Qualitative modeling
Based on the analysis of the interactions between the variables defined for this work, we used Systems Dynamics to graphically represent the operation of the emissions compensation system via forest restoration and carbon credit, as can be seen in the image below (Fig. 3).
In the model, it is possible to notice that the variable “Calculator Output” (CO) provides data that can follow three different paths: a path towards the subsystem of uncompensated emissions; another towards the subsystem of carbon credits; and a path towards the forest subsystem. Since the constant type variables “Percentage of Carbon Absorbed by Credits” (PCAC) and “Percentage of Carbon Absorbed by the Forest” (PCAF) work as a quantitative control of the variable “Calculator Output” (CO) in percentage that will be directed to the carbon credits and forest subsystems, respectively. The amount that is not directed to any of these subsystems enters the subsystem of uncompensated emissions.
The model is composed of 3 subsystems and 24 variables, where causal relationships are expressed by arrows and positive and negative signs, where the arrows indicate that there is an influence (cause) of one variable on the result of the other (effect); and the signs indicate whether the manipulated variable responds, with an increase (positive) or decrease (negative) in its value, to any increment of the variable that exerts influence on it. As an example, it is possible to notice in the model that the increase in the Average Cost of Forest Restoration per Hectare (CFRH), together with the increase in the Equivalent Forest Area (EFA), increases the Total Forest Cost (TFC), while the increase of the amount of Carbon Absorbed per Hectare (CAH) decreases the EFA and the TFC. Another example refers to increases in the Average Unit Price of Carbon Credits (AUPCC) and the Quantity of CO2 absorbed by Carbon Credits (QCACC), both of which generate an increase in the Carbon Credits Total Cost (CCTC).
Quantitative modeling
Having defined the subsystems, variables and the causal relationships between them, we set out to prepare the mathematical formulas of the model (appendix) and the subsequent exploration of potential paths for CO2 compensation through 4 different scenarios generated by changes in the percentages of emissions that will be absorbed by the forest and/or carbon credits. Scenarios for emissions generated by road and intermodal transport are presented in the topics below.
Scenario of offsetting emissions via carbon credits
It was considered in this simulation that offsets would be carried out only with carbon credits (table 4). It is possible to see in this scenario that the emissions generated by intermodal transport (1.79 tCO₂) corresponded to only 47.7% of the emissions generated by road transport (3.76 tCO₂), a pattern that was also repeated in the Carbon Credit Total Cost (CCTC), in that the cost related to intermodal transport corresponded to only 47.7% of the cost associated with road transport.
Tabela 4 - Simulação da compensação de 100% das emissões de CO2 com créditos de carbono para os dois modais.
Variables
|
Units
|
Modal
|
Road
|
Intermodal
|
Emissions (E)
|
tCO₂
|
3.76
|
1.79
|
Percentage of carbon absorbed by credits (PCAC)
|
%
|
100
|
100
|
Average unit price of carbon credit (AUPCC)
|
US$
|
17.31
|
17.31
|
Carbon Credit Total Cost (TCCC)
|
US$
|
65.09
|
30.98
|
Scenario for compensation via forest planting
In the simulation for offsetting emissions through forest planting, it was established that all quantity related to emissions was directed to the forest subsystem (table 5).
Tabela 5 - Simulação da compensação de 100% das emissões de CO2 com plantio florestal para os dois modais.
Variables
|
Units
|
Modal
|
Road
|
Intermodal
|
Emissions (E)
|
tCO₂
|
3.76
|
1,79
|
Percentage of carbon absorbed by the forest (PCAF)
|
%
|
100
|
100
|
Carbon absorbed per hectare (CAH)
|
tCO₂
|
470.74
|
470.74
|
Equivalent forest area (EFA)
|
Hectare
|
0.00798742
|
0.00380252
|
Average cost of forest restoration per hectare (ACFRH)
|
US$
|
2000
|
2000
|
Total Forest Cost (TFC)
|
US$
|
15.97
|
7.60
|
In this scenario, it is possible to perceive the percentage variation between the emissions of the two modes (52.4%); other variables, such as the Equivalent Forest Area (EFA) and the Total Forest Cost (TFC), also showed the same percentage variation (52.4%) in a comparison between modes. Regarding the last variable, it is possible to notice, in a comparison with the Carbon Credits Total Cost (CCTC) (table 4), that its values are lower in both modes, suggesting that the compensation of emissions through restoration forestry in the Atlantic Forest may be cheaper than the compensation generated by the purchase of carbon credits, at the price level used in this work.
