Quantitative evaluation of the reduction in the environmental burden during the life cycle of soy sauce

Abstract

This study focused on investigating the life cycle of soy sauce via the life cycle assessment (LCA) methodology to quantify its environmental burden. MiLCA software was used to assess the environmental impact. The production of chemicals during soybean cultivation constituted an environmental burden hotspot during the life cycle of soy sauce, as determined via the characterized and single score results. The potential for environmental burden reduction in the three impact categories was demonstrated on the basis of the introduction of organic cultivation. The sensitivity analysis results revealed that the yield of organic soybeans affected the reduction in environmental burden during the life cycle of organic soy sauce. The results showed that substituting organic soybeans as a material for soy sauce could reduce the environmental burden during the life cycle of soy sauce. Future studies, such as assessments of environmental impacts using different impact categories, should be conducted in the future.

Introduction

Recently, food shortages due to climate change and population growth have become a concern in countries worldwide, and with the adoption of the Sustainable Development Goals (SDGs) at the United Nations Summit in September 2015, it has become even more important for countries to work toward a society that balances economic, social, and environmental goals. Moreover, at the 28th Conference of the Parties to the United Nations Framework Convention on Climate Change (COP28) held in December 2023, an outcome document focusing on the shift away from fossil fuels in the energy system was adopted. Thus, in Japan, the MIDORI strategy for sustainable food systems was formulated in 2021, with the goal of establishing a sustainable food system, including the promotion of technological innovations such as carbon-neutral systems to reduce Greenhouse gases (GHGs)ⁱ⁾. Companies are required to engage in sustainable industrial activities such as environmental, social, and governance (ESG) management, and the food industry is no exception to this growing momentum.

Since Japanese food was registered as intangible cultural heritage by the UNESCO in 2013, it has become more familiar worldwide. Soy sauce is one of the most popular umami foods worldwide, and both exports to overseas and production outside the country by Japanese companies are increasingⁱⁱ⁾. The soy sauce market is projected to reach a size of $51.2 billion by 2030, indicating that the soy sauce market is expanding globallyⁱⁱⁱ⁾. There are very few studies on soy sauce that focus on assessing in detail the environmental burden needed to achieve the SDGs. Although Sharaai and Ismail (2020) conducted a life cycle assessment (LCA) of soy sauce, only the water footprint was employed as an evaluation index, whereas multiple impact categories have yet to be assessed. Therefore, this study aimed to evaluate the environmental sustainability of soy sauce and analyze scenarios for reducing its environmental burden. In addition, we adopted the life cycle assessment (LCA) method, which is an environmental burden assessment method, to quantitatively evaluate environmental burden hotspots and to investigate the emission trend within each impact category (Itsubo et al., 2007). The LCA method covers all processes from resource acquisition to disposal of the target product. By analyzing the entire life cycle from an overview perspective, this method is highly valuable as a judgment tool for developing effective strategies to reduce CO₂ emissions. It is possible to evaluate multiple impact categories (e.g., impacts on acidification and ozone layer depletion), in addition to greenhouse gases (GHGs). However, the MIDORI strategy for sustainable food systems aims to increase the area devoted to organic farming to 25 % (1 million ha) by 2050. Compared with conventional farming, organic farming could reduce the environmental burden because it does not involve the use of chemically synthesized fertilizers or pesticides. In this study, we constructed a scenario in which organic cultivation is introduced into the production of soybeans, which serve as the raw material for soy sauce, and evaluated the reduction in the environmental burden compared with that of soy sauce prepared from conventional soybeans.

Therefore, the objectives of this study were as follows: (1) to quantify the life cycle environmental burden of soy sauce via the LCA method; (2) to identify hotspots during the life cycle of soy sauce; and (3) to evaluate the potential for environmental burden reduction of the introduction of organic soybean cultivation.

Materials and Methods

System boundary and functional unit　　The system boundary is shown in Fig. 1. The life cycle of soy sauce can be divided into five stages: the raw material production stage, the soy sauce production stage, the package production stage, the transportation stage, and the disposal stage. Information on each of the stages was obtained from interviews with soy sauce store employees and from the literature. In this study, all the evaluation objects were produced and distributed domestically. Fixed assets such as agricultural machinery and facilities, transportation at the soy sauce package production stage, and storage and sales at the transportation stage were excluded from the calculation because the situation differs by production region and plant scale and because of the difficulty in collecting data. All transportation in each process was assumed to entail one-way transportation, and losses due to damage during transportation and processing were not accounted for. The purpose of the functional unit is to provide a reference unit to which the inventory data are normalized. The definition of functional unit depends on the environmental impact category and aims of the investigation and is often based on the mass of the product under study (Roy et al., 2009). Therefore, the environmental burden of 1 t of soy sauce was defined as the functional unit. Since the mass of the final product (including packages) during the life cycle was 5.2 t, the calculated life cycle environmental burden was converted to t of product on the basis of this value.

Fig. 1

System boundary.

