Quantifying the climate impacts of albedo changes due to biofuel production: a comparison with biogeochemical effects

We consider eleven LUC scenarios, comprised of five different biomass feedstocks for up to three original land uses each, as shown in table 1. Each scenario is restricted to one geographic region. The scenarios are consistent between this study and the traditional LCA study that we use as a reference (Stratton et al 2010; LUC combinations S1, S2, P1, P2, P3, H1), wherever possible. The switchgrass and rapeseed scenarios are redefined due to ambiguities in the reference LCA study. In Stratton et al (2010), switchgrass cultivation is assumed to take place on generic carbon-depleted soils. In this study we consider three possible LUC scenarios associated with carbon-depleted soils (McLaughlin et al 2002, Adler et al 2007): corn cultivation (LUC B1), soybean cultivation (LUC B2), and barren land (LUC B3) replaced by switchgrass cultivation. Furthermore, in the reference LCA rapeseed cultivation in Europe is assumed to take place on set-aside land, i.e. land areas temporarily removed from agricultural production (Stratton et al 2010). In this case we consider two LUC scenarios: corn cultivation (LUC R1) and uncultivated land (LUC R2) replaced by rapeseed cultivation. Table 1 also indicates the geographic region in which each LUC scenario is assumed to take place in the reference LCA (Stratton et al 2011a).

A minimum of four geographic locations are selected to describe each original land use type, and a minimum of eight combinations of biomass feedstock and original land use locations are used to define each of the 11 LUC scenarios shown in table 1. This multi-location approach allows for a more complete picture of the potential land conversions and the associated natural variability. The latitude and longitude of these locations are retrieved using current literature (e.g., Mosali et al 2013), satellite observations (e.g., Rhines 2008) and farming databases (e.g., FIC 2013), and are confirmed using the Moderate Resolution Imaging Spectroradiometer (MODIS) Land Cover Type database (MCD12Q1) (NASA MODIS 2013a).

Black-sky shortwave (BSW) broadband albedo coefficients (encompassing both the near infrared and visible spectra) are retrieved for each biomass feedstock and original land use pairing, i, that represents a specific LUC scenario. BSW albedo coefficients are obtained from the MODIS satellite database MCD43A3 (NASA MODIS 2013b) and vetted using a separate MODIS database (MCD43A2; NASA MODIS 2013c) Albedo data from the MODIS database are produced every 8 days and are linearly interpolated to obtain daily albedo evaluations for a full year. In case of missing or low quality data, the time interpolation is performed between the two closest acceptable observations. The average daily BSW albedo for each i is obtained for a full year by averaging the daily values retrieved for three reference years (2009, 2010 and 2011), in order to account for annual variability in local conditions. Biases in each individual albedo observation are reduced by taking the space average across the 500 m × 500 m cell where the location under investigation is found and the eight cells surrounding it. Consistency between the land-use type of these cells is verified using the MODIS Land Cover Type database (NASA MODIS 2013a). Table 2 shows the yearly-averaged BSW albedo, retrieved and processed as described, for each land use type considered in the LUC scenarios from table 1. The BSW albedos in table 2 are given as mean, minimum and maximum values among the yearly-averaged albedos retrieved for all the sample locations representing each specific land use type.

For each biomass feedstock type and original land use pairing i, representing a LUC scenario from table 1 we evaluate the albedo effect as the difference in RF induced by the conversion of the original land use to biomass feedstock cultivation. The geographical location and conditions of radiative transmittance are kept the same as that of the original land use; i.e. feedbacks between albedo changes and local weather/cloudiness conditions are not accounted for. For each i, the planetary albedo change is computed as a function of the day of the year d:

$$\begin{eqnarray}&&\label {eq1} \Delta \alpha _{i}(d)=K_{T\,\mathrm {orig},i}(d ){T}_{a} [\alpha _{\mathrm {bio},i}(d)-\alpha _{\mathrm {orig},i}(d ) ] \end{eqnarray} \tag{ 1 }$$