Scenario for offsetting 50% of emissions with carbon credits and the other 50% with forest planting
It was established in this simulation that half of the CO2 emissions would be compensated through forest planting, and the other half through carbon credits. For this purpose, the percentages of CO2 absorbed by the two routes were defined, as can be seen in Table 6, where the PCAF and PCAC for road and intermodal transport assume a value of 50%. Thus, the amount of Equivalente Forest Area (EFA) to half of the emissions generated by road transport and by intermodal transport was 0.003994 hectares and 0.001901 hectares, respectively; generating a TFC of US$7.99 and US$3.80 for road and intermodal transport, respectively. Making a comparison, the cost associated with offsetting emissions generated by road transport was 2.1 times higher than the cost associated with offsetting emissions generated by intermodal transport.
Table 6
Simulation of compensation of 50% of CO2 emissions in the carbon credits subsystem and the other 50% in the forest subsystem for both modes.
Variables
|
Units
|
Modal
|
Road
|
Intermodal
|
Emissions (E)
|
tCO₂
|
3.76
|
1.79
|
Forest subsystem
|
Percentage of carbon absorbed by the forest (PCAF)
|
%
|
50
|
50
|
Absorbed carbon per hectare (ACH)
|
tCO₂
|
470.74
|
470.74
|
Equivalent forest area (EFA)
|
Hectare
|
0.003994
|
0.001901
|
Average cost of forest restoration per hectare (ACFRH)
|
US$
|
2000
|
2000
|
Total Forest Cost (TFC)
|
US$
|
7.99
|
3.80
|
Subsystem of carbon credits
|
Percentage of carbon absolved by credits (PCAC)
|
%
|
50
|
50
|
Average unit price of carbon credit (AUPCC)
|
US$
|
17.31
|
17.31
|
Carbon Credits Total Cost (CCTC)
|
US$
|
32.54
|
19.30
|
Total offset cost (TOC)
|
US$
|
40.53
|
19.30
|
With regard to compensation via carbon credits, the values generated in this simulation for the total cost of this operation were US$ 32.54 for road transport and US$ 19.30 for intermodal transport. In a comparison, the cost for road transport was 1.7 times higher than the cost for intermodal transport, suggesting once again that the costs for offsetting CO2 emissions generated by intermodal cargo transport are lower than those for emissions generated by road transport, preserving the same settings used in the present simulation.
Comparing the TFC and CCTC, we see that the results for the first were lower than the second in both transport systems. With regard to road transport, the cost related to planting trees corresponded to only 24.6% of the cost related to the purchase of carbon credits; regarding intermodal transport, the cost related to compensation via forest restoration was equivalent to 19.7% of the cost related to the purchase of carbon credits.
Finally, the TOC, which corresponds to the sum of TFC and CCTC, was US$40.53 for road transport and US$19.30 for intermodal transport, a percentage change of 52.4%, which again suggests that the costs of offsetting CO2 emissions from intermodal transport are lower than the costs associated with offsetting emissions from road transport over the same route and at the same price level for carbon credits and average cost of forest restoration per hectare in the Atlantic Forest.
Scenario for offsetting 30% of emissions with carbon credits and 30% with forest planting
In this scenario, it was established that only 60% of emissions would be offset, and the remainder (40%) would go to the uncompensated emissions subsystem. In this simulation, it was possible to better explore the capabilities of the model, making it possible to generate more outputs than in previous scenarios for different time periods, as can be seen in the table below (Table 7).
Table 7
Simulation of compensation of 30% of CO2 emissions with carbon credits and 30% with forest planting for road transport.
