Development of spatially explicit emission scenario

Available online at www.sciencedirect.com

Procedia Environmental Sciences 64 (2011) 146–152 146–xxx

Earth System Science 2010: Global Change, Climate and People

Development of spatially explicit emission scenario from land-use change and biomass burning for the input data of climate projection Etsushi Katoa*, Michio Kawamiyaa, Tsuguki Kinoshitab, Akihiko Itoa,c a

Research Institute for Global Change, JAMSTEC, Yokohama, Japan b Collage of Agriculture, Ibaraki University, Japan c Center for Global Environmental Research, NIES, Tsukuba, Japan

Abstract In the preparatory phase of IPCC AR5 development, Representative Concentration Pathways (RCPs) have been constructed by Integrated Assessment Models (IAMs) groups as new forcing scenarios used for climate modeling and earth system modeling groups. In the process of RCP 6.0 scenario development, which has been conducted by the Asia-Pacific Integrated Model (AIM), we constructed a scenario of spatially explicit long-term aerosol emissions from biomass burning and net land-use change CO2 emissions in order to complement energy use and industrial emissions scenarios projected by socio-economic component of AIM. To estimate the emissions from biomass burning, we incorporate a vegetation fire component into the terrestrial biogeochemical process model VISIT (Vegetation Integrative SImulation Tool). For the net land-use change CO2 emission, we use gridded land-use change transition matrix data developed in the process of downscaling the RCP6.0 land use scenario for the evaluation of emission from deforestation (primary land and secondary land into cropland, pasture, and urban), and also the evaluation of regrowth absorption from abandonment of cropland and pasture. In this process, uncertainty of land emissions due to the effect of CO2 concentration and land-use scenario is examined, and reveals significant effect of CO2 fertilization effect on vegetation to the land emissions.

© 2011 Published by Elsevier Ltd. Selection under responsibility of S. Cornell, C. Downy and S. Colston. Keywords: Land-use change; biomass burning; integrated assessment model; biogeochemical model; emission scenario;

* Corresponding author. Tel.: +81-45-778-5506; fax: +81-45-778-5707. E-mail address: [email protected].

1878-0296 © 2011 Published by Elsevier doi:10.1016/j.proenv.2011.05.015

Etsushi Kato et al. / Procedia Environmental Sciences 6 (2011) 146–152 (

)

1. Introduction Constructing appropriate trajectories of atmospheric greenhouse gas concentration and aerosol emissions is a crucially important factor for reliable simulations of global warming prediction. For more intimate collaboration bHWZHHQVRFLRHFRQRPLFUHVHDUFKHUVDQGQDWXUDOVFLHQWLVWVWKH SDUDOOHODSSURDFK has recently been adopted (Moss et al. [1]). First, conventional trajectories of radiative forcing, assuming several stabilization levels, are selected. Second, reasonable trajectories of anthropogenic emissions, which are consistent with socioeconomic rules, and atmospheric GHG levels are produced using models: i.e., four Representative Concentration Pathways (RCPs). These trajectories are used to drive climate models and Earth-system models, to provide preliminary climate change scenarios for impact assessments and mitigation analyses. Then, total consistency is recursively examined for every step. Since the emission scenario of RCPs requires the whole range of emission sources categories, such as fossil fuel and biofuel emission from industry and energy use, agricultural activity, residential and commercial activity, and also emissions from land-use change and biomass burning from terrestrial ecosystem in a spatially explicit manner (0.5 degree by 0.5 degree grid cell), integrated assessment modeling groups have to set up socioeconomic models with terrestrial biogeochemical components to construct the emission scenarios. In the process of the RCP 6.0 scenario development, which has been conducted by the Asia Pacific Integrated Model (AIM) group, we developed a model to prepare spatially explicit land emissions scenario from biomass burning and land-use change consistent with the socio-economic driver from the computational general equilibrium (CGE) model (Fujino et al. [2]). The schematic view of the RCP 6.0 scenario development framework is presented in Fig. 1. This study shows the brief model description of the land emission component of RCP 6.0 scenario, and the resulting scenario of biomass burning and land-use change for the 21st century estimated by the process-based terrestrial ecosystem model.

