pascal.suer@swedgeo.se, tel fax This is the accepted author manuscript version of the paper published as Suer, P., and Andersson-Sköld, Y

(1)

BIOFUEL OR EXCAVATION? - LIFE CYCLE ASSESSMENT (LCA) OF SOIL REMEDIATION OPTIONS

Pascal Suer*, Yvonne Andersson-Sköld

Swedish Geotechnical Institute, 58193 Linköping, Sweden.

* pascal.suer@swedgeo.se, tel +46 13 201889, fax 0+46 13 201912

This is the accepted author manuscript version of the paper published as

Suer, P., and Andersson-Sköld, Y. (2011). "Biofuel or excavation - Life cycle assess- ment (LCA) of soil remediation options." Biomass & Bioenergy, 35(2), 969-981.

http://dx.doi.org/10.1016/j.biombioe.2010.11.022

ABSTRACT

The environmental consequences of soil remediation through biofuel or through dig- and-dump were compared using life cycle assessment (LCA). Willow (Salix viminalis) was actually grown in situ on a discontinued oil depot, as a phytoremediation treatment.

These data were used for the biofuel remediation, while excavation-and-refill data were estimated from experience. The biofuel remediation had great environmental advantages compared to the ex situ excavation remediation. With the ReCiPe impact assessment method, which included biodiversity, the net environmental effect was even positive, in spite of the fact that the wood harvest was not utilised for biofuel production, but left on the contaminated site. Impact from the Salix viminalis cultivation was mainly through land use for the short rotation coppice, and through journeys of control personnel. The latter may be reduced when familiarity with biofuel as a soil treatment method increases. The excavation and refill remediation was dominated by the landfill and the transport of contaminated soil and backfill.

Keywords: LCA, biofuel, remediation, contaminated soil.

1 INTRODUCTION

The soil of around 3 million sites in the EEA member countries are suspected of being contaminated, and 250 000 contaminated sites are known to require clean up in this European region [1]. At many of those sites the extent of contamination may not be suf- ficient to trigger remediation under current regulatory conditions, and there may be little economic incentive to regenerate the areas affected. The potential number of contaminated sites in Sweden is 80 000 [2] and also in Sweden known contaminated areas lie unused. Remediation is only considered for sites with high exploitation pressure and for the sites that pose the highest risk to human health or the environment [2]. At the same time, competition for land resources increases. The European target is to replace 10% of the fossil fuel with biofuel by 2020 [3]. This could require around 30 000 km² land for

(2)

biofuel production to meed the Swedish demand [4, 5]. In addition to the significant increasing demand of biofuel there also is an increasing market for other bioproducts such as bio-based plastics and fibres and bio-feedstock. Biofuel, and other non food crop, production on land that is suitable for food-crops may place an increasing stress on agricultural land and food prices [6]. By a first estimate, around 750 km² of the contaminated land in Sweden could be suitable for biofuel or other non food crop production with regard to contaminant levels, location in relation to market and infrastructural demands, topographical features etc. [7, 8].

The use of vegetation for in-situ risk reduction for contaminated soils is called phytoremediation. This can for example be designed to encourage vegetation on contaminated sites, which decreases the potential migration of contaminants through dust or through leaching, since it changes the water balance. The increased microbial flora and carbon content increase the soil quality and may contribute to increased degradation of contaminants [7]. Soil organic matter is a major sorbent for many contaminants and hence increased soil organic matter can stabilise contaminants and decrease the risk of spread- ing (phytostabilisation). On the other hand, increased leaching of soil organic matter may increase leaching of contaminants through complexation reactions [9]. If soil is to be used for biofuel production, the risks that the contamination constitutes must be managed. Contaminants may be enriched in the biofuel crop and thus removed from the soil (phytoextraction), or crop choices and clones can be made that prevent take-up of contaminants [10]. Contaminants in a biofuel crop may cause problems for grazing animals or in later steps of the biofuel production, and the decision on whether crop uptake should be encouraged or not must be made on a case-by-case basis.

When contaminated land is considered for use, remediation of the soil is always an issue to be considered. The net environmental consequences of remediation are not always positive. The cost to the environment and human health in the form of increased greenhouse gas emissions, particle emissions, use of limited resources etc may often out- weigh the gain obtained by soil remediation [11, 12]. Biofuel cultivation has a good chance of a net positive effect: the use of bio-energy in place of conventional fuels (or as an additive) results under many conditions in a net gain in the energy balance and in greenhouse gases [13]. Other environmental aspects than energy and impact on carbon dioxide (such as acidification, human health aspects and ecotoxicity) are more uncertain, less thoroughly researched, and possibly in favour of fossil fuels when compared with biofuel grown on agricultural soil. These impacts are mainly caused by harvesting and processing [13], fertilizer, pesticides, and direct emissions [14]. These impacts also occur when biofuel is grown on contaminated land, but must be set off against the impact from traditional remediation measures.

This study is concerned with a small site in Sweden where Salix viminalis (willow) has been planted on contaminated land. A life cycle assessment (LCA) has been done to compare the impacts of remediation through Salix viminalis cultivation with a tradi- tional excavation-and-landfill remediation. The LCA used the cultivation practices that have been applied to the site during the first years, but other practical results from the site were not yet available. The first objective was to investigate the extend of the environmental benefit of biomass production in comparison with the current Swedish practice of excavation. The second objective was to identify the processes that caused the major environmental impact, since if those processes can be improved, this is most likely to increase environmental efficiency.

(3)

2 METHOD

2.1 Site description

The site, a previous oil depot, was selected because Salix viminalis cultivation had stared on the site, and data on the used cultivation practices were available. The site is small (5000 m²), and therefore biofuel cultivation is not economically viable if the remediation effect is not included in the economic valuation [15]. At the studied site, the harvest was left on site to fertilise and increase the soil organic content. It may be de- cided later to grind the cuttings, but also the chippings will stay on site [16]. The Salix viminalis cultivation is expected to increase organic content and micro-flora, which in turn will increase the microbial degradation of the organic contaminants [17].

The soil was contaminated to a depth of 1 to 1.5 m, with a total contaminated volume of 6500 m³. The contamination was from mineral oil: mainly organic aliphatic compounds, with locally some aromatic compunds and BTEX. Total contamination levels were around 5 g/kg(dw) for 25 % of the soil, and around 1 g/kg(dw) for the remaining 75%

[16].