Variáveis
|
Unidades
|
Ano
|
0
|
1
|
2
|
3
|
4
|
5
|
E
|
tCO₂
|
0
|
3.76
|
3.76
|
3.76
|
3.76
|
3.76
|
TOC
|
US$
|
0
|
24.32
|
24.32
|
24.32
|
24.32
|
24.32
|
Forest subsystem
|
CAH
|
tCO₂
|
470.74
|
470.74
|
470.74
|
470.74
|
470.74
|
470.74
|
ACFRH
|
US$
|
2000
|
2000
|
2000
|
2000
|
2000
|
2000
|
EFA
|
Hectare
|
0
|
0.002396
|
0.002396
|
0.002396
|
0.002396
|
0.002396
|
PCAF
|
%
|
30
|
30
|
30
|
30
|
30
|
30
|
CSF
|
tCO₂
|
0
|
1.13
|
1.13
|
1.13
|
1.13
|
1.13
|
FCS
|
tCO₂
|
0
|
1.13
|
3.38
|
7.90
|
16.92
|
34.97
|
TFC
|
US$
|
0
|
4.79
|
4.79
|
4.79
|
4.79
|
4.79
|
Carbon credits subsystem
|
AUPCC
|
US$
|
17.31
|
17.31
|
17.31
|
17.31
|
17.31
|
17.31
|
PEAC
|
%
|
30
|
30
|
30
|
30
|
30
|
30
|
ACAC
|
tCO₂
|
0
|
1.13
|
1.13
|
1.13
|
1.13
|
1.13
|
CSC
|
tCO₂
|
0
|
1.13
|
3.38
|
7.90
|
16.92
|
34.97
|
CCTC
|
US$
|
0
|
19.53
|
19.53
|
19.53
|
19.53
|
19.53
|
Uncompensated emissions subsystem
|
UE
|
tCO₂
|
0
|
1.50
|
4.51
|
10.53
|
22.56
|
46.62
|
UETCC
|
unidade
|
0
|
1.50
|
4.51
|
10.53
|
22.56
|
46.62
|
CCUE
|
US$
|
0
|
26.03
|
78.10
|
182.24
|
390.51
|
807.06
|
UETFA
|
hectare
|
0
|
0.00319
|
0.00958
|
0.02236
|
0.04792
|
0.09904
|
CFUE
|
US$
|
0
|
6.39
|
19.17
|
44.73
|
95.85
|
198.09
|
E = emissions; TOC = total offset cost; CAH = Carbon absorbed per hectare; ACRFRH = average cost of forest restoration per hectare; EFA = equivalent forest area; PCAF = Percentage of carbon absorbed by the forest; CSF = Carbon sequestered by the forest; FCS = forest carbon stock; TFC = total forest cost; AUPCC = average unit price of the carbon credit; PEAC = percentage of emissions absorbed by credits; ACAC = amount of carbon absorbed by credits; CSC = carbon stock in credits; CCTC = carbon credits total cost UE = uncompensated emissions; UETCC = Uncompensated emissions in terms of carbon credits; CCUE = cost in credits of uncompensated emissions; UETFA = uncompensated emissions in terms of forest area; CFUE = cost in forest of uncompensated emissions.
Regarding the scenario for road transport, we can notice in the forest subsystem that, while other variables remain constant from year 1, the forest carbon stock (ECV) is varying over time, this behavior is due to the fact of its measurement is not restricted to just a period of time (year) like the others, as it makes the sum of the present value plus previous value(s). The same behavior can be seen in the scenario for intermodal transport (Table 8).
Table 8
Simulation of offsetting 30% of CO2 emissions with carbon credits; and 30%, with forest planting for road and rail transport.
