Life-Cycle Assessment Harmonization and Soil Science Ranking

Apr 17, 2017 - This study reviewed 147 life cycle studies, with 28 found suitable for harmonizing food waste management methods' climate and energy im...
0 downloads 9 Views 895KB Size
Download PDF
Policy Analysis pubs.acs.org/est

Life-Cycle Assessment Harmonization and Soil Science Ranking Results on Food-Waste Management Methods Jeffrey Morris,*,† Sally Brown,‡ Matthew Cotton,§ and H. Scott Matthews∥,⊥ †

Sound Resource Management Group, Inc., Olympia, Washington 98502, United States Ecosystem Science Division, College of Forest Resources, University of Washington, Seattle, Washington 98195, United States § Integrated Waste Management Consulting, LLC, Nevada City, California 95959, United States ∥ Department of Civil and Environmental Engineering and ⊥Department of Engineering and Public Policy, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213, United States ‡

S Supporting Information *

ABSTRACT: This study reviewed 147 life cycle studies, with 28 found suitable for harmonizing food waste management methods’ climate and energy impacts. A total of 80 scientific soil productivity studies were assessed to rank management method soil benefits. Harmonized climate impacts per kilogram of food waste range from −0.20 kg of carbon dioxide equivalents (CO2e) for anaerobic digestion (AD) to 0.38 kg of CO2e for landfill gas-to-energy (LFGTE). Aerobic composting (AC) emits −0.10 kg of CO2e. In-sink grinding (ISG) via a food-waste disposer and flushing for management with other sewage at a wastewater treatment plant emits 0.10 kg of CO2e. Harmonization reduced climate emissions versus nonharmonized averages. Harmonized energy impacts range from −0.32 MJ for ISG to 1.14 MJ for AC. AD at 0.27 MJ and LFGTE at 0.40 MJ fall in between. Rankings based on soil studies show AC first for carbon storage and water conservation, with AD second. AD first for fertilizer replacement, with AC second, and AC and AD tied for first for plant yield increase. ISG ranks third and LFGTE fourth on all four soil-quality and productivity indicators. Suggestions for further research include developing soil benefits measurement methods and resolving inconsistencies in the results between lifecycle assessments and soil science studies.

■

INTRODUCTION Food waste per capita in the United States is estimated at over 100 kg per year.1 Food waste occurs at each step in the supply chain, from farm to processer to retailer to consumer. Over 95% of food scraps generated by households and retail businesses in the United States, amounting to nearly 32 000 000 Mg, are thrown away; most is landfilled.2 If all of these food scraps were landfilled and yielded greenhouse gas (GHG) emissions at the harmonized rate calculated for lifecycle assessments (LCAs) reviewed here, annual GHG emissions would total 11 000 000 Mg carbon dioxide equivalents (CO2e), about 0.2% of the total U.S. emissions. To reduce landfilled food waste and associated GHG emissions, many municipalities and states have, or are considering, disposal bans and/or augmented diversion from disposal options such as curbside collection and processing into compost.3−5 There are a number of management options other than landfilling (LF) and landfill-gas-to-energy (LFGTE) for food waste, including aerobic composting (AC) and the use of compost as a soil amendment, anaerobic digestion (AD) followed by digestate composting or direct land application, insink grinding (ISG) and diversion to wastewater treatment facilities, home composting, and combustion.6 © XXXX American Chemical Society

The United States Environmental Protection Agency (USEPA) Waste Reduction Model (WARM) calculator for climate and energy impacts of municipal solid waste (MSW) management choices quantifies soil carbon storage as a benefit from compost use as a soil amendment and mentions additional benefits such as nutrient recycling and increased soil-water holding capacity.7 As in WARM, life cycle assessment (LCA) can be used to evaluate climate and other environmental impacts from food scraps management method, end use, and processing residue disposal choices. However, results from different LCAs can be contradictory due to, inter alia, use of differing functional units, boundary conditions, and global warming potentials (GWPs) for GHGs. Table 1 indicates how life cycle climate and energy effects vary for four food waste treatment methods applied to a kilogram of food waste, which is the functional unit for this study. The table shows nonharmonized results from the 28 LCA studies selected for harmonization. For example, 25 LCAs Received: Revised: Accepted: Published: A

December 3, 2016 March 31, 2017 April 17, 2017 April 17, 2017 DOI: 10.1021/acs.est.6b06115 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

■

MATERIALS AND METHODS Methods for Harmonization of Climate and Energy Impacts. A total of 147 recent LCA type studies published through the first half of 2014 were reviewed to select a subset providing enough data and transparency to facilitate harmonization. A total of 108 of the 147 studies provided information that assisted with selection of parameters and assumptions requiring harmonization. However, only 28 were suitable for harmonization.8 Table 2 lists the criteria used to select out studies lacking in some aspect required for climate and energy impacts

Table 1. Means and Ranges for Non-Harmonized Climate and Energy Impacts per Kilogram of Food Wastea treatment method AC AD ISG LF

GWP (kg CO2e) 0.01 −0.15 0.19 0.44

(−1.12 to 0.47) [25] (−0.48 to 0.03) [10] (0.00 to 0.44) [5] (−0.26 to 0.91) [15]

energy (megajoules) 1.15 −2.25 0.49 0.19

Policy Analysis

(0.18 to 3.63) [10] [1] (0.19 to 0.81) [3] (−2.15 to 1.20) [5]

a Note: Numbers in parentheses show ranges; numbers in brackets [n] indicate number of LCAs providing estimates. Adapted with permission from ref 8.