Inventory analysis　　Primary data was obtained via interviews with employees from the Asanuma soy sauce shop and Yamazaki Ironworks Company in Iwate Prefecture. Moreover, secondary data were obtained from inventory data in the life cycle inventory (LCI) database IDEA ver. 3.1 (National Institute of Advanced Industrial Science and Technology (AIST), the Japan Environmental Management Association for Industry (JEMAI)) and from soy sauce statistical data for 2022^iv). Details of inventory analysis are provided below.

1. (i) Raw material production stage　　An inventory for the raw material production stage is provided in Table 1. The ingredients of soy sauce are defatted soybeans and wheat. On the basis of interviews with Asanuma soy sauce shop employees, the input amounts of defatted soybeans and wheat were set to 0.9 t each. In this study, the raw materials were assumed to be grown in the Tokachi region of Hokkaido, which is a major production area for soybeans and wheat in Japan. The materials were transported by coastal vessels between the ports of Tomakomai and Hachinohe, which are the main import and export ports for major agricultural products in Japan, over a distance of 423 km. The distance between Hachinohe port and the manufacturing plant in Morioka city was assumed as 140 km for 2-t trucks, and the loading rate was set to 58 %, which is the default value in the inventory database. Each transportation distance was obtained from a map application (Google Map). The environmental burden of the transport of soybean and wheat fertilizers was small and was therefore excluded from our assessment.
   Table 1. Material and energy inputs in the raw material production of soy sauce stage.

   Process	Input
   Defatted soybeans production	0.9 t
   Wheat production	0.9 t
   Shipping (Tomakomai port to Hachinohe port) *	476.0 tkm
   Transportation (Hachinohe port to Morioka city) *	157.6 tkm

   *  The loading rate in IDEA ver. 3.1 was set to the average value of 58 %.

2. (ii) Soy sauce production stage　　The inventory data for the processing stage are presented in Table 2, and the breakdown of the associated power consumption is provided in Table 3. This stage can be divided into four stages: koji generation, preparation and pressing, fire quenching, and packaging. On the basis of the conducted interviews, the amount of brine used after koji generation was approximately 4 t, with a 23 % concentration. The fuel for the wheat roaster is heavy oil A. The fuel consumption and machine price were estimated at 3.5 million Japanese yen (JPY) on the basis of interviews with employees from Yamazaki Ironworks Co., Ltd. The number of filter cloths used for pressing was 240 (1 × 1 m), which are washed in a washing machine after each pressing and discarded every three years. Finally, the mass of the product was 4.9 t. The residues generated after soy sauce pressing and heating were treated as plant residues derived from soybeans and they accounted for 10 % of the mass of soy sauce (after pressing: 0.61 t, after fire quenching: 0.55 t). The total floor area of the manufacturing plant of the Asanuma soy sauce shop was 1640.3 m², as measured via a map application (Google Earth).
   Table 2. Material and energy inputs in the soy sauce production stage.

   Process	Input
   Koji generation
   Salt	1.2 t
   Water	3080.0 kg
   Soybean steamer	9.3 × 10−4 unit
   Wheat roasting machine	3267.6 JPY
   Wheat crusher	9.3 × 10−4 unit
   Pressure cooker	1.2 × 10−3 unit
   Koji molding equipment	1.2 × 10−3 unit
   Power consumption	489.2 kWh
   Heavy oil	42.6 L
   Preparation and pressing
   Bloewr	2.2 × 10−3 unit
   Filter press	1.2 × 10−3 unit
   Filter cloth	3.6 m2
   Washing machine	3.7 × 10−3 unit
   Water	1572.0 L
   Effluent treatment	1572.0 L
   Tank of fermented soy sauce mash	13433.9 JPY
   Power consumption	316.6 kWh
   Fire quenching
   Continuous sterilization of liquid	3361.0 JPY
   Heating tank	1.24 × 10−3 unit
   Power consumption	0.11 kWh
   Packaging
   Air conditioner	1.9 × 10−3 unit
   Filling machine	1.1 × 10−3 unit
   Power consumption	72.8 kWh
   Manufacturing plant	0.5 m2
   Recycling into livestock feed	99.8 %*

   *  After soy sauce production, 99.8 % of the plant residues were recycled and used as livestock feed.

   Table 3. Breakdown of proportional power consumption of machinery used in soy sauce production stage.

   	Machinery	Usage time (h)	Proportion (%)	Power consumption (kWh)
   Koji molding	Wheat crusher	1.12	0.8	7.2
   	Soybean steamer	3.0	2.2	19.2
   	Pressure cooker	0.5	0.4	3.2
   	Koji generation equipment	72.0	52.3	459.6
   Preparation and pressing	Blower	9	6.5	57.5
   	Filter press	30.6	22.2	195.4
   	Washing machine	10.0	7.3	63.9
   Pasteurization	Continuous sterilization of liquid	0.02	0.01	0.1
   Packaging	Air conditioner	5.7	4.1	36.4
   	Filling machine	5.7	4.1	36.4

For the machines listed in each inventory table, the number of machines per t of product manufactured was determined as follows:

  

