Do the Different Exergy Accounting Methodologies Provide Consistent or Contradictory Results? A Case Study with the Portuguese Agricultural, Forestry and Fisheries Sector

Abstract

Three exergy accounting approaches are used to evaluate exergy efficiency: the Energy Resources Exergy Accounting (EREA), the Natural Resources’ Exergy Accounting (NREA) and the Extended Exergy Accounting (EEA). To test the consistency of the results provided by these methodologies, we apply them to evaluate the Portuguese agricultural, forestry and fisheries (AFF) sector, from 2000 to 2012. EREA shows an increase of 30% in the efficiency of the Portuguese AFF sector, while NREA and EEA methodologies increases of 27% and 43%, respectively. Although the results are consistent for the AFF sector, the same does not happen in the fisheries subsector, whose exergetic efficiency increases 14% with the EREA but decreases 42% with the NREA approach. The ratio of output to useful exergy reveals that a better thermodynamic efficiency is not translated into a higher energy service efficiency because fishing vessels have to travel more to get the same fish. Thus, results provided by the EREA and NREA approaches are complementary and both are needed to provide a realistic picture of exergy efficiency. On the other hand, results obtained by the EEA approach are dominated by capital and environmental impacts, revealing the disproportionality between material and immaterial inputs in this methodology.

1. Introduction

Sustainable development is the desirable long term goal of bringing humankind to a compromise between its welfare and the protection of our planet. According to Pezzey [1] “A temptation when writing on ‘defining sustainability’ is to try to distill, from the myriad debates, a single definition which commands the widest possible academic consent. However, several years spent in fitful pursuit of this goal have finally persuaded me that it is an alchemist’s dream, no more likely to be found than an elixir to prolong life indefinitely”. Still, the most consensual definition of sustainable development is provided by the Brundtland Report [2]: “Sustainable development is development that meets the needs of the present without compromising the ability of future generations to meet their own needs”.

The sustainability of society is critically dependent on the agricultural, forestry and fisheries (AFF) sector because all mass and energy flows that maintain our society come either directly from the environment or from the energy and mass surpluses obtained in this sector and the resource extraction and energy transformation sectors. The AFF sector provides all the food, feed and fiber and is also a major source of renewable energy resources. According to Committee on Twenty-First Century Systems Agriculture [3], the sustainability of the AFF sector should take into account the environmental, social and economic dimensions:

• Satisfy human food, feed, and fiber needs, and contribute to biofuels needs.
• Enhance environmental quality and the resource base.
• Sustain the economic viability of agriculture.
• Enhance the quality of life for farmers, farm workers, and society as a whole.

Several efforts have been made in measuring the sustainability of the AFF sector, but the focus is, in most cases, only on one of the pillars of sustainability. The complementary research of both social and economic sustainability dimensions with efficiency and environmental extraction is both scarce and complex since no common unit is easily found and accepted. Also, success in one dimension does not imply the same for the others. The economic viability of an agricultural sector is not directly correlated with the environmental efficiency of the same sector nor the wellbeing of farmers.

Exergy analysis is an assessment methodology that has been used to account for efficiency at several scales, which can be society as a whole, economic sectors or industries. The term efficiency usually reflects a narrow approach that considers only one type of output and input flows. However, an efficiency measure of the AFF sector that would take into account several types of flows, such as the extraction from the environment of renewable and non-renewable resources, undesired output flows such as waste and pollution and the capital and labour flows involved, could present useful insights about its sustainability. Further assessment of other aspects, such as the economic viability of agriculture or the quality of life of farmers, would still be needed for a complete evaluation of sustainability.

Exergy is the maximum amount of work that can be obtained from a system (a resource or a flow) when it is brought to equilibrium with the surrounding environment through a reversible process [4]. One of the key properties of exergy is that it allows the conversion of all inputs and outputs of a process into a common unit, allowing comparisons otherwise impossible.

The first methodology to quantify exergy flows was the energy resources’ exergy accounting (EREA) methodology that started with energy carriers only [5]. Later, natural resources were included [6] in the natural resources’ exergy accounting (NREA) methodology and more recently capital, labour and environmental externalities [7] in the extended exergy accounting (EEA) methodology.

EREA is an input assessment method that measures the exergy embedded (or intrinsic) in the energy carriers used by the transformation devices (inputs) and the useful exergy (outputs) (Figure 1). With EREA is possible to determine how well the system converts the final exergy to useful work from the available energy carriers and transformation devices; this knowledge can then be used to propose measurements to improve the efficiency of the system.

EREA quantifies the useful exergy that is available for each end use, but does not inform on how well that exergy service is being used. To illustrate, consider two similar gasoline vehicles on a highway, the first with only one passenger and the second with five people on board. When assessing both systems by EREA the useful exergy is the same since the energy carrier-end use pairs are the same (gasoline-transportation). However, when measuring the system output, in this case the number of passengers times distance, the performances are widely different.

