|No 52 Em Português|
Text for Discussion:
Carbon Balance Concerning the Emissions of Greenhouse Effect in the Energy Use and Transformation in Brazil: Comparison between the Extended Top-Down and Bottom-Up Methodologies – Analysis of Results and Conclusions.
Carlos Feu Alvim, Frida Eidelman and Omar Campos Ferreira
The Economy and Energy Organization has made together with the Ministry of Science and Technology a study about carbon balance in the emissions resulting from the use and transformation of energy. The dissemination of the results of this study has been made by the e&e periodical. The following results have already been published:
● Carbon Balance in the Production, Transformation and Use of Energy in Brazil – Methodology and Results of the Top-Bottom Process in the period 1970 – 2002 (e&e N 48).
● Carbon Balance in the Energy Transformation Centers (e&e N 50).
● Results corresponding to the adopted accounting process that includes the extended Top-Down approach and the use of coefficients calculated in the national inventory for the period 1990-1994 for estimating emissions from 1970 to 2002 by the Bottom-Up process (e&e Nº 51).
In the present issue the results of both methods are compared and some deviations found are pointed out; they should bring about corrections in the inventory of carbon balance and emissions inventory. Suggestions regarding the corrections are presented and they will be the object of a complementary analysis.
The benemis program made for calculating emissions permits to obtain synthetic tables grouping energy sources and economical sectors. In the case of the benemis_c_eee version the contained carbon data, emissions using the two processes and their comparison can be obtained for each year.
Table 1 shows the values of contained carbon without discounting emissions of the main consuming sectors and energy sources grouped by origin.
In the following tables the emissions calculated by the Top-Down (Table 2) and Bottom-Up (Table 3) methods are compared in the aggregated form. Table 4 illustrates the procedure used for the comparison: the discrepancies relative to these two methods are indicated by colors with limits fixed by percents of deviations (white for differences below 0.1% or zero values, green, between 0.1% and 10%, yellow, between 10% and 30% and red, above that value). Concerning the aggregated tables, there is an additional difficulty namely the aggregation criterion of fuel by origin. In the case of gases, for example, residential gas had different origins along time and in the usual BEN’s structure it is presented together with coke plant gas. In the program’s present representation by origin it is recorded as mineral gas.
Table1: Carbon Contained in Fuels Used in Gg/year, Year 1990
Table 2: Carbon Emissions in Gg/year (1990) – Top Down Method
Table 3: Emissions by Sector and by Group of Fuels - Year:1990 - Gg /year Bottom-Up
(*) Excludes Transformation
Identifying the problems is easier when one examines the difference by fuel and when the accounts have a larger disaggregated form. This will be carried out in what follows. Preliminarily, it should be observed that in Table 4 the red cells for mineral coal identify problems concerning the fuel by origin, mainly gas. Some deviations pointed out for biomass are due to difficulties already detected in transformation.
Figure 1 shows the emissions by sector and by fuel by origin obtained using coefficients obtained in the Bottom-Up process. The Transport Sector is the largest sector responsible for carbon emissions from fossil sources.
The following tables illustrate the results obtained from calculating the carbon balance (year 1990) and are used in the analysis of the existing problems.
The first two tables (Table 5 and Table 6) show the original carbon content in the fuels used in transformation and consumption. In transformation, the negative masses show (as in BEN) the absorption of an energy source that is transformed into another one and recorded as positive input on the same line. For the transformation centers where emissions are not calculated (Petroleum Refineries, Natural Gas Plants, Gasification Plants, Coke Plants, Distilleries and Other Transformations), the “Total” column on the right points out the faults in the carbon balance. Later on, using the results of the following tables it will be possible to complete the carbon balance of the remaining transformation units.
Comparison of Results of the two Processes.
Figure 1: Values of carbon emitted by sector and by fuel by origin.
Table 7 presents the emissions obtained through the reconstruction of the Bottom-Up process and Table 8 presents those obtained through the Top-Down process. Table 9 presents the analysis of results and the percent deviations found between the two processes (values relative to the Bottom-Up value).
Table 5 –Carbon Content by Activity and by Energy Source
(Final Consumption) – Year 1990 – Gg/year
Table 6: Carbon Content by Activity and by Energy Source
– Year 1990 - Gg /year (Transformation)
Table 7: Emissions by Activity and by Energy Source (Final Consumption
and Transformation) – Bottom-Up Method – Year 1990 - Gg /year
Table 8 Emissions by Activity by Energy Source (Top-Down Method)
1990 - Gg /year
(values relative to the 2nd)
The comparison regarding the balance uses the already described colors code. In Table 9 the colors indicate the magnitude of discrepancies between the values calculated by the two approaches. The analysis of the transformation centers could be completed using the collected data.
Considering that our aim is to make a diagnosis and not a revision of the coefficients, some energy sources were picked up as a help for this purpose. In order to make this analysis easier, we have calculated the relationship between the emission coefficients and the carbon content in the carbon mass conservation of gasoline, firewood and fuel alcohol. We have also picked up the cases of diesel oil and charcoal to allow a comparison with the three fuels identified as those that deserve more attention.
The carbon balance results for the transformation centers are shown in Table 10.
Table 10: Carbon Balance in the Transformation Units for the Year1990 calculating the emissions by Bottom-Up and Top-Down Processes
The emission calculated by the two methodologies is included in Table 10, the carbon balances present satisfactory results for most transformation units. In the autoproducers power plants some important differences were detected in the results of the two methodologies that may be assigned to emissions calculation in the use of “other primary” and tar (see Table 9). In the charcoal plants the emissions are underestimated but the bigger problem seems to be in the Bottom-Up methodology. In an attempt to equate the matter, firewood will be analyzed in what follows as it presents deviations relative to the two methods. The case of distilleries has already been previously commented and there are problems in the carbon mass/ energy coefficients both for alcohol (that will be analyzed in what follows) and the raw material (sugarcane juice and molasses) for which a generic coefficient is applied.
