Search

Spildevandsslam er restproduktet dannet ved rensning af spildevand fra husstande. Siden implementeringen af strengere krav til rensning af spildevand i den Europæiske Union i 2005 er produktionen af slam steget markant, hvilket øger efterspørgslen for mere effektiv behandling og genanvendelse af slam.I Danmark er en af de mest anvendte genanvendelsesstrategier for spildevandsslam ud-bringning på landbrugsjord, da dette giver mulighed for at recirkulere næringsstoffer og mikroelementer, og derved erstatter brugen af handelsgødning. I Danmark behandles spil-devandsslam konventionelt via mekanisk afvanding. I slutningen af 1980'erne blev der imidlertid indført en alternativ slambehandlingsmetode, biologiske slamanlæg (BSA). I 2016 var der i Danmark mere end 100 BSA i drift. Behandling af slam i BSA betragtes ofte som mere miljøvenlig i forhold til konventionelle slambehandlingsmetoder. Der er imidlertid kun udført få undersøgelser med formål at vurdere de miljømæssige effekter ved brug af BSA ift. andre slambehandlingsmetoder. Grundet et sparsomt datagrundlag for BSA er resultaterne af allerede udførte miljøvurderinger desuden behæftet med væsentlig usikkerhed.Formålet med projektet var at udføre en miljøvurdering af behandling af spildevandsslam i BSA, og at sammenligne denne med mekanisk behandling på centrifuge og efterfølgende oplagring. Projektet fulgte Erhvervs Ph.d. Programmet, udbudt af Innovationsfonden, og foregik som et samarbejde mellem Danmarks Tekniske Universitet (DTU) og den danske miljøingeniørvirksomhed Orbicon A/S. Resultatet af projektet var et metodespecifikt datasæt for BSA til brug i livscyklusvurderinger, og en livscyklusvurdering af BSA og mekanisk behandling, baseret på danske forhold.For at opnå pålidelige resultater var målet at generere nye data repræsentative for behandling af slam i BSA. Tre fokusområder blev valgt: Kvantificering af biologiske gasemissioner fra selve behandlingsprocessen i BSA, kortlægning af massestrømme gennem behand-lingsprocessen og en undersøgelse af dynamikken i omdannelse og udvaskning af kulstof- og nitrogenforbindelser i det færdigbehandlede slamprodukt i forbindelse med udbringning på landbrugsjord. For at gøre vurderingen af de to behandlingsmetoder så præcis som muligt, blev der for de samme fokusområder også indsamlet data repræsentative for mekanisk behandling af slam. Felt- og laboratoriearbejde udgjorde en væsentlig del af arbejdsprocessen. Data blev ind-samlet fra tre danske BSA kendt for at være veldrevne og for at levere et færdigt slamprodukt af god kvalitet. Endvidere blev data, repræsentative for mekanisk behandling af slam og for de samme fokusområder, også indsamlet. For at gøre data for de to behandlingsteknologier så sammenlignelige som muligt, blev størstedelen af data indsamlet på et renseanlæg, som anvender både BSA og mekanisk afvanding på centrifuge.Livscyklusvurderingen inkluderede 14 miljøpåvirkningskategorier. De miljømæssige bidrag blev normaliserede til at repræsentere de miljømæssige påvirkninger som følge af behandling af 1000 kg slam (vådvægt). Livscyklusvurdering fulgte de internationale standarder for livscyklusvurderingsprincippet. En attributionel tilgang til vurderingen blev valgt, hvilket betyder at sammenligningen tager udgangspunkt i teknologiernes aktuelle formåen. For alle påvirkningskategorier blev bidragene normaliserede til personækvivalenter (PE), hvor én PE repræsenterer det årlige bidrag produceret af én gennemsnitlig person. Tre scenarier for slambehandling blev opstillet: 1) mekanisk behandling, efterfulgt af oplagring udbringning på landbrugsjord (S-CEN), 2) behandling i BSA efterfulgt af udbringning på landbrugsjord (S-STRB) og 2) behandling i BSA efterfulgt af efterbehandling på omlasteplads og udbringning på landbrugsjord (S-SPA).Målsætningen om at producere et datasæt for BSA brugbart i livscyklusvurderinger, samt et datasæt, repræsentativt for mekanisk behandlet slam, blev nået. Biologiske slamanlæg viste sig at klare sig tilsvarende eller bedre end mekanisk slambehandling. Miljøpåvirkningen forårsaget af de to scenarier baseret på BSA (S-STRB og S-SPA) var stort set ens. I forhold til toksikologiske effekter var miljøpåvirkningen den samme for alle tre scenarier, svarende til 2.010-2 PE for påvirkningskategorierne Human Toksicitet – Ikke-kræftfremkaldende stoffer og Økotoksicitet, og 5.0 10-4 PE for kategorien Human Toksicitet – Kræftfremkaldende stoffer. Toksikologiske effekter blev primært forårsaget af metaller, hvilke for alle tre scenarier blev opkoncentreret i det færdigbehandlede slam, og derved ultimativt udbragt på landbrugsjord. Emissionsrater for CO2, CH4 og N2O fra biologiske processer i slam under behandling i BSA, blev målt