NREA is an input-output assessment where all energy and matter that enters and leaves the system is accounted for (Figure 1). Contrary to EREA, no information on how each energy carrier is used is needed but more data is required to account for all fluxes across system boundaries. By accounting both the input and output fluxes, the NREA approach measures the overall physical performance of the transformation process.

Finally, EEA is similar to NREA but also introduces capital and labour as inputs and environmental impact as a virtual input, integrating the social, economic and environmental pillars of sustainability. Translating capital, labour and environmental impact externalities into an exergetic cost is a key feature of EEA as it reflects the real thermodynamic system’s efficiency and the usually hidden resource costs [8]. The production factors are no longer just energy resources (EREA) or energy and matter (NREA) but also immaterial flows which are essential to assess the system’s behaviour (Figure 1). The exergy of inputs is no longer the physical and chemical exergies but the cumulative exergy introduced by [9] in the cumulative exergy consumption methodology (CExC). The latter takes into account the sum of all the exergetic content of resources that are consumed in a production chain. Consequently, the main focus of EEA is to find the real exergy embodied into the output flows under study, adding not only all exergy necessary to bring all inputs to the specific space and time domains but also all labour, capital and environmental impacts.

First exergetic studies focusing exclusively on agriculture adopted an EREA approach and included conversion in agriculture machinery, such as tractors, and electric pumps. Agricultural sectors of Saudi Arabia [10], Turkey [11], Jordan [12], Iran [13] and finally Malaysia [14] were studied longitudinally (except for Jordan) in order to verify the evolution of efficiency. The efficiencies of devices vary widely among studies.

First exclusive NREA approaches to agriculture also included virtual environmental impact as an input to integrate pollutant emissions [15]. The input-output ratio was approximately 10 (ten times more exergy in inputs than outputs) but solar radiation was responsible for 90 to 95% of the inputs with an estimated photosynthetic efficiency of 4.5%. A combination of EEA and cumulative exergy extraction from the natural environment was proposed by [16] to measure the agricultural performance of 29 countries from 1990 to 2003. As important findings, the study highlighted the three types of resources that agricultural production most extracts from the natural environment: organic content in top soil, feed and water. Finally, it concluded that the exergy efficiency in the livestock sector is much lower than in the crop sector.

Some societal exergy studies that use the EREA approach, also included, in detail, the AFF sector [17,18]. These studies have allocated all energy carriers (e.g., diesel) used in the AFF sector to end-uses (e.g., mechanical drive). The AFF sectors of Sweden [19,20], Japan [21] and Italy [22] were assessed by the NREA methodology where the major energetic and material inputs were considered along with all the production from the sector. EEA studies also addressed the AFF sector. Efficiencies estimated by the EEA approach for the province of Siena, Italy [23]; the U.K. economy [24]; Chinese society [25,26,27,28,29], and the economy of Nova Scotia, Canada [30] were all below 100%. These EEA studies did not take into account solar radiation, but considered crops and wood as extractions from the environment and not as productions.

All three methodologies have been used to assess the performance of society [31] and of the AFF sector. However, analyses typically focus on results provided by only one methodology without justifying the use of that specific methodology in detriment of others. Also, there are major differences in the exergy inputs and outputs that are accounted for, across published NREA and EEA analyses of the agricultural sector. This is problematic if results provided by the three methodologies or even by the same methodology are inconsistent driving to different policy recommendations.

The present study evaluates the performance of the agricultural sector in Portugal, between 2000 and 2012, applying three exergy methodologies: the energy carrier’s exergy accounting approach, the natural resource’s exergy accounting approach and the extended exergy accounting.

2. Methodology and Data

2.1. Exergy Accounting of Energy Resources

Appendix A briefly revisits the exergy concept. More detail can be found in [32]. EREA measures the useful exergy services delivered to each end use. EREA estimates the system’s useful exergy by multiplying the intrinsic exergy applied to the system by the specific real efficiency for each energy carrier in its dominant end use. The efficiencies reflect the actual performance of the machinery or device that uses the energetic carrier in each end use (Figure 2). The accounting methodology is done in four separate steps [17,32]. First, all final energy that enters the system is converted into final exergy, multiplying each carrier by its exergy factor. Second, end-uses are allocated to each energy carrier. Third, a real exergy (or second law) efficiency is attributed to each carrier-end use pair based on the actual final to useful transformation devices involved. Finally, all useful exergy obtained from each end use category are added to discover the overall system’s useful exergy delivery and calculate an output-input ratio to assess the efficiency of the transformation process.