Table 2. Exclusion Criteria and Counts for Excluded Studiesa examining the AC treatment for food wastes show net CO2e emissions ranging from a reduction of 1.12 kg to an increase of 0.47 kg of CO2e/kg of food waste. The climate impact reducing LCA included offsets for increased soil carbon storage and reduced use of synthetic fertilizers. The high-climate-impact LCA did not include credits for soil carbon storage, fertilizer displacement, or any other potential benefit from the use of compost. As a second example, results for 15 LCAs covering landfilling diverge due in part to differing assumptions on landfill methane capture efficiencies and carbon footprints for energy offsets. Climate impacts range between reductions of 0.26 kg of CO2e to increases of 0.91 kg CO2e per kg food waste landfilled. The LCA showing most climate benefit assumed 66% landfill methane capture and 100% utilization of captured methane for power generation that displaced electricity generated by a mix of 72.5% coal and 27.5% natural gas. The LCA indicating most climate harm assumed 45% methane capture and just 12% utilization of captured methane to generate power. That power displaced electricity with a much-smaller carbon footprint than from a 72.5% coal−27.5% gas mix. Further discussion on Table 1 and other results herein is available in OR DEQ (2014).8 The Supporting Information provides additional detail on pre- and post-harmonization sample sizes for climate impacts. The divergent results exhibited in Table 1 provide one purpose for this study: to review and harmonize LCA results on food waste management methods. The other major purpose is to rank management methods based on results found in the literature on soil productivity and soil quality. Some LCAs evaluate increased soil carbon storage and displacement of synthetic fertilizers and pesticides associated with the use of composts, digestates, or biosolids. However, LCA literature does not have impact categories that directly address soil quality and soil productivity. Yet the benefits of adding organic matter to soils are well quantified. For example, soil’s ability to hold and store water, to transform wastes and nutrients, to store carbon, and to support plant growth increases as the organic matter concentration of soils increases.9−16 Previous work also has shown that adding organic-waste-based composts or biosolids to soils is a rapid way to increase soil carbon reserves.10,11,16−18 Although impacts from adding organic matter vary on the basis of the health of the receiving soil, beneficial results from amending soils with organic material outputs produced by AC, AD, and ISG are likely, especially because much of the urban and agricultural soil in the United States is degraded.19,20

criteria for excluding LCA from harmonization

LCA study counts

reviewed studies not an LCA review study of LCAs no assessment of targeted food waste treatments functional unit not food-waste-based system boundary not clear input−output data not detailed duplicative studies harmonized

147 40 16 8 36 0 12 7 28

a

Reprinted with permission from ref 8.

harmonization. These criteria were applied sequentially. Once a study matched an exclusion criterion, it was dropped from scrutiny for other criteria. Hence, the counts do not reveal the number of LCAs that would be excluded by any particular criterion other than the first. After all exclusions, 28 LCAs remained for harmonization of climate and energy impacts. These LCAs use a wide range of assumptions and estimates for critical parameters such as nitrogen content of soil amendments produced by AC, AD, and ISG; LF methane capture rate and amount of captured methane burned to generate electricity; and the extent to which soil amendments were used in ways that actually reduce the use of synthetic fossil-fuel-based fertilizers or increase soil carbon storage. Note that resource acquisition and manufacturing impact reductions from displacement of synthetic fertilizers or peat are included in harmonization calculations. Comparison of soil amendments versus fertilizers or peat on impacts of transport to land application sites or of climate pollutant emissions such as nitrous oxide (N2O) from land applications are not. Harmonization of these assumptions and parameter estimates follows the general approaches of LCA meta-analysis pioneered by Heath and others at the United States Department of Energy’s National Renewable Energy Laboratory (NREL) for energy generation systems.21−23 Specific adjustments and harmonizations for assumptions and parameter estimates are detailed in the following subsections. In general, harmonization proceeds by averaging estimates for a specific impact from a specific management method across all of the 28 studies that provide an estimate. That average is applied as the impact estimate in studies that did not cover that specific impact and also is the harmonized value for that specific impact. Harmonized impacts totals for each management method are pair-wise compared statistically using the t test of significance. B

DOI: 10.1021/acs.est.6b06115 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Policy Analysis

Environmental Science & Technology Table 3. General Harmonization Results for Climate and Energy Impactsa climate impact (kg CO2e/kg food waste) activity

aerobic composting

anaerobic digestion

in-sink grinder

LCA sample size

(25)

(10)

collection and transport processing carbon storage fertilizer displacement peat displacement electricity displacement total impact (net)

0.04 0.11 −0.12 −0.05 −0.07

0.02 0.09 −0.08 −0.02 −0.01 −0.19 -0.20

a

-0.10

energy impact (MJ/kg food waste) landfill

aerobic composting

anaerobic digestion

in-sink grinder

landfill

(5)

(15)

(10)

(1)

(3)

(5)

0.05 0.23 −0.03 −0.05

0.03 0.61 −0.12

0.68 0.66

0.68 0.58

0.39 0.54

0.51 0.57

−0.14 −0.06

−0.22

−0.11 0.10

−0.14 0.38

−0.08 −0.09 −0.83 0.27

1.14

−1.03 -0.32

−0.68 0.40

Adapted with permission from ref 8.

