Ewen Coxworth, Consultant, 1332 10th Street East, Saskatoon, Saskatchewan, S7H 0J3.Report to the Canadian Agricultural Energy Use Data and Analysis Centre, Department of Agricultural Economics, University of Saskatchewan, Saskatoon, SK S7N 0W0.October, 1997. Revised, November, December, 1997.

1. Introduction.

There have been a number of reasons for interest in energy use in the economy, and in agriculture in particular. In the early 1970s, large rises in the price of crude oil and its products, such as gasoline and diesel fuel, raised concerns about the effects of increased fuel costs on the profitability of farming. In the 1990s, the price of fuel has come down (in constant dollars), but concerns about the effects of burning large amounts of fossil fuels on the atmosphere and on climate have risen.

Burning of fossil fuels releases carbon dioxide into the atmosphere, carbon dioxide originally sequestered by plants millions of years ago. This has the effect of increasing the concentration of carbon dioxide in the atmosphere. Carbon dioxide is a greenhouse gas (GHG), in the sense that increasing concentrations in the atmosphere can reduce heat loss from the planet and change world climates. Other gases also can cause a change in climate if their concentration increases in the atmosphere. These include methane and nitrous oxide, both of which are released into the atmosphere by human activities, including nitrogen fertilization of crops (nitrous oxide) and production of animals (methane released from manure and from enteric fermentation in ruminant animals).

Such increases in GHGs are predicted to have substantial effects on climate, not all of which would be beneficial to agriculture (Environment Canada, 1997). Predictions indicate the effects on climate from a doubling of greenhouse gases could be: - increased severity and frequency of droughts, - increased variability of crop yields, - increased severity and frequency of storms, - changes in disturbances and increased susceptibility to pests and diseases.

The period since 1990 has seen a considerable increase in greenhouse gas emissions in the Canadian economy (Environment Canada, 1997; Jaques et al., 1997). Most of Canada's greenhouse gas (GHG) emissions are related to the use of energy; thus the increase in GHGs indicated an increase in energy use in the Canadian economy. Total GHG emissions (81` due to carbon dioxide, 5t nitrous oxide, 12t methane, 1.5` other greenhouse gases) increased by 9` in the period of 1990 to 1995 inclusive. In 1990 total GHG emissions, reported as carbon dioxide equivalents, were 567 Mt (million metric tonnes). They rose to 619 Mt in 1995. Analysis of the results indicated that the increase was due to population growth and an expanding economy. There was an improvement (reduction) in energy intensity, indicating an improvement in energy efficiency, among the various factors affecting energy intensity. Environment Canada (1997) calculated that Canada's GHG emissions would have been 3.5% higher if it had not been for the reduction achieved in energy intensity during the period 1990 to 1995. Energy intensity could refer to the amount of energy required to produce a unit of some output, such as nitrogen fertilizer, or the energy required to produce a unit of grain.

Some studies on GHG emissions from agriculture have focussed on emissions from soils and from livestock and their wastes (e.g., Jaques et al., 1997). These emissions are mainly nitrous oxide emitted from soils as a result of the use of nitrogen fertilizers and the inclusions of legumes in crop rotations (Mosier et al., 1996) and methane from digestion processes in animals and the decomposition of manure (Jaques et al., 1997). Soils can act both as a source of carbon dioxide emissions (a decline in soil organic matter) and as a sink for carbon dioxide (an increase in soil organic matter). The present study focusses on the energy used, and indirectly on the carbon dioxide emissions, from the inputs used for food and feed production (e.g., liquid fuels for cars, trucks, tractors and combines, heating fuels and electricity, fertilizers, pesticides, and the energy required to manufacture and repair machinery and buildings). Both direct and indirect uses of energy are included.

What has happened to energy use in agriculture in Canada in the period 1990 to 1996? Energy use (directly as fuels or indirectly as the fuels used to make fertilizers and machinery, for example) is a significant component of agriculture's GHG emissions. Energy use contributed mainly to agriculture's carbon dioxide emissions. As mentioned earlier, other sources of GHG emissions included (a) nitrous oxide emissions from soils as a result of treatment with fertilizers, manure or legumes, and, (b) methane emissions from livestock manures and from enteric fermentation in ruminant animals. Environment Canada (1997) calculated that nitrous oxide was the primary GHG in agriculture in 1991, accounting for 45` of the GHG emissions from the total agriculture sector, followed by methane (30~) and carbon dioxide (25~). Although the total weight of nitrous oxide emitted from agricultural activities is much less than the weight of carbon dioxide emitted, a molecule of nitrous oxide has 310 times the global warming potential of a molecule of carbon dioxide (100 year time horizon). Thus small emissions of nitrous oxide from agriculture have a large global warming effect.