2.1.1 No action

In the “No action” alternative the site would have been left as it is. Natural degradation is likely to be very slow due to the poor quality of the soil, so it must be expected that the soil after 20 years will be contaminated to a similar level as today.

2.1.2 Biofuel remediation

The Salix viminalis cultivation has been managed by personnel from a nearby garden centre. The soil has been ploughed with a tractor, and planted by hand with shoots transported from Svalöv in the south of Sweden. Fertilising (100 kg NPK-fertiliser), irriga- tion and weeding have been done by hand. The Salix viminalis shoots have been cut twice so far, using a brush saw. The Salix viminalis is expected to stay on the site for 20 years, with cuttings every four years. This necessitated in total 14 journeys of the garden centre personnel to the site, of 10 km each.

The groundwater table is usually 0.5 m below the surface. 4 groundwater observation wells have been installed. Sampling and observation of the wells has been done from Gothenburg, and 17 journeys have been planned in total [16].

2.1.3 Excavation and refilling

Excavation, landfilling of the soil, and refilling with pristine material was considered as the alternative option. Life cycle assessment can be useful also to regard other alternative treatments, but in this study we selected excavation and refilling for two reasons: 1) it is the usual practice for smaller contaminated sites in Sweden, and 2) it was considered as alternative option in the discussions between site owner and the competent au- thority and would have been used if exploitation pressure on the site had been higher [16].

Excavation could have been accomplished with an excavator in ca 40 days, and the excavated soil (6500 m³ or 11700 ton(ww)) likely be transported by truck to the nearest landfill (Djupdalen), which is a sanitary landfill 22 km from the site [18]. Treatment at the landfill by composting was not an option: the organic content was so low that an unreasonable amount of organic matter would have to be added. Refilling of the site would likely have been done with pristine soil [16].

(4)

Excavation of contaminated oil depots commonly includes at least one controller, who measures the contamination levels in the excavated soil, in order to assess when the excavation has reached the clean soil. Controllers are housed locally during the week, but travel home on weekends.

2.2 Life cycle inventory

The life cycle assessment (LCA) was done in order to determine whether the actual remediation through biofuel cultivation, or the possible alternative option of excavation and refill was the environmentally preferable option and to investigate strategies that would reduce impacts within each system. The life cycle was cradle-to-gate: The LCA stops when the remediation has led to a clean soil, except for the no action alternative.

This results in a 20 year land occupation for the biofuel remediation and for no action, and a 40 day land occupation for the excavation-and-refill remediation. The assumption is that after a remediation the site ceases to pose an environmental risk. The quality of the soil at the end points will differ: after the biofuel remediation an agricultural soil possible for any use, without any action the soil remains contaminated, and after the excavation-and-refill remediation an organic-poor sand that would only be acceptable for buildings (residential or other wise). The soil would need further improvement if used for anything else after the excavation-and-refill remediation. Similarly, removal of Salix viminalis roots was not included for the biofuel remediation alternative.

The assessment was done using the SimaPro software [19]. One site (5000 m²) was used as functional unit.

The ecoinvent database [20] was preferred for inventory information since the data is often European and acceptably updated. This insured that the inventory was consistent and the processes comparable with each other. Some processes were created for this study by the authors. Karlstad Salix viminalis cultivation, excavated soil and refilled soil were major process and are shown in Table 1 and Table 2. The other processes from outside the ecoinvent database were: tractor on road, brush saw, and groundwater monitoring well. Data for these processes is available as supplementary content. In short, the process tractor on road used the diesel consumption by a Swedish tractor, 0,35 l/km [21, 22], as input to the process “diesel used in tractor” from [23]. The brush saw was created using the ecoinvent power-saw module and replacing CO, NOx, HC and CO₂ emissions with brush saw values from [24]. The groundwater monitoring well combined HDPE pipes [20] with drilling of a hole. Particulars of diesel use for groundwater well drilling were taken from [25] and used for the module “diesel burned in building machine” [20].

2.2.1 No action

The “No action” alternative was LCA-modelled as occupation of the site as an industrial area, Corine land class 121, [26]. The Corine industrial land class 121 includes aban- doned industrial sites where buildings are still present [26]. An alternative interpretation of the land classes is that since the site is small, it should be included in the surrounding site land class, and be classed as urban. Urban, industrial and dump site land classes all have the same characterisation factors [27]. The occupation was assumed to last 20 years in order to be directly comparable to the biofuel remediation.

2.2.2 Biofuel remediation

The LCA model of the biofuel remediation is shown in Table 1. The site was transformed from industrial to arable, since free agriculture should be possible on the site

(5)

after the remediation. The tillage process included 0.5 km transport from farm to site, but since the present site was 10 km from the supplier of machines, the “tractor on the road” process was added to account for the extra distance.

European average planting stock inventory was taken from [28]. 14000 pieces are commonly used per ha in Sweden [29]. The planting stocks were transported by van from Svalöv to the case site, a distance of 482.7 km [30]. The fertiliser amounts in Table 1 add up to 100 kg NPK 20-3-5. The operation of the lorry is to move the drilling rig for the groundwater monitoring wells to the site. The upper structure process “transport, passenger car” shows the 17 journeys of the controller from Gothenburg. The pas- senger car within the Salix viminalis cultivation process is to move the personnel from the garden centre to the site and back.

Table 1: SimaPro model of biofuel remediation. Upper structure processes are marked in bold

Process Amount Unit

Salix cultivation, site K/PaS* 5000 m²

Transformation, from industrial area 5000 m²

Transformation, to arable land 5000 m²

Occupation, forest, intensive, short-cycle 5000*20 m²a

Tractor on road/ PaS* dist_equip_site^#*2 km

Tillage, ploughing/CH S 0,5 ha

Operation, van < 3,5t/RER S 482,7*2 km

Planting stocks, short-rotation wood, at field/p/RER U 14000*0,5 p Ammonium nitrate, as N, at regional storehouse/RER S 20-5,38 kg Potassium sulphate, as K2O, at regional storehouse/RER S 12 kg Diammonium phosphate, as P2O5, at regional storehouse/RER S 13,75 kg Diammonium phosphate, as N, at regional storehouse/RER S 13,75/46*18 kg Transport, passenger car/RER S dist_equip_site^#*14 personkm

Brush sawing /PaS* 5*6 hr

GW monitoring well/PaS* 8 m

Operation, lorry >32t, EURO3/RER S 2*dist_equip_site^# km

Transport, passenger car/RER S 17*2*255 personkm

* Process created for this case study

# Dist_equip_site is distance of equipment to site = 10 km

2.2.3 Excavation and refill

The LCA model of the excavation-and-refill remediation is shown in Table 2. Transport in the passenger car under “excavated soil, site K” and refilled soil, site K” concerns the daily movement of excavator, truck, and loading shovel operators to the site. The same process in the top structure (in bold in Table 2) concerns the controller, who is assumed to travel home to Gothenburg for the weekends, 255 km, but to stay locally during the weeks, 10 km. Lorry transport concerns the transport of soil, while the lorry operation is to move the excavator and the loading shovel to the site.