2.1.1. End Use Categories

From all energy carriers presented in the statistical data, natural gas, biofuels, heat production and fuel oil were predominantly used in low temperature heat processes while liquefied petroleum gas and kerosene were mainly used to support medium temperature heat devices. Gasoline and diesel were prevalent in motors in fisheries and agricultural machinery, being used mainly for transportation in the fisheries subsector and mechanical drive in general (including transportation) in crops, husbandry and forestry [17]. Electrical energy was used for several end uses: 26.7% for low temperature heat (water and process heating), 51.7% for mechanical drive (irrigation, material handling and other process use), 5.0% for lighting, 4.2% for cooling (AC), 10.0% for refrigeration (industrial cooling) and 2.4% for other electrical equipment like phones, computers and network devices [33].

2.1.2. Conversion Efficiencies

Efficiencies vary by end use and by energy carrier since the technologies involved are different. Low temperature heat (LTH) efficiencies considered were 10% for solid biofuels and biodiesel, 13% for natural gas and fuel oil, 15% from electricity and 18% from derived heat. Medium temperature heat (MTH) achieved better efficiencies since higher temperature flows have higher exergetic content. Both liquefied petroleum gas and kerosene were considered to have a 19% transformation efficiency. Gasoline and diesel were used for transportation and mechanical drive (MD) with 11% and 13% efficiencies, respectively. Also electricity was used to drive mechanical appliances with an increasing efficiency from 85% in 2000 to 88% in 2012 (Figure 2). All efficiency values were obtained from [34,35].

2.2. Natural Resources’ Exergy Accounting Approach

The NREA approach consists in quantifying all energy and matter that enters and leaves the system in the common unit of exergy. Contrary to EREA where the system is analysed in order to understand how the energy resources are used, in this methodology the system is evaluated only by quantifying the inputs and outputs, without any knowledge about the system’s internal behaviour (Figure 3). The ratio between outputs and inputs will reflect how well energy and matter are being used and offers insights on the sustainability of the agricultural sector. Although the amount of data is usually a lot bigger than in EREA, the methodology is simpler to apply since we only have to address the exergetic content for each flux across the system’s boundaries.

In this study, there is a clear distinction between the Agricultural, Forestry and Fisheries Sector and the environment (see Figure 3) focusing on the efficiency of anthropogenic activities. We consider crops, wood removals and fish from aquaculture as outputs because they are a product of an anthropogenic activity. In contrast with other studies [23,24,25,26,27,28,29,30] we do not count them as an extraction from the environment because they do not appear spontaneously. In contrast, we consider fish catch as a natural input from the environment.

2.3. Extended Exergy Accounting Approach

Extended exergy analysis adds, on top of natural NREA, the immaterial inputs that are crucial to run and maintain the system working and also adds the environmental impact as a virtual input (Figure 3). Labour and capital exergy estimations are based on two postulates elaborated by Sciubba [36], defending that the global influx of exergy resources are mainly used to sustain workers and, that the capital exergy flux is proportional to the labour exergy flux. The proposed equation to evaluate the exergy used to sustain all workers is (exergy of labour): $E_L=f\times e_{surv}\times N_h,$ where $f$ is an amplification factor that represents the society consumption over the survival mode, $e_{surv}$ is the minimum exergy required to maintain healthy metabolic needs (2500 kcal/day per person or 1.05 × 10⁷ J/day·person) and $N_h$ is the total population. The amplification factor is estimated using the Human Development Index (HDI) [37] which, based on three social dimensions, reflects the development of a country, $f=\frac{HDI}{HD{I}_0}$, being HDI₀ the Human Development Index of a pre-industrial society ($HD{I}_0\approx 0.055$) [36]. Labour equation implies that $E_L$ is linearly dependent on the HDI and independent of the fraction of workers in the society and the type of work. The exergy equivalent of Labour $\left(e{e}_L\right)$ for the society, or economic sector, can be calculated dividing the exergy of labour by the total hours worked $\left(N_{wh}\right)$, $e{e}_L=\frac{E_L}{N_{wh}}$.

Seckin et al. [38] followed by Rocco et al. [8] slightly changed the equation proposed by Sciubba [36] for the exergy of the capital flux from $E_K=\frac{M_2}{S}\times E_L$ to $E_K=\frac{M_2-S}{S}\times E_L$ where S is sum of all wages and M₂, defined by the European Central Bank [39], is an intermediate monetary aggregate which reflects the currency under circulation plus the liquid deposits (maturity up to 2 years and redeemable up to 3 months). The new formulation considers M₂-S because labour (reflected as the wages in the previous formula) was already internalized as an exergetic flux. The exergy equivalent of capital is obtained by dividing the total exergy of the capital flux by the monetary circulation, $e{e}_K=\frac{E_K}{M_2-S}$. Capital equation assumes that the monetary aggregate M₂ minus wages is eligible to the productive sector and represents the effective capital that drives all economic sectors. To translate money into exergy, the capital equation assumes the relationship developed for labour: $e{e}_K=\frac{E_K}{M_2-S}=\frac{E_L}{S}$.