International protocols for reporting GHGs requires that some of the GHG emissions from inputs into agriculture are reported as part of energy production, or as part of manufacturing, e.g., pesticides and machinery manufacture (Jaques et al., 1997). In the present study, the focus is on the total impact of agricultural activities on energy use in the Canadian economy. Thus it includes most of the inputs into agriculture.

2. The Present Study.

Sources of Energy Use The present study examined energy use trends from the direct and indirect uses of energy in agriculture for all of Canada for the period 1990 to 1996 inclusive. Energy use in agriculture included the following:

Direct use of energy:

- all fuels (gasoline and diesel fuel) for transportation (trucks

and cars used for farm business) and operation of field equipment

(tractors and self propelled machinery such as combines and


- fuels used for heating, such as natural gas, propane (natural

gas liquids) and fuel oils,

- electricity for lighting,


Indirect use of energy:

- fossil fuels and electricity used in the production of

fertilizers, either as sources of energy for the processing or as

raw materials (e.g., natural gas used as a raw material in the

manufacture of ammonia),

- fossil fuels and electricity used in the manufacture and

maintenance of farm machinery,

- fossil fuels used in the manufacture of farm buildings,

- fossil fuels and electricity used in the manufacture of

pesticides, including the fossil fuels used as feedstocks.

heating and operation of powered

Energy used in the production of seed (used for seeding) and feeds for livestock was not included. It was not clear whether inclusion of such data might lead to double counting. Some provincial studies have included such indirect energy costs (e.g., Stirling and Kun, 1995).

Trends in agricultural use of input energy (in the forms outlined previously) were analyzed for the period 1990 to 1996 inclusive.


3. Materials and Methods.

Sources of Data. Data on energy use in agriculture for a number of different energy sources was obtained from yearly data presented in Statistics Canada Catalogue 57-003: refined petroleum products, principally gasoline and diesel fuel, natural gas, natural gas liquids (propane, etc.), and electricity. Data in this Catalogue were also obtained for the two main regions of Eastern Canada (Newfoundland, Nova Scotia, New Brunswick, Prince Edward Island, Quebec and Ontario) and Western Canada (British Columbia, Alberta, Saskatchewan and Manitoba).

Farm Energy Use Survey 1996. The data in Statistics Canada Catalogue 57-003 reports on the total amounts of energy consumed on the farm; it does not distinguish between energy used for crop and animal production and energy used for personal purposes by the farm family (Korol, 1997). The Farm Energy Use Survey (FEUS) for the crop year 1996 (FEUS, 1997) was used to estimate the percentage of total

agricultural energy (fuels, electricity) reported in Statistics Canada Catalogue 57-003 which was actually used for farm business (production of crop and animal products), the rest being used for personal uses, e.g., heating the farm home, using cars for buying groceries in town, etc.

A large sample of farmers across the country filled out a matrix form which asked for details of energy expenses as follows:

Sources of energy expenses: gasoline, diesel fuel, LPG, natural gas, electricity, stove and furnace oil, wood purchased for fuel, and other sources of energy.

For each energy source, farmers filled in details about the expenditures, amounts purchased, and percents used for personal use, and, for farm business use, the percents used in trucks and automobiles, mobile farm machinery (tractors, etc.), building heating and lighting, and other uses such as crop drying, motors, irrigation pumps.

The only previous survey was conducted in 1981 (FEUS, 1983). There has been no interim surveys between these two dates. As a rough estimate of changes (trends) in the percentage of personal uses of energy, it was assumed that changes in the percent of total energy use (for a particular energy source) assigned to farm business were linear with time. The total percentage change in farm business use of gasoline (for example) between the 1981 and the 1996 surveys was divided by 15 to obtain an estimate of percent change per year. This annual change was used to calculate the amount of energy per year used for business purposes.


The FEUS did not ask questions of farmers about the percentage of farm costs for machinery and building repairs and depreciation which was attributable to farm business operations, as distinguished from personal costs, e.g., repairs to the farm house. In the absence of a way to separate actual farm business costs from personal costs, it was assumed in the present study that all costs reported for farm machines and buildings in Statistics Canada catalogues were related to farm business. It is believed that most of the machinery costs are associated with farm business activities.

