Indirect energy input of agricultural machinery in bioenergy production

Renewable Energy 35 (2010) 23–28

Contents lists available at ScienceDirect

Renewable Energy journal homepage: www.elsevier.com/locate/renene

Review

Indirect energy input of agricultural machinery in bioenergy production Hannu J. Mikkola*, Jukka Ahokas Department of Agrotechnology, University of Helsinki, P.O. Box 28, (Koetilantie) 3, FI-00014 Helsinki, Finland

a r t i c l e i n f o

a b s t r a c t

Article history: Received 22 December 2008 Accepted 7 May 2009 Available online 21 June 2009

Sustainability of bioenergy products should be evaluated by means of an energy analysis that takes into account all relevant direct and indirect energy inputs. Direct energy input is viewed as the major energy consuming factor, and is quite easy to measure. Indirect energy input, however, has received relatively scant attention, so it is likely to be insufficiently analysed and possibly underestimated. This paper reviews the data available and suggests the type of research that would be needed to get a better understanding of the indirect energy input. The analysis addresses questions about the use of energy to produce and maintain agricultural machinery, the allocation of energy to different bioenergy products, and the real use and lifetime of machinery. Ó 2009 Elsevier Ltd. All rights reserved.

Keywords: Bioenergy Energy input Energy analysis

1. Introduction

2. Indirect energy input

Indirect energy input for farm machinery comprises numerous, small energy items, which appear in material products and nonmaterial services. Fig. 1 illustrates energy flows related to the manufacturing and use of machinery. Direct energy input is easy to identify and analyse, while indirect energy input is also relatively easy to identify but more difficult to analyse. This is a common problem in energy analysis. Single indirect energy items are often small so they are considered insignificant and are neglected. Although they are significant as a whole, there is no easy way to analyse them. A life cycle assessment (LCA) [1,2] or corresponding procedure is often considered too laborious for this purpose. Thus indirect energy input is usually assessed either not at all, or only as a percentage of the total energy consumption. For example, a common procedure is to calculate repair and maintenance costs as a percentage of purchase costs, and then apply the same percentage to the manufacturing energy input to estimate the maintenance energy input. This paper reviews the data available and suggests the type of research that would be needed to allow a better understanding of the indirect energy inputs. This paper also suggests how the indirect energy consumption could be allocated to different bioenergy products.

Manufacturing energy and energy for repair and maintenance (R&M) are the main items of the indirect energy input [3]. These energy items are considered in Sections 2.1 and 2.2. Delivery includes the energy for transport from the factory to the farm either directly, or indirectly via a dealer. This operation happens once in the lifetime of a machine and Loewer et al. [4] reported the energy cost to be 8.8 MJ/kg. Agricultural machines are normally stored in low-priced, unheated buildings or outdoors. Storing outdoors is free of direct financial and energy inputs but it is not recommended, because the continuous exposure to the weather generally increases repair and maintenance costs. Energy costs of buildings can be defined in the same way as that for machines and the costs are allocated for machines according to the floor area or the space needed.

* Corresponding author. Tel.: þ358 9 191 58483; fax: þ358 9 191 58491. E-mail addresses: hannu.j.mikkola@helsinki.fi (H.J. Mikkola), jukka.ahokas@ helsinki.fi (J. Ahokas). 0960-1481/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.renene.2009.05.010

2.1. Manufacturing energy A systematic energy analysis created especially for manufacture of agricultural machines is missing and this is why the study of Berry & Fulton Fells [5] of the car industry is widely used as a reference e.g. [6–9]. Berry & Fulton Fells [5] made an energy analysis of the manufacture, discard and reuse of the automobile and its component materials. Their data originated from the automobile industry in Chicago in 1967. Pimentel et al. [6] referred to the study of Berry & Fulton Fells [5] in their well-known article ‘Food production and the energy crisis’. That article seems to be a much more popular reference for manufacturing energy than the original article of Berry & Fulton Fells [5].

