem modeling. We considered abandoned agriculture as land that was previously used for crop or pasture but is now abandoned to these uses and has not been converted to forest or urban areas. We used historical land use data from the History Database of the Global Environment 3.0 (HYDE)10 which consists of a time series of global maps of permanent crop and permanent pasture for each decade between 1700 and 2000. Shifting agriculture, which also contributes to abandoned agriculture lands18, is not included in these maps. For comparison with this HYDE-based analysis, we considered areas of abandoned crop from the Center for Sustainability and the Global Environment (SAGE) land use database (0.5 degree resolution)11,26.
In order to exclude forest and urban areas we used a MODIS satellite map of the current forest and urban land cover (3 minute resolution, MODIS/Terra Land Cover Types MOD12C1)12. Spatial analysis between HYDE and MODIS data should introduce uncertainty due to the different methods used in the creation of these data.
We developed a high estimate for the potential plant production on abandoned lands using natural plant production from the Carnegie-Ames-Stanford Approach (CASA) ecosystem model (1 degree resolution)13. At the global scale, primary production on agricultural lands, determined from harvest statistics, is about 65% of the modeled production from natural vegetation on the same lands27. Estimating biomass yield as the potential primary production of the natural vegetation on a site reflects local constraints from climate. It also allows that total plant production from biomass agriculture may be significantly higher than that for current agriculture, at the global scale. Our production estimates do not account for degradation on the available lands, which could decrease yields, or for irrigation, which could increase them.
*NPP is total (above-ground + below-ground)
Global terrestrial annual plant growth is more than five times the 8 billion tons of carbon released to the atmosphere in fossil fuel combustion. In principle, diverting a small fraction of total plant growth into biomass energy could satisfy the majority of global energy needs.
*Abandoend Crop 270 Mha. Concentration in Eastern US where crop moved to Midwest. Two data setst agreeAbandoned pasture largely in Midwest where crop replaced pasture this not counted towrads total abandoned agCorrection for forest and urban excludes areas in eastern US where forest regrowthSimilar with different land cover estiamtesData driven approach, spatially explicit result 385-472 Mha
Tran: Next overlay areas with maps of plant production
Based on the HYDE data we found that 269 Mha of crop lands have been permanently converted to land uses other than cropping (Fig. 1b), while 479 Mha of pasture lands have been converted to land uses other than pasture (Fig. 1c) at some point in the last 300 years. This HYDE-based abandoned crop area is somewhat higher than the 210 Mha of abandoned crop area from the SAGE crop data (Fig. 1a)11. The abandoned crop areas from HYDE and SAGE data had the highest concentrations over the Eastern U.S. as a result of the relocation of cropland from the eastern to the Midwestern region of North America. The most extensive area of abandoned pasture was over the Midwestern region of North America where HYDE data indicates that cropland has replaced pasture land. High concentrations of pasture abandonment also were found in Australia where pasture areas peaked in the mid-1970s and have steadily declined.
We estimate that the total abandoned agriculture (crop and pasture) ranges from 474 Mha to 579 Mha globally. These estimates exclude abandoned agriculture areas arising from the conversion of crop to pasture or pasture to crop. These estimates do not exclude abandoned areas arising from agriculture to forest or agriculture to urban transitions. In order to exclude forest and urban areas we used a MODIS satellite map of the current forest and urban land cover (3 minute resolution, MODIS/Terra Land Cover Types MOD12C1)12. Our estimate of abandoned agriculture that excludes forest and urban areas is 377 Mha to 472 Mha (Fig 1d).
summed to 66% to 110% of the range of areas assumed in previous bioenergy assessments
Spatial analysis between HYDE and MODIS data should introduce uncertainty due to the different methods used in the creation of these data. For example, HYDE agriculture areas were spatially distributed by human population at the sub-administrative level. Since the HYDE spatial distribution may be biased towards urban areas, the exclusion of MODIS urban areas may overcorrect our abandoned area estimate. However, the abandoned agriculture areas were only reduced by 3% using the MODIS urban areas. Our application of the MODIS forest map appears to have correctly excluded forest regrowth in the eastern U.S. where abandoned agriculture has transitioned to secondary forests28.
Crop 1,445 MhaPasture 3,321 Mha****Not the case for Brazil currently*Marginal Abatement Cost CurvesFor rural isolated communities in NicaraguaBased on substittuion for diesel power generation (f diesel fuel at US$1.06 per liter leads to 0.54 $/kWh compared to national grids that are 0.10$/kWh)
Previous work not accounting for BC impact of diesel generators! We can do this and this should bring down the $/tCO2 significantly!*Why this is the right place to look? Can we find win-win solutions?What is missing is the natural uptake by plants**