Emission Coefficients and Carbon Conservation
The emission coefficients used by the benemis program utilize the results of the emission calculations carried out by the Bottom-Up process. Since the carbon mass is conserved, these coefficients must have a relationship so that, for example, if an automobile emits less carbon monoxide (CO), the CO2 quantity or other carbon compounds must increase. The relationship regarding the coefficients is shown in the following box.
Analysis of Gasoline and Diesel Oil Emissions
Table 11: Gas Emissions and Masses of Contained Carbon for Gasoline (year 1990) and carbon/ implicit energy factor in the emission coefficients used.
The carbon mass/energy factor for gasoline should be 27.1tC/TJ in order to correspond to the emission factor, instead of IPCC’s 18.9 that was adequate for carbon balance in refineries.
The difference found is assigned to a procedure suggested by IPCC where in the case of CO2 the total carbon mass obtained would be indicated. Even then it seems convenient in the future to avoid this double counting of emissions. It would be prudent, at least, to make explicit the procedure adopted and warn that all emitted carbon mass is considered in the CO2 emissions. An additional observation is that in a Bottom-Up process it would be expected that emissions would be based on experimental values (in the case, measurements that represent the fleet).
In the future, a modification will be introduced in the benemis program so that one can have emission coherent with the carbon balance (avoiding double counting).
The same emission evaluation was carried out for diesel oil (Table 12). The results were compared with those of the inventory. In the Top-Down methodology the contained carbon (17789 Gg in Table 6) and the carbon emissions (17531 Gg in Table 8) were evaluated and they are not much different from those calculated in the Bottom-Up process (17955 Gg in Table 7). The calculated fc value (20,7 tC/TJ) is not much different from the value recommended by IPCC (20,2 tC/TJ) as well, and the small difference could be caused by the relative importance of the small emissions that only correspond to 2.3% of the total emission.
Table 12: : Gas Emissions and Masses of Contained Carbon for Diesel Oil (year 1990) and carbon/ implicit energy factor in the emission coefficients used.
Firewood is another energy source (besides gasoline and alcohol) in which the carbon monoxide in the emissions is important and its analysis is of interest.
The contained carbon value from the Top-Down method (17210 Gg) is lower than that from the Bottom-Up method (19341 Gg), as in the case of gasoline. However, the coefficients used are less trustworthy than those of gasoline and consequently the calculation of the carbon mass is also less trustworthy. The fc value (25.5 tC/TJ) is lower than that indicated by the IPCC and used in the Top-Down approximation (fc=29.9 tC/TJ). Its calculation is shown in Table 13.
Table 13: Emissions for firewood (year 1990) and calculation of the fc coefficient (carbon mass / energy) corresponding to emissions
Regarding the values for charcoal (Table 14) it has a carbon mass/energy factor compatible with that recommended by IPCC (fc=29,9 tC/TJ) and presents a 2.6% difference in carbon emissions.
Table 14: Emissions for charcoal (year 1990) and calculation of the fc coefficient (carbon mass / energy) corresponding to emissions
Analysis of Alcohol Emissions
In the case of alcohol the carbon mass / energy coefficient in the Top-Down method is incoherent with the value obtained using pure ethanol data. The emission values were transferred to Table 15. The calculated carbon mass in emissions (4333 Gg) exceeds the mass calculated in the Top-Down process (3652 Gg) that must be underestimated due to the C mass/ energy used (14.81 tC/TJ), also underestimated.
However, the carbon mass is coherent with the value expected from the carbon content in ethanol. In fact, the alcohol mass (anhydrous +hydrated) is 9063 thousand t of alcohol that corresponds to about 8900 thousand t of pure ethanol. From the chemical formula of ethanol and from the atomic masses involved we conclude that 24/46 of the ethanol mass is made up of carbon. Therefore we have a mass of approximately 4600 thousand t of this element in alcohol consumption in 1990. This estimate is also coherent with the carbon mass in the emitted gases. That is, in this case the Bottom-Up approximation seems trustworthy and the fc value found (17.9 tC/TJ) is coherent with the expected value for ethanol.
Table15: Emissions for alcohol (year 1990) and calculation of the fc coefficient (carbon mass / energy) corresponding to emissions
List of Annexes to the MCT Report, available only in Portuguese at the Internet: http://ecen.com
Contained Carbon, Equivalent Energy and Energy Balance 49 X 46 -
Annex 2: Tables of Contained Carbon in Fuels in Selected Years
Annex 3: Results Of Carbon Balance in Selected Years
Annex 4: Goal 1 Report of the Carbon Balance Project
 For the transformation centers the masses follow BEN’s standards where the values are presented as negative when used in transformation and positive when produced. Presentation of aggregated results does not make sense in this case.
 Consumption in 1990 was 9543 thousand m3 of automotive gasoline and 63 thousand of aviation gasoline. Considering the respective specific masses (0.740 and 0.720 t/m3, respectively) we have a mass of 7041 thousand t of gasoline.
Bases for calculating the emission of green house effect gases.
Omar Campos Ferreira
 The conversion values in the IPCC recommendations regarding the HHV and LHV are based on a very simplified hypothesis that does not take into account the carbon content in each type of fuel.
Graphic Edition/Edição Gráfica:
Tuesday, 11 November 2008.