24

H.J. Mikkola, J. Ahokas / Renewable Energy 35 (2010) 23–28

Indirect energy input

Direct energy input

Manufacturing

Fuel, electricity

Delivery Repair, maintenance

Storage

Allocation of energy input

Agricultural usage, J/ha, J/h, J/kg, ...

Other usage

Fig. 1. Direct and indirect energy input for agricultural machinery – energy use and allocation.

Berry & Fulton Fells [5] broke the fabrication process of an automobile into the major steps of primary materials recovery, materials finishing and fabrication. They calculated that 37 275 kWh of primary energy was needed for the fabrication of an average car of 1653 kg, i.e. 81.2 MJ/kg. Some fifteen years later a life cycle assessment made at Carnegie Mellon University showed that 86.6 MJ kg1 was needed to manufacture a 1990 Ford Taurus [10]. A third example from the car industry is an environmental report of the Volkswagen company, which showed that nearly the same amount of primary energy was needed to fabricate a Volkswagen Golf [11] in 2002. The energy consumption for the fabrication of the petrol model was 80.8 MJ kg1 and 74.9 MJ kg1 for the diesel model. There may be minor differences in the methods of analysis and assumptions, but the order of magnitude was the same. Steel and iron components accounted for 94% of the mass of an automobile in the energy analysis of Berry & Fulton Fells [5] and respectively 60–61% in the LCA-analysis of Schweimer & Levin [11]. This difference illustrates a major change in car manufacture. The energy use in U.S. steel and iron production has decreased by 27% from 35.6 GJ Mg1 in 1958 to 25.9 GJ Mg1 in 1994 [12]. Bo¨rjesson [13] used the value of 24.0 GJ Mg1 for Swedish steel production in his own energy analysis, and Farla & Blok [14] reported still lower consumption figures of 22.5 GJ Mg1 for steel and iron derived from ore, and 8.5 GJ Mg1 for recycled steel and iron. Against this background the energy use in car manufacture should be clearly lower than that for 40 years ago, but this is not the case. There are many possible reasons for this. Plastics, fibreglass and aluminium are more energy intensive materials than the replaced steel. Cars today have safety, anti pollution, and facility equipment, which make them more complex. Complexity means more work phases and more energy. One possible reason for increased energy use is that current manufacturing processes are analysed more accurately. For all the stated reasons, the energy analysis of agricultural machinery should not be made on the grounds of energy consumption in steel production. A change in use of materials and the increased technical sophistication of modern machinery should be better taken into account. A comparative analysis of the energy use in the car industry and in the agricultural machinery industry has not been made. It is necessary that this is done because there are essential structural differences between cars and tractors. Tractors are made of many heavy cast iron components, while the body of a car is invariably made of steel sheet. The proportion of synthetic materials has