Variáveis
|
Unidades
|
Ano
|
0
|
1
|
2
|
3
|
4
|
5
|
E
|
tCO₂
|
0
|
1.79
|
1.79
|
1.79
|
1.79
|
1.79
|
TOC
|
US$
|
0
|
11.58
|
11.58
|
11.58
|
11.58
|
11.58
|
Forest subsystem
|
CAH
|
tCO₂
|
470.74
|
470.74
|
470.74
|
470.74
|
470.74
|
470.74
|
ACFRH
|
US$
|
2000
|
2000
|
2000
|
2000
|
2000
|
2000
|
EFA
|
Hectare
|
0
|
0.001141
|
0.001141
|
0.001141
|
0.001141
|
0.001141
|
PCAF
|
%
|
30
|
30
|
30
|
30
|
30
|
30
|
CSF
|
tCO₂
|
0
|
0.54
|
0.54
|
0.54
|
0.54
|
0.54
|
FCS
|
tCO₂
|
0
|
0.54
|
1.61
|
3.76
|
8.06
|
16.65
|
TFC
|
US$
|
0
|
2.28
|
2.28
|
2.28
|
2.28
|
2.28
|
Carbon credits subsystem
|
AUPCC
|
US$
|
17.31
|
17.31
|
17.31
|
17.31
|
17.31
|
17.31
|
PEAC
|
%
|
30
|
30
|
30
|
30
|
30
|
30
|
ACAC
|
tCO₂
|
0
|
0.54
|
0.54
|
0.54
|
0.54
|
0.54
|
CSC
|
tCO₂
|
0
|
0.54
|
1.61
|
3.76
|
8.06
|
16.65
|
CCTC
|
US$
|
0
|
9.30
|
9.30
|
9.30
|
9.30
|
9.30
|
Uncompensated emissions subsystem
|
EU
|
tCO₂
|
0
|
0.72
|
2.15
|
5.01
|
10.74
|
22.20
|
UETCC
|
unidade
|
0
|
0.72
|
2.15
|
5.01
|
10.74
|
22.20
|
CCUE
|
US$
|
0
|
12.39
|
37.18
|
86.76
|
185.91
|
384.21
|
UETFA
|
hectare
|
0
|
0.001521
|
0.004563
|
0.010647
|
0.022815
|
0.047151
|
CFUE
|
US$
|
0
|
3.04
|
9.13
|
21.29
|
45.63
|
94.30
|
E = emissions; TOC = total offset cost; CAH = Carbon absorbed per hectare; ACRFRH = average cost of forest restoration per hectare; EFA = equivalent forest area; PCAF = Percentage of carbon absorbed by the forest; CSF = Carbon sequestered by the forest; FCS = forest carbon stock; TFC = total forest cost; AUPCC = average unit price of the carbon credit; PEAC = percentage of emissions absorbed by credits; ACAC = amount of carbon absorbed by credits; CSC = carbon stock in credits; CCTC = carbon credits total cost UE = uncompensated emissions; UETCC = Uncompensated emissions in terms of carbon credits; CCUE = cost in credits of uncompensated emissions; UETFA = uncompensated emissions in terms of forest area; CFUE = cost in forest of uncompensated emissions.
With regard to the carbon credits subsystem, in both scenarios we can also note the presence of only one stock variable, namely the variable carbon stock in credits (CSC). The emissions that are being offset via carbon credits are being stored there. The value of these variables in each period is equal to the value of the forest carbon stock in the same period in their respective scenarios, because the percentage of emissions that will be offset in both subsystems is the same (30%).
In both scenarios, the Total Forest Cost (TFC) and Carbon Credits Total Cost (CCTC) were different, with a percentage variation of 75.5% between them between years 1 and 5, since costs remained constant between these years old. In both scenarios, costs related to forest restoration (TFC) were lower than costs related to offsetting via carbon credits (CCTC) between years 1 and 5.
As 40% of CO2 emissions entered the subsystem of uncompensated emissions each year, outputs were generated in both scenarios between years 1 and 5 referring to the quantification of these uncompensated emissions in terms of forest area (UETFA) and credits (UETCC), together with their respective costs, CFUE and CCUE (Tables 7 and 8).
In this subsystem there is only one inventory variable (UE), but there is a peculiarity, all the other variables also vary over time, but not because they are of the inventory type, but because they are varying due to the only variable: existing inventory .
Regarding uncompensated emissions in terms of forest area (UETFA), we have a variation over time that represents the increase in the forest area necessary to offset accumulated emissions over time. There is a percentage variation of 52.4% between UETFA generated by road transport and by intermodal transport between years 1 and 5. The highest values were measured for road transport. The same percentage variation (52.4%) was also found for the values of UETCC between years 1 and 5, once again the values related to road transport were greater than the values related to the intermodal transport.