In addition, the value of 89.3 denotes the annual production times of the factory producing soy sauce and is an approximate estimate derived from the annual production volume of 366 kL (439 t). The same calculation method was adopted for the price (JPY) and area (m²). The amount of electricity consumption (kWh) was calculated on the basis of the annual electricity bill of 2 million JPY for the manufacturing plant of the Asanuma soy sauce shop and the average electricity bill of 25.5 JPY/kWh from the average of energy consumption for June 2020 to April 2022 for Tohoku Electric Power Co., which resulted in an annual consumption of approximately 78431 kWh. This value was divided by the annual production times, 89.3, and the energy consumption per cycle was 878.7 kWh. A breakdown of the electricity consumed by each machine and the lifetime of each machine and facility are listed in Table 3. The electricity consumption distribution was based on the value of 878.7 kWh consumed over the entire life cycle, which was proportionally distributed on the basis of the number of running hours of each machine. The washing machine was assumed to provide a capacity of 10.0 kg (IAW-T1001, IRIS OHYAMA Inc., Miyagi, Japan) and was operated for a total of 12 times per filter cloth washing cycle. The filling machine (LS3000DII, Kyowa Automatic Machinery Co., Ltd., Kanagawa, Japan) was assumed to require 5.7 h to finish filling, on the basis of a throughput of 400 bottles/h. The lifetime was determined on the basis of the lifetime of the major depreciable assets^v).

According to Japan Soy Sauce Association (2011), it has been reported that almost 99.8 % of soy sauce waste after pressing is recycled as livestock feed or fertilizer, so the recycling into livestock feed process have also been considered in this study as a 99.8 % substitution rate.

1. (iii) Package production stage　　The inventory data for this stage are listed in Table 4, and an overview of the product packaging during transportation is shown in Fig. 2. On the basis of the actual shipping configuration of the manufacturing plant, the final product was packaged in 1.8-L polyethylene terephthalate (PET) bottle containers (48 g/bottle) and boxed in cardboard boxes (0.4 kg/piece, 33.5 × 33.0 × 22.0 cm), with 6 bottles in each. The most common container type for soy sauce in Japan is the 1- to 1.8-L PET bottle container^iv). The weight of each container was measured on a digital platform scale (FW-100k, A&D Co., Ltd., Tokyo, Japan). Finally, the mass of the packages was 0.3 t.
   Table 4. Material and energy inputs in the package production stage.

   Process	Input
   Manufacture of PET bottles	0.1 t
   Manufacture of cardboard boxes	0.2 t
   Transportation*	14.2 tkm

   *  This is the total transport distance from the plastic bottle and cardboard box plants to the Asanuma soy sauce shop.

   

   Fig. 2

   Overview of the product packaging during transportation.

2. (iv) Transportation stage　　The inventory data for this stage are provided in Table 5. It was assumed that the goods were transported to Tokyo, which was the model area in this analysis. The products were delivered to a major retailer in Tokyo via a distribution center in Saitama Prefecture. The distance was set to 482 km (from the manufacturing plant to the distribution center) and 68 km (from the distribution center to the retail store) on the basis of Google Maps. Since the mass of the product, including packages in IDEA ver. 3.1, was approximately 5.2 t, the input value was set to 10 t for truck transportation with a loading ratio of 50 %. As the unit is tkm, the value was calculated by multiplying the mass of the product (5.2 t) by the transportation distance.
   Table 5. Material and energy inputs in the transportation stage.

   Process	Input
   Transportation (Asanuma soy sauce shop to distribution center)	2499.5 tkm
   Transportation (Distribution center – retail)	352.6 tkm

3. (v) Disposal stage　　The inventory data for the disposal stage are provided in Table 6. Disposal at this stage refers to the incineration and landfill disposal of waste generated after fire quenching process. The waste transportation process at this stage was excluded from the calculation because of the uncertainty involved in identifying the specific process.
   Table 6. Disposal stage of residues after fire quenching process

   Process	Input
   Industrial waste disposal of plant residues	0.6 t

Impact categories and impact assessment methodology　　MiLCA ver.3.1. (Sustainable Management Promotion Organization (SuMPO), Japan), which is one of the most common software programs in Japan for LCA, was employed to calculate the environmental burden. The life cycle impact assessment method based on endpoint modeling (LIME2) was adopted in this study (Itsubo and Inaba, 2010). The calculated outputs are expressed as characterized results and single score results. The characterization and weighting indices of the single score results were determined via the LIME2 method. The results of the environmental loads are shown as characterized results and single score results and are calculated with the following equation (Itsubo et al., 2007).

  

where CI_i is the characterized result for impact category CF_i,s is an index of characterization (contribution of substance “s” to the environmental loads for impact category “i”). LCI_s are inventory data (amounts of substance “s”).