The refilling transformed the site into urban land use, since the soil will not be suitable for other purposes without further soil improvement.

(6)

Table 2: SimaPro model of excavation and refill. Upper structure processes are marked in bold

Excavated soil, site K/PaS* 6500 m³

Occupation, construction site 5000*20/365 m²a

Transformation, from industrial area 5000 m²

Excavation, hydraulic digger/RER S 6500 m³

Excavation, skid-steer loader/RER S 6500 m³

Transport, passenger car/RER S 20*3*10 Personkm

Transport, lorry >32t, EURO3/RER S

6500*soil_density^&

*22*2 Tkm Disposal, inert material, 0% water, to sanitary landfill/CH S 6500*soil_density^& Ton

Refilled soil, site K/PaS* 6500 m³

Occupation, construction site 5000*20/365 m²a

Transformation, to urban, discontinuously built 5000 m²

Sand, at mine/CH S 6500*soil_density^& Ton

Transport, lorry >32t, EURO3/RER S

6500*soil_density^&

*dist_site_quarr⁺ Tkm

Excavation, skid-steer loader/RER S 6500 m³

Transport, passenger car/RER S 20*2*10 Personkm

Operation, lorry >32t, EURO3/RER S 2*dist_equip_site^# *4 Km

Transport, passenger car/RER S 39*2*10 Personkm

Transport, passenger car/RER S 3*2*255 Personkm

* Process created for this case study

# Dist_equip_site is distance of equipment to site = 10 km

& Soil_density 1.8 t/m³

+ Dist_site_quarr is distance from site to quarry = 30 km

2.2.4 Omissions

Emissions from the excavation itself (not the excavating machine) such as dust particles and emissions to air as the contaminants become more available have been excluded, as have emissions and leaching of contaminants from the contaminated soil under present conditions an and any changes in leaching and emissions to air due to the Salix vimi- nalis cultivation.

Risk assessment and laboratory testing were not included. The excavation alternative would have required analysis of soil samples, while the biofuel remediation is conducted with groundwater analyses.

The gross caloric value of the biomass and the uptake, retention and possible reemission of CO₂ by the biomass were not included, because the yield of the Salix viminalis culti- vation was unknown. The net effect of this omission is to overestimate, possibly largely, the global warming impact of the biofuel remediation [31].

2.3 Impact assessment methods

Two impact assessment methods were used in the evaluation of the environmental impact: ReCiPe 2008 [27] and the environmental product declaration (EPD, [32]). ReCiPe was selected because land use is included and because of the high acceptance of the models it builds on, i.e. the Eco-indicator99 and CML2001 [33] . The default ReCiPe endpoint method, hierarchist version was used. Normalisation values for Europe and the average weighting set was used to arrive at single scores. ReCiPe used three main damage categories: human health, ecosystem and resources. Human health included climate

(7)

change-human health, ozone depletion, human toxicity, photochemical oxidant formation, particulate matter formation, and ionising radiation (expressed in disability ad- justed life years, DALY). Ecosystems included climate change-ecosystems, terrestrial acidification, freshwater and marine eutrophication, terrestrial, freshwater and marine ecotoxicity, agricultural and urban land occupation, and natural land transformation (expressed in species×yr). Resources included metal depletion and fossil depletion, expressed in $ [27].

EPD is supported by the Swedish government and was selected due to its special sig- nificance to Sweden. Characterisation factors from version 1.0, 2008 were used [32], except for Gross Calorific Value (GCV). The SimaPro adaption of the draft version EPD of June 2007 was used for the GCV because the substances in the IEC report [32]

were closer to application and further from raw materials, and therefore The EPD method does not aggregate categories, so no weighting or normalisation was included.

The EPD damage categories are global warming (GWP, in kg CO2 eq), ozone layer depletion (in kg CFC-11 eq), photochemical oxidation (in kg C₂H₄), acidification (in kg SO2 eq), eutrophication (in kg PO4 eq) and gross caloric values (in MJ eq) [32].

2.4 Sensitivity

Sensitivity analyses were performed by four alternations of the base case scenario conditions:

• Excavated soil to an inert landfill instead of a sanitary landfill

• Commercial fertiliser amounts instead of the low amount of the case study

• Field emissions added to the Salix viminalis cultivation

• Land use transformation to short-cycle forest instead of arable

The landfill dominated the environmental impact for the excavation remediation (see results). In the base case the soil was deposited on a sanitary landfill, the ecoinvent process closest to the most likely destination for the excavated soil from the case site.

The analysis was repeated with landfilling on an inert landfill instead of a sanitary landfill (also from the ecoinvent database). However, if the material had shown sufficiently low contamination for an inert landfill, it would likely not have needed remediation. No settling and decomposition is expected for the soil, and therefore data from landfilling of an inert material was used, both for the sanitary as well as for the inert landfill.

Fertiliser is often important for the environmental impact [34-38], but was not so in this case, likely because the fertiliser amount was so low (see results). Fertiliser amounts of 100 kg N-fertiliser 15 times a year are usual in commercial Salix viminalis cultivation in Sweden [15]. An impact assessment was made with 1500 kg-N ammonium nitrate instead of 15 kg.

Field emissions, such as nitrate from fertilisation and terpenes from the plants themselves, are often omitted from LCA because of the difficulty of defining data, and are usually uncertain. We chose to do the principal comparison without direct field emissions. To determine the magnitude of the consequences of this choice, field emissions of Salix viminalis cultivation from [28] were added. Pesticide emissions were not included, since no pesticides were used in the case study. The field emissions were adapted to the case by assuming that the emissions were constant per ha. The resulting emissions for the case site are shown in Table 3. This included also some uptake of heavy metals from the soil. [28] is concerned with commercial Salix viminalis cultivation, with higher fer-

(8)

tiliser amounts and more mechanical processing of the land than that of the case study here. Therefore the use of field emissions from [28] may be regarded as considerably higher than those occurring on the case site.