for alle fire årstider. Resultaterne viste, at årstidsvariationer giver anledning til væsentlige udsving i emissionerne af de nævnte gasarter, og derfor bør inddrages, når gennemsnitlige årsrater beregnes. For mekanisk behandlet og efterfølgende oplagret slam var emissionsraten af CO2 meget lavere end for BSA, hvilket afspejler mindre biologisk aktivitet i mekanisk behandlet slam. Den procentvise andel af kulstof og nitrogen omdannet til de potente klimagasser CH4 og N2O var derimod lavere for BSA, hvilket betød at de miljømæssige bidrag til påvirkningskategorien Klimaforandringer var lige store for BSA (S-STRB og S-SPA) og mekanisk behandling (S-CEN), begge 9.0 10-4 PE, på trods af højere biologisk aktivitet i BSA.Som følge af den lavere biologiske aktivitet i oplagret, mekanisk behandlet slam viste mas-sestrømsanalysen at koncentrationerne af kulstof- og nitrogenforbindelser i det færdigbe-handlede slamprodukt, produceret af denne teknologi, var højere end i det færdigbehandlede slamprodukt fra BSA. Derfor var bidragene til påvirkningskategorier relateret til eutrofiering og forsuring højere fra mekanisk behandlet slam, især for kategorien Marin Eutrofiering, hvor det samlede bidrag fra mekanisk behandlet slam (S-CEN) udgjorde 8.0 10-4 PE, mens det for slam, behandlet i BSA (S-STRB og S-SPA), udgjorde 3.0 10-4 PE.Behandlingsprocessen knyttet til BSA havde et lavere forbrug af abiotiske ressourcer, ho-vedsageligt på grund af at den mekaniske behandlingsproces kræver et input af polymer-masse, hvilket BSA ikke gør. Desuden giver mekanisk behandlet slam ofte anledning til lugtgener, mens slam, behandlet i et velfungerende BSA, er uden lugt. Dette betyder, at slam behandlet i BSA typisk hurtigt bliver afsat til landbrug i lokalområdet, mens mekanisk afvandet slam må transporteres over længere afstande til landbrug, der er villige til modtage det, hvilket resulterer i et højere forbrug af brændstof i forbindelse med transport.I fremtiden vil det være relevant at bruge de nygenererede data for BSA til at udføre livs-cyklusvurderinger, der sammenligner teknologien med andre ofte anvendte slambehand-lingsmetoder. Det vil desuden være relevant at skabe lignende datasæt for BSA beliggende i andre klimazoner, og at udvide datasættende med detaljer om de økonomiske aspekter af behandlingsprocessen. Sewage sludge is generated from the treatment of domestic wastewaters at wastewater treatment plants. Since the implementation of stricter requirements for wastewater treatment in the European Union in 2005, the amount of sludge produced has increased, creating the demand for more effective treatment and recycling. In Denmark, the application of sludge on agricultural land is an often-used recycling strategy, as it returns nutrients and microelements to the soil, which can substitute for commercial fertilisers. Conventionally, sludge produced in Denmark is dewatered with mechanical devices; however, in the late 1980s, sludge treatment reed bed (STRB) systems were intro-duced in Denmark and in 2016, more than 100 STRB systems were operating in the country. Sludge treatment in STRB systems is often considered more environmentally friendly compared to mechanical sludge treatment technologies, albeit only a few life cycle assess-ments (LCAs) comparing the environmental performances of sludge treatment technologies include STRB systems. Furthermore, as data on the STRB system technology suitable for LCA are scarce, the results of these LCAs are unreliable.The project aimed at generating data on the STRB system technology that would be useable for LCA. Based on identified knowledge gaps, research focused on three areas; quantification of gas emissions directly related to treatment, establishment of substance flows through the technology and the fate of carbon and nitrogen-based compounds in treated sludge when applied to the land. The overall goal of the project was to perform an LCA comparing the environmental performance of the STRB system technology with a conventional technology based on mechanical dewatering of sludge on a decanter centrifuge and subsequent storage. Geographically, the project focused on Denmark, and was carried out as a collaborative effort between the Technical University of Denmark (DTU) and the Danish environmental consultancy Orbicon A/S. The outcome of the project was a dataset on the STRB system technology usable for LCA, and an LCA comparing the environmental profiles of the STRB system technology and a mechanical treatment technology, constituting a basis for decision-making in relation to choice of technology.A major part of the project involved performance of fieldwork and laboratory work. Data were collected at three Danish, well-operated