The 1996 FEUS also asked farmers in the prairie region questions about their tillage practices on summerfallow and cropped land, basically whether they used zero till, minimum till or conventional till. These data for Saskatchewan were compared with similar data collected in the 1991 Agricultural Census to obtain an idea of trends in tillage practices in Saskatchewan.

Energy Units Employed. Energy units (MJ/kg or MJ/L) were the same as those used in Statistics Canada Catalogue 57-003. Many studies of energy use in agriculture include a factor to account for the energy used to extract a particular fossil fuel source from the ground, refine it, and transport the final product to the end user (e.g., Coxworth et al., 1995). Southwell and Rothwell (1977) refer to this factor as the Energy Resource Depletion Value (ERD). These factors have been deleted in the present study to minimize differences from procedures used in national GHG studies (e.g., Jaques et al., 1997). In GHG studies, these fossil fuel resource extraction and refining costs would be assigned to the energy sector of the national economy. Thus in the present study, energy costs for the fossil fuels and electricity used as such in agriculture, and used to manufacture fertilizers, pesticides, machinery and buildings, do not contain energy resource depletion factors.

Petroleum and natural gas products. Data on the energy content of fuels are reported in Statistics Canada Catalogue 57-003 and in the present report as the direct higher heating values. The higher heating values of fuels were generally the same as those used by Jaques (1992). As stated in the previous paragraph, no factors were included to account for the energy used to extract fossil fuels from the ground, refine them into final products (such as gasoline), or to transport the raw materials to refineries or transport final products to their point of use on the farm.

Electricity. Electricity consumption is reported in Statistics Canada Catalogue 57-003 using the conversion of 1 GWh = 3.6 TJ (tera joule = 1012 joules). This conversion factor was used in the present study. This conversion factor does not take into account the thermal energy, such as in the form of coal, required to generate the electricity. About three joules of coal thermal energy may be required to generate one joule of electrical energy. Electricity generated in different provinces may have very different carbon dioxide emission consequences, if electricity in one province is generated principally from coal (e.g., Alberta) and in another province principally from hydra power (e.g., Manitoba or Quebec). Hydroelectric or nuclear generation of electricity does not produce significant amounts of carbon dioxide emissions.

Fertilizers. The energy to produce different nitrogen, phosphorous and potassium fertilizers was calculated using similar methods to those described previously (Coxworth et al., 1994, 1995). These methods were based on the report of Bhat et al. (1994), which in turn was based on detailed reports of energy use in the modern mix of fertilizer plants in North America. Previous calculations of the energy used in the manufacture of fertilizers (Coxworth et al., 1994, 1995) included energy resource depletion factors for all fuels and electricity employed. To avoid double counting in national and provincial studies of energy use in the economy, these factors were removed in the present study. As stated earlier, in national and provincial studies, energy use in the energy industries would be reported as part of the activity of the energy industry (e.g., Jaques, 1992), rather than being transferred to an end product, such as fertilizer. The use of factors allows one to estimate the total impact of a change in energy use in agriculture; removal of the factors was considered more appropriate for estimating trends in energy use in agriculture as part of the whole economy. Included in the energy use calculations for fertilizers was the energy used to transport fertilizer raw materials to processing centres (e.g., phosphate rock), package final products and move products to retail centres for distribution to farms.

Data on fertilizer consumption by province, by eastern and western regions, and for the whole country, per year, was obtained from the yearly reports of Agriculture and Agri-Food Canada entitled: Canadian Fertilizer Consumption, Shipments and Trade, published by the Farm Input Markets Unit, Agriculture and Agri-Food Canada, Ottawa, Ontario, K1A 0C5.

Machinery and Buildings. Energy is required to produce the steel and rubber used in farm machinery and maintain and repair the machinery throughout its useful life. In similar fashion, energy is required to produce the concrete, steel and wood used in the manufacture of farm buildings. The energy for manufacture is divided by the estimated lifetime of the equipment to obtain an average yearly energy cost. A detailed study was conducted by Stirling (1979) and Stirling and Kun (1995) of the energy used in farm machinery and buildings in Saskatchewan, and trends in energy use for the period 1976 to 1990. Stirling and Kun (1995) used energy factors (ERDs, see discussion on page 5) for all the energy inputs, such as gasoline and natural gas, used in the manufacture of machinery and buildings. It was assumed that an average ERD factor was 1.20 across all energy inputs used. Thus the total energy used for machinery manufacture and maintenance in 1990 was 21.22 PJ, according to Stirling and Kun (1995). This was reduced to 17.68 PJ when estimated factors (ERDs) were removed.