increased in the car industry, but similarly tractors today have very large and heavy wheels, cabin surfaces have been upholstered, fuel tanks, and bonnets and wings are mainly made of plastic or fibreglass. In the same way modern tractor implements contain many parts made of synthetics, e.g. tanks, cover plates, gear wheels, hoses, mouldboards, feeding wheels in seed drills and so on. Considering the energy consumption over the whole lifetime of a car, the fuel energy accounted for 70–73% of the energy invested in a Volkswagen Golf if it was used for 10 years and the mileage was 150 000 km [11]. The manufacturing process accounted for 9% and the production of materials for 12%. The remainder originated from fuel production. The mileage of 150 000 km corresponds with the use of 2500–3000 h (at an average speed of 50–60 km/h). A tractor with a 75 kW engine weighs 3–4 times more than a Volkswagen Golf. If the expected use of a tractor is 10 000 h, it is four times more than that of a car, and thus the fuel consumption of a tractor should represent roughly the same share in the lifetime energy consumption as that of a car. According to this comparison, at least 20% of the lifetime energy use of a tractor should originate from manufacturing and materials. This estimation is very well in line with Fluck’s [3] studies, which suggest a share of 21% for the energy sequestered in manufacture and distribution of agricultural machinery. 2.2. Repair and maintenance (R&M) Repair and maintenance of tractors and agricultural machinery cover fabrication, storage and transport of spare parts, service and repair facilities, and service and repair activities done by dealers and farmers. Service and repair facilities mean that service entrepreneurs and farms often have workshops of their own, equipped with a set of hand tools and other repair shop equipment. Energy consumption in R&M consists of many small items. There are certainly differences between machine categories, production sectors, farms and locations resulting from different economic, social and climate conditions. Fluck & Baird [15] reported that the energy needed for R&M was significant compared with the energy needed for manufacturing. Later Fluck [3] presented two models, which could be used for analysing the energy requirement. These models were The Industry Costs Model and The Lifetime Machine Repair Cost Model. His results left some space for discussion, because the Industry Costs Model indicated that 55% of the energy needed for manufacturing was needed for R&M. The Lifetime Machine Repair Costs model suggested 138%. In the Industry Costs Model Fluck used statistics of the sales value of agricultural machinery, spare parts, and service business to define the use of energy. Energy consumption was calculated by multiplying the monetary value of the after-sales services by their energy intensities. Energy intensity is a concept that relates the energy used in a specific sector of a society to the monetary value of the production of that sector. The Industry Costs Model was fitted to a real life situation by making assumptions that, for instance, farmers mounted 70% of the spare parts themselves, farmers acquired 20% of the spare parts from shops other than agricultural machinery sales companies, the one-way travel to acquire spare parts was 35.4 km and farmers used 67 h annually to repair their machines. These assumptions are also relevant today and they were made on the grounds of statistics and corresponding sources. The Lifetime Machine Repair Costs Model resulted the specific lifetime energy for R&M for 14 items of agricultural machinery. In this model Fluck [3] used ratios of accumulated repair and maintenance costs to initial costs over the lifetime of the equipment

H.J. Mikkola, J. Ahokas / Renewable Energy 35 (2010) 23–28

presented by Kepner et al. [16]. The energy was on average 1.38 times that needed for manufacturing. It was more than double that according the Industry Costs Model and much more than in any earlier analysis. A cutter bar mover was an extreme example of a very high energy requirement. The energy for R&M was 3.6 times the energy needed for manufacturing. Fluck [3] himself regarded the energy costs of this model to be high and suspected that the high costs were a consequence of the following limitations of the model:  The ratios of the lifetime R&M costs to initial costs were too high, because: 1. Older agricultural machinery may be used despite inadequacies in repair and maintenance. 2. Older equipment may be repaired using more salvaged as opposed to new parts, due to parts unavailability and obsolescence. 3. The rate of R&M of a machine is lower near the end of its useful life. There are usually many un-repaired faults when a machine is finally scrapped. 4. Repairs in latter stages of the life of equipment may consume more labour and fewer parts.  Each of the fourteen items of equipment was treated in the same way insofar as purchased versus farmer services and parts versus services are concerned.  Single energy intensity was assumed for all fourteen different machines though variations between machines undoubtedly exist.  The final result was an unweighted average of fourteen machines though the energy sequestered in R&M is relatively more important for some machines than for others. As the main result of the Lifetime Machine Repair Costs Model Fluck [3] considered that there was substantial variance in the R&M costs between different machines. Because this model showed a much higher energy need than the Industry Costs Model, Fluck’s [3] final conclusion was that the energy needed for R&M was, on average, at least 55% of the energy needed for manufacturing. Fluck’s studies [3] so far seem to be the most thorough attempt to analyse the energy requirement of R&M. Seven years later Bowers [7] presented Fluck’s ratios [3] of R&M energy and manufacturing energy for the same fourteen machines. He had multiplied the machine specific ratios by the value of 0.4, resulting in 0.55 as the average ratio for all fourteen machines. Because 0.55 seems to be the minimum ratio, higher ratios should perhaps be used in order to avoid overoptimistic energy yields in bioenergy analyses. Fluck’s studies [3] were made 23 years ago and it is time to update his results. At the end of his report Fluck [3] summarised the annual energy consumption for agricultural machines. Percentages were as follows: fuel 67%, manufacturing and distributing 21%, repairs and maintenance 12%. Though the share of R&M, or the indirect energy as a whole, is not the most important energy input, it is a significant and possibly underestimated factor.