Table 3: Field emissions for site K, 5000 m². Adapted from [28]

Emissions to air

Ammonia 26,2 kg

Dinitrogen monoxide 10,9 kg

Isoprene 386 kg

Terpenes 19,3 kg

Nitrogen oxides 2,29 kg Emissions to water

Nitrate 177 kg

Phosphate 1,26 kg

Phosphate 7,14 kg

Phosphorus 0,129 kg

Emissions to soil

Copper -0,189 kg

Lead 0,0137 kg

Mercury -0,00117 kg

Nickel 0,00866 kg

Zinc -5,22 kg

The biofuel remediation is expected to transform the land from industrial to free use.

The increased quality of the soil could even make agriculture possible. Continuation of short rotation wood on the site is not likely, since the site is too small to allow for finan- cial profit from the Salix viminalis cultivation. Since the land use caused the major im- pact (see results), the effect of transformation to short rotation wood instead of to arable was tested in the LCA model of the biofuel remediation.

3 RESULTS

3.1 No action

Since the “no action” alternative only consisted of occupation of an industrial area, there was no impact according to EPD 2008. The impact according to ReCiPe 2008 was 4.42·10³ points, all of urban occupation, or 0.002 species.yr (potentially disappearing species × years). Contaminated sites may, however, develop high biodiversity if no action is taken [39], but in the present case the soil is of poor quality in addition to the contamination. The ReCiPe therefore may be relevant as site recovery is expected to take very long time if no actions are taken.

3.2 Biofuel remediation

The single score (ReCiPe) of the biofuel remediation showed that the Salix viminalis cultivation had the dominant environmental impact in this alternative (Fig 1). This was mainly due to occupation of the land used for the Salix viminalis cultivation (see sup- porting content). This had a damaging impact of 3.5 × 10³ points, or 0.0015 species.yr.

The planting stocks contributed a benefit to biodiversity, mostly through the transformation of land to use for short-cycle forest. This benefit was 0.71 × 10³ points, or 0.0003

(9)

species.yr, which was detracted to arrive at the total score. Other processes contributed in a minor way, and the total impact according to ReCiPe 2008 was in total 3.0·10³ points, somewhat lower than for the no action alternative. Biodiversity impact was 0.0012 species·yr, also of the same magnitude but slightly less than the loss of biodiversity and the industrial or urban use in the “no action” alternative. Occupation of the site by short cycle wood resulted in less loss of biodiversity than the industrial or urban use in the “no action” alternative, so the impact of the biofuel remediation was slightly less.

For remaining (other than land use related) impact categories, the cause of the impact was the Salix viminalis cultivation and the journeys by car of the controller, both of similar magnitude. The groundwater observation wells were unimportant (Fig 1 to Fig 6). The extent of the controller’s journeys for the biofuel remediation is highly case- specific: due to lack of familiarity with biofuel growth on contaminated soil a distant expert was preferred to a local controller.

ammonium nitrate, as N,

at regional tillage,

ploughing/ha /CH

transport, passenger car/personk

operation, van <

3,5t/km/RER

planting stocks, short-rotatio

brush sawing Biofuel

remediation

GW monitoring

well /PaS

Tractor on road

Salix cultivation,

site K/PaS

Fig 1: Single score of environmental impact of biofuel remediation, ReCiPe. Processes contributing more than 0.1% are shown (cut-off 0.1%). Contribution of planting stocks to Salix viminalis cultivation was positive to the environment.

at regional

diammonium phosphate, as P2O5, at potassium

sulphate, as K2O, at tillage,

ploughing/ha/

CH

transport, passenger car/personkm

operation, van

<

3,5t/km/RER

planting stocks, short-rotation

remediation

GW monitoring

well /PaS

Tractor on road

Salix cultivation,

site K/PaS

Fig 2: Impact of biofuel remediation on human health, ReCiPe. Impact on resources was similar. Cut-off 1%

(10)

at regional

diammonium phosphate, as

P2O5, at tillage,

ploughing/ha/C H

transport, passenger car/personkm/

operation, van

< 3,5t/km/RER

operation, lorry

>32t, EURO3/km/RER

wood, at

remediation

GW monitoring well /PaS

Tractor on road

Salix cultivation, site

K/PaS

Fig 3: Impact of biofuel remediation on ecosystems, ReCiPe. Cut-off 0.015%

ammonium nitrate, as

N, at

diammonium phosphate, as P2O5, at tillage,

ploughing/ha /CH

operation, van <

3,5t/km/RER

operation, lorry >32t, EURO3/km/R

planting stocks, short-rotatio

remediation

GW monitoring

well /PaS

Tractor on road

Salix cultivation,

site K/PaS

Fig 4: Impact of biofuel remediation on global warming, EPD. Ozone layer depletion, acidification, and gross caloric value were similar. Cut-off 0.85%

ploughing/ha /CH

operation, van <

3,5t/km/RER

remediation

GW monitoring

well /PaS

Tractor on road

Salix cultivation,

site K/PaS

(11)

Fig 5: Impact of biofuel remediation on photochemical oxidation, EPD. Cut-off 0.3%

ammonium nitrate, as N, at

regional

ploughing/ha/CH

transport, passenger car/personkm/

operation, van

< 3,5t/km/RER

operation, lorry

>32t, EURO3/km/RER

wood, at

remediation

GW monitoring well /PaS

Tractor on road

Salix cultivation, site

K/PaS

Fig 6: Impact of biofuel remediation on eutrophication, EPD. Cut-off 0.3%

Within the actual cultivation process, the operation of the van for transporting the planting stocks, the cutting by brush saw and the planting stocks were important contribu- tors to different environmental problems, and the phosphate fertiliser had an impact on eutrophication (Fig 1- Fig 6). Ecosystem damage in the ReCiPe assessment method was mainly concerned with land use for the site itself and for the production of planting stocks (Fig 2). The planting stock had a negative land transformation damage (i.e. positive to biodiversity) because the transformation to short cycle wood is expected to increase the number of species compared to the reference, woodland [33].

Salix viminalis cultivation generally has its highest impacts in the fertiliser, harvesting, and the land use itself, if land use is considered [34-38, 40]. The results from the present data were in agreement with this except for the influence of the fertiliser. The fertiliser dose in the case study was low, since the harvest will remain on site instead of being removed to an application. Thus nutrients are kept on the site. This is further discussed in the sensitivity analysis.