To calculate trends in energy use an economic model was used by Stirling and Kun (1995), in that energy use was assumed to be proportional to economic costs (depreciation and repair for machinery and depreciation and repair for buildings) each year, in constant dollars. Stirling and Kun's data for 1990 were used in the present study as the starting point for calculation of trends and calculations of energy use in farm machinery and buildings across Canada for the period 1990 to 1996. It was assumed that machinery and building energy costs across Canada were proportional to Saskatchewan energy costs, and changes each year were proportional to relative economic costs. Economic costs for each province and for the whole country are reported in Statistics Canada Catalogue 21-603. The price indices for machinery repair and depreciation and building repair and depreciation were used to adjust yearly costs to 1990 values. The price indices for machinery (depreciation and repair), and buildings (depreciation and repair) are reported in Statistics Canada Catalogue 62-004.

Example: Machinery repair and depreciation costs in Saskatchewan in 1990 were 17.68 PJ (Stirling and Kun, 1995, corrected to remove ERD factors on fossil fuels). According to Statistics Canada Catalogue 21-603, machinery repair and depreciation costs were $1,001,000,000 for Saskatchewan in 1990 and $3,910,000,000 for Canada as a whole. Therefore, it was estimated that machinery energy costs were 69.06 PJ for Canada. In 1996, machinery depreciation costs were $2.970 billion for all of Canada. The price index was 134.0 compared to 103.7 for 1990 (Statistics Canada uses 1986 as the base year). Thus corrected costs for 1996, adjusted to the 1990 base year, would be $2.298 billion (103.7/143.0 X $2.970 billion). Adjusted repair costs were $1.516 billion. Total machinery costs for 1996 were therefore $3.814 billion. By proportionality to 1990 energy costs and economic costs, energy costs in 1996 were 67.36 PJ (3.814/3.910 X 69.06).


In an earlier study, it was estimated that the total amount of

herbicides used by Saskatchewan farmers in 1990 (average year

1989 to 1991) was 8,800,000 kg of active ingredients (Coxworth et

al., 1995). This is quite close to the value of 9,246,000 kg

active ingredients of herbicides reported, in a 1990 survey, for

the province of Saskatchewan for all users (agriculture, forestry, home and garden)(Agriculture Canada and Environment Canada, 1991). Insecticide and fungicide use in Saskatchewan was small (417,000 kg active ingredient, all users). The 1990 survey did not separate agricultural use from other users. As a rough approximation, we estimated total pesticide use by agriculture in Saskatchewan at 9 million kg of active ingredients in 1990.

Based on an earlier study (Coxworth et al., 1995), 8.8 million kg of herbicide had an energy cost of 1.72 PJ (including ERD factors), or 196 MJ/kg. Costs of pesticides in 1990 in Saskatchewan were $194,086,000 (Statistics Canada Catalogue 21603). Thus 9 million kg would use 1-76 PJ, or 0.907 PJ/$100 million. If ERD factors add 20` to the energy costs, then energy use, without factors, would have been 0.756 PJ/$100 million.

It was assumed that pesticide costs for the whole of Canada were proportional to Saskatchewan costs, and that the average energy cost of pesticides, per kg, did not change significantly from year to year. The pesticide price index (Statistics Canada Catalogue 62-004) was used to adjust costs for years 1991 to 1996 to 1990 levels. Energy costs for Canada for each year were then calculated in a similar fashion to the method described for machinery and buildings.

4. Results.

4.1. Amounts of Energy Used on the Farm for Business Purposes. Table 1 shows the percentages of total energy used on Canadian farms in 1980 and 1996 which were employed for farm business purposes.

Table 1. Percentages of energy used on farms employed for farm business purposes. Comparison between 1981 and 1996.

Energy item Percent farm business Percent change per

1981 1996 year, 1981-96

Gasoline 85.2 74.1 -0.74

Diesel fuel 99.1 88.4 -0.71

NGLs (LPG) 75.6 48.4 -1.81

Natural gas 71.7 57.3 -0.96

Electricity 66.8 66.1 -0.047

Fuel oils 40.8 14.4 -1.76

All sources of energy showed a decline in the percentage used in 1981 for farm business compared to the percentage used in 1996. Readers should note that the questions used in the 1981 survey may have been prone to some ambiguity in interpretation; therefore the percentages reported for farm business uses in 1981 may only be approximate (Korol, 1997). Nevertheless, the trends seem to show a decline in the percentages of all energy sources (with the possible exception of electricity) which were used for farm business purposes between 1981 and 1996.