Any material stream that is not physically confined and represents a discharge of the system to the environment is considered a pollutant. The environmental impact in EEA consists in developing a state-of-the-art process to bring the pollutant wastes and emissions to equilibrium with the reference environment. The total amount of exergy used in the innovative process is accounted for as an environmental externality and a virtual input to the economic sector. The equivalent exergy of the environmental impact portrays all exergetic consumption to run such a process (real or virtual) namely energy, matter, capital and labour having no relation with the pollutant physical exergy.

2.4. Reference Environment

The reference environment should be carefully chosen since the resource flow has exergy due to the imbalance (lack of equilibrium) with this reference state. The flow will have zero exergy in complete equilibrium with the reference state. In this study, the pressure P₀ = 101325 Pa and temperature T₀ = 298.15 K were adopted as well as the standard chemical composition of the natural environment. This is the reference environment used by Kotas [40] to estimate chemical exergy values that were used in this study.

2.5. Time Window and Control Volume

Recent exergy studies of countries or large regions focus on the seven economic sectors of society plus the environment and abroad [24,30,41] (see

Table 1. Description of Economic Sectors, Environment and Abroad.

Extraction (Ex)	Extraction of minerals and ores, coal mining, oil and gas extraction, refining and fuel manufacturing.
Conversion (Co)	Power, cogeneration and heat plants, based on renewables and non-renewables.
Agriculture, Forestry and Fisheries (AFF)	Crop and animal production, forestry and logging, fisheries and aquaculture. Includes animal products like milk, honey and eggs.
Industry (In)	All manufacturing industry excluding refining.
Tertiary (Te)	Public and private sector services, includes schools, hospitals, construction, trade and commerce, tourism, real estate and finance.
Transportation (Tr)	Private and commercial transportation of people and goods.
Domestic (Do)	Includes households and population.
Environment (E)	Lithosphere, the hydrosphere and the atmosphere.
Abroad (A)	Other countries or regions outside de economic system boundaries which allow energy and mass transfers.

).

This study focuses on just one economic sector, the AFF, because of the relevance and preponderance in sustaining human life. Figure 4 exhibits the agricultural sector within the region’s economy and specifies inputs and outputs considered for this study.

Time and regional boundaries were chosen in accordance with the available data, since few data were available posterior to 2012 and there was a lot of missing data prior to 2000.

2.6. Data

The statistical office of the European Union, Eurostat, provides in its database [42] the quantities of energy carriers, pesticides, fertilizers, harvested crops, meat slaughtered, fish catch, incubation eggs, milk collection, fish aquaculture, wood removals, atmospheric emissions, population, human labour and economic information. The energy of energy carriers was assumed to be the lower heating value (LHV) and the exergy factors for the energy carriers are the ratio of the standard chemical exergy of the organic fuels to the LHV, these factors being obtained from [40,43,44]. However, the consumption matrix of energy carriers is only available in two branches, being the first the agricultural plus the forestry subsectors and the second is the fisheries subsector. Also the data for pesticides and fertilizers is not disaggregated between agriculture and forestry subsectors. Due to this limitation and not wanting to apply some educated division between these carriers on both subsectors, all aggregated results will be presented in the conjunction of the agriculture and forestry subsectors.

Crop seeds, feed, produced eggs and honey carriers and some nutritional energy values for products not intended for human consumption were obtained from the statistics division of the Food and Agriculture Organization of the United Nations (FAOSTAT) [45]. Nutritional energy values assigned to each food element, in its raw state, are from USDA (United States Department of Agriculture) National Nutrient Database [46]. Moisture content of all food items was considered to be equal between USDA and Eurostat databases. All economic values such as compensation for employees, agricultural gross value added (GVA), monetary aggregate M2 and gross domestic product (GDP) were obtained at current prices and converted to constant euro GDP prices through a GDP price deflator obtained from the economic and financial affairs of the European commission (Ameco) [47]. The GDP price deflator is referenced to 2010 and measures the ratio between real GDP and the nominal GDP, providing a measure of inflation over the period.

Agricultural outputs were obtained in mass units, except wood which was available in volume units, and converted to exergy units using the energetic nutritional value presented in the USDA database [46] or in FAO (Food and Agriculture Organization of the United Nations) nutritive factors table [48], for food compounds.