STRB systems; furthermore, data required to represent the mechanical treatment technology were collected alongside data on STRB sys-tems. Most of the data collection was undertaken at a wastewater treatment plant housing both technologies, thereby making it possible to make the two datasets as comparable as possible.Fourteen environmental impact categories were included in the LCA, and the environmental loadings and saving provided by the sludge treatment technologies normalised to represent the treatment of 1000 kg wet weight of sludge. The life cycle inventory and the choices underlying the life cycle impact assessment were based on international acknowledged standards and recommendations. An attributional LCA approach was chosen, and the loadings and savings for all impact categories were normalised to people equivalents (PE) (the annual loadings and savings provided by one average person). Three sludge treatment scenarios were defined: 1) mechanical treatment on centrifuge, followed by storage and finally land application, 2) treatment in an STRB system and finally land application (S-STRB), and 3) treatment in an STRB system, followed by post-treatment on a stockpile area (SPA) and finally application (S-SPA).The project succeeded in generating data on STRB systems, which could form the basis for a LCA, and comparable data related to mechanical sludge treatment. The results of the LCA revealed that STRB systems performed comparable to or better than mechanical treatment. The two scenarios based on the STRB system technology (S-STRB and S-SPA) performed comparable which only minor differences.According to toxic impact categories, which for both technologies were mainly impacted by metals contained by treated sludge applied on land, the three scenarios performed com-parable. Indeed, the substance flow analyses revealed that the metals held by sludge subjected to treatment for all scenarios were accumulated in the final sludge product. For all scenarios, the net-loadings for the impact categories Human Toxicity – Non-Carcinogenic and Ecotoxicity corresponded to 2.010-2 PE, and for Human Toxicity – Carcinogenic to 5.0 10-4 PE.Emission rates of CO2, CH4 and N2O related to biological processes in sludge subjected to treatment in STRB systems were measured during all four seasons of the year. The results revealed that seasonal variations were considerable, and should be taken into account when calculating annual, average emission rates. The emission rate of CO2 measured from external storage of mechanically treated sludge was much lower compared to those measured for STRB systems, reflecting a lower microbial activity in the mechanical dewatered sludge. As the emission rates of the potent greenhouse gasses CH4 and N2O were larger for mechanical dewatered sludge, the net environmental loadings provided to the impact category Climate Change by this technology (S-CEN) and the STRB system technology (S-STRB and S-SPA) ended up being equally sized (9.010-4 PE), despite of higher biological activity in the STRB systems.As a consequence of the lower microbial activity in mechanically treated sludge, the con-centration of carbon and nitrogen-based compounds in the final sludge product produced by this treatment technology was higher compared to the final sludge product produced by treatment in STRB systems. Hence, the loadings affecting impact categories related to eu-trophication and acidification were higher for the mechanical treatment technology, espe-cially in relation to the category Marine Eutrophication, the net-loadings to this category being 8.0 10-4 PE for mechanical treatment (S-CEN) and 3.0 10-4 PE for STRB systems (S-STRB and S-SPA).The STRB system technology consumed fewer abiotic resources, due mainly to the fact that the mechanical treatment process requires an input of polymer coagulant, while a STRB system does not require this contribution. Furthermore, as mechanically treated sludge often have a stronger odour compared to sludge treated in STRB systems, the latter is often claimed by the local land application sites, while mechanically treated sludge often must be transported longer distances to land application sites willing to apply it. Hence, the STRB system technology required a lower input of fuel for transportation.In the future, it would be relevant to use the obtained data on STRB systems to compare the technology with other sludge treatment technologies commonly used. Furthermore, it would be relevant to generate a comparable dataset on representing the performance of the technology in other climate zones, and to expand the data set with more data related to economics, making it possible to make more detailed economical assessments.