The need for controller journeys is specific to contaminated land. They may be signifi- cantly reduced when more experience with Salix viminalis cultivation on contaminated land is available. The specialist in the current case travelled 255 km per single journey in order to check up on the site and take groundwater samples. This may be performed by local personnel when Salix viminalis cultivation on contaminated land is a more fa- miliar process.

3.3 Excavation and refill

For the remediation process of excavation-and-refill, the ReCiPe single score was dominated by landfilling of the soil (disposal inert material on sanitary landfill, Fig 7), which had a large impact on human health (human toxicity, climate change - human health and particulate matter in Fig 9). The landfill constituted more than half of the impact according to EPD, but soil transport constituted a considerable impact as well.

Fig 8 shows global warming; the other categories were very similar.

(12)

Excavation as remediation method is often dominated by transport of the soil [11, 12, 41, 42], while the landfill has not previously been identified as dominant [11, 12]. This may be due to previous lack of data for landfills, which is now gradually improving.

Controllers’ journeys were not important for the excavation-and-refill remediation, in contrast to the biofuel remediation (Fig 7 and Fig 8). The total length of the daily local journeys of the excavation controller were about a quarter of the total length for the biofuel controller.

sand, at mine/kg/CH excavation,

hydraulic digger/m3/R

excavation, skid-steer loader/m3/RE

disposal, inert material, 0%

water, to

transport, lorry >32t, EURO3/tkm/

Excavation

Excavation and refill remediation

Refilling

Fig 7: Single score impact of excavation-and-refill, ReCiPe. Cut-off 0%

excavation, skid-steer loader/m3/RE

disposal, inert material, 0%

water, to

Excavation

Excavation and refill remediation

Refilling

Fig 8: Impact of excavation-and-refill remediation on global warming, EPD. All other categories similar. Cut-off 0%

(13)

Climate change Human Health Climate change Ecosystems Ozone depletion terrestrial acidification freshwater eutrophication marine eutrophication

human toxicity photochemical oxidant formation

particulate matter formation terrestrial ecotoxicity freshwater ecotoxicity marine ecotoxicity

ionising radiation agricultural land occupation urban land occupation natural land transformation

water depletion metal depletion

fossil depletion Excavated

soil, site

Refilled soil , site K/PaS

Operation , lorry >32

Transport , passenge

kPt

55 50 45 40 35 30 25 20 15 10 5 0

Fig 9: Contribution of impact categories to the single score for excavation-and-refill remediation, ReCiPe.

3.4 Comparison of biofuel with excavation-and-refill remediation Biofuel remediation caused lower damage to the environment then the traditional excavation-and-refill remediation according to both evaluation methods (Fig 10, Fig 11, Fig 12). Only the agricultural land use impact in ReCiPe was higher for the biofuel remediation. The single score impact indicated that the higher land use was well compensated by the lower impact in the other impact categories (Fig 12).

(14)

Comparing 1 p 'Biofuel remediation' with 1 p 'Excavation and refill remediation'; Method: EPD 2008 V1.00 / characterization Biofuel remediation Excavation and refill remediation

Global war ming (GW

Ozone lay er depleti

Photoche mical oxid

Acidificati on

Eutrophic ation

Gross Cal orific Valu

%

120 110 100 90 80 70 60 50 40 30 20 10 0

Fig 10: Comparison of impacts from biofuel remediation (green) with excavation-and- refill (red), EPD.

Comparing 1 p 'Biofuel remediation' with 1 p 'Excavation and refill remediation'; Method: Recipe Endpoint (H) V1.01 / Europe Recipe H/A / normalization Biofuel remediation Excavation and refill remediation

Human Health Ecosystems Resources

100 90 80 70 60 50 40 30 20 10 0

Fig 11: Comparison of impacts from biofuel remediation (green) with excavation-and- refill (red), ReCiPe, normalised damage.

(15)

fossil depletion

Biofuel remediation Excavation and

refill remediation

kPt

65 60 55 50 45 40 35 30 25 20 15 10 5 0

Fig 12: Comparison of impacts from remediation by excavation and Salix viminalis cultivation, single score ReCiPe 2008

3.5 Sensitivity

When the sanitary landfill was replaced with an inert landfill, environmental impact was decreased, especially with regard to human health impacts (Fig 13). Contribution of the landfill to the total environmental impact was now on a level with transport of the soil and the effects of the sand mining (Fig 14). The excavation-and-refill remediation remained more environmentally costly than the biofuel remediation (Fig 13).

The accepted doctrine is that transport of soil should be minimised, both in order to minimise cost as well as environmental impact. This has been found for the present case in a previous study with older data as well [31]. The present results suggest that an environmental gain may be achieved using the most inert possible landfill even if the transport distance is increased. A sanitary landfill requires more resources for drainage, geo- textiles, treatment of leaching water etc; the increased control over the waste causes more environmental impact. A decrease in such landfill-related impacts may make some increase of transport-related impacts acceptable.

(16)

human toxicity photochemical oxidant formation particulate matter formation terrestrial ecotoxicity

freshwater ecotoxicity marine ecotoxicity ionising radiation agricultural land occupation urban land occupation natural land transformation

fossil depletion Biofuel bas

e case

Biofuel ferti liser increa

Biofuel field emissions

Biofuel land transform

E & r landfil l inert

Excavation and refill

No action

kPt

60 50 40 30 20 10 0 -10 -20

Fig 13: Sensitivity analysis: environmental impact of LCA models for biofuel, excava- tion and no action. Single score ReCiPe results.

excavation, skid-steer loader/m3/R

disposal, inert waste,

5% water,

Refilling E & r landfill

inert

Excavated soil, site K to

inert landfill

Fig 14: Environmental impact for excavation-and-refill remediation with inert landfill, ReCiPe 2008 single score. Cut-off 0%. Comparable with Fig 7.

When the actual fertiliser amount was replaced with the amount necessary for commer- cial Salix viminalis cultivation where the harvest is removed from the site, the environ- mental impact was increased, but the main conclusion remained unaffected (Fig 13).

The increased fertiliser now had a considerably environmental impact, especially through the increased effect on climate change and depletion of fossil resources (Fig 15).

The inclusion of field emissions in the Salix viminalis cultivation resulted in a small increase of climate-change related impact (Fig 15).

(17)

The transformation of the site to forest instead of the arable lead to a total negative impact in the ReCiPe single score, i.e. the gains outweighed the costs to the environment.