With regard to costs, we have two variables: one related to possible costs to compensate, through restoration planting, emissions not yet compensated (CFUE); and another variable associated with the likely costs of buying carbon credits to offset emissions not yet offset (CCUE). There was variation in the values of these two variables in the two scenarios between years 1 and 5, and within each scenario there was a percentage variation of 75.5% between CCUE and CFUE (Fig. 5AB). Between scenarios there was a percentage variation of 52.4%, both for the CCUE and CFUE values between years 1 and 5 (Fig. 5CD).
Figure 5 - Comparisons between the values assumed over time by the variables Cost in credits of uncompensated emissions (CCUE) and Cost in forest of Uncompensated Emissions (CFUE) in the two scenarios (road and rail).
The present work presents a singularity in relation to the cited works in terms of the outputs of the model. While the works by Machado et al. (2013), Du et al. (2019), Chen (2020) and Tang et al. (2020) aimed at simulating GHG emissions or carbon stock, this work focused on measuring emissions related to the Atlantic Forest area and/or carbon credits, that is, while for the present work the emissions are inputs in the model, in the cited works the emissions are outputs in the models.
With regard to the comparison of data referring to road and rail modes, it is evident that intermodal emits less GHG; and, consequently, generates less compensation costs. This phenomenon is due to the fact that rail transport is more energy efficient, that is, trains consume less fuel per ton loaded than trucks, which, in turn, have the advantage of greater operational flexibility and greater delivery capacity, that is, with trucks it is possible to travel on different types of roads and deliver goods to most destinations (SUN, YU & HUANG, 2021; CNT, 2022).
Since the cargo transport matrix in Brazil is still very dependent on road transport, intermodality appears, a priori, as a more viable option if we only take into account the environmental aspect, but when the economic aspect is brought into focus, there are important obstacles to its adoption as the high costs of investments in the expansion of the railway network and the maintenance of existing, but underutilized, railway lines (PINTO et al, 2018).
In regions where there are rail networks widely used for the transport of agricultural cargo and ore, such as in the Southeast region of Brazil, higher maintenance costs can be mitigated thanks to the large volume of cargo that can be transported in a single trip, in addition to the reductions in significant effects of environmental externalities (MARCHETTI & WANKE, 2018).
Regarding the choice between the purchase of carbon credits or forest planting, it does not seem to be an easy decision, even the results showing, a priori, that the cheapest option is forest planting, as there are several cost drivers that are difficult to quantify or predict in forest restoration projects, requiring contingency funds for corrective actions (BRANCALION ET AL, 2019; ZANINI ET AL, 2021).
It is known that forests are important carbon sinks, but the accumulation of carbon in the form of plant biomass varies due to factors such as climate, soil, topography, plant species, among others; it is extremely important to know these factors and their impacts on the local native vegetation, together with the socio-ecological context of the area, in order to choose the type of forest restoration appropriate for the location, and thus be able to more adequately scale the costs and results of the project with regard to carbon storage (SHIMAMOTO ET AL, 2014; BRANCALION ET AL, 2021).
As it is an easier message for the consumer in ecomarketing initiatives, the planting of trees has been adopted by numerous companies around the world in projects to offset GHG emissions, but the results in the field do not always correspond to the goals previously communicated to consumers., falling, in many cases, short of what was promised, which is configured as greenwashing ("green washing", in a literal translation), that is, an act of forging a false image of "carbon neutral" to its processes, products and/or services; this very common failure in forest restoration projects is caused by errors in choosing the species and/or seedlings to be planted, lack of monitoring of planting, insufficient financial resources, among other causes (BOSSHARD ET AL, 2021).
If a company chooses to offset emissions via carbon credits from REDD+, in order, for example, not to assume the risks inherent in forest planting, it must pay attention to some points so that the purchase of these certificates is safe and responsible. It is necessary to verify the history, technical capacity and suitability of the entity responsible for the REDD + project; it is also necessary to make sure that the REDD + project and its respective credits have been registered in a recognized independent platform, such as the Verified Carbon Standard (VCS); it is important to know what are the strategies and activities adopted by the institution responsible for the REDD + project regarding the good relationship with the local community and the conservation of the standing forest, in the face of the pressure of local deforestation; and make sure that the REDD + project in which are provides materials and content necessary for communication with the public, such as photos, videos, availability for visits to the place where the project is being implemented and contact for information (ALIANÇA BRASIL NBS, 2022).