  

where SI_e is the single score result. IF_s,i,e denotes an index of weighting for the single score results (contribution of substance “s” to the environmental loads for impact category “i” and endpoint “e”). The indices of characterization and weighing for the single score results were followed by LIME2. The unit of the single score result is expressed as the social cost based on economic value (JPY) in LIME2; social costs are the costs associated with damage to health and the environment (e.g., the extermination of species) related to the production of a product or provision of service (Itsubo and Inaba, 2010). The impact categories investigated in this study were as follows: climate change (CC), photochemical oxidation (PO), resource consumption (RC), acidification (AC), waste (WA), ozone depletion (OD), eutrophication (EU), urban air pollution (UAP), and transportation noise (NO) (Table 7). Since the characterized results exhibit different units for each impact category, it is not appropriate to simply compare impact categories on the basis of the corresponding numerical values. Therefore, since it is common to express the amount of environmental burden per functional unit given by each process as a percentage and to express the results as a contribution (%), the characterized results were also expressed as the contribution of each process.

Table 7. Impact categories.

Impact category		Unit
Climate change	CC	kg-CO2 eq
Photochemical oxidation	PO	kg-C2H4 eq
Resource consumption	RC	kg-Sb eq
Acidification	AC	kg-SO2 eq
Waste	WA	m3
Ozone depletion	OD	kg-CFC11 eq
Eutrophication	EU	kg-PO43- eq
Urban air pollution	UAP	kg-SO2 eq
Noise	NO	J

Scenario analysis for reducing the environmental burden during the life cycle of soy sauce　　We examined scenarios for reducing the environmental burden of introducing organic soybean cultivation during the life cycle of soy sauce. In this study, organic cultivation involved the application of organic fertilizers instead of chemically synthesized fertilizers. The definition of organic cultivation according to the Law Concerning the Promotion of Organic Agriculture in Japan^vi) is as follows: (1) No chemically synthesized fertilizers or pesticides are used. (2) No genetic modification technology is employed. (3) The burden on the environment derived from agricultural production is reduced as much as possible. Organic fertilizers are animal and vegetable fertilizers such as fish meal, animal dreg powder, bone meal, and vegetable oil cake (Mihara et al., 2007). The total amount of chemical fertilizers (nitrogenous, phosphate, and potassium) included in the inventory data for soybean cultivation during conventional soy sauce was 153 kg. Thus, the evaluation was conducted under the assumption that an equivalent amount of organic fertilizer could be substituted for chemical fertilizers. The inputs for organic soybean cultivation are listed in Table 8. The inputs of organic fertilizers and a breakdown of the amount of each component are provided in Table 9. According to IDEA database, the amount of input for each fertilizer (including basal and additional fertilizer), was defined from the ratio of the production of each fertilizer in the data from Association of Agriculture & Forestry Statistics (AAFS) (2008), the input for chemical fertilizer in Table 9 have therefore been converted to values per 1 t of soy sauce. For organic fertilizers, the contents of N, K and P in organic fertilizers in the IDEA database were not identified, therefore, we calculated average values based on the nutrient contents of organic fertilizers published by Hokkaido government^vii) (2020).

Table 8. Material and energy inputs per mass-based functional unit in the organic soybean cultivation scenario in IDEA ver.3.1.

Process	Input
Agricultural water use	72.6 m3
Water supply	2.0 × 10−2 m3
Power consumption	17.4 kWh
Effluent treatment	2.0 × 10−2 m3
Industrial waste disposal of scrap metal	2.4 × 10−3 kg
Industrial waste disposal of waste plastics	4.5 × 10−1 kg
Combustion of heavy oil	1.4 × 10−2 L
Agricultural water use	159.6 m3
Combustion of petrol	5.5 L
Paraffin combustion	0.5 L
Combustion of paraffin	0.3 kg
Industrial waste disposal of waste oil	1.9 × 10−5 kg
Industrial waste treatment of glass, concrete and ceramic waste	2.9 × 10−6 kg
Combustion of lite oil	21.1 L
Soft plastic films (less than 0.2 mm thick)	2.3 kg
Soft plastic films (less than 0.2 mm thick)	0.1 kg
Soft plastic films for packaging (less than 0.2 mm thick)	8.7 × 10−3 kg
Organic fertilizers	481.0 kg

Table 9. The inputs for organic fertilizers and a breakdown of the amount of each component using a mass-based functional unit (t-soy sauce).

Data	Input	N	P	K	Source
Chemical fertilizers	29.5	4.7	12.7	12.0	IDEA ver.3.1 Database AAFS, 2008
Organic fertilizers	209.3	10.7	6.9	1.3	IDEA ver.3.1 Database Hokkaido government, 2020

Each value is an average value of several years.

Although several studies have revealed that introducing organic cultivation can effectively reduce the environmental burden compared with that of conventional cultivation, other studies have noted that the burden per product weight can exceed that of conventional cultivation because of the lower yields (Hayashi, 2008; Knudsen et al., 2010; Venkat, 2012; Schrama et al., 2018; Lee and Choe, 2019). Thus, the environmental burden of organic cultivation is not necessarily lower than that of conventional cultivation in LCA of agriculture. The number of chemicals used in agriculture is relatively large. Therefore, it is difficult to properly measure their life cycle impacts. Since obtaining estimation results from literature data could introduce uncertainty into the study results, the sensitivity analysis application is crucial for ensuring the reliability of the data (Longo et al., 2017). Hence, in this study, we conducted the sensitivity analysis to evaluate the impact of changes in the soybean yield caused by organic cultivation on the environmental burden during the life cycle of soy sauce. The hotspot analysis was conducted in accordance with the Product Environmental Footprint Category Rules Guidance Version 6.3 (European Commission 2018). Thus, a hotspot in this study was determined based on the definition of the European Commission. The most relevant impact categories were all impact categories that cumulatively contributed to at least 80 % of the total environmental impact.