where α_{bio, i} and α_{orig, i} are the daily cloud-free shortwave albedo coefficients for biomass feedstock cultivation and the original land use, respectively, obtained from the MODIS database (NASA MODIS 2013b) and averaged in time and space as previously described. K_{T orig, i} is the mean daily all-sky clearness index for the original land use point, and T_a is the transmittance factor, as in Bright et al (2012). The calculation of planetary albedo differences from clear-sky surface albedo values in equation (1) follows a procedure widely reported in the literature (Lenton and Vaughan 2009, Muñoz et al 2010, Bright et al 2012, Cherubini et al 2012). Local mean daily values of K_{Torig, i} are constant for each month and are retrieved from the NASA Atmospheric Science and Data Center (ASDC) database, which provides monthly 22-year averages of the all-sky clearness index (including maximum and minimum bounds) (NASA ASDC 2013). The transmittance factor T_a is chosen as a global annual average of 0.854, consistent with previous findings and modeling comparisons (Lenton and Vaughan 2009, Muñoz et al 2010, Cherubini et al 2012, Bright and Kvalevåg 2013).

The planetary albedo change due to biomass feedstock cultivation on land originally used for some other purpose, found in (1), alters the radiative balance of the Earth which can be quantified as a radiative forcing. This is equal to the time integral of the product of daily albedo variation (1) and daily radiative flux at the top of the atmosphere R_{TOA, i}(d), calculated at the original land use location (Bright et al 2012, Cherubini et al 2012). For each biomass feedstock and original land use pairing,i, the yearly global RF (measured in W m⁻²) is therefore:

$$\begin{eqnarray}&&\label {eq2} \Delta {\mathrm {RF}}_{\mathrm {global},i}=\left \{ -\frac {1}{365}\sum _{d=1}^{365} [{{R}}_{\mathrm {TOA},i}(d )\cdot \Delta \alpha _{i}(d ) ] \right \}\cdot \frac {{A}_{a}}{{A}_{\mathrm {earth}}} \end{eqnarray} \tag{ 2 }$$

where A_a is the reference area subject to the albedo change, and A_earth is the total area of the earth. The RF associated with each of the LUC scenarios in table 1, ΔRF_LUC, is calculated as the average of all the global yearly radiative forcings ΔRF_{global, i} found for all of the pairings, i representing the same LUC case.

CO₂e emissions per unit energy of biofuel produced (gCO₂e/MJ) is a common metric adopted in LCA studies (Larson 2006, Adler et al 2007, Stratton et al 2010, 2011a). To establish direct comparison between albedo change effects and biogeophysical effects for the same LUC scenario, the global RF associated with each one of the LUC scenarios in table 1 is converted into CO₂e. The correspondence between the RF induced by albedo changes and CO₂e emissions is well established in the literature (Betts 2000, Bird et al 2008, Muñoz et al 2010, Joos et al 2013). First, RF is converted into a change in atmospheric carbon concentration ΔC by using a logarithmic relation with the background carbon concentration (Betts 2000) (linearized for small perturbations). Positive RF (induced by a decrease in the land albedo) corresponds to carbon emissions, while negative RF corresponds to carbon sequestrations. The concentration ΔC (in parts per million, ppm) is then converted into an equivalent carbon emission ΔC_T per unit area subject to albedo change:

$$\begin{eqnarray}\label {eq3} \Delta {\rm C}_{T}&=& \left (\frac {1}{{\mathrm {AF}}_{\mathrm {TH}=100}} \right )\Delta {\rm C} {{m}}_{\mathrm {atm}} \left (\frac {{{M}}_{{\rm C}}}{{{M}}_{\mathrm {air}}} \right )\nonumber \\ &&\times \ \left (\frac {1}{{{A}}_{{a}}} \right ) \frac {1}{{10}^{6}}\ ({\mathrm {tC}}\ {\mathrm {ha}}^{-1}) \end{eqnarray} \tag{ 3 }$$

where m_atm is the total mass of the atmosphere (in tons), M_C and M_air are the molecular weights of carbon and air respectively (in g mol⁻¹), A_a is expressed in hectares and AF_TH=100 is the airborne emission fraction of CO₂ for a time horizon (TH) of 100 years, consistent with the reference biogeochemical impacts assessment by Stratton et al (2010). Finally, using data about the biomass yield, mass conversion factor, and specific energy conversion efficiencies, the carbon emission per unit area in (3) is converted into a CO₂e emission per unit energy of the fuel produced. In order to evaluate the albedo effects under the same assumptions as the biogeochemical impacts, the resulting emissions are distributed over 30 years, as in the reference LCA (Stratton et al 2010).