4.2. Liquid Fuels. Gasoline and diesel fuel constitute the majority of the uses of refined petroleum products in Canadian agriculture, although some provinces use considerable amounts of heating oils. Total refined petroleum use and gasoline and diesel were all plotted year by year, and for western and eastern regions separately to determine trends. Gasoline and diesel fuel were plotted together to determine trends in total liquid fuels used for transport and powered field equipment. The amount of energy attributable to farm business activities was also determined.

Gasoline, diesel fuel and gasoline and diesel fuel use combined are shown in Tables 2a,2b,2c,2d and 3a, 3b and are displayed graphically in Figures la,lb,2a,2b,2c and 2d (figures are placed at the end of the report).


Table 2a. Gasoline use (PJ) in Canadian agriculture (business and personal). 1990 to 1996.

Canada or region 1990
















Eastern region









Western region









Note: numbers may not add exactly to totals because of rounding.


Table 2b. Gasoline use (PJ) in Canadian agriculture (farm business only). 1990 to 1996.

1990 1991 1992 1993 1994 1995 1996

Canada 44.0 37.9 31.4 27.4 25.3 30.2 32.4



Table 2c. Diesel fuel use (PJ) in Canadian agriculture (business and personal). 1990 to 1996.



















Eastern region









Western region









Table 2d. Diesel fuel use (PJ) in Canadian agriculture. Business use only. 1990 to 1996.

1990 1991 1992 1993 1994 1995 1996


Canada 66.2 61.8 62.5 69.3 75.9 79.0 84.6


Table 3a. Gasoline and diesel fuel use (PJ) combined. Farm business and personal use combined. 1990 - 1996.

Canada or region




































Table 3b. Gasoline and diesel fuel use (PJ) combined. Farm business use only. 1990 to 1996.




















Gasoline use declined over the period as a percentage of transportation fuels used in agriculture. Gasoline use declined until 1994, and increased in 1995 and 1996 (Tables 2a,2b and Figures 2b,2d). Diesel fuel use declined slightly from 1990 to 1991 and then increased to 1996 (Table 2c,2d and Figures 2a,2c). Total liquid fuels (gasoline and diesel fuel combined) declined from 1990 to 1992 and then rose to 1996 (Tables 3a,3b and Figures la,lb). Effects were most pronounced in western Canada.

Historical data on liquid fuel use (farm business and personal uses combined) in agriculture since 1978 is shown in Figures 3 and 4 (data obtained from Statistics Canada Catalogue 57-003). Over this period, fuel use was highest in 1979 and then declined substantially until 1984. (Data for western Canada and for all of Canada for 1983 is anomalously high, suggesting some sort of data collection problem, and is not shown). The long term trends show a decline in the use of gasoline and an increase in the use of diesel fuel.

The trends in diesel fuel use in agriculture from 1990 to 1996 followed a similar Pattern to total diesel fuel use in Canada.from that total uses time only slightly from 1,002 Diesel fuel use for road transport, urban transit and retail pump sales increased from 241.6 PJ in 1990 to 329.1 PJ in 1995 (data Statistics Canada Catalogue 57-003). It is also noteworthy agricultural use of diesel fuel is a significant fraction of l diesel duel use in Canada (compare 329.1 PJ total transport with 88.7 PJ for agriculture in l99S, see Table 2c). In the period 1990 to 1995 retail gasoline sales in Canada changed PJ in 1990 to 1,121 PJ in 1995. Thus the largest increase in transportation fuel use has been in diesel fuel, both in the Canadian economy as a whole and in agriculture.

In table 4a is shown total energy used in refined petroleum products employed in Canadian agriculture (business and personal) for the period 1990 to 1996. The main components (gasoline, diesel fuel and light fuel oil) are also shown. Table 4b shows the same type of data but refers to farm business use only. Minor uses of refined petroleum products (kerosene, etc.) were assumed to be entirely farm business.

Table 4a. Agricultural energy use (PJ) of refined petroleum products for the period 1990 to 1996. Business and personal uses combined.

