8 Sensitivity analysis

Life Cycle Assessment of Biogas from Separated slurry

8.1 Overview of sensitivity analysis performed
8.2 Results of the sensitivity analysis

8.1 Overview of sensitivity analysis performed

While building the life cycle inventory for this study (Annexes F to I), the uncertainty regarding the choice of some values and methodologies has been highlighted. Different values or methodologies could have been used and this would have affected the outcome.

A sensitivity analysis consists to answer such “what if” questions by changing some of the inputs in the model and assess the effect this has on the overall results.

In this study, different sensitivity analyses have been performed, as enumerated through the inventory annexes. The sensitivity analyses performed are summarised in table 8.1. As it can be seen in table 8.1, most of these sensitivity analyses refer to parameters that are common for all Scenarios. In such a case, the sensitivity analysis was only performed for one scenario.

Table 8.1. Summary of sensitivity analysis performed for the 4 biogas scenarios assessed

Parameter on which a sensitivity analysis is performed	Description of the parameter		Annexes concerned	Annex for which the sensitivity analysis is performed
Parameter on which a sensitivity analysis is performed	Used in this study	Alternative for sensitivity analysis	Annexes concerned	Annex for which the sensitivity analysis is performed
Soil type	JB3 (sandy)	JB6 (clay)	F, G, H, I	F
Horizon for C in the field^[a]	10 years	100 years	F, G, H, I	F, G, H, I
Electricity avoided from biogas production	Mix marginal (see table 2.1)	• 100 % coal • 100 % natural gas	F, G, H, I	F
Heat avoided from biogas production	Heat from coal	Heat from natural gas	F, G, H, I	F
Biogas utilisation	Used for co-generation of heat and power	Injected in the natural gas grid. It is considered that 100 % biogas = 100 % natural gas replaced, which is a rather rough assumption.	F, G, H, I	F
Heat from biogas: Amount of heat that is “usable” (i.e. not lost)	Only 60 % of the heat surpluses can be used	It is impossible to use the heat surpluses, so no heat surpluses can be used to fulfil a heat demand.	F, G, H, I	F
Long term CO₂ emissions from anaerobic digestion residues	Modelled as for untreated slurry	Higher retention time	F, G, H, I	Discussion
Avoided N: N substitution values	Based on rule a) and b), see section F.28.2 of Annex F	Based on rule c), see section F.28.2 of Annex F	F, G, H, I	F

[a] This is already included in all scenarios

8.2 Results of the sensitivity analysis

8.2.1 Soil type

Figure 8.1 presents the sensitivity analysis for the different soil types. As it can be seen from figure 8.1, the soil type only affects two impact categories;

Global warming
N eutrophication

For this reason, a larger scale presenting the whole impact range for “non-renewable energy” is not presented as this is not affected by the soil type.

The difference between soil type JB3 and JB6 has significance for the actual value of the result, especially regarding nitrogen eutrophication (caused by N leaching). However, the goal of this study is to compare the “biogas scenario” with the reference scenario (the soil type being fixed and the same in both scenarios), and on that perspective, the differences between a given biogas scenario and the reference scenario are not changed if the soil type is changed.

Figure 8.1 is for Scenario F. The calculations and graphs have also been made for scenario G, but as these provided no additional information, they have not been shown in this report.

Figure 8.1 Sensitivity analyses, illustrating the difference between soil type JB3 and JB6
Overall environmental impacts for the selected impact categories – scenario F vs scenario A. Fattening pig slurry management. 10 years time horizon for global warming and for aquatic eutrophication (N).

Click here to see Figure 8.1.

8.2.2 Horizon considered for C during field processes

In all scenarios, the impacts affected by different horizon for C during field processes, namely global warming and N-eutrophication, were presented for both 10 year and 100 year horizon values, see figures 4.2, 5.2, 6.2 and 7.2.

The concept of 10 year horizon and 100 year horizon regards the time for the turnover of C and N in the field, i.e. “after 10 years, how much of the carbon has been transformed into CO₂ and how much of the nitrogen has leached” and the same after 100 years at the field.

As can be seen from figures 4.2, 5.2, 6.2 and 7.2, the time perspective has influence on the actual results for each scenario, but for the comparison between the biogas scenario and the reference scenario, the time horizon does not change the overall conclusions. The contributions to global warming (CO₂ emissions from the field) and the nitrogen leaching is higher over a 100 years perspective – but the relative difference between the biogas scenarios and the reference scenario is at the same magnitude, and most important: The conclusion does not change whether a 10 years or 100 years perspective is taken into account.