Accounting Adjustments. The first step of the harmonization process was to adjust published results to attain accounting consistency and correct for computational errors. Most LCAs used 2007 Intergovernmental Panel on Climate Change (IPCC) 100 year GWPs; those that did not were adjusted to 2007 IPCC GWPs. There also were discrepancies in accounting for the energy value of displaced electricity. Most LCAs used the energetic value of electricity, 3.6 MJ/kWh. However, several studies used the energetic value of fuels used to produce power. This is electricity’s primary energy demand and depends on fuel types used and the energy-conversion efficiency of power-generation facilities. Converting the energy value for displaced electricity to the energy value of fuels used to generate that electricity for each LCA requires data transparency that was not available for every LCA. Hence, energy offsets for displaced electricity were all converted to the 3.6 MJ/kWh. Collection and Collection Container Impact on Harmonization. All treatment options except ISG require the collection of food scraps. For studies that did not include climate or energy impacts from collection, we added in the average impact from studies that did provide these impact estimates. Only one LCA examined AD energy impacts, and that study did not provide an estimate for collection energy usage. In this case, mean AC collection energy was used as an estimate of AD collection energy. Collection energy is not likely to depend on whether the processor is an AC or AD facility. Note that the three ISG LCAs evaluating energy impacts all included estimates for grinder electricity use. Very few studies included impact estimates for collection bags or bins. Hence, available estimates were not robust, and collection-bag and -bin impacts were excluded. Facility Construction and Equipment Manufacturing Impact on Harmonization. Each of the food waste treatment options may require treatment facility construction. Of the 28 studies harmonized, 3 assessed AC facility construction climate impacts, and 4 assessed AC facility construction energy impacts. Estimates for climate and energy impacts differed by more than 6 and 3 orders of magnitude, respectively. There was only one assessment for AD facility construction climate impacts. There were three estimates for climate and energy effects of landfill cell construction. These also differed by more than 3 orders of magnitude, and one climate estimate was essentially zero. Due to the highly uncertain nature of these assessments, facility construction was removed as a process included within the boundaries of the harmonized LCAs. Furthermore, solid waste management LCAs typically find facility construction and

equipment manufacturing impacts per unit of waste to be de minimus.24,25 The LCAs chosen for harmonization did not include production impacts for collection vehicles, roads, or wastewater conveyance pipes. All of the ISG LCAs provided climate and energy assessments for grinder production. None of the ISG LCAs provided assessments for wastewater treatment facilities. To be consistent with the lack of impact assessments for production of food waste collection vehicles for the AC, AD, and LF options, ISG grinder production impacts were removed from consideration in harmonized results. Harmonizations for Soil Carbon Storage, Fertilizer Displacement, Peat Displacement, and Electricity Displacement Impacts. Benefits from treatment outputs such as soil amendments or energy were added wherever they were missing in the LCAs. For soil amendments, benefits harmonized included carbon storage and displacement of synthetic fertilizers and peat. Soil carbon storage can occur when land application of a food waste derived amendment results in a stable increase in the receiving soil carbon content. Previous work has shown that soil carbon increases following the use of AC compost, AD digestate, or biosolids generated from ISG soil amendments.11,18−20,26,27 A portion of the carbon in food waste that is landfilled will also persist.7,28 LCAs lacking carbon storage estimates were harmonized by adding carbon storage for a particular treatment based on average carbon storage benefit estimated in LCAs on that treatment that included carbon storage. For fertilizer displacement, credits were added to LCAs that did not include these offsets. Credits are based on average climate and energy displacement estimates from the subset of harmonized LCA studies that did estimate these displacement benefits. For synthetic nitrogen (N), significant quantities of energy are required to transform gaseous N to mineral N. Mined phosphate has to be processed from phosphate rock into a plant available form. Thus, displacement of synthetic fertilizers by soil amendments produced from food wastes or other organic wastes may conserve energy and reduce carbon emissions.9,12 For the displacement of peat in growth media, AC and AD LCAs were harmonized by adding climate and energy benefits to the studies that did not include one or both of these benefits. The additions for a particular treatment are based on the average climate or energy benefit estimated in studies on that treatment that included climate or energy offsets for peat substitution. None of the ISG LCAs assessed climate or energy benefits from peat displacement with ISG biosolids. While C

DOI: 10.1021/acs.est.6b06115 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Policy Analysis

Environmental Science & Technology Table 4. Oregon Specific Harmonization Results for Climate and Energy Impactsa climate impact (kg CO2e/kg food waste) activity

aerobic composting

anaerobic digestion

in-sink grinder

LCA sample size

(25)

(10)

collection and transport processing carbon storage fertilizer displacement peat displacement electricity displacement total impact (net)

0.04 0.11 −0.12 −0.03 −0.04

a

-0.05

energy impact (MJ/kg food waste) landfill

aerobic composting

anaerobic digestion

in-sink grinder

landfill

(5)

(15)

(10)

(1)

(3)

(5)

0.02 0.09 −0.08 −0.02

0.05 0.23 −0.03 −0.05

0.03 0.69 −0.12

0.68 0.66

0.68 0.58

0.39 0.54

0.51 0.57

−0.09 −0.03

−0.15

−0.22

−0.17 -0.17

−0.18 0.03

−0.08 0.52

−0.83 0.29

−1.03 -0.32

1.23

−0.48 0.60

Reprinted with permission from ref 8.