The food (and feed) energy content was considered equal to the metabolizable energy as defined by [46]. The energy content (combustible or gross energy) of the ingested food should be measured by a bomb calorimetry. However, foods are not fully digested and absorbed by the organism. From the ingested or gross energy, some is lost to faeces (faecal energy) and gases (combustible gas), the rest is the digestible energy. Subtracting from the digestible energy, the energy that is lost is urine and heat results in metabolizable energy [49]. Data available from the United States Department of Agriculture for food energy is based on the Atwater system which is equivalent to the metabolizable energy [46]. The choice to use the metabolizable energy was mainly due to the lack of a complete database of combustible or gross energy.

The aggregate of food energy values presented are not fully ingested by society. Food supply chains cover five steps: agricultural production, postharvest, handling and storage, processing and packaging and finally distribution and consumption. In each step food is lost or wasted and the total in Europe reaches 31%. For Europe, at the producer level, food losses and waste accounts 2% for cereals, 3.1% for meat, 3.5% for milk, 9.4% for fish and seafood, 10% for oilseeds and pulses and 20% for roots, fruits and vegetables [50]. Such losses are not edible for societal consumption.

To fill in the gaps for a given resource or product, in a year or more, data was interpolated (between known values) or extrapolated (in extremes of the data set) by a linear regression made from all the other known values.

Fluxes inside the sector were not considered and only fluxes between sectors that have economic relevance were accounted for. Although seeds, feed and incubations are produced by the sector, they were considered to leave the agricultural sector to other economic activities (ex. industry or tertiary) and returned as an input, essential to the production. However straw, green fodder and manure (used as a fertilizer) that are internal inputs and outputs of activities within the sector were not accounted for.

Solar radiation, vital for crop production, and water, essential for crops and livestock, were not included in the study. The sun radiation is not an anthropogenic controlled flux and it can be seen as a flux from the environment with no economic value. Water from rain follows the same thinking as it is not an anthropogenic activity but irrigation water should have been an input although the lack of trustworthy data prevented its use in the study.

Animal skins production data (including wool) is only available for goats and sheep but the absence of a specific exergetic value for skins lead us to consider only the wool production flux that has a specific exergy of 5850 kJ/kg [51]. However, its exergetic value is very low ranging from 51 TJ in 2000 to 35 TJ in 2012. Other animal products such as lanolin or cow’s leather were not accounted for, due to lack of data. For lanolin and cow’s leather there are no specific exergetic values while for cow’s leather there is also no production data.

Wood removals are considered as under bark and the exergy values were obtained from Dewulf et al. [52]. Dry matter exergy values are 22.11 MJ/kg for softwoods and 22.01 MJ/kg for hardwoods and densities of 450 kg/m³ and 650 kg/m³ respectively. Moisture content was assumed to be 20%.

Environmental remediation exergy costs were estimated by multiplying the pollutant air emission (obtained in mass units) by the specific extended cost of removing the pollutant from the atmosphere. Studying the transportation sector for Turkey, Seckin et al. [38] also created a virtual process to find the environmental extended exergy cost of three pollutants: carbon dioxide (57,600 kJ/kg), methane (322,400 kJ/kg) and nitrous oxide (10,600 kJ/kg). Dai et al. [53] studying the transportation sector in China determined the specific environmental remediation exergy cost of carbon monoxide (11,800 kJ/kg), nitrogen oxides (3610 kJ/kg) and sulphur oxides (5890 kJ/kg). For ammonia, the chemical exergy was used as the extended exergy cost of removing the pollutant (since no study is available in the literature) and valued by 19,841 kJ/kg.

3. Results

3.1. Energy Resources’ Exergy Accounting

3.1.1. Energy Resources

Energy resources are essential as they fuel all the machinery applied in the AFF sector. In the agriculture and forestry subsectors, energy use decreased drastically, more than 50%, mainly due to the decrease in the use of diesel to one third of the initial value (Figure 5).

In contrast to agriculture and forestry, the consumption of energy resources in fishery has almost tripled due to the diesel carrier (Figure 6). In this subsector, LPG (Liquefied Petroleum Gas) and gasoline were also replaced by biodiesels and electricity. Notably, electricity grew from an 8% share in 2000 for agriculture and forestry to 23% in 2012 while in fishery and aquaculture (just fishery subsector from now on) it increased from 2.6% to 7.1% in 2012.

3.1.2. End-Uses

Mechanical drive (e.g., tractors, irrigation pumps) and transportation (ex. boat engines) consume the biggest share of useful exergy in the sector. The decrease of 56% in the exergetic input of energy carriers in agriculture and forestry, from 2000 to 2012, was reflected in a 39% decrease in the amount of useful exergy applied to both subsectors (Figure 7). Fisheries increase energy carriers’ consumption by 291% and obtained a 320% increase of useful exergy (Figure 8). Both improvements are associated to an increasing final-to-useful efficiency and are mostly due to the increased use of electricity, namely in irrigation.

