Application of probability distributions to the modeling of biogenic CO2 fluxes in life cycle assessment
Journal article, Peer reviewed
MetadataShow full item record
Original versionGlobal Change Biology Bioenergy. 2012, 4 (6), 784-798. 10.1111/j.1757-1707.2011.01156.x
In life cycle assessment (LCA), the same characterization factors are conventionally applied irrespective of when the emissions occur (the same importance is given to emissions in the past, present, and future). When the assessment is constrained by fixed timeframes, the appropriateness of this paradigm is questioned and the temporal distribution of emissions becomes of relevance. One typical example is the accounting for biogenic CO2 emissions and removals. This article proposes a methodology for assessing the climate impact of time-distributed CO2 fluxes using probability distributions. Three selected wood applications, such as fuel, nonstructural panels, and housing construction materials are assessed. In all the cases, CO2 sequestration in growing trees is modeled with an appropriate forest growth function, whereas CO2 emissions from wood oxidation are modeled with different probability distributions, such as the delta function, the uniform distribution, the exponential distribution, and the chi-square distribution. The combination of these CO2 fluxes with the global carbon cycle provides the respective changes caused in CO2 atmospheric concentration and hence in the radiative forcing. The latter is then used as basis for climate impact metrics. Results demonstrate the utility of using emission and removal functions rather than single pulses, which generally overestimate the climate impact of CO2 emissions, especially in presence of short time horizons. Characterization factors for biogenic CO2 are provided for selected combinations of biomass species, rotation periods, and probability distributions. The time discrepancy between biogenic CO2 emissions and capture through regrowth results in a certain climate impact, even for a system that is carbon neutral over time. For the oxidation rate of wooden products, the use of a chi-square distribution appears the most reliable and appropriate option under a methodological perspective. The feasibility of its adoption in LCA and emission accounting from harvested wood products deserves further scientific considerations.