A survey of Oregon AC facilities was conducted to determine current end uses of composts generated at these facilities. Survey responses indicate that 25% of compost produced with some food waste inputs is sold to agriculture users, 62% to households, 12% to landscapers, and 1% to nurseries.8 It is assumed that all agriculture users land apply compost as a fertilizer substitute, nurseries use compost as a peat substitute, and landscapers do not use compost as a substitute for either fertilizer or peat. Survey results for Danish household home gardening indicate that 18% of compost is used to replace synthetic fertilizers, 11% manures, and 20.5% peat; 50.5% does not displace any soil amendments.31 Assuming that Oregon households behave similarly, 36% of AC compost displaces fertilizers, and 14% displaces peat. The Oregon facilities survey did not cover AD digestate or ISG biosolids utilizations. However, a large portion of biosolids generated in OR are applied to agricultural fields.32 Assuming that all AD digestate and ISG biosolids in Oregon are used in agriculture, the utilization rate for these outputs would be 100% for fertilizer displacement. Regardless of use, it is assumed that soil carbon storage in Oregon is attained at the rates estimated for general harmonization from incorporation into agricultural or home garden soils: 0.12 kg of CO2e for AC compost, 0.08 kg of CO2e for AD composted digestate, and 0.03 kg of CO2e for ISG biosolids/kg of food waste, as indicated in Table 3. Methods for Assessing Soil Productivity. A total of 80 recent studies on impacts of residuals-based soil amendments use on plant and soil productivity were reviewed.8 From these studies a qualitative ranking of potential soil productivity benefits from food waste treatment options was developed based on results reported in the literature, e.g., Table S2. Response of different amendments will vary on the basis of soil type, cropping system, and characteristics of the amendments. Because of this inherent variability, it is not possible to develop a quantitative ranking for soil response that would apply across end use sites. The categories ranked include soil carbon storage and sequestration, fertilizer replacement, water conservation, and yield increase.10−12,27,33 Each of these benefits has the potential for impacts on climate and energy use, indicating that LCA results and soil productivity rankings are necessarily interrelated. For example, soil carbon sequestration removes carbon from the atmosphere, and storage in soil of previously sequestered carbon prevents or reduces its biodegradation to atmospheric pollutants. Fertilizer replacement avoids the energy use and emissions associated with the manufacture of synthetic fertilizers (and associated

studies have shown the potential to use biosolids for potting mixtures, this practice is far from mainstream.29 It also requires additional materials and additional processing. On this basis, offsets for peat displacement were not added to ISG LCAs. For grid electricity displacement, harmonized AD LCAs all included these offsets. LCAs for ISG and LF that did not include climate and energy offsets for grid electricity displacement were harmonized by adding benefits based on means from studies that did include these offsets. The exception was for the one ISG study lacking an energy benefit. Its adjustment was based on the 0.3 kWh electricity production estimate used in that study to calculate its climate benefit from electricity production. An important caveat for the ISG treatment is that, while ISG can generate energy during anaerobic digestion at the wastewater treatment plant (WWTP), the prior treatment (secondary aeration) requires significant amounts of energy and may lower methane generation from digestion. Harmonization did not attempt to adjust for these uncertainties regarding ISG treatment. Oregon Specific Adjustments. The Oregon Department of Environmental Quality (DEQ) funded a significant portion of the research for this article. DEQ requested that the general harmonization results reported in Table 3 be adjusted to Oregon specific conditions for grid electricity, landfill methane capture efficiency, and the uses of soil amendments by Oregon households and businesses. The Oregon specific harmonization results are reported in Table 4. The Oregon grid’s carbon footprint is based on the USEPA’s 2010 eGRID data for Oregon nonbaseload electricity. It is 0.61 kg CO2 e/kWh based on 2007 IPCC global warming potentials,30 which approximately represents a lower bound in the context of the United States. The nonbaseload footprint is for power-generation facilities used to meet peak power demand. As such, they more closely measure the marginal rather than the average climate impact for power generation in Oregon. The marginal (i.e., last to be used), as opposed to baseload, power sources are more apt to be displaced by electricity generated at AD, ISG, or LF facilities. Oregon landfill methane capture efficiency and use for power generation are based on DEQ staff estimates that Oregon landfill gas capture averages 62%, in-landfill oxidation of fugitive methane averages 10%, and 72.6% of captured methane is used to generate electricity, with the remainder flared. This compares with 67% capture and 10.5% oxidation averages and 100% utilization of captured methane for power generation from the general harmonization of 28 LCAs. D

DOI: 10.1021/acs.est.6b06115 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Policy Analysis

Environmental Science & Technology Table 5. Rankings from Harmonization and Qualitative Assessment of Food Waste Treatmentsa

a

treatment

climate

energy

soil carbon

fertilizer replacement

water conservation

yield increase

AC AD ISG LF

2 1 3 4

4 2 1 3

1 2 3 4

2 1 3 4

1 2 3 4

1 1 3 4

Reprinted with permission from ref 8.

Applying the lower Oregon specific utilization rates for soil amendments from AC, AD, and ISG treatments mostly increases climate and energy impacts. This is because the nonharmonized LCAs, and the general harmonization results, assumed utilization rates of either 100% for fertilizer substitution or 50% each for fertilizer and peat when both fertilizer and peat substitutions were considered. In contrast, the Oregon user survey results indicated a substantial portion of uses for organic outputs from food waste treatments that did not displace either fertilizer or peat. Soil Productivity Results. Table 5 displays soil productivity rankings along with rankings for harmonized climate and energy impacts. A ranking of 1 is highest, and 4 is lowest. The soil productivity rankings are qualitative. Soil Carbon. Land application of soil amendments may increase soil carbon reserves. Several studies have shown increased soil carbon with as little as a single application of animal manure compost. However, most studies show significant increases with repeated applications over time or with a single high rate application of compost or biosolids.7 The relative potential for carbon storage may vary on the basis of the type of stabilization used for the residual material prior to land application. One study saw slightly higher soil carbon storage for composts compared with anaerobically digested municipal biosolids.11 However, this study did not include sideby-side comparisons of the two types of amendments. Induced carbon sequestration will also vary across soil types, land use, and climate.9 Note that AC involves aerobic stabilization of organics, whereas AD and ISG involve anaerobic stabilization. It is likely that material that is anaerobically stabilized will undergo additional carbon mineralization when it is in an aerobic soil environment. However, some research has indicated that formation of mineral-organic complexes during AD will, in fact, increase the stability of C in the land applied amendments from AD. For ISG the secondary aerobic digestion process in wastewater treatment facilities was considered to result in additional carbon loss. While a portion of the carbon in LF food waste will not decompose and thus can be considered sequestered, there is no potential for enhanced plant growth and the increases in soil carbon associated with increased net primary productivity. For this reason, LF carbon-storage potential was ranked lowest of the alternatives considered. We ranked the potential for soil carbon storage highest for AC, followed by AD and then ISG. Carbon emissions from soils as a result of using composts, digestates, biosolids, fertilizers, or peat as soil amendments may also be important to consider in evaluating climate impacts. However, these impacts are difficult to quantify in general. For example, a default emission factor for N2O from fertilizer may be 1% of total applied nitrogen, but actual emissions will vary on the basis of soil texture, climate, and crop factors. As a result, reported emissions may be substantially above or below such default factors.36