3.1.3. Sector and Subsector Efficiencies

All subsectors have increased their efficiency (Figure 9) with agriculture and forestry showing a better performance. The increase in the share of electricity and the reduction of the share of diesel (mostly associated to the transition from diesel-powered to electrically-powered irrigation pumps) allowed an increase in performance of 38% (from 16% to 22%) for agriculture and forestry and 30% (from 16% to 21%) overall. Fishery also increased their conversion efficiency from energy carriers to useful exergy from more than 14% to 16%.

To test whether the increase in efficiency was only due to technological improvements we estimated EREA subsectorial efficiencies assuming the carrier end-use pairs efficiencies were constant from the year 2000 onwards (Figure 10). In these conditions, the useful exergy per final exergy would have increased 35% for agriculture and forestry (from 16% to 22%), 8% for fishery (from 15% to 16%) and 27% overall (from 16% to 21%), showing that the overall increase in efficiency was due not only to better technologies but also reflect a wiser choice of energy carriers for the desired end-use.

3.2. Natural Resources’ Exergy Accounting

3.2.1. Total Outputs

About two thirds of all agricultural, forestry and fisheries exergetic output is from wood removals. The changes in the overall annual production that ranges between 140 PJ to 160 PJ are controlled by changes in wood production (for a more detailed description see Appendix B). The aggregated production intended for food and feed decreased 14% mainly due to a 20% decrease in crops harvested despite a 20% increase in meat production (Figure 11).

The production matrix for the Portuguese AFF sector has remained almost constant, being forestry the major producer and increasing its share. Fishery nutritional values are neglected while energetic values for animal husbandry are constant and the share of vegetables and fruits have decreased, mainly after 2004 (Figure 12).

3.2.2. Total Inputs

Overall input to the agricultural sector is decreasing (24% over the 13 years) mainly due to fewer amounts of energy resources which alone represented 29% of all inputs in 2012 (Figure 13 and Figure 14). Seeds, fertilizers, pesticides and eggs for incubations were almost negligible in the sector’s consumption matrix while feed is by far the biggest input (for a more detailed description see Appendix C).

3.2.3. Sector and Subsector Efficiencies

The agricultural sector relies heavily on environmental conditions to sustain production. Water and radiation are offered freely by our natural environment and do not constitute a societal activity. Their contribution leads to efficiencies higher than 100% for agriculture and forestry. This subsector reached a 255% efficiency in 2012, a 36% increase since 2000, contrasting with fishery, which decreased its efficiency by 53% from a 43% efficiency in 2000 to 20% in 2012 (Figure 15).

3.3. Extended Exergy Accounting

The inclusion of three new production factors offers a new perspective of the AFF sector and allows comparisons between material and immaterial inputs. Exclusively for the three immaterial inputs of the EEA approach, the agriculture and forestry subsectors are quantified separately to allow comparisons between them.

3.3.1. Labour

Total exergy allocated to labour was calculated by the labour equation with HDI values from the United Nations Development Programme—Human Development Reports [37] and total population and number of workers data from the Eurostat database [42]. These values as well as other labour and capital statistics are available in Appendix D. To measure the associated labour exergy to the AFF sector and subsectors, the total societal labour exergy was multiplied by the fraction of workers in this sector and subsectors. This assumes that all workers have the same labour exergy independently of the sector or even the working hours, which is consistent with the equation used.

The number of workers and respective labour exergy is far superior for agriculture over forestry and fishery (Figure 16). Total labour exergy remained constant due to an increase in the HDI values and to a decrease in the number of workers in the AFF sector. Overall equivalent exergy of labour increased rapidly from 51.5 MJ/h in 2000 to 89.0 MJ/h in 2012 which means that the necessary societal exergy to produce one hour work increased 73% in 13 years. Although the cost of labour was similar, in the beginning of the century, for all three subsectors, in 2012 the necessary exergy to supply one hour work to agriculture was 30% higher than to forestry and fishery. This increase was due to the decrease in the number of hours worked per year by farmers from about 2000 to less than 1500 h in 2012.

3.3.2. Capital

In addition to labour, capital is a fundamental key for a productive system and should also be introduced as an input. To discover the monetary aggregate associated to the AFF sector, a direct relation was made between M2 and total gross value added (GVA) of the country (which in case of Portugal is almost identical). The monetary aggregate of the sector and subsectors under analysis was obtained assuming a linear relationship with the GVA of the AFF sector and subsectors allowing the calculation of different equivalent exergy of capital (ee_k) for each of the three subsectors. Capital exergy relates to labour exergy through the ratio $\frac{{\mathrm{M}}_{2\mathrm{S}}-{\mathrm{S}}_{\mathrm{S}}}{{\mathrm{S}}_{\mathrm{S}}}$ which is higher than 1, reaching 8 for forestry. The S subscript in the previous equation means that the considered values are sector or subsector related.