Product 1990 1991 1992 1993 1994 1995 1996

All refined

Petroleum products 139.8 132.5 157.5 126.8 131.1 145.3 155.5

of which:

Gasoline 56.1 48.4 40.7 35.9 33.4 40.4 43.7

Diesel ftl 71.4 67.2 68.5 76.6 84.5 88.7 95.7

Light fuel

oil 10.8 14.7 45.1 11.9 10.8 14.3 14.0


Table 4b. Agricultural energy use (PJ) of refined petroleum products for the period 1990 to 1996. Farm business uses only.

Product 1990 1991 1992 1993 1994 1995 1996

All refined


products 114.4 105.1 106.8 101.4 105.5 113.4 121.1

Of which:

Gasoline 44.0 37.7 31.4 27.4 25.3 30.2 32.4

Diesel f'1 66.2 61.8 62.5 69.3 75.9 79.0 84.6

Light fuel oil 2.7 3.4 9.7 2.3 1.9 2.3 2.0

The light fuel oil use in 1992 appears to be unusually high compared to data from other years. The reasons for this have not been investigated.

4.2.1. Effects of changes in tillage practices and reduction in summerfallow area in Saskatchewan on fuel use. Data provided from the 1997 Farm Energy Use Survey (1996 crop year)(Agriculture and Agri-Food Canada, Ottawa, Ontario, provided by M. Korol, Economist, Policy Branch) indicated an increase in land cropped or fallowed using zero tillage techniques, compared to tillage systems in 1991 using more minimum or conventional tillage (Table 5). This shift might be expected to decrease fuel usage, the opposite effect to what was revealed by the statistics for western Canada and for Saskatchewan. Between 1990 and 1996, Saskatchewan agriculture increased consumption of gasoline and diesel fuel (combined) from 40.0 PJ to 42.6 PJ, an increase of 2.6 PJ. But between 1990 and 1996, the area in summerfallow decreased from 5.87 million ha to 4.45 million ha, a decrease of 1.42 million ha (Saskatchewan Agriculture Statistics). These two effects (less tillage, less summerfallow) may work in opposite directions as far as fuel use is concerned.

Table 5. Changes in tillage practices on Saskatchewan farms between 1991 and 1996.

Crop or fallow and tillage system

Percent of total land in fallow or crop

1991 1996


Chemical fallow (no till) 4.1 8.7

Tillage only 56.8 57.4

Tillage and chemicals

(herbicides) 39.1 33.9


Several tillage passes 63.9 51.4

Till just before seeding 25.7 30.2

No till before seeding 10.4 18.4


For example, a four-year study of tillage systems at Indian Head, in the thin Black soil zone, found that fuel use was influenced by tillage systems as shown in Table 6. The data reported by Lafond et al. (1993) recorded fuel used by tractors only. To thi was added the fuel used by heavy trucks for hauling grain and fertilizer on the farm (Rutherford and Gimby, 1987).


Table 6. Effects of tillage systems and fallow on fuel usage for spring wheat production (adapted from Lafond et al., 1993).

Tillage system Fuel use (L per hectare per two years)

Crop yr. Two stubble crop

+ fallow yr. years

Zero tillage 26.1 42.7

Minimum tillage 34.2 55.3

Conventional tillage 49.3 69.3


The main reduction in summerfallow area in Saskatchewan since 1990 has been reported to be in the Dark Brown and Brown soil zones (Johnston, 1997). In these areas tillage systems for fallow or crop are likely to be minimum tillage, i.e., a combination of tillage and herbicides used to control weeds. Thus a change from minimum tillage crop-fallow (34.2 L per hectare per two years) to zero tillage continuous cropping (42.7 L per hectare per two years) would actually increase fuel use (see Table 6). Furthermore, transportation costs of grain from farm to elevator would be expected to increase with more crop to haul (less summerfallow, more land in crop).

It is known that shifting from conventional tillage systems to zero tillage systems increases soil water storage (Lafond et al., 1992). Thus a move to zero tillage might be expected to encourage more continuous cropping in the Dark Brown and Brown soil zones (more water stored for a following crop) which is what seems to have occurred. Further evidence is needed.

4.3. Fertilizers. Energy used to produce various fertilizers is shown in Table 7. These data do not include the use of ERD factors on the fossil fuels employed as energy sources or as feedstocks.

Table 8 shows actual sales of fertilizer nutrients for Canada and the eastern and western regions for 1990 to 1996.