Results and Discussion

Hotspots of environmental burden　　Fig. 3 shows the single score results for the life cycle of soy sauce. The impact category with the greatest environmental burden was eutrophication (EU) (76.1 %), followed by climate change (CC) (10.2 %) and urban air pollution (UAP) (8.0 %). Given that these three impact categories accounted for 94.3 % of the total impact categories, we considered EU, CC, and UAP the three most relevant impact categories in our study. According to the EC guideline^viii), at least three relevant impact categories shall be considered to identify the most relevant impact categories. The contribution of each stage to the various impact categories per 1 t of soy sauce is listed in Table 10. According to Orikasa et al. (2023), in the LCA for apple exported from Japan to Taiwan, the cultivation stage accounted for 99 % of the impact category of marine eutrophication. Additionally, several studies reported that the highest environmental burden during some fruit juice production can be attributed to the raw material production stage and juice production stage (Dwivedi et al., 2012; Aganovic et al., 2016). A similar trend was observed in the results of these studies. In this study, the concentration of the environmental burden at the production stage was confirmed in five of the nine impact categories (CC, PO, WA, EU, and UAP). The Japan Soy Sauce Association (2020) reported that the recycling rate of soy sauce cake, which is generally produced after mash pressing, reaches over 99 %^ix), ^x). Although the soy sauce production stage in this study contributed to - 3 % of the environmental burden in the EU impact category, this finding may be due to the reduction in the environmental burden associated with the soy sauce cake production for recycling into livestock feed. In addition, the main factor contributing to the environmental burden at the raw material production stage in each impact category was the production of defatted soybeans, which serve as the raw material for soy sauce (Fig. 4). As shown in Table 1, raw material production stage consists of 4 stages: defatted soybeans production, wheat production, shipping (Tomakomai port to Hachinohe port), transportation (Hachinohe port to Morioka city). Furthermore, according to the characterized results from MiLCA software, 93 % of the environmental burden of defatted soybeans was attributed to soybean cultivation, so characterization was conducted on the process (Fig. 5). In the EU impact category, the production of pesticides, nitrogenous, and fertilizers accounted for 97 % of the total burden. In the CC and UAP impact categories, the production of phosphate and nitrogenous fertilizers contributed significantly to the total burden. These results suggest that the cultivation process of soybeans for soy sauce is an important component of the life cycle environmental burden. Several previous studies have indicated that fertilizers and pesticides are key factors of the environmental burden contributing to the cultivation process of agricultural products (Knudsen et al., 2010; Lee and Choe, 2019; Orikasa et al., 2023), with similar trends.

Fig. 3

Single score results of life cycle of soy sauce using a mass-based functional unit (t-soy sauce).

CC; climate change, PO; photochemical oxidation, RC; resource consumption, AC; acidification, WA; waste, OD; ozone depletion, EU; eutrophication, UAP; urban air pollution, NO; noise.

Table 10. Contribution of each stage to the impact category per mass-based functional unit (t-soy sauce).

Impact categories	Raw material production	Soy sauce production	Package production	Transportation	Disposal
CC	47.8	19.1	19.5	13.2	0.4
PO	58.3	12.5	26.0	3.2	0.0
RC	31.5	14.2	16.4	37.9	0.0
AC	19.4	8.3	26.1	0.0	46.2
WA	69.6	12.7	15.0	2.8	0.0
OD	30.5	53.9	15.6	0.0	0.0
EU	102.8	-2.9	0.1	0.0	0.0
UAP	32.2	19.8	21.4	26.5	0.0
NO	3.5	0.0	0.3	96.2	0.0

Fig. 4

Characterized results of the CC, EU and UAP of the raw material production stage using a mass-based functional unit (t-soy sauce).

Shipping refers to the process from Tokachi region in Hokkaido to Hachinohe port, and transportation refers to the process from Hachinohe port in Aomori to Morioka city in Iwate prefecture.

Fig. 5

Characterized results of the CC, EU and UAP of the soybean cultivation process in the raw material production stage using a mass-based functional unit (t-soy sauce).

CC; climate change, EU; eutrophication, UAP; urban air pollution.