25

assessment of energy intensities on the databank of the U.S. Department of Commerce. The databank contained data from 368 branches of the economy. The net energy analysis minimises the effort required to obtain a predetermined degree of accuracy in the result and the net energy analysis offers better possibilities to manage indirect energy input than process analysis or input– output analysis alone [18]. Fluck [3] stated that there was only a little variance between the energy intensities of spare part sales for different machines. This is why the most dominant factors affecting the ratio of the energy of R&M to the energy of manufacturing were ratios of costs of R&M to the costs of manufacturing reported by Kepner et al. [16]. A main advantage of Fluck’s methods [3] was that the energy needed for R&M could be analysed without calculating energy items related to producing every single R&M service or goods. A disadvantage is that it may be difficult to get disaggregated statistical data to fix correct energy use to correct sectors and to avoid registering the same energy item many times to different sectors. Outside the U.S. lack of statistical material has presumably prevented the use of this method. Table 1 compares the ratio of R&M energy to the manufacturing energy from different sources. It seems evident that the earliest assumptions have been underestimated. The big difference between Fluck’s two models [3] would need further study. Fluck [3] himself criticised the results of the Lifetime Machine Repair Cost Model. Perhaps the substantial difference with earlier evaluations made him suspect his own results. 4. Lifetime of agricultural machinery There is no exact lifetime for agricultural machinery. It depends on usage, level of service, and speed of technical and economical development. If development goes fast, machines become obsolete or uneconomical sooner than they are technically worn out. This is a well-known phenomenon discussed in numerous books e.g. [24–26]. ASABE standards provide some estimates for the lifetime of agricultural machinery. ASAE D497.5 [27] gives a 12 000 h lifetime for a 2-wheel drive tractor and 16 000 h for a 4-wheel drive tractor and crawler. Estimated lifetimes seem to be high compared with typical usage and the economic lifetime. For instance, in Denmark [28] the average annual usage was over 200 h. In Canada tractor operating costs are calculated according to annual use values of 200, 400 and 600 h [29]. Schnitkey & Lattz [30] based calculations on an annual use of 300 h in USA. Although these cost calculations are based on evaluated usage, they should reflect the real use. Results from Finnish farms [31] support the Danish results. The newest tractors are used more than the average during the first 5–7 years, but after that period usage declines. After 15 years the usage is only 100 h. In order to achieve the estimated technical lifetime

Table 1 The ratio of energy needed for repair, service and manufacturer to the manufacturing energy [3].

3. Methods of analysis

Source

Repair, service and maintenance energy percentage of the manufacturing energy

Analysis of indirect rather than direct energy input is more challenging from the methodological point of view. Machinery repair and maintenance consist in the quantity of partial processes, which vary from case to case. Fluck [3] solved this problem by utilising energy intensities in his two models. Energy intensities resulted from a net energy analysis, which is a combination of process analysis and input–output analysis developed by Herendeeen & Bullard [17] and Bullard et al. [18]. Bullard et al. [18] based

Pimentel et al. [6] Smil et al. [19] Foster et al. [20] Van Hecke [21] Doering, 1977, [22] Leach [23] Fluck [3] Industrial Costs model Lifetime Machine Repair Costs model

6 8 10 (repair parts fabrication only) 20 32 53 55 138

26

H.J. Mikkola, J. Ahokas / Renewable Energy 35 (2010) 23–28

usage tractors should be running for 40–60 years, while the estimated economic lifetime is only 10–15 years. It would be important to know the real lifetime usage because there is a large difference depending on whether the manufacturing energy is allocated for 3000 h or 16 000 h. This same problem relates to implements and agricultural buildings as well. As long as there are no exact data for the lifetime and annual use of machinery, manufacturing energy should be allocated to the real usage during the economic lifetime.