The exergy of capital has decreased 42% from 2000 to 2012, mainly due to a strong decrease in AFF’s GVA (Figure 17). The equivalent exergy of capital has been constant for all subsectors with the agriculture subsector presenting the highest value. The explanation is that the wage per hour in agriculture subsector is more than five times lower than in fishery or forestry, although there was a 73% increase in the wage per hour in agriculture (in 13 years) compared to an 18% increase for forestry and 12% for fishery (at constant 2010 prices).

3.3.3. Environmental Impact

The EEA considers that the exergy value of the waste or pollutant should not be his chemical exergy but the extended exergy cost to avoid the pollutant or bring the pollutant to the dead state. Such extended exergy costs are the sum of all physical exergy required to run the cleaning process plus the labour and capital exergy fluxes imperative to run the process [54]. Carbon dioxide is mainly emitted from the combustion of fossil and renewable fuels; emissions from biomass (not used as fuel) were not accounted for because in a life cycle perspective biomass is neutral. Emissions by animals comprise carbon dioxide and methane from enteric fermentation in the ruminants’ digestive system. The methodology doesn’t account carbon dioxide from animal husbandry but accounts methane emissions, being the total amount almost exclusively from livestock production (Figure 18). The total exergy needed for environmental clean-up from atmospheric emissions is the biggest input of the system showing a decrease of 10% over the years. Carbon dioxide represents almost two thirds of the equivalent exergy and methane the remaining part, being the exergy from the other pollutants residual. Methane emissions follow the trend line of animal husbandry production (almost constant) and carbon dioxide follows the decreasing trend line of energy resources.

3.3.4. Total Inputs

All three immaterial inputs addressed by EEA are quantitatively superior to NREA inputs. The biggest share of energetic and material inputs reached 14% in 2012, revealing the discrepancy between immaterial and material inputs and pointing out capital and labour as the main factors of production. Environmental impact proved to be the strongest concern and quantitatively exceeded all other inputs (Figure 19). Overall values dropped 25% due to the decrease in energy resources, capital and air emissions.

3.3.5. Sector and Subsector Efficiencies

The introduction of three highly valuable production factors to the input matrix dramatically lowered the efficiency of all AFF subsectors. The agriculture and forestry subsector increased their efficiency by 33% (from 27% in 2000 to 36% in 2012) while fishery increased its efficiency by 32% (from 2.7% in 2000 to 3.6% in 2012) (Figure 20). This improvement was mainly due to a decrease of the capital exergetic input.

4. Discussion and Conclusions

Results obtained for the Agriculture, Forestry and Fisheries (AFF) sector and subsectors, regarding the ratio of exergy outputs to inputs, using the three exergy accounting methodologies are synthesized in Figure 21. Results for the AFF sector are similar among methodologies. With the EREA approach there is a 30% increase in the overall efficiency due to a better match between the energy carrier and the end uses. NREA presented a similar growth in overall efficiency representing the sector’s return over the applied energy and matter. The EEA methodology showed the same overall tendency regarding the evolution of efficiency, although the results are clearly dominated by the capital and environmental impact. For the fisheries subsector, the results are contradictory among methodologies.

Energy resources’ exergy accounting (EREA) only addresses one type of inputs—energy carriers—to estimate the available useful exergy for its end uses and the overall efficiency. Although it estimates the actual amount of exergy that is delivered to a system function, no information regarding the output is introduced in the analysis. For Portugal, the overall efficiency reached 21% and is rapidly increasing over time due to a bigger share of the use of electricity whose efficiency is substantially higher, mainly for its predominant use in mechanical drives, mostly associated with irrigation systems. In fisheries, diesel fuel continues to be preponderant in fishing vessels, although with lower efficiencies. This explains the different performance levels of the two subsectors.

The natural resources exergy accounting (NREA) approach uses all energetic and physical inputs and outputs to describe a system’s efficiency without any knowledge about the system’s processes and resources utilization. In the present analysis, all inputs were accounted for only by their intrinsic exergies values, not by their cumulative exergy values. The exergy that has been used to transform and bring the inputs to the boundaries of the AFF sector are integrated in the other economic sectors like industry for the fertilizers and pesticides, extraction and conversion for energy carriers and tertiary and transportation for almost all. The exergy input has decreased 24% since 2000 (Figure 13) with feed for animal husbandry representing the biggest share (68% in 2012), followed by the energy resources. The exergy of feed is on average 140% superior to the exergy of all crops harvested, which means that Portuguese agriculture does not produce enough to support its animal husbandry. Feed does not consider green fodder, pasture, meadows and straw because these are internally produced and consumed in the sector. The output-input ratio of animal husbandry (with feed and incubation eggs as inputs, only) evolved from 26%, in 2000, to 31% in 2012. This positive evolution indicates either a better use of feed in livestock production or a bigger share of non-accounted feed. The exergy output from the sector is mostly embedded in wood and harvested crops with wood exergy reaching a share of 71% in 2012 (Figure 11).