In order to represent the magnitude of natural variability, we examine low, baseline, and high cases for each LUC scenario in table 1. The baseline cases utilize the mean of the RFs calculated for all the biomass feedstock and original land use pairings representing the same LUC scenario. Low and high cases utilize the maximum and minimum RFs calculated among the pairings used to simulate the same LUC scenario, and the upper and lower estimates of the relevant all-sky clearness index from the ASDC database (NASA ASDC 2013). Variability in meteorological conditions is therefore accounted for.

The geographic locations and relevant physical parameters for all sample locations are given in the Supporting Information (SI available at stacks.iop.org/ERL/9/024015/mmedia). A more detailed derivation of the albedo-emission conversion model and the yields and energy efficiencies of each biomass feedstock type are also discussed in the SI (available at stacks.iop.org/ERL/9/024015/mmedia).

The albedo change due to replacement of corn cultivation, soy cultivation and barren land with switchgrass (scenarios B1–B3) leads to a negative RF in the baseline results, the equivalent of a sequestration of CO₂e. The effect is stronger when switchgrass replaces corn (LUC B1, −22 gCO₂e/MJ) or soy (LUC B2, −13 gCO₂e/MJ) than it is for barren land (LUC B3, −9 gCO₂e/MJ). A negative RF (or equivalently a cooling effect) for the conversion of land to switchgrass cultivation has also been computed by Georgescu et al (2011) (− 0.0053 W m⁻²) and Anderson-Teixeira et al (2012) (− 50 Mg CO₂e ha⁻¹ 50 yr⁻¹) however the disparate methodologies, metrics and LUC assumptions used in those studies do not allow a quantitative comparison with this work. The biogeochemical lifecycle impacts of renewable MD fuel production and use from switchgrass replacing carbon-depleted soils are estimated by Stratton et al (2010) to be nearly carbon neutral (−1.6 gCO₂e/MJ). In absolute terms, the albedo impact of switchgrass cultivation for MD production is therefore 1380%, 813% and 563% greater than the biogeochemical lifecycle effects of corn, soybean, and barren land conversion, respectively. The net sum of albedo and biogeochemical effects is −23 gCO₂e/MJ, − 15gCO₂e/MJ and −11 gCO₂e/MJ for corn replacement (LUC B1), soy replacement (LUC B2) and barren land replacement (LUC B3), respectively. It should be noted that the high emission cases (high whisker bar limits) demonstrate that a net positive RF is also possible for these three LUC scenarios.

In figure 1, LUC S1 and S2 show that the albedo effect of replacing Brazilian Cerrado and tropical rainforest with soybean cultivation results, in both cases, in a negative RF, equivalent to a cooling effect (− 146 and − 161 gCO₂e/MJ, respectively). Soybean cultivation in Brazil generally exhibits a higher reflectivity (average albedo of 0.175, see table 2) than the savannah land of the Brazilian Cerrado (average albedo of 0.133) or than the dense dark tropical rainforest in the Amazon region (average albedo of 0.120). An albedo-induced CO₂e reduction due to the establishment of soy cultivation in Brazil has also been found by Anderson-Teixeira et al (2012) (− 70 MgCO₂e ha⁻¹50 yr⁻¹). The green bar in figure 1 for LUC S1 shows that, considering both the effects of albedo change and of biogeochemical lifecycle GHG emissions calculated by the reference LCA (Stratton et al 2010), the aggregate climate impact of the conversion of cerrado grassland to soybean cultivation is equivalent to a sequestration of 50 gCO₂e/MJ. By adding the albedo effect to the biogeochemical results from the reference LCA, renewable MD from soybean cultivated on land that was previously Cerrado grassland exhibits a decrease in climate impact of 156% with respect to conventional MD (from 90 to − 50 gCO₂e/MJ). This reverts the results of the reference LCA which suggests that soybean derived MD has greater climate impact, in terms of lifecycle GHG emissions, than conventional MD (97 gCO₂e/MJ versus 90 gCO₂e/MJ) (Stratton et al 2010). In the case of tropical rainforest replacement with soybean cultivation (LUC S2), inclusion of the albedo effects does not revert the findings of the reference LCA: the direct GHG emission calculated by the reference LCA (569 gCO₂e/MJ) (Stratton et al 2010) is 453% larger than the impact attributable to the increase in albedo (− 161 gCO₂e/MJ).