Fig. 1. A schematic representation of the RCP 6.0 scenario development framework by Asia Pacific Integrated Model (AIM). Each component of model used in RCP 6.0 development is represented by a rounded box.

2. Method 2.1. Vegetation model description and forcing data A process-based model of the terrestrial ecosystem, Vegetation Integrative SImulator for Trace gases (VISIT), is adopted in this study. The model is the successor of terrestrial carbon cycle model Sim-

147

148

Etsushi Kato et al. / Procedia Environmental Sciences 6 (2011) 146–152

CYCLE (Ito and Oikawa [3]), and evaluates biogeochemical cycles of carbon and nitrogen explicitly (Ito [4]). Model simulations in this study are conducted with two separate time sequences; one is for the 20th century (1901-2000) as validating historical and base year emissions, and the other is for the 21st century run to produce the land emissions scenario after the historical validation. The spatial resolution used in the simulations is half degree grid cells globally, and the time resolution is one month. For the climate forcing, temperature, precipitation, cloud cover, and vapor pressure from CRU TS2.1 data (Mitchell and Jones [5]) is used in the historical run. For the scenario run, since RCPs scenarios need to eliminate climate feedback in the emission data, the average climatology data for 1991-2000 of CRU TS2.1 is used in the simulation. CO2 concentration scenario forced in the 21st century run is derived from AIM CGE and simple climate model in the preparation of the RCP 6.0 emissions, which stabilizes at 629.5 ppmv by 2100. As a reference scenario, 21st century run with CO 2 concentration reaching 779 ppmv by 2100 is also conducted. To incorporate the effects of anthropogenic land-use change, historical land use transition dataset for &0,3+XUWWHWDO>@ RI /8+DBXWY W\SHLVXVHGLQWKHKLVWRULFDOUXQRIWKHWKFHQWXU\,QWKH GDWDVHW XW W\SH GHQRWHV WKH ODQG XVH GDWD KDYLQJ XUEDQ FDWHJRU\ LQ DGGLWLon to the basic four categories; primary land, secondary land, cropland, and pasture, and having an assumption of expansion of secondary land started from 1700. For the 21st century scenario run, transition matrix data of land-use categories among primary land, secondary land, cropland, pasture, and urban is prepared by land use model of RCP 6.0 in the process of downscaling a regional land-use scenario derived from AIM CGE model. As well as the CO2 concentration forcing described above, a reference land-use change scenario is also prepared and used for the reference scenario simulation. 2.2. Burned area calculation and emissions from biomass burning Fire season and burned fraction of the grid cells are calculated after Thonicke et al. [7], using fuel load and soil moisture estimated by the VISIT model, with calibrating burned fraction for each of the 14 regions defined in GFEDv2 (van der Warf et al. [8]). In addition, anthropogenic burned area is assigned using historical and scenario land-use change transition matrix data, by considering annual deforested area of primary land and secondary land, which implicitly includes shifting-cultivation, in tropical to temperate regions are assumed to be burned. Emissions of CO 2, BC, CH4, CO, N2O, NH3, NMVOC, NOX, OC, and SO2 are estimated using combustion completeness after Hoelzemann et al. [9] and emission IDFWRUVDIWHUWKHEDVH\HDUGDWDRI5&3VHPLVVLRQGDWDEDVHZKLFKDUHEDVHGRQ*)('Y VHPLVVLRQIDFWRU In the historical validation run, global burnt area for 1997-2000 is estimated as 3.46 ± 0.08 (mean ± 1SD) 106 km2 by the model and emission of black carbon (BC) by biomass burning in that period is estimated as 3.05 ± 0.24 (mean ± 1SD) Tg BC year -1, which is in good agreement with the estimates by GFEDv2 at the same period, 3.52 ± 0.16 106 km2 burnt area and 3.08 ± 0.75 Tg BC year-1 emission of BC. For the 21st century run to construct RCP emission scenario, soil moisture change caused from CO 2 fertilization effect on plant biomass is excluded in order to remove indirect climate feedback effect in the emission scenario. In addition, two extra experiments, one with fixed CO 2 concentration at the year 2000 level, and the other with scenario CO2 concentration run with the effect of soil moisture change on burnt area calculation, are conducted as a sensitivity analysis to see the effect of CO2 fertilization effect on biomass burning emission for both RCP 6.0 scenario and reference scenario simulations. To harmonize with the base year data at 2000 of the official historical emission dataset for CMIP5 experiment, bias correction for the gridded model output is conducted to produce the final biomass burning emission scenarios.