Table 9 shows the calculated energy costs of fertilizer nutrients sold in Canada for the years 1990 to 1996, based on the data provided in Tables 7 and 8.


Table 7. Energy used for the production of fertilizers.

Fertilizer and nutrient

Energy use (GJ/tonne nutrient)

Nitrogen Ammonia 57.62

Urea 76.14

Ammonium nitrate 68.55

Urea-ammonium nitrate (N solutions) 65.12

Ammonium sulfate 60.42

Monoammonium phosphate (N only) 59.50

Diammonium phosphate (N only) 59.50

Other nitrogen fertilizers 58.22










Phosphate (P2Os)

Monoammonium phosphate (P only) 12.13

Diammonium phosphate (P only) 12.40

Triple super phosphate 15.05

Other phosphate fertilizers 14.99


Potash (K2O)


Potash 11.15



The energy used for single super phosphate production was assumed to be the same as for triple super phosphate (only small amounts of single super phosphate were sold).

Nitrogen fertilizer consumption has increased substantially in the time period of 1990 to 1996, whereas phosphate and potash consumption has changed very little, whether measured by tonnes or-by PJ of energy expended (Tables 8 and 9, Figures 5 and 6). However, there were striking differences between east and west. Western Canada and eastern Canada showed very different patterns of change in fertilizer consumption. Eastern Canada has shown a decline in fertilizer sales, whereas western Canada has seen a substantial increase in the same time period. The major increase in fertilizer sales has been in Saskatchewan, as discussed later.

Eastern Canada: In eastern Canada, historical data (Canadian Fertilizer Consumption and Trade, 1995/96, various figures in section 2) showed that fertilizer sales tN,P and K) peaked about 1985 and have declined somewhat since that time. For example, nitrogen fertilizer sales were 339,153 tonnes in 1985; in 1996 they were 288,321 tonnes, a decline of 15 percent. The effects were most pronounced in Ontario; there, nitrogen fertilizer sales dropped from 237,409 tonnes in 1985 to 173,884 tonnes in 1996, a decline of 27 percent. During this same time period there was a drop in corn grain and corn silage hectarage from 1,116,000 ha in 1985 to


Table 8. Amounts (thousands of tonnes) of nitrogen, phosphate and potash fertilizer nutrients sold in Canada, eastern Canada and western Canada in the period 1990 to 1996.

Canada or region

1990 1991 1992 1993 1994 1995 1996




Nitrogen 307.6 290.0 290.9 283.6 275.0 284.4 288.3

Phosphate 192.6 189.2 189.4 184.5 170.2 160.1 149.3

Potash 279.2 262.8 246.1 243.8 241.2 219.1 225.0



Nitrogen 888.7 867.8 962.4 1022.2 1131.0 1164.0 1287.9

Phosphate 420.9 389.0 402.8 431.4 471.0 468.3 509.1

Potash 80.6 75.1 64.1 84.0 86.8 90.8 108.2


Nitrogen 1196.3 1157.8 1253.3 1305.8 1405.9 1448.4 1576.2

Phosphate 613.6 578.2 592.2 615.9 641.2 628.5 658.4

Potash 359.8 337.9 310.2 327.6 328.0 309.9 333.3

Phosphate as P2Os; potash as K2O .


862,000 ha in 1996 (Statistics Canada, Field Crop Reporting Series, various volumes). During the same time period soybean hectarage increased from 425,000 ha to 777,000 ha. Based on a nitrogen fertilizer rate for corn of 150 kg N/ha (Southwell and Rothwell, 1977) and a nitrogen fertilizer rate of 11 kg N/ha for soybeans (a legume crop needing very little nitrogen fertilizer), a drop in demand for nitrogen fertilizer of 34,380 tonnes would have been predicted. This could explain about 54 percent of the actual decline. Further investigations into the reasons for the decline in N,P and K fertilizer consumption in eastern Canada, especially Ontario, seems warranted.


Western Canada: Between 1990 and 1996, nitrogen fertilizer sales increased from 888,678 t to 1,287,881 t, an increase of 399,203 t. Thus 1996 sales were 45% higher than 1990 sales. The largest increase was seen in Saskatchewan, where nitrogen fertilizer sales increased from 274,194 t in 1990 to 517,829 t in 1996. The increase was 243,635 t or an 89` increase. In the same time period, the area in summerfallow in Saskatchewan decreased from 5.87 million ha in 1990 to 4.45 million ha in 1996. This decrease in summerfallow indicated an increase in cropped area of 1.42 million ha. If all this new cropped area was seeded to a non-legume crop and 50 kg N/ha was applied, the new demand for N fertilizer would be 71,000 t or about 29% of the actual increase. This suggest that rates of nitrogen fertilizer on other cropped lands were also being increased in the period from 1990 to 1996.