8.2.3 Electricity, heat and biogas utilisation

Figures 8.2.A and 8.2.B present the sensitivity analysis for the following possibilities for scenario F:

Sensitivity 1: Avoided electricity is 100 % coal;
Sensitivity 2: Avoided electricity is 100 % natural gas;
Sensitivity 3: Avoided heat is 100 % natural gas;
Sensitivity 4: Biogas is injected in the natural gas grid, and accordingly allow to avoid a corresponding amount of natural gas with regard to the same amount of energy content (it is assumed that 1 MJ energy content in the biogas replace 1 MJ as natural gas);
Sensitivity 5: No avoided heat from biogas (for biogas plants situated in remote areas it might not be possible to utilise the heat for district heating and accordingly, the heat might just be wasted).

From figures 8.2.A and 8.2.B it can be seen that:

The choice of which electricity production (sensitivity 1 and 2) that is avoided when production heat and power from biogas changes the contribution to almost all impact categories, but the overall conclusions are not changed. The sensitivity analysis with 100% coal gives the largest reductions when comparing the biogas scenarios with the reference scenarios.
Changing the source for “avoided heat” (sensitivity 3) does not have significant influence on the overall results
The possibility of injecting biogas directly into the natural grid (a future possibility, which requires some treatment of the biogas) in sensitivity 4 does change the overall conclusions somewhat, as no electricity production is avoided which is especially significant for the contributions to global warming and the consumption of non-renewable resources. Furthermore, in this sensitivity analysis, it is assumed that the biogas plant then have to buy natural gas to for the heating requirements at the biogas plant (in the biogas scenarios this heat is taken from the heat produced from the biogas and is in this way “free”). It should be noted that if this possibility is seriously considered – to inject biogas directly into the natural gas grid – a more thorough Life Cycle Assessment is required in order to investigate the consequences of this. This sensitivity analysis only provides a very rough estimate.
In sensitivity analysis 5, the heat produced by the biogas plant is not utilised. As can be seen from figures 8.2.A and 8.2.B this has influence on all impact categories, especially on global warming and consumption of non-renewable resources. The reductions gets smaller, however, the conclusion remains: The biogas production reduces the contributions to global warming and the savings of non-renewable resources.

The calculations and graphs have also been made for scenario G, but as these provided no additional information, they have not been shown in this report.

Figure 8.2.A Sensitivity analysis: avoided heat and electricity, biogas utilisation and amount of usable heat.
Scenario F vs scenario A. Fattening pig slurry management. Soil type JB3. 10 years time horizon for global warming and for aquatic eutrophication (N). Axis ranging from -180 to 120.

Click here to see 8.2.A

Figure 8.2.B Sensitivity analysis: avoided heat and electricity, biogas utilisation and amount of usable heat.
Scenario F vs scenario A. Fattening pig slurry management. Soil type JB3. 10 years time horizon for global warming and for aquatic eutrophication (N). Axis ranging from -950 to 200.

Click here to see 8.2.B

8.2.4 Long term CO₂ emissions from anaerobic digestion residues

Regarding the long term (decades to a century) emission of CO₂ from various residues from biogas production, there is a lack of empirical data to support the model findings. It is a well-established fact that C from animal manure has a higher long-term retention than C from plant residues (e.g. Stemmer et al., 2000), which is supported by many independent long-term field trials. The same issue has, to our knowledge, not to date been investigated for biogas residues.

The present model simulations are based on the conservative assumption that the long-term retention equals that of C from animal manure. From a theoretical viewpoint, it may be argued that the retention is a bit higher for biogas residues, but this is not verifiable, neither can there be put a possible coefficient on this hypothetical effect.

If the retention should be higher than presumed here, it all in all would mean that the biogas system is slightly more favourable than estimated. But only results from possible, future long-term experiments can disclose that.

8.2.5 Avoided N fertiliser

Sensitivity analysis for alternative methods for calculating the replaced amount of mineral N fertiliser has been performed by a combination of rule a, c and d as mentioned in section F.28.2 of Annex F. The calculations are shown in tables 8.2-8.5 below:

In table 8.2, rule c) has been applied (“The producer of the degassed biomass (i.e. the biogas plant staff) sets the “mineral fertiliser replacement value” for the degassed biomass based on representative measurement of samples of the degassed biomass.”). It is assumed that the mineral fertiliser value of the degassed biomass corresponds to 90% of the N content in the degassed biomass. Rule a) still applies for the separation (the outgoing amount is identical to the ingoing amount).
In table 8.3, rule d) has been applied, which is rather similar to the calculations in table 8.2 but with 75% instead of 90% for the fertiliser value of the degassed biomass.
In table 8.4, rule d) has been applied, but it is assumed that all values are based on measured values of the N content instead of Norm Data values (as in table F.35 of Annex F). The measured values are assumed to be identical to the N content in the fractions.
Table 8.5 gives an overview of the various sensitivity analysis and yield changes.