supply chain emissions). Increasing soil carbon though the use of residual-based nutrients will increase soil water holding capacity and improve soil structure. This will in turn increase plant yield and quality, resulting in additional increases in soil organic matter.9,15,20 Increased efficiencies in production, including reduced water use per unit of food output and increased quantity grown per hectare, will impact the carbon and energy associated with growing food. Anaerobic digestion of food wastes is most commonly done in municipal wastewater treatment facilities. There is limited literature on characteristics of AD digestate from food scraps. Hence, literature on land application of municipal biosolids was used as a surrogate for ranking soil productivity impacts of AD digestate and ISG biosolids.34,35

■

RESULTS LCA Harmonization Results. Table 3 displays general harmonization outcomes, i.e., results prior to adjustments for Oregon specific conditions. Climate impacts per kilogram of food waste range between −0.20 kg of CO2e for AD and 0.38 kg of CO2e for landfill disposal. AC performs second-best, and ISG ranks third. Harmonization reduced climate impacts for all four treatments from the nonharmonized averages shown in Table 1. Harmonized energy impacts per kilogram of food waste range between −0.32 MJ for the ISG treatment to 1.14 MJ for AC. AD and landfill rank second and third, respectively. As the only end-use option that typically does not produce an energy output, AC has a comparatively large energy footprint. The energy impact of AC is about the same as its nonharmonized average. Harmonization decreased the energy impact of ISG substantially due to removal of manufacturing energy for in-sink grinders and inclusion of offsets for electricity generated from methane produced at wastewater treatment plants. Harmonization doubled the energy impact of landfill disposal due to adjusting offsets for displaced grid electricity down to their direct electrical energy equivalence. Landfill disposal energy also increased due to the addition of collection impacts. Harmonization increased the energy impact of AD substantially due to the accounting adjustment for displaced grid electricity as well as the addition of food waste collection energy use. Table 4 displays harmonization results after adjustments to Oregon specific conditions. Substantial changes from results shown in Table 3 include fertilizer and peat substitution and energy generation offsets, as indicated by comparing line item values in Tables 3 and 4. For example, the low carbon footprint for electricity in Oregon reduces the climate benefits associated with generating power for AD slightly and for ISG and LFGTE substantially. As another example, lower methane (CH4) capture and use in Oregon result in higher LF climate impacts from fugitive methane and lower climate and energy offsets for displaced electricity. E

DOI: 10.1021/acs.est.6b06115 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Policy Analysis

Environmental Science & Technology

Furthermore, if the footprint for displaced electricity were to be based on solar power (lower carbon footprint), then AC would displace AD in being best for the climate. If the footprint were based on coal-fired power (higher carbon footprint), then ISG would move into second place behind AD, and AC would fall to third. The climate-impact ranking for LFGTE is not as sensitive to the carbon footprint for displaced electricity due to landfill’s relatively low electricity output compared to its high climate impact from fugitive methane generated by biodegrading food scraps. In contrast, the energetic value of electricity is not sensitive to local, regional, or national energy resource uses. Using primary energy demand, rather than the energetic value of electricity, to measure energy impacts of displaced power would potentially increase energy offsets for electricity for AD and LFGTE by enough for both to show reductions in overall energy use. ISG’s energy reduction benefits would likely increase substantially as well. However, multiplying electricity displacements for AD, ISG, and LFGTE by the same factor would not change their relative ranks for energy impacts. In addition, even if all electricity were low-carbon from solar or wind power, AC likely would still rank last for net energy impacts. However, energy impact differences between the four treatment options likely would not differ substantially or significantly in that situation. Oregon specific climate impact rankings for AC and AD are sensitive to the utilization rate for amendments as a substitute for fertilizer or for peat. Oregon utilization rates for fertilizer and peat displacement are 36% and 14%, respectively, for AC harmonization, but 100% is the fertilizer substitution rate for AD. Increasing AC compost utilization rates and decreasing AD digestate utilization could switch AC into first place for the smallest climate impact. Climate rankings are not very sensitive to estimates of soil carbon storage rates from use of soil amendments produced by AC, AD, or ISG. For example, the carbon storage factor for AC compost would have to double for AC to be similar to AD in both the general and Oregon specific climate rankings. Rankings may be sensitive to outliers that substantially skew average climate or energy impacts of a unit process. For example, one of the five LFGTE LCAs that estimated LF energy use has an estimate for landfill processing energy that is very high. Removing that outlier reduces LF processing energy by more than half. This would lower LFGTE total Oregon specific energy impact below AD, pushing AD into third place and moving LFGTE into second in the energy-usage ranking.