5. Tractor fuel consumption and a suggestion to allocate indirect energy input to field crops The direct energy demand (fuel consumption) of a tractor is measured under test conditions as l h1 or kg h1. The specific fuel consumption (g kW h1) is calculated on the grounds of the measured power and fuel consumption. Fuel consumption varies depending on the work and the corresponding engine load. In an energy analysis made for a specific crop, fuel consumption has been normally given in l ha1. This is a way to account for the capacity of machines of different size. If the tractor is optimal in relation to the power requirement of the implement, the size of the tractor and implement has only a minor impact on the fuel consumption per hectare [32]. For this reason it would easier to tie the indirect energy input i.e. energy for manufacturing, repair and maintenance, to the fuel consumption. In order to do this the total fuel consumption of a tractor during its lifetime needs to be estimated. The real use and lifetime of tractors should produce realistic results. When indirect energy input is divided by the lifetime use, it can be added to the fuel consumption. Tractor fuel consumption depends on the engine load and on its working point. A good fuel economy is typically achieved with a high torque and relatively low engine speed i.e. 55–85% of the nominal engine speed. The specific fuel consumption curve (g kW h1) shows the efficiency of the tractor engine (Fig. 2). In order to estimate the fuel consumption for the tractor’s lifetime we need to know the mean engine power during its lifetime, which depends on the work done. Under northern European conditions the mean engine power has been 25–35% of the nominal power and the engine speed 50–70% of the nominal speed [34–36]. This corresponds typically with 225–240 g kW h1 specific consumption. The lifetime fuel consumption can be now calculated with equation (1).

Vf ¼ qs $Tl $Pm

(1)

In equation (1) Vf is the total fuel consumption during the lifetime of the tractor, qs is the specific mean consumption, Tl is the lifetime in hours and Pm is the mean engine power. Manufacturing energy is normally given per unit weight (MJ kg1) and it can be changed to energy per unit power by multiplying it by the weight to power ratio (WPR). The same procedure can be used for R&M energy. Fig. 3 presents the WPR for tractors on the Finnish market in 2005 [37]. The most popular power class in 2002 was 81–100 kW [38] it corresponds to about 57 kg kw1 WPR. The mass-based manufacturing and R&M energy EmW can now be changed to engine power based energy EmP by using equation (2).

Emp ¼ Emw $WPR

(2)

When the fuel consumption over a tractor’s lifetime is known the manufacturing and R&M energy per unit fuel consumption qmE can be calculated according to equation (3). Pn is the nominal power of the tractor.

EmP $Pn Vf

qmE ¼

(3)

Now qmE can be added to the fuel consumption to give the sum of the direct and indirect energy. A corresponding procedure can be applied to calculate the energy needed for manufacturing and R&M of tractor implements. The units in this case are J ha1. The energy per hectare is conditional on the following factors: the mass of the implement, the field capacity, the lifetime of the machine and the coefficient defining the ratio of R&M energy to the manufacturing energy. Equation (4) can be used. Table 2 shows the energy per hectare for some implements.

Eha ¼

EMþR&M *M=W T*C

(4)

In equation (4) Eha is the energy for manufacture and lifetime repair and maintenance of a tractor implement allocated to one hectare field area, J ha1, EMþR&M is the energy for manufacture and lifetime repair and maintenance of a tractor implement, allocated to the mass of the implement, MJ kg1, M is mass of the implement, kg, W Working width, m, T is Iifetime, h, and C is capacity per one metre working width, ha h1 m1.