Table 8.2 Sensitivity analysis for the replaced amount of mineral N fertiliser in Annex F. All calculations per 1000 kg slurry ex-animal. Application of rule c).

Calculations
Step 1: Substitution value for fibre fraction to biogas plant Identical to Step 1 in table F.35. N in fibre fraction = 0.97017 kg N per 1000 kg slurry ex-animal.
Step 2: Acknowledging the above, make the weighted sum of the substitution values (liquid and fibre). For raw pig slurry, the substitution value is 75 %. Identical to Step 2 in table F.35. Mineral fertiliser replacement value for the liquid fraction (at the farm): 2.47135 kg N
Step 3: Substitution values for the materials leaving the biogas plant. In this calculation, it is assumed that the producer of the degassed biomass (i.e. the biogas plant staff) sets the “mineral fertiliser replacement value” for the degassed biomass based on representative measurement of samples of the degassed biomass. The N content of the degassed biomass is 8.722 kg N per 1000 kg degassed biomass (table F.22). As can be seen in figure F.1, there is 319.84 kg degassed biomass. The substitution value for degassed biomass set to 90% in these calculations, accordingly the mineral fertiliser replacement value for the degassed biomass is 8.722 kg N per 1000 kg degassed biomass * 319.84 kg degassed biomass * 90% = 2.5107 kg N. This value is used for the further calculations.
Step 4a: Use a substitution value of 50% for the fibre fraction of the degassed material from the biogas plant (like step 1) Amount of degassed fibre fraction: 77.272 kg (see figure F.1). N in degassed fibre fraction: 7.65 kg per 1000 kg fibre fraction (see table F.26). Substitution value: 50% * 7.65 kg per 1000 kg fibre fraction * 77.272 kg fibre fraction / 1000 kg = 0.2956 kg N per 1000 kg slurry ex-animal. Mineral fertiliser replacement value the degassed fibre fraction: 0.2956 kg N
Step 4b: Calculation of the substitution value for the liquid fraction as “the rest”. Here, rule (a) applies again: “The sum of the “mineral fertiliser replacement value” of the outgoing fractions shall be the same as the “mineral fertiliser replacement value” of the ingoing slurry before separation”. Total substitution value out of biogas plant = total substitution value in biogas plant, as calculated in step 3: 2.5107 kg N. Substitution value for the liquid fraction = total from biogas plant – fibre fraction (from step 4a) = 2.5107 kg N - 0.2956 kg N = 2.2151 kg N Mineral fertiliser replacement value for the degassed liquid fraction (after the biogas plant: 2.2151 kg N
Total amount of substituted mineral N fertiliser in the system 2.47135 kg N + 0.2956 kg N + 2.2151 kg N = 4.982 kg N

Table 8.3 Sensitivity analysis for the replaced amount of mineral N fertiliser in Annex F. All calculations per 1000 kg slurry ex-animal. Application of rule d), norm data values.