Scenario analysis for reducing the environmental burden of soy sauce　　The environmental burden hotspot during the life cycle of soy sauce is the production of pesticides, fungicides, and fertilizers applied in the soybean cultivation process. The introduction of organic cultivation is one way to reduce the use of these chemicals. Therefore, we examined a scenario in which organic cultivation was introduced as a method to reduce the environmental burden at the raw material production stage and compared it with that in the conventional soy sauce process. The overall reduction in the life cycle environmental burden of organic soy sauce reached 60.0 %, as indicated by the single score results (Fig. 6). The results of the conventional soy sauce are the sum of the single score results in Fig. 3. In each impact category, the reduction was 71.3 % (EU), 31.4 % (CC), and 16.8 % (UAP) from the characterized results (Table 11). Hokazono et al. (2011) reported that the environmental burden of organic soybean cultivation using organic fertilizers was lower than that of conventional soybean cultivation in all impact categories, including EU and CC. In addition, Nemecek et al. (2011) reported that organic cultivation of arable crops reduced CO₂, CH₄, and NH₃ emissions originating from N₂O and phosphate fertilizers, as well as the global warming potential per kg of dry matter yield. Furthermore, a high correlation between N₂O application and NO_X emissions has been suggested (Bouwman et al., 2002; Lu et al., 2006). Similar to previous studies, the reduction in NH₃ and NO_X generation due to the decreased production and application of N₂O and phosphate fertilizers during organic cultivation in this study may have contributed significantly to the reduction rate in each impact category.

Fig. 6

Single score results of conventional soy sauce and organic soy sauce using a mass-based functional unit (t-soy sauce) in the whole life cycle.

The conventional results are the sum of the single score results in Fig. 3.

The sensitivity analysis results are shown in Fig. 7. The advantage of organic soy sauce was maintained despite the change in soybean yields, and this trend was also observed in the EU impact category. Moreover, when the soybean yield decreased to less than 69 % for CC and 84 % for UAP, the environmental burden of organic soy sauce production exceeded that of conventional soy sauce production. Hokazono et al. (2011) estimated that the advantage of organic soybeans is maintained even if the yield decreases to 80 % of the conventional level. Therefore, the obtained result was considered reasonable because of the similar trends in previous studies (Hokazono et al., 2010; Nemecek et al., 2011; Venkat, 2012).

Fig. 7

Sensitivity analysis of life cycle environmental burden of organic soy sauce associated with changes in soybean yield using a mass-based functional unit (t-soy sauce).

(a) represents the entire life cycle of soy sauce, (b) represents the EU, (c) represents the CC, and (d) represents the UAP. The straight line shows the original value of the environmental burdens of conventional soy sauce. The yield of conventional soybean was 170.9 kg/10 a. The environmental burden of conventional soy sauce was the original value calculated in the yield of 170.9 kg/10 a.

Limitations of this study　　In this study, we focused on the soybean cultivation process, which is an environmental burden hotspot in the life cycle of soy sauce. However, there is a range of input and N, P and K components for chemical and organic fertilizers which also differ due to farming methods (Onodera and Nakamoto, 2007; Knudsen et al., 2010; Tanifuji et al., 2018; Schrama et al., 2018). Future studies should also consider the consistency of input values for each component. It is also necessary to evaluate the effect of adopting organic cultivation on the wheat cultivation process, which exhibits the second highest environmental burden. Unlike wheat, which has been reported to be less effective under organic cultivation^xi), there are reports that the best way to reduce nitrogenous fertilizer application is to adopt organic cultivation. (Nemecek et al., 2011). Several studies suggest that more than half of food losses could occur at the distribution and consumption stages in middle- and high-income countries (Kummu et al., 2012; Gustavsson et al., 2011; Wunderlich and Martinez, 2018). Therefore, losses during transportation and consumption, which were not considered in this study, should also be examined in the future.

In addition, it is necessary to consider the impact of different functional units on the LCA results. The functional unit, whether a product-based (t) or area-based (ha) unit, may reverse the relative dominance relationship between conventional and organic farming (Hayashi, 2008; McBride and Greene, 2009). Longo et al. (2017) reported that the environmental burden of growing organic apples was lower per area and year, whereas organic apples exhibited a greater burden per product than did conventional apples on the basis of a comparison of 34 LCA cases for organic and conventional agricultural products by Meier et al. (2015). Thus, the LCA analysis results depend on the functional unit, and future studies should aim to clarify the impact of different functional units on the interpretation of the obtained LCA results. In addition, MiLCA assumes that products are domestically produced in Japan, so careful attention should be paid when assuming a mixture of imported products because imported chemical fertilizers tend to have higher CO₂ emissions per kg than domestic ones, if imported chemical fertilizers are used (Hayashi et al., 2012).

Conclusions

We performed LCA of soy sauce and evaluated environmental burden hotspots. The highest burden could be attributed to soybean cultivation, with the three main categories (EU, CC, UAP) accounting for 94 % of the total burden. The production of pesticides, fungicides and agrochemicals constituted a hotspot in the soybean cultivation process. The ability of organic cultivation (with the use of organic fertilizers) to reduce the life cycle environmental burden was therefore investigated, and the burden of organic soy sauce production was 17–31 % lower than that of conventional soy sauce production. In addition, in the case of soy sauce, new findings indicated that the soy sauce production stage did not cause significant exposure, contrary to the trends observed in several studies (Dwivedi et al., 2012; Aganovic et al., 2016). These results are extremely valuable for considering the production of more environmentally friendly soy sauces. However, considering the effect of the reduction in soybean yields due to the adoption of organic cultivation on the life cycle environmental burden by the sensitivity analysis, reductions were not observed in all impact categories. Future studies should not only focus on assessing the different impact categories but also focus on conducting a detailed multifaceted analysis involving different functional units and scales of production.