Although exergy is degraded, the output-input ratio of the agricultural and forestry subsectors can be higher than one (contrary to other economic sectors except extraction and conversion based on renewables) because solar radiation is not taken into account. Considering NREA, the sector’s overall ratio increased more than 27% in this period being close to 2.5 in 2012, which means that for every exergy Joule that entered in the AFF sector, two and a half Joules were harvested. Agriculture and forestry performed much better than fishery (Figure 15) whose efficiency decreased 53% since 2000 (from 43% in 2000 to 20% in 2012). This decreasing efficiency verified for fisheries and aquaculture is exclusive for the NREA methodology. While in EREA we observe a better match between the energy carrier and the end use, in NREA the total amount of nutritional values from fish caught is decreasing for each exergy unit spent on the fishery activities.

The explanation for this puzzle is revealed by looking at the ratio of the output exergy (measured by the NREA methodology) to the useful exergy (measured by the EREA methodology) which is a proxy for the efficiency of the energy services in the AFF sector (see Figure 22). Contrary to agriculture and forestry which improved the efficiency of the exergy services by 59% (from 27 to 47), fishery reduced theirs to 31% of the initial value (from 5 to 1.5). While in the agriculture subsector each Joule of useful exergy produced 47 Joules of food in 2012, in fishery only 1.5 Joules were caught. The agriculture and forestry subsectors are less dependent on energy resources.

The efficiency of the exergy services in fisheries is the amount of fish catch per unit of useful exergy. To explain this better, we will focus on the diesel carrier, which is responsible for almost 90% of the consumption matrix (in 2012) and used for transportation services. The fish catch per useful exergy ratio is equal to the fish catch per distance travelled times the thermodynamic efficiency of the boat engine (distance travelled per unit of useful exergy). In fisheries a higher thermodynamic efficiency is not translated into higher energy services efficiency presumably because fish population is decreasing leading fishing vessels to travel more (or more often) to get the same amount of fish.

Extended exergy accounting (EEA) allows us to compare the weight of different inputs. The exergy embodied in labour, which is less than an average daily work in other sectors (1500 h per year and per worker in 2012), has the same magnitude of all other material and energetic inputs taken together. Capital exergy is predominant over labour and although it is rapidly decreasing over time, it is larger than the overall exergy of outputs. Figure 18 also shows that actually, the bigger input in EEA is the environmental impact from the sector, being all three immaterial inputs seven times superior to the material ones. These results raise issues regarding the proportionality between material and immaterial inputs in the EEA methodology. By considering that the cost of labour and capital is a cumulative societal exergy and, by accounting all other inputs in its intrinsic form, the study reveals the disproportionality that results from this inconsistency. Also, measuring the capital input to the sector by all currency money (M2) minus wages creates an additional problem, because some of this money is used to buy all energetic and material inputs that entered the agricultural sector, and that were taken already into account for by their intrinsic exergies. Following the proposed methodology leads to double counting all natural resources inputs because both intrinsic exergy plus the currency money necessary to buy them are taken into account.

So, are the different exergy accounting methodologies providing consistent or contradictory results? Results provided by the EREA and NREA approaches can be contradictory but they are complementary because the EREA is focused on the thermodynamic efficiency while the NREA extends the boundary allowing the evaluation of the efficiency of the energy service. Both approaches are important in providing a realistic picture of the sectorial exergy efficiency. On the other hand, results obtained by the EEA approach are completely dominated by the exergies of capital and environmental impacts, revealing the disproportionality between material and immaterial inputs in this methodology. Improving the estimation of EEA externalities, labour and capital, for sectorial studies, remains a challenge for future studies.

We propose and apply to our case studies of the AFF sector, using the NREA and EEA approaches, methodological options to account for the renewable flows obtained from the environment, that are more adequate for sustainability assessments. While former NREA and EEA studies [19,21,22,23,24,25,26,27,28,29,30] account all crops, wood and fisheries as an input to the agricultural sector from the environment, this study acknowledges that all these flows are renewable and an output of human activities. Fish catch (excluding fish farming) is considered an extraction because no human activity helps to feed and raise the fish and excessive catches together with pollution have caused a decrease in fish population [55]. For a better assessment of sustainability, it would be important to include and correctly evaluate in NREA and EEA studies, the environmental positive impact of the carbon dioxide sequestered from atmosphere by the AFF sector because, although there is a substantial fraction of the carbon content in vegetables, fruits, crops and wood that is released after consumption, some carbon remains sequestered in living beings and wood used for veneer and saw logs.