Due to their leaf characteristics and plantation density, palms are characterized by the lowest average shortwave albedo (0.088, see table 2) among the biomass feedstock types considered in this study. Furthermore, palm is not harvested, meaning that the low albedo coefficient is nearly constant throughout the year. This accounts for the smaller cooling effect shown in figure 1 for the case of tropical rainforest replaced by palm plantations (LUC P2, − 25 gCO₂e/MJ) compared to the cooling effect occurring for soybean cultivation replacement of tropical rainforest (LUC S2, − 161 gCO₂e/MJ). The low albedo of palm also accounts for the positive RF induced by conversion of previously logged-over forest (LUC P1), equivalent to a GHG emission of 14 gCO₂e/MJ. Palm replacement of peat land rainforest (LUC P3) yields a relatively small sequestration of − 4 gCO₂e/MJ. For the aggregate climate impact of LUC P1, indicated by the green bar in figure 1, the baseline case of 55 gCO₂/MJ remains below the conventional jet fuel baseline even if the albedo effect is included. If the albedo effect is included, the high emission case (high whisker bar limit, 129 gCO₂e/MJ) is 43% larger than conventional MD, whereas the high emission case for the biogeochemical effects does not exceed the conventional MD reference. For LUC P2, while the baseline warming effect remains higher than that of conventional MD even if the albedo effect is included, the aggregate climate impact in the low emission case (77 gCO₂e/MJ) is below that of conventional MD, which was not the case in the reference LCA. For LUC P3, the biogeochemical effects from the reference LCA are so large (705 gCO₂e/MJ) (Stratton et al 2010), that the inclusion of albedo effects does not change the total climate impact significantly.

LUC R1 and R2 show the albedo effect of replacing corn or uncultivated land with rapeseed cultivation in Europe. In both cases a small cooling effect is found, − 3gCO₂e/MJ for LUC R1 and − 10 gCO₂e/MJ for LUC R2. According to the reference LCA, the biogeochemical effects of renewable MD fuel from rapeseed cultivated on previously set-aside land, yields a biogeochemical effect of 96 gCO₂e/MJ (Stratton et al 2010). The contribution of the albedo effect is negligible for this set of LUC scenarios.

LUC H1 considers salicornia cultivation on land that was previously desert. The albedo of the desert (average of 0.326 from table 2) is much larger than the albedo of the salicornia cultivations (average of 0.149), since deserts are composed of smooth, clear sand, and are highly reflective of incoming solar radiation (Pielke and Avissar 1990). The large desert albedo accounts for the relatively large positive RF induced by LUC H1, equivalent to 222 gCO₂e/MJ. In comparison, the direct GHG emissions computed by the reference LCA are only 6 gCO₂e/MJ (Stratton et al 2010). This result shows that production and use of renewable MD using salicornia grown on desert lands can have a larger warming effect than producing and using conventional MD.

Figure 2 shows a comparison between albedo effects and GHG emissions from LUC, excluding all the other stages accounted for in the reference LCA (Stratton et al 2010) (i.e. biomass cultivation and transport; feedstock to fuel conversion; and biofuel transport and combustion). This comparison is instructive since it shows the relative magnitude of the biogeochemical and biogeophysical effects that exclusively stem from (direct) alterations in land use. Both effects are evaluated as instantaneous CO₂e emissions, not distributed across the reference 30-years time span as in figure 1 (see equation S19 in the SI available at stacks.iop.org/ERL/9/024015/mmedia) but only across the first year of land use change. This is because the albedo effect is obtained using albedo differences and transmittance parameters averaged over a full year of variation, and carbon sequestration due to LUC is evaluated after a full year of vegetation replacement (Stratton et al 2010). Therefore, the results for albedo effects (blue bars) are directly proportional to the ones shown in figure 1. Results for LUC emissions (red bars) are instead proportionally smaller than the ones reported in figure 1, since they exclude steady-state transport, production and combustion emissions. The results indicate that albedo effects are on the same order of magnitude as the traditional direct biogeochemical LUC emissions for most of the LUC scenarios considered.

Zoom In Zoom Out Reset image size