90

Weight/Power, kg/kW

80 70 60 50 40

y = -0,1027x + 66,692

30 20 10 0

0

50

100

150

200

250

Power, kW Fig. 2. A typical chart for tractor engine performance [33].

Fig. 3. Weight per unit power ratio (WPR) of agricultural tractors on the Finnish market 2005 [37].

H.J. Mikkola, J. Ahokas / Renewable Energy 35 (2010) 23–28

27

Table 2 The energy needed for manufacture, repair and maintenance (R&M) of implements allocated per hectare. Implement

Plough,4 furrows Chisel plough S-tine harrow Combined drill Direct drill Roller Field sprayer, mounted a b c d

Mass,a kg

Working width, m

Mass per 1 m working width

Typical working speed, km h1

Capacity per 1 m working width, ha h1

Lifetime,b h

Ratio of the lifetime R&M energy to manufacturing energyc

Energy for manufacturing þ R&M, MJ kg1d

Energy for manufacturing þ R&M, MJ ha1

936

1.60

585

7

0.7

2000

0.97

180

75.2

1296

3.00

432

8

0.8

2000

0.51

140

43.2

1353

3.80

356

8

0.8

2000

0.55

143

31.8

1375 3192 1824 396

2.50 3.00 4.00 12.00

550 1064 456 33

8 10 6 7

0.8 1.0 0.6 0.7

1500 1500 2000 1500

0.55 0.55 0.55 0.37

143 143 143 119

65.5 101.4 54.3 15.1

The mass is typical for the type and size of the machine in question on the Finnish market in 2005 [39]. [27]. Coefficients are presented by Bowers [7], who edited Fluck’s [3] original coefficients so that the mean of 14 machines is 0.55. Energy for manufacturing 86.7 MJ/kg [7] þ energy for transportation from plant to farm 8.8 MJ/kg [4] þ energy for R&M [5,7].

In Table 2 a field sprayer requires the lowest energy input (MJ ha1) because of it low mass and high capacity. A relatively short lifetime of 1500 h has a contrary influence, but in this case it cannot overcome the influence of low mass and high capacity. A direct drill had the highest energy input because of high mass and relatively low capacity (working width  speed). The ratio of the lifetime R&M energy to the manufacturing energy is an important factor if the R&M energy input is high. It nearly doubles the energy input of a plough. On the other hand, this ratio causes uncertainty because there is no state-of-the-art study that addresses this subject.

6. Conclusions Results of the energy needed to manufacture agricultural machinery are mainly based on studies done in the car industry or for steel production. Energy use in steel production has decreased, but examples from the car industry show that the energy requirements have remained nearly constant for 40–50 years. This is why assumptions about energy use in production of agricultural machinery should not be made solely based on energy consumption during steel production. Indirect energy needed for manufacture, repair and maintenance is equivalent to 30% of the total lifetime energy input of a tractor. Results on the annual use of tractors show that very few tractors are technically worn out during their economic lifetime of 10–15 years. Most tractors are used longer, but the real lifetime and the use during the lifetime are unknown. This makes it difficult to allocate the energy needed for manufacturing, repair and service to the whole lifetime. If this energy is allocated to usage during the economic lifetime there is no danger of underestimating indirect energy input. Repair and maintenance consist of many small energy items and this is why the determination of sequestered energy based upon monetary costs is commonly practised. It would be too laborious to trace all energy flows by means of a process analysis. Monetary cost of repair and maintenance is transformed into an energy requirement by multiplying monetary cost by an appropriate value of energy intensity. Energy intensities are derived from national statistics of energy use and production. The easiest way to take into account the indirect energy input of self-propelled machines is to include it in the fuel consumption. The lifetime fuel consumption is estimated and the lifetime indirect