)

2.3. Net land-use change carbon emission estimate To estimate net carbon emission from anthropogenic land-use change, deforested biomass and its emission is calculated in VISIT model using land-use transition matrix data among primary land, secondary land, cropland, pasture, and urban following a protocol of deforested and production biomass decay (McGuire et al. [10]) with taking into account of the CO 2 uptake on regrowing secondary land by land abandonment of cropland and pasture on each ecosystem type (Houghton et al. [11]). In the historical validation run, VISIT estimates cumulative net land-use change carbon emission for 1901-2000 as 126.0 Pg C, and mean annual emission in 1990s as 1.14 Pg C yr -1, which is in the range of uncertainty (±0.5 Pg C year-1 RIWKHHVWLPDWHVE\+RXJKWRQ>@ VXSGDWHGQHWODQG-use change emission by bookkeeping method (116.9 Pg C and 1.57 Pg C yr-1 respectively). 3. Results and Discussion Global BC emission from biomass burning increases up to 3.2 Tg BC yr -1 at 2100 for the RCP 6.0 scenario run (Fig. 2). In total, BC emission from grassland fire at 2100 is 1.7 Tg BC yr -1, increased about 0.2 Tg BC yr-1 from 2005. For forest fire, the emission at 2100 is 1.5 Tg BC yr -1, increased about 0.4 Tg BC yr-1 from 2005, showing a large increase in biomass burning from forested area (Fig. 3). From a sensitivity analysis in regards to direct and indirect CO2 fertilization effects on biomass burning emissions, the indirect effect has large potential increases in the future biomass burning emissions.

Fig. 2. Time series of projected global BC emission (Tg BC year-1) from biomass burning. Solid thick line represents RCP 6.0 scenario run.

149

150


)

Fig. 3. Spatial distribution of BC emission (ton BC year-1 per 0.5 degree grid cell) from biomass burning. Upper panels show decadal average at 2000 and lower panels show that of 2100. Left figures show emissions from grassland fire and right panels show emission from forest fire.

Fig. 4. Global land use change scenarios (a. RCP 6.0 scenario, b: reference scenario) denoting the fraction of primary land, secondary land, cropland, pasture, and urban of total land, and c: corresponding net land-use change carbon emissions (Pg C year-1).