Acknowledgements　　This work was supported by JSPS KAKENHI, Grant Number JP23K23736 (Grant-in-Aid for Scientific Research(B)).

Conflict of interest　　There are no conflicts of interest to declare.

References

• AAFS. (2008). “Pocket Agriculture, Forestry and Fisheries Statistics (2008 edition).” MAFF, Tokyo.
•  Aganovic,  K.,  Smetana,  S.,  Grauwet,  T.,  Toepfl,  S.,  Mathys,  A.,  Van Loey,  A., and  Heinz,  V. (2017). Pilot scale thermal and alternative pasteurization of tomato and watermelon juice: an energy comparison and life cycle assessment. J. Clean. Prod., 141, 514–525. doi: 10.1016/j.jclepro.2016.09.015
•  Bouwman,  A.F. and  Boumans,  L.J.M. (2002). Emissions of N₂O and NO from fertilized fields: summary of available measurement data. Global Biogeochem. Cycles, 16, 1058. doi:10.1029/2001GB001811
•  Dwivedi,  P.,  Spreen,  T., and  Goodrich,  R.M. (2012). Global warming impact of Florida’s not-from-concentrate (NFC) orange juice. Agric. Syst., 108, 104–111. doi: 10.1016/j.agsy.2012.01.006
•  Gustavsson,  J.,  Cederberg,  C.,  Sonesson,  U.,  Van Otterdijk,  R., and  Meybeck,  A. (2011). “Global food losses and food waste – extent, causes and prevention.” FAO, Rome.
•  Hayashi,  K. (2008). Is organic farming environmentally superior? what agricultural LCA reveals. J. Life Cycle Assess. Jpn. (Nihon LCA Gakkaishi), 4, 112–118. doi: 10.3370/lca.4.112 (in Japanese)
•  Hayashi,  K.,  Makino,  N., and  Shobatake,  K. (2012). Influence of fertilizer inventories on LCA results of agricultural production systems. Presented at 7th Meeting of the Institute of Life Cycle Assessment, Noda, March 7–9, p. 154. (in Japanese)
•  Hokazono,  S.,  Sato,  M., and  Hayasi,  K. (2010). Sustainability assessment of organic rice farming. Kanto Tokai J. Farm Manage. (Kanto Tokai Nougyou Keiei Kenkyu), 100. 15–26. (in Japanese)
•  Hokazono,  S.,  Hayashi,  K., and  Sato,  M. (2011). Comparative life cycle assessment of organic and conventional soybean production in paddy fields under rotational cropping. Presented at 6th Meeting of the Institute of Life Cycle Assessment, Sendai, March 2–4, p. 250. (in Japanese)
•  Itsubo,  N.,  Tahara,  K., and  Narita,  N. (2007). “Introduction to LCA.” JEMAI. Tokyo.
•  Itsubo,  N. and  Inaba,  A. (2010). “Life-cycle impact assessment method based on endpoint modeling” JEMAI. Tokyo.
•  Knudsen,  M.T.,  Yu-Hui,  Q.,  Yan,  L., and  Halberg,  N. (2010). Environmental assessment of organic soybean (Glycine max.) imported from China to Denmark: a case study. J. Clean. Prod., 18, 1431–1439. doi:10.1016/j.jclepro.2010.05.022
•  Kummu,  M.,  de Moel,  H.,  Porkka,  M.,  Siebert,  S.,  Varis,  O., and  Ward,  P.J. (2012). Lost food, wasted resources: global food supply chain losses and their impacts on freshwater, cropland, and fertilizer use. Sci. Total Environ., 438, 477–489. doi:10.1016/j.scitotenv.2012.08.092
•  Lee,  K.S. and  Choe,  Y.C. (2019). Environmental performance of organic farming: evidence from Korean small-holder soybean production. J. Clean. Prod., 211, 742–748. doi: 10.1016/j.jclepro.2018.11.075
•  Longo,  S.,  Mistretta,  M.,  Guarino,  F., and  Cellura,  M. (2017). Life Cycle Assessment of organic and conventional apple supply chains in the North of Italy. J. Clean. Prod., 140, 654–663. doi: 10.1016/j.jclepro.2016.02.049
•  Lu,  Y.,  Huang,  Y.,  Zou,  J., and  Zheng,  X. (2006). An inventory of N₂O emissions from agriculture in China using precipitation-rectified emission factor and background emission. Chemosphere, 65, 1915–1924. doi:10.1016/j.chemosphere.2006.07.035
•  McBride,  W.D. and  Greene,  C. (2009). The profitability of organic soybean production. Renewable Agric. Food Syst., 24, 276–284. doi: 10.1017/S1742170509990147