energy input is divided by the fuel consumption. Each consumed litre of fuel per hectare is added to the indirect energy input. For implements, indirect energy input can be tied to hectares. Indirect energy input is calculated by summing the lifetime indirect energy input and dividing it by the lifetime capacity. In bioenergy production one hectare is the most practical metric of capacity. Every time a hectare of field has been worked the process is charged by the hectare-specific energy input. The weight, capacity and lifetime of the implement have the greatest impact on the hectare-specific energy input. There are no up-to-date studies concerning the indirect energy input for agricultural machinery, i.e. manufacturing energy and the energy needed for repair and maintenance. This would require research efforts in order to reduce uncertainty related to energy analyses for different bioenergy crops. There is a danger that energy analyses give too positive results if indirect energy input is underestimated. This same problem is associated with agricultural buildings and machinery used inside buildings. References [1] Sfs-En Iso 14040. Environmental management. Life cycle assessment. Principles and framework. Finnish Standards Association. 48 p. [2] Sfs-En Iso 14044. Environmental management. Life cycle assessment. Requirements and guidelines. Finnish Standards Association. 96 p. [3] Fluck RC. Energy sequestered in repairs and maintenance of agricultural machinery. Trans ASAE May–June 1985;28(3). [4] Loewer Jr. OJ, Benock G, Gay N, Smith EM, Burgess S, Wells LG, et al. Beef: production of beef with minimum grain and fossil energy inputs, I, II, III. Report to NSF. Washington, DC; 1977. [5] Berry RS, Fulton Fels M. The production and consumption of automobiles. An energy analysis of the manufacture, discard and reuse of the automobile and its component materials. A report to the Illinois institute for environmental quality. Department of Chemistry, University of Chicago; July 1972. 74 p. [6] Pimentel D, Hurd LE, Bellotti AC, Forster MJ, Oka IN, Sholes OD, et al. Food production and the energy crisis. Science 2 November 1973;182(4111):443–9. [7] Bowers W. Agricultural field equipment. In: Stout BA, Fluck RC, editors. Energy in world agriculture. Energy in farm production, vol. 6. Elsevier; 1992. p. 117–29. [8] Conforti P, Giampietro M. Fossil energy use in agriculture: an international comparison. Agric Ecosyst Environ 1997;65:231–43. ˜ avate J, Hernanz JL. Energy for biological systems. In: Kitani O, [9] Ortiz-Can Jungbluth D, Peart RM, Ramdani A, editors. Energy and biomass engineering, vol. v. CIGR – The International Commission of Agricultural Engineering; 1999. p. 13–24. CIGR handbook of agricultural engineering vol. I–V. American Society of Agricultural Engineers, ASAE. [10] Maclean H, Lave L. A life-cycle model of an automobile. Environmental Policy Analysis 1998;3(7):322A–30. [11] Schweimer G, Levin M. Life cycle inventory for the Golf A4. VW environmental report 2001/2002, http://www.mobilitaet-und-nachhaltigkeit.de/_download/ sachbilanz_golf_a4_englisch.pdf; 2002 [Read 24.5.07].

