Past and future carbon fluxes from land use change, shifting cultivation and wood harvest

Figures & data

Fig. 1 Schematic illustration of simulating gross versus net land use change (LUC). Under a scheme for gross LUC (upper row), cropland is claimed from primary land (‘prim’) and abandoned to secondary land (‘secd’) in parallel within one grid cell. In the model, croplands and pasture areas undergo gross LUC in areas of shifting cultivation. Under a scheme for net LUC (lower row), only the difference of claimed minus abandoned undergoes a transition to use as cropland and no separate secondary land is formed. In the latter scheme, a smaller grid cell area fraction is affected by LUC.

Table 1. Split of C and N mass into slash (on-site litter pools) and product pools of different turnover times (0 yr, 2 yr, 20 yr) after deforestation (in% of deforested and harvested tree biomass)

		Product pools
	Slash	0 yr	2 yr	20 yr
Roots	100	0	0	0
Leaves	100	0	0	0
Sapwood	0	37.5	37.5	25
Heartwood	0	37.5	37.5	25

Table 2. Generated transitions priorities

→	Croplands	Pasture	Built-up
Primary	1	6	10
Secondary	2	5	9
Croplands	–	7	12
Pasture	3	–	11
Built-up	4	8	–

Numbers represent priority of transitions executed to satisfy net land use area changes (transition priorities associated with land turnover are in bold). Cropland and pasture area abandoned due to land turnover is always transferred to secondary (not shown here).

Fig. 2 Historical annual LUC fluxes. Splined annual fluxes (thick colour lines) and year-by-year data for ‘gross, incl. wood harvest’ (thin grey line) are shown. The dashed lines (‘land turnover contribution’ and ‘wood harvest contribution’) are the differences between the respective curves (‘land turnover contribution’=‘gross, no wood harvest’ –‘net’; ‘wood harvest contribution’ =‘gross, incl. wood harvest’ –‘gross, no wood harvest’).

Table 3. Cumulative LUC emissions (E, [GtC]) and decadal average annual LUC fluxes (e, [GtC yr⁻¹]

	E [GtC]		e [GtC yr^−1]
	−1850	1850–2004	1980s	1990s	2000s
Total LUH	72	171	1.55	1.57	1.21
Total GNT	66	165	1.53	1.63	1.28
Primary LUH	70	154	1.48	1.27	0.99
CO2 effect		+21	+0.36	+0.41	+0.43
Climate effect		−4	−0.27	−0.11	−0.19
Climate & CO2		+17	+0.07	+0.30	+0.22
Wood harvest	+7	+23	+0.26	+0.27	+0.23
Land turnover	+20	+25	+0.21	+0.26	+0.23

Individual effects from CO₂, climate, wood harvest, and land turnover (shifting cultivation) are quantified only for the LUH dataset. Effects from climate are zero before 1850 by design (constant climate before 1901) and are not provided for CO₂ (constant atmospheric concentration prescribed before 1765).

Fig. 3 Annual LUC fluxes by region (black curves) and their splined time series (red curve). The regions are illustrated by the map in the middle and correspond to the delineation used in IPCC AR5 (Ciais et al., 2013), except that ‘Eurasia’ is separated into Europe (everything west of 60°E), and Russia and the Former Soviet Union (FSU) (everything east of 60°E).

Fig. 4 Spatial distribution of cumulative historical (1850–2004) land use change (LUC) emissions and its components. (a) Total (gross, incl. wood harvest) cumulative LUC emissions in kgC/m². (b) Cumulative emissions due to wood harvest. (c) Effect of land turnover (including the introduction of secondary land) as the difference between the run with gross land use (LU) transitions and a run with net LU transitions; both runs do not consider wood harvest. Note the different colour scales in the upper and the lower two panels.

Fig. 5 Annual land use change (LUC) fluxes for 1900–2100 and the different RCPs. Thick black and coloured curves are splined time series of annual data, which is shown by the thin solid curve for the past. Dashed and dotted lines represent annual LUC fluxes, diagnosed from a simulation without wood harvesting and from a simulation where only net LUC is simulated, respectively.

Fig. 6 Spatial distribution of cumulative future (RCPs) land use change (LUC) emissions and its components. Left: Total (gross, incl. wood harvest) cumulative LUC emissions in kgC/m² middle: Cumulative emissions due to wood harvest. Right: Effect of land turnover (including the introduction of secondary land) as the difference between the run with gross land use (LU) transitions and a run with net LU transitions; both runs do not consider wood harvest. Note the different colour scales in the panels.

Table 4. Cumulative LUC emissions (E, [GtC]) for 2005–2099; combined effects from CO₂ and climate; and individual effects from wood harvest, and land turnover (shifting cultivation)

	RCP 2.6	RCP 4.5	RCP 6.0	RCP 8.5
Total	105	27	97	151
Primary	86	25	86	122
Climate & CO2	+19	+2	+12	+29
Harvest	+26	+37	+46	+53
Land turnover	+16	+13	+13	+13

Fig. 7 Simulated (map) and observational (dots) vegetation carbon density (kgC/m²). The map represents total vegetation C density in the primary land use class at present day, diagnosed from a simulation without any anthropogenic LUC and wood harvest. Observational data are from Luyssaert et al. (2007) and Keith et al. (2009) and also represent primary forests.

Ciais P , Sabine C , Bala G , Bopp L , Brovkin V , co-authors . Chapter 6: carbon and other biogeochemical cycles. Climate Change 2013: The Physical Science Basis. Working Group I Contribution to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. 2013. Final Draft, 7 June 2013.

 Google Scholar

Luyssaert S , Inglima I , Jung M , Richardson A. D , Reichstein M , co-authors . CO2 balance of boreal, temperate, and tropical forests derived from a global database. Glob. Change Biol. 2007; 13(12): 2509–2537.

 Web of Science ®Google Scholar

Keith H , Mackey B. G , Lindenmayer D. B . Re-evaluation of forest biomass carbon stocks and lessons from the world's most carbon-dense forests. Proc. Natl. Acad. Sci. U. S. A. 2009; 106(28): 11635–11640. [PubMed Abstract] [PubMed CentralFull Text].

 Google Scholar