Calculations
Step 1: Substitution value for fibre fraction to biogas plant Identical to Step 1 in table F.35. N in fibre fraction = 0.97017 kg N per 1000 kg slurry ex-animal.
Step 2: Acknowledging the above, make the weighted sum of the substitution values (liquid and fibre). For raw pig slurry, the substitution value is 75 %. Identical to Step 2 in table F.35. Mineral fertiliser replacement value for the liquid fraction (at the farm): 2.47135 kg N
Step 3: Substitution values for the materials leaving the biogas plant. In this calculation, it is assumed that the producer of the degassed biomass (i.e. the biogas plant staff) sets the “mineral fertiliser replacement value” for the degassed biomass based on representative measurement of samples of the degassed biomass. The N content of the degassed biomass is 8.722 kg N per 1000 kg degassed biomass (table F.22). As can be seen in figure F.1, there is 319.84 kg degassed biomass. The substitution value for degassed biomass set to 75% in these calculations, accordingly the mineral fertiliser replacement value for the degassed biomass is 8.722 kg N per 1000 kg degassed biomass * 319.84 kg degassed biomass * 75% = 2.0922 kg N. This value is used for the further calculations.
Step 4a: Use a substitution value of 50% for the fibre fraction of the degassed material from the biogas plant (like step 1) Amount of degassed fibre fraction: 77.272 kg (see figure F.1). N in degassed fibre fraction: 7.65 kg per 1000 kg fibre fraction (see table F.26). Substitution value: 50% * 7.65 kg per 1000 kg fibre fraction * 77.272 kg fibre fraction / 1000 kg = 0.2956 kg N per 1000 kg slurry ex-animal. Mineral fertiliser replacement value the degassed fibre fraction: 0.2956 kg N
Step 4b: Calculation of the substitution value for the liquid fraction as “the rest”. Here, rule (a) applies again: “The sum of the “mineral fertiliser replacement value” of the outgoing fractions shall be the same as the “mineral fertiliser replacement value” of the ingoing slurry before separation”. Total substitution value out of biogas plant = total substitution value in biogas plant, as calculated in step 3: 2.0922 kg N. Substitution value for the liquid fraction = total from biogas plant – fibre fraction (from step 4a) = 2.0922 kg N - 0.2956 kg N = 1.7966 kg N Mineral fertiliser replacement value for the degassed liquid fraction (after the biogas plant: 1.7966 kg N
Total amount of substituted mineral N fertiliser in the system 2.47135 kg N + 0.2956 kg N + 1.7966 kg N = 4.56355 kg N

Table 8.4 Sensitivity analysis for the replaced amount of mineral N fertiliser in Annex F. All calculations per 1000 kg slurry ex-animal. Application of rule d), measured values.

Calculations
Step 1: Substitution value for fibre fraction to biogas plant Identical to Step 1 in table F.35. N in fibre fraction = 0.97017 kg N per 1000 kg slurry ex-animal.
Step 2: Acknowledging the above, make the weighted sum of the substitution values (liquid and fibre). For raw pig slurry, the substitution value is 75 %. The calculations are identical to the calculations in table F.35 except for that instead of using the Danish Norm Data for the N content of the raw pig slurry, the measured values has been used, i.e. 4.80 kg N per 1000 kg slurry ex storage (table A. 1) instead of 5.00 kg N per 1000 kg slurry ex storage (table A. 1). The calculations follow the calculations in table F.35: For the system, the mineral fertiliser substitution value is then: 4.80 kg N per 1000 kg slurry ex storage * 1086 kg slurry ex storage / 1000 kg slurry ex animal * 75% = 3.9096 kg N per 1000 kg slurry ex-animal. However, there is only 845.064 kg slurry being separated (see figure F.1), i.e. 3.9096 kg/1000 kg * 845.064 kg = 3.30386 kg N. Of this 3.30386 kg N, 0.97017 kg N belongs to the fibre fraction (as calculated in step 1). The difference i.e.: 3.30386 kg N – 0.97017 kg N = 2.33369 kg N belongs to the liquid fraction. Mineral fertiliser replacement value for the liquid fraction (at the farm): 2.33369 kg N
Step 3: Make a weighed sum of the substitution values for the materials entering the biogas plant. Rule (b): “Mass balance in and out of Biogas Plant – i.e. the “mineral fertiliser replacement value” of the outgoing biomass is calculated in accordance with the ingoing biomass”. The raw slurry going directly to biogas plant (without separation) has a mineral fertiliser replacement value of 3.9096 kg N per 1000 kg slurry (as described under step 2 above – 75% of 4.80 kg N ex storage). The amount of this raw slurry is 154.936 kg (see figure F.1). Its mineral fertiliser replacement value is: 3.9096 kg N per 1000 kg slurry * 154.936 kg slurry/1000 kg = 0.60574 kg N per 1000 kg slurry ex-animal. This is the substitution value for the raw slurry into the biogas plant. At the plant, a biomass mixture is made from this raw slurry and the fibre fraction from step 1, so the substitution value for this input mixture is: 0.97017 kg N (fibre fraction, step 1) + 0.60574 kg N (raw slurry, see above) = 1.57591 kg N. This is the substitution value for the input biomass mixture going into the biogas plant, and accordingly also the substitution value for the degassed biomass mixture coming out of the biogas plant – i.e. the degassed biomass before separation. This value is used for the further calculations.
Step 4a: Use a substitution value of 50% for the fibre fraction of the degassed material from the biogas plant (like step 1) Amount of degassed fibre fraction: 77.272 kg (see figure F.1). N in degassed fibre fraction: 7.65 kg per 1000 kg fibre fraction (see table F.26). Substitution value: 50% * 7.65 kg per 1000 kg fibre fraction * 77.272 kg fibre fraction / 1000 kg = 0.2956 kg N per 1000 kg slurry ex-animal. Mineral fertiliser replacement value the degassed fibre fraction: 0.2956 kg N
Step 4b: Calculation of the substitution value for the liquid fraction as “the rest”. Here, rule (a) applies again: “The sum of the “mineral fertiliser replacement value” of the outgoing fractions shall be the same as the “mineral fertiliser replacement value” of the ingoing slurry before separation”. Total substitution value out of biogas plant = total substitution value in biogas plant, as calculated in step 3: 1.57591 kg N. Substitution value for the liquid fraction = total from biogas plant – fibre fraction (from step 4a) = 1.57591 kg N - 0.2956 kg N = 1.28031 kg N Mineral fertiliser replacement value for the degassed liquid fraction (after the biogas plant: 1.28031 kg N
Total amount of substituted mineral N fertiliser in the system 2.33369 kg N + 0.2956 kg N + 1.28031 kg N = 3.9096 kg N