Arable land has a low biodiversity since it is based on measurements of monocultures, while forests, even short rotation forests, have higher biodiversity. Biodiversity is the main consideration in ReCiPe when considering the occupation and transformation of land [33]. But arable land as a limited resource is closer to the issue when the increased needs of land for biofuel production is discussed and the competition of biofuel with food agriculture. The view on land use has a major effect in the case of contaminated sites. Risk assessment has prevented humans from exploiting sites with low or interme- diate contamination, leaving the way free for redevelopment of nature. Remediation of the soil may therefore decrease biodiversity and at the same time increase the limited land resources in the area.

fossil depletion Biofuel bas

e case

Biofuel ferti liser increa

Biofuel field emissions

Biofuel land transform

No action

kPt

4 2 0 -2 -4 -6 -8 -10 -12 -14 -16 -18 -20

Fig 15: Sensitivity analysis: environmental impact of LCA models for biofuel and no action. Single score ReCiPe results

4 DISCUSSION

The excavation-and-refill remediation scored high in all traditional impact evaluation categories, and the contribution from different sub-processes was very similar. The impacts of transport or landfilling were much the same on global warming, fossil resources, acidification, ozone formation, etc. Biofuel remediation impacts varied somewhat over the categories, but global warming potential, ozone depletion potential, acidification and gross caloric value were very similar. That left only photochemical oxidant formation and eutrophication to differ from the others. The many categories contributed little to our understanding of the environmental impact and the possibilities for improvement in this case study.

(18)

These emerged clearer from the summarised damage control of ReCiPe, where the impact categories have been summarised to human health, ecosystem quality, and resources (Fig 11). The impacts of the two remediation alternatives were on very different environmental problems. The excavation-and-refill remediation showed a primary impact in the traditional categories of global warming etc. The biofuel remediation showed large importance of land occupation and biodiversity. Thus, in this comparison environmental effects do occur on very different environmental problems. Inclusion of land use issues is an active research area [43-45], and our results demonstrate again the importance of further development in that area.

The similarities of the emerging impact patterns using well established categories (Fig 8) may be partly due to a bias in the available inventory data. These focus on the emissions caused by use of energy and by production of capital goods [20]. Due to data dif- ficulties we omitted ‘use-phase’ emissions for both the biofuel remediation and the excavation-and-refill. For the latter e.g. dust and contaminant emissions and noise were excluded. These data depend more on conditions at the site and are less well known than fossil fuel emissions. The issue of field emissions for the biofuel remediation is ad- dressed in the sensitivity analysis. Dust emissions from contaminated site excavation have been found not to lead to health risks in children [46]. Occupational health risks were omitted for all processes in the case study, including excavation.

The biofuel remediation has an impact on global warming through the uptake of carbon dioxide in the Salix viminalis. This carbon either contributes to the soil carbon pool, or is reemitted through degradation. In either case carbon dioxide is temporarily stored and reduces the global warming impact. However, the flows of carbon cannot at present be accurately estimated [47, 48]. More research is required into the carbon balance inven- tory of Salix viminalis cultivation if the effect on global warming is to be assessed cor- rectly.

In the present case study the harvested Salix viminalis remained on site to improve soil conditions and accelerate degradation of the remaining contamination. Other use of the harvest was prevented by a number of barriers [7], which may decrease in the future.

Utilisation of the harvest would have caused a further positive environmental effect, in that the harvest from the contaminated site would partly replace conventionally grown Salix viminalis. Cultivation on contaminated sites likely will improve the soil regardless of removal of the harvest in the betterment of the micro fauna and organic content due to fertilising and the crops themselves.

With biodegradable contaminants there is no conflict between usefulness and treatment of the soil. Metal contaminated soil necessitates a conscious balance between usefulness now and cleaning effect. Metals cannot be destroyed but the destination of metals may be controlled through choice of Salix viminalis clones, which can be accumulating or non-accumulating, or be directed to selected rest fractions such as fly ash [7].

There is a third option for the treatment of this contaminated site, No action. The site could be left as it is. The fence would need some maintenance, but the environmental impact would be very low. However, natural degradation is likely to be very slow due to the poor quality of the soil, so it must be expected that the soil after 20 years will be contaminated to a similar level as today. And during this time the site would be occu- pied without further use to society.

(19)

5 CONCLUSION

The biofuel remediation affected the environment mainly through the controller’s jour- neys, transport of planting stocks, land use for Salix viminalis cultivation, and harvest- ing. Fertilisers had a minor impact. The controller journeys may be reduced when familiarity with biofuel as a soil treatment method increases. When land use was considered, ecosystem quality dominated the impact categories. Depending on the land use function and definition the impact could be either negative or positive.

The excavation-and-refill remediation affected the environment mainly through the landfill and the transport of soil and backfill. The selection of the type of landfill was important to the outcome. Excavation of soil should be avoided as far as possible to minimize damage to the environment, since it leads to both transport and landfilling.

The environmental impacts of the biofuel remediation were negligible in comparison with the excavation-and-refill impacts, except for land use. The higher land use was well compensated by the other impacts of the excavation-and-refill.

Transports were an important cause of impact in the assessment. Transport of contaminated soil and backfill for the excavation remediation, transport of the controller and the planting stocks for the biofuel remediation. There may be an effect of an inventory bias regarding transport: the importance of transportation has been long known, emissions are included in the databases and the impact assessment methods.

The level of knowledge and the availability of data affects the results. Inventory data for e.g. landfills need enlargement. Further development of impact assessment methods is necessary, especially with regard to biodiversity and land surface as a limited resource.

In summary, the impacts of the two remediation alternatives excavation-and-refill ver- sus biofuel remediation were on very different environmental problems. The excavation-and-refill remediation showed a primary impact in the traditional categories of global warming etc. The biofuel remediation showed large importance of land occupation and biodiversity. Thus, in this comparison environmental effects do occur on very different environmental problems and geographical scales. Inclusion of land use issues is an active research area and our results demonstrate again the importance of further development in that area.

6 ACKNOWLEDGEMENTS

The authors are grateful to Sonja Blom for background information regarding the case study, and for the Rejuvenate project group for their discussions on biofuel as presented in [10]. The study was initially financed through the project Rejuvenate, under the um- brella of an ERA-Net Sustainable management of soil and groundwater under the pressure of soil pollution and soil contamination (SNOWMAN), by the Department for En- vironment Food and Rural Affairs and the Environment Agency (England), FORMAS (Sweden), SGI (Sweden) and Bioclear BV (Netherlands), and throughout the work process complementary and additional funding has been received from the Swedish Geotechnical Institute (SGI).