Due to higher application rates, use of soil amendments may result in differential impacts due to transport fuel use. For example climate-changing emissions for land applying 910 dry kg of biosolids (25% solids) to meet the nitrogen needs of a crop have been estimated as 7.6 kg.12 Fertilizer Replacement. A significant fraction of the total nitrogen in food scraps is lost to volatilization during stabilization for each of the food waste treatments evaluated. The dilution of total nutrient compost in the majority of AC as a result of the use of high-carbon feedstocks (e.g., leaves, branches, or wood chips) in the compost blends can result in lower nutrient content in the final product. In extreme cases, composts can induce nutrient deficiencies or provide very limited fertility to soils.33 In contrast, anaerobic digestion conserves nutrients while reducing the volatile solids content of the material, resulting in higher total and available nutrient concentrations.37 For this reason, AD was considered to have higher nutrient displacement potential. ISG is considered to be lower than direct AD or AC due to the likelihood of nitrogen loss during secondary treatment at the wastewater plant. Landfill ranks last because there is no potential to reduce use of synthetic fertilizers. Water Conservation. Rankings based on water conservation benefits follow soil carbon rankings because higher carbon concentration in soils enhances water storage. For higher-clay soils, increased soil carbon will reduce bulk density, and that will, in turn, increase water infiltration rates. In the case of sandier soils, higher carbon will increase water holding capacity. For these reasons, increased soil carbon improves soil water conservation across the range of soils.10,11 In fact many properties including fertilizer status of the soil are associated with increased soil carbon concentrations. However, as traditional agronomic management has not focused on these linkages, these are considered separately in this analysis. Higher nutrient status of the material can effectively reduce potential for increased water storage. Lower-nutrient composts are often applied at high rates as a soil conditioner. Composts are also not subject to application based on nutrient requirements as ISG biosolids currently are and AD digestates may be. For changes in water potential to be realized, high rates of organic amendment are required, either as multiple applications over time or a high single application. Yield Increase. Crop yield increase can be a benefit of land application of soil amendments produced by AC, AD, and ISG. Use of compost or digestate provides soil conditioning and micronutrients in addition to substituting for conventional fertilizers. Yield increases have been observed in multiple cases where organic amendments are used instead of synthetic fertilizers.8 If material is being added to soil to grow agricultural or commercial crops, there will be a clear dollar value associated with increased yield. The market may also recognize a dollarvalue increase from better nutrient availability due to quality increases for food products, such as higher protein in wheat. If material is added to landscaped areas, the financial benefits are more difficult to quantify. Both AC compost and AD digestate typically improve yields and tie for top ranking on yield. ISG is again lower because of volume loss during aerobic digestion. Sensitivity of Harmonization Results. Climate rankings for AC, AD, and ISG are particularly sensitive to the carbon footprint of displaced electricity. The Oregon specific harmonized results shown in Table 4 are based on Oregon’s nonbaseload electricity footprint, which is 10−15% higher than the carbon footprint of natural gas fired power generation.

■

DISCUSSION Following general harmonization, the average climate impact for AD is statistically significantly lower than for AC, ISG, or LFGTE. This is due to both energy creation and land application benefits associated with AD. AC is better for the climate than ISG as a result of lower energy use during treatment; however, this result has a lower confidence level than for the AD comparisons because there were five or fewer LCAs providing climate and energy impact estimates for ISG. Both AC and ISG are significantly better for the climate than LF. The large difference in mean climate impacts between ISG and LF more than compensates for the paucity of ISG LCAs and their resultant large standard deviations for climate and energy-impact estimates. For general harmonization’s energy impacts, AD has a sample size of one and cannot be statistically compared to other F

DOI: 10.1021/acs.est.6b06115 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Policy Analysis

Environmental Science & Technology

digestate from AD. This lack of data suggests that there is a potential for differences in characteristics of the two products and how they are used. If AD digestion of food scraps at municipal treatment plants continues to expand, this is less of a concern. While there is a national regulatory structure for municipal biosolids, as of yet, there is no similar system for regulating food waste digestates. Soil studies suggest that AD soil amendments provide greater potential for fertilizer substitution than AC compost. However, the LCA studies selected for harmonization had mean estimates of fertilizer substitution potential that were higher for AC compost than for soil amendment outputs of either AD or ISG. This suggests that soil quality research may not be adequately reflected in LCA studies on food waste management systems. Finally, methane has a residence time of approximately 12 years in the atmosphere, much shorter than that for carbon dioxide. Its global warming potential for shorter time frames is much higher than for the 100 year time frame used in most LCAs and in the harmonization results here. It is possible that if a shorter time frame had been used (say, 20 years or less), impacts such as methane emissions from virgin peat mining, from AC piles that go anaerobic for a period of time, or from process stages of the other treatments examined could change relative rankings. Future LCA studies using shorter time frames would help with the understanding of this uncertainty.