•  Meier,  M.,  Stoessel,  F.,  Jungbluth,  N.,  Juraske,  R.,  Schader,  C., and  Stolze,  M. (2015). Environmental impacts of organic and conventional agricultural products – are the differences captured by life cycle assessment?. J. Environ. Manage., 149, 193–208. doi: 10.1016/j.jenvman.2014.10.006
•  Mihara,  M.,  Chen,  Y.,  Gopal,  T.,  Kaneko,  A.,  Okazawa,  H.,  Sang-Arun,  J.,  Sirwattananon,  L., and  Yamamoto,  N. (2007). “Sustainable agriculture with organic fertilizer.” ERECON, Tokyo, pp. 1–125.
•  Nemecek,  T.,  Dubois,  D.,  Huguenin-Elie,  O., and  Gaillard,  G. (2011). Life cycle assessment of Swiss farming systems: integrated and organic farming. Agric. Syst., 104, 217–232. doi:10.1016/j.agsy.2010.10.002
•  Onodera,  M. and  Nakamoto,  H. (2007). Open culture of vegetable crop without chemical fertilizer using organic fertilizers and compost in Hokkaido. Jap. J. Soil Sci. Plant Nutr., 78, 611–616. doi:10.20710/dojo.78.6_611 (in Japanese)
•  Orikasa,  T.,  Goosesns,  Y. and  Geeraerd  A. (2023). Life cycle assessment of apple exported from Japan to Taiwan and potential environmental impact abatement. Eng. Agric. Environ. Food., 16, 1–14. doi:10.37221/eaef.16.1_1
•  Roy,  P.,  Nei,  D.,  Orikasa,  T.,  Xu,  Q.,  Okadome,  H.,  Nakamura,  N., and  Shiina,  T. (2009). A review of life cycle assessment (LCA) on some food products. J. Food. Eng., 90, 1–10. doi:10.1016/j.jfoodeng.2008.06.016
•  Schrama,  M.,  de Haan,  J.,  Kroonen,  M.,  Verstegen,  H., and  Van der Putten  W.H. (2018). Crop yield gap and stability in organic and conventional farming systems. Agric. Ecosyst. Environ., 256, 123–130. doi:10.1016/j.agee.2017.12.023
•  Sharaai,  A.H. and  Ismail,  N.D. (2020). The water footprint of soy sauce. J. Wastes Biomass Manage., 2, 1–4. doi:10.26480/jwbm.01.2020.01.04
•  Tanifuji,  K.,  Fueki,  N.,  Yasuoka,  S.,  Miyake,  N.,  Onodera,  K.,  Shimizu,  M., and  Shirahata,  M. (2018). Characteristics of soil fertility trends, crop yields, pest outbreaks in organic field rotations and weed suppression measures. Hokunou, 85, 232–240. (in Japanese)
•  Venkat,  K. (2012). Comparison of twelve organic and conventional farming systems: a life cycle greenhouse gas emissions perspective. J. Sustainable Agric., 36, 620–649. doi: 10.1080/10440046.2012.672378
•  Wunderlich,  S. and  Martinez,  N. (2018). Conserving natural resources through food loss reduction: production and consumption stages of the food supply chain. Int. Soil Water Conserv. Res., 6, 331–339. doi:10.1016/j.iswcr.2018.06.002
• i)  https://www.maff.go.jp/j/kanbo/kankyo/seisaku/midori/#Midorisennryaku (Sept. 26, 2024)
• ii)  https://www.soysauce.or.jp/knowledge/data (Oct. 9, 2024)
• iii)  https://www.datamintelligence.com/research-report/soy-sauce-market (Oct. 9, 2024)
• iv)  https://view.officeapps.live.com/op/view.aspx?src=https%3A%2F%2Fwww.soysauce.or.jp%2Fwp-content%2Fuploads%2F2018%2F06%2Ftoukei2022.xlsx&wdOrigin=BROWSELINK (Oct. 9, 2024)
• v)  https://www.nta.go.jp/taxes/shiraberu/taxanswer/shotoku/pdf/2100_01.pdf (Oct. 9, 2024)
• vi)  https://www.maff.go.jp/j/seisan/kankyo/yuuki/ (Oct. 10, 2024)
• vii)  https://www.pref.hokkaido.lg.jp/fs/1/1/0/1/5/2/0/7/_/VI%E4%BB%98%E5%B8%AF%E8%B3%87%E6%96%99.pdf (Dec. 24, 2024)
• viii)  http://ec.europa.eu/environment/eussd/smgp/pdf/PEFCR_guidance_v6.3.pdf (Dec. 20, 2024)
• ix)  https://www.soysauce.or.jp/environment/disposal (Oct. 9, 2024)
• x)  https://www.maff.go.jp/j/council/seisaku/syokusan/recycle/haiki_h23_04/pdf/111202_data2-4.pdf (Oct. 8, 2024)
• xi)  https://www.nature.com/articles/nature.2012.10519 (Oct. 7, 2024)