(20)

7 REFERENCES

[1] EEA. Progress in management of contaminated sites (CSI 015) - Assessment published Aug 2007. EEA, European Environment Agency; 2007.

http://themes.eea.europa.eu/IMS/IMS/ISpecs/ISpecification20041007131746/IAssessm ent1152619898983/view_content

[2] Swedish EPA. Läget i landet - efterbehandling av förorenade områden (in Swed- ish). In: Swedish Environmental Protection Agency, editor. Stockholm: Naturvårds- verket; 2009. http://www.naturvardsverket.se/sv/Verksamheter-med-

miljopaverkan/Efterbehandling-av-fororenade-omraden/Laget-i-landet--efterbehandling/

[3] Commission of the European communities. Impact assessment. Document accom- panying the package of implementation measures for the EU's objectives on climate change and renewable energy for 2020. Proposals for Directive of the European parlia- ment and of the council amending Directive 2003/87/EC so as to improve and extend the EU greenhouse gas emission allowance trading system. Commission staff; 2008.

http://ec.europa.eu/energy/climate_actions/doc/2008_res_ia_en.pdf

[4] Rydberg T. Dagens analysmeteoder skapar övertro på bioenergy (Todays methods for assessing values overestimates bioenergy). Formas fokuserar 2007:137.

[5] Semelsberger TA, Borup RL, Greene HL. Dimethyl ether (DME) as an alternative fuel. Journal of Power Sources 2006;156:497.

[6] Bardos P, Andersson-Sköld Y, Keuning S, Pachon, Track T, Wagelmans, et al.

Special session 1: Brownfields, Bioenergy and Biofeedstocks & Special session 2:

Green Remediation. ConSoil 2008. Milano, Italy: F&U confirm; 2008, p. Sps 2.

[7] Andersson-Sköld Y, Enell A, Blom S, Rihm T, Angelbratt A, Haglund K, et al.

Biofuel and other biomass based products from contaminated sites - Potentials and barriers from Swedish perspectives. Gothenburgh: Swedish Geotechnical Institute (SGI);

2009. www.swedgeo.se

[8] Enell A, Wik O, Andersson-Sköld Y. Arable area of potential contaminated sites in Sweden. Swedish Geotechnical Institute; 2009. www.swedgeo.se

[9] Sparks DL. Elucidating the fundamental chemistry of soils: past and recent achievements and future frontiers. Geoderma 2001;100:303.

[10] Bardos P, Andersson-Sköld Y, Keuning S, Polland M, Suer P, Track T. Rejuvenate - Final Research Report. 2009. http://www.snowman-

era.net/downloads/REJUVENATE_final_report.pdf

[11] Suer P, Andersson-Sköld Y, Andersson J. Local gain, global loss - the environmental cost of action. In: Singh A, Kuhad RC, Ward OP, editors. Advances in applied bioremediation, Soil Biology 17. Berlin Heidelberg: Springer-Verlag; 2009.

[12] Suer P, Nilsson-Påledal S, Norrman J. LCA for site remediation: A literature review. Soil and Sediment Contamination: an International Journal 2004;13:415.

[13] von Blottnitz H, Curran MA. A review of assessments conducted on bio-ethanol as a transportation fuel from a net energy, greenhouse gas, and environmental life cycle perspective. Journal of Cleaner Production 2007;15:607.

[14] Zah R, Böni H, Gauch M, Hischier R, Lehmann M, Wäger P. Life Cycle Assess- ment of Energy Products: Environmental Impact Assessment of Biofuels Executive Summary of Ökobalanz von Energiprodukten: Ökologishe Bewertung von Biot- riebstoffe. Bern: Empa; 2007. http://www.bfe.admin.ch/themen/00490/00496/ index.html?%20lang=en&dossier_id=01273

[15] Lantmännen, Gustafsson J, Larsson S, Nordh N-E. Manual för salixodlare.

Örebro: Lantmännen Agroenergi AB/Salix; 2007.

[16] Blom S. Personal communication

(21)

[17] Spriggs T, Banks MK, Schwab P. Phytoremediation of polycyclic aromatic hydro- carbons in manufactured gas plant-impacted soil. J Environ Qual 2005;34:1755.

[18] Larsson P. Karlstads Energi AB. 2008-10-07. Personal communication [19] PRé Consultants bv. SimaPro 7. 7.1.8 ed: PRé Consultants bv,; 2008.

[20] Ecoinvent Centre. Ecoinvent data v2.0. Dübendorf: Swiss Centre for Life Cycle Inventories; 2007. www.ecoinvent.org

[21] Lundborg M. Effects of speed and road conditions on fuel consumption of agricultural tractor (in Swedish). Institutionen för lantbruksteknik. Uppsala: Swedish Agricul- tural Univeristy; 2003. http://www.bt.slu.se/lt_old/Meddelande/Me2003-04/Meddel.pdf [22] Lindgren M, Pettersson O, Hansson P-A, Norén O. Engine load pattern and engine exhoust gas emissions from off-road vehicles and methods to reduce fuel-consumption and engine exhaust gas emissions (in Swedish). JTI Institutet för jordbruks- och mil- jöteknik; 2002.

[23] Jungbluth N, Frischknecht R, Tuchschmid M, Faist Emmenegger M, Steiner R, Schmutz S. Life Cycle Assessment of BTL-fuel production: Final report. RENEW: Re- newable fuels for advanced powertrains. Uster, Switzerland; 2008.

http://www.esu-services.ch/cms/index.php?id=50

[24] SMP. Jämförande emissionsmätningar av tvåtaktsbränslen (in Swedsih). Svensk Maskingprovning AB; 2005.

[25] Landesanstalt für Umweltschutz Baden-Württemberg. Umweltbilanzierung von Altlastensanierungsverfahren. Landesanstalt für Umweltschutz Baden-Württemberg;

1999.

[26] Bossard M, Feranec J, Otahel J. CORINE land cover techical guide - Addendum 2000. Copenhagen: EEA; 2000. http://www.eea.europa.eu/publications/tech40add [27] ReCiPe. ReCiPe 2008 method, version June 30th 2009. 2008.