food-waste treatments. ISG has statistically significantly lower energy use than AC or LFGTE, although the harmonized studies did not account for energy requirements of secondary aeration for ISG. LFGTE has significantly lower net energy use than AC. For both general and Oregon-specific harmonizations, grid electricity displacement tends to result in the largest reduction in both GHG emissions and energy use among the four outputs (carbon storage, fertilizer displacement, peat displacement, and grid electricity displacement), except in the case of AC, which has no power output. For AC, carbon storage in soils amended with compost has the highest GHG reduction benefit. LFGTE carbon storage and electricity displacement are similar, especially for the general harmonization results. Energy benefits from fertilizer displacement exceed peat displacement benefits under current compost utilization patterns in Oregon for AC, AD, and ISG. Under general harmonization results, this result holds for AC and ISG, but AD energy benefits for fertilizer and peat displacement are similar. Across the different treatments, energy consumption for collection and transport and for treatment is similar for each option considered. Climate impacts of processing are much larger for each treatment than collection and transport. This is the case for LFGTE in particular due to fugitive methane emissions. However, this disparity is somewhat exaggerated by a very high estimate for fugitive methane emissions in one of the five LF LCAs. Excluding that study’s estimate reduces the Oregon specific processing average energy use for landfills to 0.26 MJ/kg of food waste landfilled. A further comment on harmonization as well as soil productivity results is that AC and AD both have treatment method output benefits of sufficient magnitude to more than offset the climate impacts of collection, transport, and processing. For AD, the climate offset for electricity is most important, whereas for AC, it is carbon storage that provides the largest benefit. Overall, AC and AD are highest in five of six categories of environmental impact rankings shown in Table 5. ISG ranks third in five of the ranking categories. Landfill is always last, except for energy use, for which AC ranks last. Based on information in the 28 harmonized LCA studies and the 80 reviewed soil studies, relative preferences for the six environmental impacts would determine whether AC or AD is the best treatment option for food waste. High preference for energy production would favor AD. Strong preferences for soil carbon storage and water conservation would favor AC. ISG and LFGTE rank below AC and AD for five of the six environmental criteria which were examined in this review and harmonization in order to generate the rankings shown in Table 5. Energy usage is the only variable in which the preference for AC or AD over ISG does not hold. Generalization of results and conclusions from harmonization and soil productivity literature review are constrained by a general lack of LCA studies on food waste processing systems. There was only one AD LCA that had harmonizable results for energy impacts. More than 10 studies suitable for harmonization were available only for AC and LFGTE climate impacts. There is at least one more recent LCA not reviewed and harmonized that could help redress this lack.38 However, LCA literature still does not seem to evaluate impact categories that directly address soil quality and soil productivity. Another source of uncertainty is that wastewater treatment plant biosolids were selected as a proxy for the soil impacts of AD digestate due to a lack of research studies on soil impacts of

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.est.6b06115. A table showing detail and sources supporting the yield increases that are discussed as a soil productivity benefit in the Yield Increase subsection of the Soil Productivity Results section of the article. (PDF) A table showing pre- and post-harmonization sample sizes and impact estimates for three of the components of climate change offsets. (XLSX)

■

AUTHOR INFORMATION

Corresponding Author

*Phone: +1-360-867-1033; e-mail: jeff[email protected]. ORCID

Jeffrey Morris: 0000-0002-5815-389X Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors acknowledge the Oregon State Department of Environmental Quality (DEQ) for funding much of the research and analysis for this article. We also acknowledge helpful review and suggestions from DEQ staff during conduct of the project that led to this article as well as valuable comments from three anonymous peer reviewers. All errors and omissions remain the responsibility of the authors. We declare no conflict of interest in any of the analyses or results discussed in this article.

■

REFERENCES

(1) Buzby, J. C.; Hyman, J. Total and per capita value of food loss in the United States. Food Policy. 2012, 37, 561−570. (2) U.S. Environmental Protection Agency (EPA). Office of Resource Conservation and Recovery. Advancing Sustainable Materials

G

DOI: 10.1021/acs.est.6b06115 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Policy Analysis

Environmental Science & Technology Management: Facts and Figures 2013, Assessing Trends in Material generation, Recycling and Disposal in the United States; EPA530-R-15− 002; EPA: Washington, DC, June 2015. (3) Clark, N. Vermont Districts Closing Loop on Food Waste Diversion. BioCycle. 2013, 54 (7), 38; https://www.biocycle.net/ 2013/07/18/vermont-districts-closing-loop-on-food-waste-diversion/ (4) California Department of Resources Recycling and Recovery (CalRecycle). Mandatory Commercial Organics Recycling, AB 1826 Chesbro (Chapter 727, Statutes of 2014). http://www.calrecycle.ca. gov/Recycle/Commercial/Organics/ (accessed October 13, 2016). (5) Massachusetts Executive Office of Energy and Environmental Affairs. Commercial Food Waste Disposal Ban, Compliance, Regulations and Guidance. http://www.mass.gov/eea/agencies/ massdep/recycle/reduce/food-waste-ban.html (accessed October 13, 2016). (6) Parry, D. Municipal food waste management options: climate and economic impacts. In Brown, S.; McIvor, K.; Hodges Snyder, E., Eds.; Sowing Seeds in the City - Ecosystem and Municipal Services; Springer: New York, 2016, pp 107−123. (7) U.S. Environmental Protection Agency (EPA) Office of Resource Conservation and Recovery. Documentation for Greenhouse Gas Emission and Energy Factors Used in the Waste Reduction Model (WARM); Prepared by ICF International; EPA: Washington, DC, 2015. (8) State of Oregon Department of Environmental Quality (DEQ). Evaluation of Climate, Energy, and Soils Impacts of Selected Food Discards Management Systems; DEQ: Portland, OR, 2014. (9) Amundson, R.; Berhe, A. A.; Hopmans, J. W.; Olson, C.; Sztein, A. S.; Sparks, D. L. Soil and human security in the 21st century. Science 2015, 348 (6235), 647−653. (10) Brown, S.; Cotton, M. Changes in soil properties and carbon content following compost application: results of on-farm sampling. Compost Sci. Util. 2011, 19, 87−96. (11) Brown, S.; Kurtz, K.; Bary, A.; Cogger, C. Long-term effects of organic amendments on soil carbon storage and physical properties. Environ. Sci. Technol. 2011, 45 (17), 7451−7458. (12) Brown, S.; Carpenter, A.; Beecher, N. Calculator tool for determining greenhouse gas emissions for biosolids processing and end use. Environ. Sci. Technol. 2010, 44, 9509−9515. (13) Clothier, B.; Hall, A.; Deurer, M.; Green, S.; Mackay, A. Soil ecosystem services: sustaining returns on investment into natural capital. Proceedings of the OECD Workshop on Sustaining Soil Productivity in Response to Global Climate Change-Science, Policy and Ethics; Wiley Blackwell: Hoboken, NJ, 2009. (14) Costanza, R.; d’Arge, R.; de Groot, R.; Farber, S.; Grasso, M.; Hannon, B.; Limburg, K.; Naeem, S.; O’Neill, R. V.; Paruelo, J.; Raskin, R. G.; Sutton, P.; van den Belt, M. The value of the world’s ecosystem services and natural capital. Nature 1997, 387, 253−60. (15) Doran, J. Soil health and global sustainability: translating science into practice. Agric., Ecosyst. Environ. 2002, 88, 119−27. (16) Robinson, D. A.; Jackson, B. M.; Clothier, B. E.; Dominati, E. J.; Marchant, S. C.; Cooper, D. M.; Bristow, K. L. Advances in soil ecosystem services: concepts, models, and applications for earth system life support. Vadose Zone J., 2013, 12(4), 10.2136/ vzj2013.01.0027. (17) Spargo, J. T.; Alley, M. M.; Follett, R. F.; Wallace, J. V. Soil carbon sequestration with continuous no-till management of grain cropping systems in the Virginia coastal plain. Soil Tillage Res. 2008, 100, 133−140. (18) Stewart, C. E.; Follett, R. F.; Wallace, J.; Pruessner, E. G. Impact of biosolids and tillage on soil organic matter fractions: implications of carbon saturation for conservation management in the Virginia Coastal Plain. Soil Sci. Soc. Am. J. 2012, 76, 1257−1267. (19) Wuest, S. B.; Reardon, C. L. Surface and root inputs produce different carbon/phosphorus ratios in soil. Soil Sci. Soc. Am. J. 2016, 80, 463−471. (20) Montgomery, D. R. Soil erosion and agricultural sustainability. Proc. Natl. Acad. Sci. U. S. A. 2007, 104 (33), 13268−13272.