Table 8.5 Sensitivity analyses for the replaced amount of mineral N fertiliser in Annex F. All calculations per 1000 kg slurry ex-animal. Overview of all results.

Description of the sensitivity analysis	Amount of replaced mineral N fertiliser kg N per 1000 kg slurry ex animal
Base case (table F.35, Annex F)	4.0725 kg N
Sensitivity 1 (Table 8.2): Rule c) has been applied (“The producer of the degassed biomass (i.e. the biogas plant staff) sets the “mineral fertiliser replacement value” for the degassed biomass based on representative measurement of samples of the degassed biomass.”). It is assumed that the mineral fertiliser value of the degassed biomass corresponds to 90% of the N content in the degassed biomass.	4.982 kg N
Sensitivity 2 (Table 8.3): Rule d) has been applied, which is rather similar to the calculations in table F.37 but with 75% instead of 90% for the fertiliser value of the degassed biomass.	4.564 kg N
Sensitivity 3 (Table 8.4): Rule d) has been applied, but it is assumed that all values are based on measured values of the N content	3.9096 kg N

The different methods for how the amount of avoided N fertiliser is calculated do not change the overall conclusions, which is illustrated in figures 8.3.A and 8.3.B, where the results for Scenario A, Scenario F, sensitivity 1 and sensitivity 3 are compared. It can be seen that if the amount of avoided N fertiliser is increased it will just reduce the yield, which will to some extend counteract for the increased amount of avoided N.

Or, to explain this further: When applying pig slurry, the N in the slurry replace 75% mineral fertiliser, which means that if applying 100 kg N in slurry, the farmer has to apply 75 kg mineral N fertiliser less (Gødskningsbekendtgørelsen (2008), paragraph 21). For example, if the farmer has a field with winter barley, and the soil type is JB3, the farmer has a “Nitrogen quota” for that field at 149 kg N per ha (Plantedirektoratet, 2008). If the farmer applies 100 kg N per ha as pig slurry, this accounts for 75 kg N per ha, which means that the farmer is allowed to apply the remaining 149 kg N per ha – 75 kg N per ha = 74 kg N per ha as mineral N fertiliser. In consequence, the farmer is interested in that the “mineral fertiliser replacement values” of the slurry are as small as possible as this will increase the yield.

When the yield is increased, the extra yield is subtracted from the system.

The increase of a crop yield of is assumed to replace winter wheat produced somewhere else in Denmark. This is a very simplified assumption. The consequences of increased crop yield probably replace another crop type somewhere else in the world. It is beyond the frame of this project to identify the avoided crop as a consequence of the increased crop yield. In this report, it is assumed that the increased crop yield replace winter wheat, using data from the process “Wheat, conventional, from farm“ from LCA-food data base (modified with the updated data for production of fertilisers as described in Annex A in Wesnaes et. al (2009)).

Figure 8.3.A Sensitivity analysis: Replaced amount of mineral N.
Scenario F vs scenario A. Fattening pig slurry management. Soil type JB3. 10 years time horizon for global warming and for aquatic eutrophication (N). Axis ranging from -180 to 120.

Click here to see Figure 8.3.A

Figure 8.3.B Sensitivity analysis: Replaced amount of mineral N.
Scenario F vs scenario A. Fattening pig slurry management. Soil type JB3. 10 years time horizon for global warming and for aquatic eutrophication (N). Axis ranging from -900 to 200.

Click here to see Figure 8.3.B