[28] Jungbluth N, Chudacoff M, Dauriat A, Dinkel F, Doka G, Faist Emmenegger M, et al. Life Cycle Inventories of Bioenergy.: ESU-services, Uster, CH 2007.

http://www.esu-services.ch/cms/index.php?id=53

[29] Bioenergiportalen. Bioenergiportalen. 2008. www.bioenergiportalen.se [30] hitta.se. 2009. www.hitta.se

[31] Suer P, Andersson-Sköld Y, Blom S, Bardos PR, Track T, Polland M. Environ- mental impact assessment of biofuel production on contaminated land - Swedish case studies. Swedish Geotechnical Institute; 2009. www.swedgeo.se

[32] IEC. EPD - supporting annexes for environmental product declarations, EPD, version 1.0. The international EPD cooperation; 2008.

http://www.environdec.com/pageId.asp?id=163

[33] Goedkoop M, Heijungs R, Huijbregts M, De Schryver A, Struijs J, van Zelm R.

ReCiPe 2008 - A life cycle impact assessment method which comprises harmonised category indicators at the midpoint and the endpoint level. Report I: Characterisation.

Ruimte en Milieu - Ministerie van Volkshuisvesting, Ruimtelijke Ordeningn en Milieu- beheer; 2009. http://www.lcia-recipe.net/

[34] Goglio P, Owende PMO. A screening LCA of short rotation coppice willow (Salix sp.) feedstock production system for small-scale electricity generation. Biosystems En- gineering 2009;103:389.

[35] Heller MC, Keoleian GA, Volk TA. Life cycle assessment of a willow bioenergy cropping system. Biomass & Bioenergy 2003;25:147.

[36] Zah R, Böni H, Gauch M, Hischier R, Lehmann M, Wäger P. Life Cycle Assess- ment of Energy Products: Environmental Assessment of Biofuels. Switzerland: Swiss Federal Institute of Materials Science and Technology Research (EMPA); 2007.

www.bioenergywiki.net/images/8/80/Empa_Bioenergie_ExecSumm_engl.pdf

(22)

[37] Jungbluth N, Frischknecht R, Faist Emmenegger M, Steiner R, Tuchschmid M.

Life Cycle Assessment of BTL-fuel production: Life Cycle Impact Assessment and interpretation. RENEW: Renewable fuels for advanced powertrains. Uster, Switzerland;

2007. http://www.esu-services.ch/cms/index.php?id=50

[38] Börjesson P. Livscykelanalys av salixproduktion. IMES/EESS. Lund: Dept of En- vironmental and Energy System Studies, Lund University; 2006.

[39] Ravilious K. Despite mutations, Chernobyl wildlife is thriving. National Geo- graphic News 2006;April 26.

[40] Keoleian GA, Volk TA. Renewable energy from willow biomass crops: Life cycle energy, environmental and economic performance. Critical Reviews in Plant Sciences 2005;24:385.

[41] Ribbenhed M, Wolf-Watz C, Almemark M, Palm A, Sternbeck J. Livscykelanalys av marksaneringstekniker för förorenad jord och sediment. Stockholm: IVL Svenska Miljöinstitutet AB; 2002, p. 108.

[42] ScanRail Consult, HOH Water Technology, NIRAS, Revisorsamvirket/PKF. Envi- ronmental/economic evaluation and optimising of contaminated sites remediation - Evaluation of demonstration projects. Copenhagen: DSB, Banestyrelsen and Miljøsty- relsen, Denmark; 2000, p. 99.

[43] Canals LMI, Bauer C, Depestele J, Dubreuil A, Knuchel RF, Gaillard G, et al. Key elements in a framework for land use impact assessment within LCA. Int J LCA

2007;12:5.

[44] Kløverpris J, Wenzel H, Banse M, Milà i Canals L, Reenberg A. Global land use implications of biofuels: state-of-the-art - Conference and workshop on modelling global land use implications in the environmental assessment of biofuels. Int J LCA 2008;13:178.

[45] Schenck RC. Land use and biodiversity indicators for Life Cycle Impact Assess- ment. Int J LCA 2001;6:114.

[46] Khoury GA, Diamond GL. Risks to children from exposure to lead in air during remedial or removal activities at Superfund sites: A case study of the RSR lead smelter Superfund site. J Expo Anal Environ Epidemiol 2003;13:51.

[47] Rytter RM. Biomass production and allocation, including fine-root turnover, and annual N uptake in lysimeter-grown basket willows. For Ecol Manage 2001;140:177.

[48] de Neergaard A, Porter JR, Gorissen A. Distribution of assimilated carbon in plants and rhizosphere soil of basket willow (Salix viminalis L.). Plant Soil 2002;245:307.

(23)

FIGURE CAPTIONS

Fig 1: Single score of environmental impact of biofuel remediation, ReCiPe. Processes contributing more than 0.1% are shown (cut-off 0.1%). Contribution of planting stocks to Salix viminalis cultivation was positive to the environment.

Fig 2: Impact of biofuel remediation on human health, ReCiPe. Impact on resources was similar. Cut-off 1%

Fig 3: Impact of biofuel remediation on ecosystems, ReCiPe. Cut-off 0.015%

Fig 4: Impact of biofuel remediation on global warming, EPD. Ozone layer depletion, acidification, and gross caloric value were similar. Cut-off 0.85%

Fig 5: Impact of biofuel remediation on photochemical oxidation, EPD. Cut-off 0.3%

Fig 6: Impact of biofuel remediation on eutrophication, EPD. Cut-off 0.3%

Fig 7: Single score impact of excavation-and-refill, ReCiPe. Cut-off 0%

Fig 8: Impact of excavation-and-refill remediation on global warming, EPD. All other categories similar. Cut-off 0%

Fig 9: Contribution of impact categories to the single score for excavation-and-refill remediation, ReCiPe.

Fig 10: Comparison of impacts from biofuel remediation (green) with excavation-and- refill (red), EPD.

Fig 11: Comparison of impacts from biofuel remediation (green) with excavation-and- refill (red), ReCiPe, normalised damage.

Fig 12: Comparison of impacts from remediation by excavation and Salix viminalis cul- tivation, single score ReCiPe 2008

Fig 13: Sensitivity analysis: environmental impact of LCA models for biofuel, excavation and no action. Single score ReCiPe results.

(24)

Fig 14: Environmental impact for excavation-and-refill remediation with inert landfill, ReCiPe 2008 single score. Cut-off 0%. Comparable with Fig 7.

Fig 15: Sensitivity analysis: environmental impact of LCA models for biofuel and no action. Single score ReCiPe results