(21) Brandão, M.; Heath, G. A.; Cooper, J. S. What can metaanalyses tell us about the reliability of life cycle assessment for decision support? J. Ind. Ecol. 2012, 16 (s1), S3−S7. (22) Heath, G. A.; Mann, M. K. Background and Reflections on the Life Cycle Assessment Harmonization Project. J. Ind. Ecol. 2012, 16 (s1), S8−S11. (23) Zumsteg, J. M.; Cooper, J. S.; Noon, M. S. Systematic Review Checklist. J. Ind. Ecol. 2012, 16 (s1), S12−S21. (24) Sheltair Group. Environmental Life Cycle Assessment of Solid Waste Management: Evaluation of Two Waste Disposal Scenarios for the Metro Region; Sheltair Group: Vancouver, BC, 2008. (25) Morris, J. Bury or burn North American MSW? LCAs provide answers for climate impacts & carbon neutral power potential. Environ. Sci. Technol. 2010, 44 (20), 7944−7949. (26) Tian, G.; Granato, T. C.; Cox, A. E.; Pietz, R. I.; Carlson, C. R.; Abedin, Z. Soil carbon sequestration resulting from long-term application of bio- solids for land reclamation. J. Environ. Qual. 2009, 38, 61−74. (27) Trlica, A.; Brown, S. Greenhouse gas emissions and the interrelation of urban and forest sectors in reclaiming one hectare of land in the Pacific Northwest. Environ. Sci. Technol. 2013, 47, 7250− 7259. (28) Brown, S. Greenhouse gas accounting for landfill diversion of food scraps and yard waste. Compost Sci. Util. 2016, 24 (1), 11−19. (29) Hummel, R. L.; Cogger, C.; Bary, A.; Riley, R. Marigold and pepper growth in container substrates made from biosolids composted with carbon-rich organic wastes. Hor Technol. 2014, 24 (3), 325−333. (30) U. S. Environmental Protection Agency. Emissions & Generation Resource Integrated Database (eGRID) data for 2010. https://www.epa.gov/energy/egrid (accessed October 13, 2016). (31) Andersen, J. K.; Christensen, T. H.; Scheutz, C. Substitution of peat, fertilizer and manure by compost in hobby gardening: User surveys and case studies. Waste Manage. 2010, 30 (12), 2483−2489. (32) North East Biosolids and Residuals Association (NEBRA). A National Biosolids Regulation, Quality, End Use & Disposal Survey - Final Report, July 2007; NEBRA: Tamworth, NH, 2007. (33) Recycled Organics Unit. Life Cycle Inventory and Life Cycle Assessment for Windrow Composting Systems; University of New South Wales: Sydney, Australia, 2006. (34) Parry, D. L. Sustainable Food Waste Evaluation; WERF: Alexandria, VA, 2012. (35) U. S. Environmental Protection Agency. Turning Food Waste into Energy at the East Bay Municipal Utility District (EBMUD). https://www3.epa.gov/region09/waste/features/foodtoenergy/ wastewater.html (Accessed October 13, 2016). (36) Gaillard, R.; Duval, B. D.; Osterholz, W. R.; Kucharik, C. J. Simulated effects of soil texture on nitrous oxide emissions factors from corn and soybean agroecosystems in Wisconsin. J. Environ. Qual 2016, 45 (5), 1540−1548. (37) Metcalf & Eddy, Inc. Wastewater Engineering: Treatment & Reuse, 4th ed.; McGraw-Hill: New York, 2003. (38) Hodge, K. L.; Levis, J. W.; DeCarolis, J. F.; Barlaz, M. A. Systematic evaluation of industrial, commercial, and institutional food waste management strategies in the United States. Environ. Sci. Technol. 2016, 50, 8444−8452.

H

DOI: 10.1021/acs.est.6b06115 Environ. Sci. Technol. XXXX, XXX, XXX−XXX