For the net land-use change carbon emission, emission continues to drop early in the 21st century in the both RCP 6.0 and reference runs, showing the continued trend of decline in net land-use emission of 1990s (Houghton [12]). With RCP 6.0 land-use scenario, shrinkage of pasture area around 2020s and 2030s causes further decrease in net land-use change carbon emission and leads to negative emission (i.e. net absorption of CO2) due to the expansion of secondary recovering land (Fig 4.). The emission increases after that period due to the carbon storage of deforested biomass, and stabilized to 0.18 Pg C yr-1 at the end of 21st century in RCP 6.0 scenario run. 4. Conclusion Using CO2 concentration and land-use scenario derived from the socio-economic component of AIM, the trace gas emissions from biomass burning and net land-use change carbon emission in the RCP 6.0 stabilization scenario have been developed in a spatially explicit manner by the process based terrestrial ecosystem model VISIT. Using this model, biogeochemically consistent emissions from terrestrial ecosystem can be evaluated along with the scenario of energy use and industrial fossil use, that can be used as a consistent forcing data for the climate models in the future climate change projection experiments. Acknowledgements 7KLV VWXG\ ZDV VXSSRUWHG E\ WKH *OREDO (QYLURQPHQWDO 5HVHDUFK )XQG 6-5 Integrated Research on Climate Change Scenarios to Increase Public AwareQHVV DQG &RQWULEXWH WR WKH 3ROLF\ 3URFHVV RI WKH Ministry of the Environment, Japan. References [1] Moss RH, Edmonds JA, Hibbard KA, Manning MR, Rose SK, van Vuuren DP, Carter TR, Emori S, Kainuma M, Kram T, Meehl GA, Mitchell JFB, Nakicenovic N, Riahi K, Smith SJ, Stouffer RJ, Thomson AM, Weyant JP, Wilbanks TJ. The next generation of scenarios for climate change research and assessment. Nature 2010; 463:747 756. [2] Fujino J, Nair R, Kainuma M, Masui T, Matsuoka Y. Multi-gas mitigation analysis on stabilization scenarios using AIM global model. The Energy Journal 2006;3 Special Issue:343 353. [3] Ito A, Oikawa T. A simulation model of the carbon cycle in land ecosystems (Sim-CYCLE): a description based on dry-matter production theory and plot-scale validation. Ecological Modelling 2002;151:143 176. [4] Ito A. Changing ecophysiological processes and carbon budget in East Asian ecosystems under near-future changes in climate: implications for long-term monitoring from a process-based model. Journal of Plant Research 2010;123:557 588. [5] Mitchell TD, Jones PD. An improved method of constructing a database of monthly climate observations and associated highresolution grids. International Journal of Climatology 2005;25:693 712. [6] Hurtt G, Chini1 LP, Frolking S, Betts R, Feddema J, Fischer G, Goldewijk KK, Hibbard K, Janetos A, Jones C, Kindermann G, Kinoshita T, Riahi K, Shevliakova E, Smith S, Stehfest E, Thomson A, Thornton P, van Vuuren D, Wang YP. Harmonisation of global land-use scenarios for the period 1500 2100 for IPCC-AR5. iLEAPS Newsletter 2009;7:6 8. [7] Thonicke K, Venevsky S, Sitch S, Cramer W. The role of fire disturbance for global vegetation dynamics: coupling fire into a dynamic global vegetation model. Global Ecology and Biogeography 2001;10:661-677. [8] van der Werf GR, Randerson JT, Giglio L, Collatz GJ, Kasibhatla PS, Arellano Jr AF. Interannual variability in global biomass burning emissions from 1997 to 2004. Atomospheric Chemistry and Phisics 2006;6:3423-3441. [9] Hoelzemann JJ, Schultz MG, Brasseur GP, Granier C. Global wildland fire emission model (GWEM): Evaluating the use of global area burnt satellite data. Journal of Geophysical Research 2004;109:doi:10.1029/2003JD003666. [10] McGuire AD, Sitch S, Clein JS, Dargaville R, Esser G, Foley J, Heimann M, Joos F, Kaplan J, Kicklighter DW, Meier RA, Melillo JM, Moore III B, Prentice IC, Ramankutty N, Reichenau T, Schloss A, Tian H, Williams LJ, Wittenberg U. Carbon

151

152


balance of the terrestrial biosphere in the twentieth century: Analyses of CO2, climate and land use effects with four processbased ecosystem models. Global Biogeochemical Cycles 2001;15:183-206. [11] Houghton RA, Hobbie JE, Melillo JM, Moore B, Peterson BJ, Shaver GR, Woodwell GM. Changes in the carbon content of terrestrial biota and soils between 1860 and 1980: a net release of CO2 to the atmosphere. Ecological Monographs 1983;53:235-262. [12] Houghton RA. Carbon Flux to the Atmosphere from Land-Use Changes: 1850-2005. In TRENDS: A Compendium of Data on Global Change. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, U.S. Department of Energy, Oak Ridge, Tenn., U.S.A; 2008.