28

H.J. Mikkola, J. Ahokas / Renewable Energy 35 (2010) 23–28

[12] Worrell E, Price L, Martin N. Energy efficiency and carbon dioxide emissions reduction opportunities in the US iron and steel sector. Energy 1999;26(2001):513–36. [13] Bo¨rjesson PII. Energy analysis of biomass production and transportation. Biomass and Bioenergy 1996;11(4):305–18. [14] Farla JCM, Blok K. The quality of energy intensity indicators for international comparison in the iron and steel industry. Energy Policy 2001;29:523–43. [15] Fluck RC, Baird CD. Agricultural energetics. Westport, CT: AVI Publishing Co., Inc.; 1980. 192 p. [16] Kepner RA, Bainer R, Barger EL. Principles of farm machinery. 3rd ed. Westport, CT: AVI Pub. Co., Inc.; 1978. 527 p. [17] Herendeen R, Bullard C. CAC document 140. Centre for Advanced Computation, University of Illinois; 1974. [18] Bullard CW, Penner PS, Pilati DA. Net energy analysis. Handbook for combining process and input–output analysis. Resour Energ 1978;1:267–313. [19] Smil V, Nachman P, Long TV. Energy analyses and agriculture. Boulder, CO: Westview Press, Inc.; 1983. 191 p. [20] Foster P, Flemming J, Wichelns D. Energy accounting: the case of farm machinery in Maryland. J Agr Resource Econ 1980;12:189–92. [21] Van Hecke E. A regional approach to the analysis of Belgian agricultural energy use and production. Energ Agr 1982;1(2):117–30. [22] Doering III O.C., Considine TJ, Harling CE. Accounting for tillage equipment and other machinery in agricultural energy analysis. Agr. Exp. Sta., Purdue University, W. Lafayerre, IN, 1977. 4 p. [23] Leach G. Energy and food production. Guildford, Surrey, England: IPC Science and Technology Press, Ltd; 1976. 137 p. [24] Bainer R, Kepner RA, Barger EL. Principles of farm machinery. John Wiley & Sons, Inc..; 1965. [25] Culpin C. Profitable farm mechanization. London: Crosby Lockwood & Son, Ltd; 1968. 321 p. [26] Srivastava AK, Goering CE, Rohrbach RP, Buckmaster DR. Engineering principles of agricultural machines. 2nd ed. American Society of Agricultural and Biological Engineers; 2006. 588 p.

[27] ASAE D497.5 FEB2006. Agricultural machinery management data. American Society of Agricultural and Biological Engineers; 2006. [28] Bak F, Jensen MG, Hansen KF. Forurening fra traktorer og ikkevejgående maskiner i Danmark. Teknologisk Institut; 2003. Miljøprojekt nr. 779. 70 p. [29] Farm machinery cost guide. Alberta Government [www-document], http:// www1.agric.gov.ab.ca/$department/deptdocs.nsf/all/inf10589; 2005 [Read 12.6.07]. [30] Schnitkey G, Lattz D. Machinery cost estimates: tractors. University of Illinois; 2005. Farm business management handbook. [pdf-document], http://www. farmdoc.uiuc.edu/manage/pdfs/Mach_tractors_2005.pdf Read 15.6.2007. [31] Profitability book-keeping data of Finnish agriculture (Farm Accountancy Data Network – FADN). Agrifood Research of Finland, Economics Research; 2005. [32] Rinaldi M, Erzinger S, Stark R. Treibstoffverbrauch und Emissionen von Traktoren bei landwirtschaftlichen Arbeiten. FAT-Schriftenreiche; 2005. 65 p. [33] Kunze RF. Das Neue Traktorlexikon. Landwirtschaftsverlag GmbH, 48084 Mu¨nster, 4. Auflage. 175 p. [34] Traktoreiden ryhma¨koetus. [Group test of tractors]. Valtion maatalouskoneiden tutkimuslaitos. Koetusselostus [Test report] 1132; 1984. 66 p. [35] Traktoreiden ryhma¨koetus [Group test of tractors]. Valtion maatalouskoneiden tutkimuslaitos. Koetusselostus [Test report] 1262; 1988 51 p. [36] Lindgren M. A methodology for estimating annual fuel consumption and emissions from non-road mobile machinery. The Swedish University of Agricilture, Department of Biometry and Engineering; 2007. Rapport – miljo¨, teknik och lantbruk 01. 43 p. [37] Traktoreiden ryhma¨esittely. [Group presentation of tractors for sale 2005]. Koneviesti 2005;4:53. [38] Maatalous - ja metsa¨koneiden myynti 2000–2002. [Sales of agricultural and forestry machines in Finland 2000–2002] [www-document], https://portal. mtt.fi/portal/page/portal/www/Palvelut/Vihdin%20palvelut/Mittaus%20ja% 20testaus/TILASTOKOKO.pdf [Read 21.2.08]. [39] Ryhma¨esittely, keva¨ttyo¨koneet. [Group presentation of field machinery for sale]. Koneviesti 2005;1:53.