Saskatchewan exports a large portion of all annual crops grown. Doyle and Cowell (1993) calculated the amount of fertilizers which would have to be applied to replace the nutrients included in the crops exported from the Canadian prairies. Using their data on nutrients in crops exported, a rough estimate is that 587,000 tonnes of nitrogen were removed from Saskatchewan, via crops exported, in the 1996 crop year (crop production data from Agricultural Statistics, Saskatchewan Agriculture and Food, 1997). Thus the rapid rise in nitrogen fertilizer sales since 1990 is starting to approach the amount of nitrogen exported from the province.

What would be the situation if western Canada greatly increased livestock production, especially of hogs, in the future? Such an increase is starting to take place already. Some portion of the crops now exported (e.g., barley and peas) would be fed locally to animals, and the manure from the livestock recycled to the soil. Would this imply decreased demands for nitrogen, phosphorous and potassium fertilizer? A study of the fertilizer demand implications and of the GHG implications of a substantial increase in livestock production in western Canada would be warranted.

Table 9. Energy used (PJ) in fertilizers sold in Canada. 1990 1996.

Nutrient 1990 1991 1992 1993 1994 1995 1996


nutrients 91.2 87.6 93.9 98.1 104.5 107.7 116.3

Nitrogen 79.5 76.6 83.0 86.8 93.0 96.5 104.4

Phosphate 7.6 7.2 7.4 7.7 7.9 7.8 8.2

Potash 4.0 3.8 3.5 3.7 3.5 3.5 3.7

It can be seen from Table 9 and Figure 7 that nitrogen fertilizer use dominated the total energy consumed in fertilizer use, and also was responsible for most of the increase in energy expended in fertilizer consumption. The results seem to be explainable by the increase in cropping intensity (more land in crop, less land in fallow) and fertilizer use intensity (more kg of fertilizer per hectare) in western Canada, particularly in Saskatchewan.

An increase in cropping intensity (less summerfallow), increased use of fertilizers and/or legumes and a reduction in tillage would be expected to increase the amount of carbon stored in soils as soil organic matter (Smith et al., 1997, Campbell et al., 1997). A recent Environment Canada study of GHG emissions in Canada calculated that net emissions of carbon dioxide from soils decreased from 7,090,000 tonnes carbon dioxide per year in 1990 to 2,480,000 tonnes carbon dioxide per year in 1995 (Environment Canada, 1997, Table 3.1). This reduction in carbon dioxide emissions would be expected to offset much of the carbon dioxide emitted from the manufacture of fertilizers used in the same time period. However, soils would eventually reach a new equilibrium, with a higher organic matter content. After that time, the carbon dioxide emissions from the use of fertilizers would not be offset by a continued increase in soil organic matter.

4.4. Machinery. The energy expended in the manufacture (measured via the depreciation costs) and repair of farm machinery is shown in Table 10. The data indicate only small changes in the energy invested in the manufacture and repair of machinery used in agriculture.

Table 10. Energy expended (PJ) in the manufacture and repair of farm machinery in Canada. 1990 - 1996.


Energy expenditures (PJ)

1990 69.1

1991 66.6-

1992 66.2

1993 66.1

1994 66.8

1995 67.0

1996 67.4

4.5. Buildings.

The energy expended in building manufacture and repair (measured

via the depreciation and repair costs) is shown in Table 11.

There do not appear to have been major changes in the time period

from 1990 to 1996.

Table 11. Energy expended (PJ) in the manufacture of farm buildings in Canada. 1990 - 1996.

Year Energy expenditures (PJ)

1990 35.9

1991 37.2

1992 36.0

1993 34.2

1994 33.2

1995 33.9

1996 35.7


This study did not examine differences between provinces in energy expenses for buildings compared to machinery. Data in Stirling and Kun's report (1995) indicate that farm building energy in Saskatchewan was relatively less important than machinery energy than is the case for all of Canada, as measured in the present study.

4.6. Pesticides. This project did not succeed in obtaining much data on the weight of pesticide active ingredients sold in Canada for agricultural uses, nor the breakdown into the amounts of different herbicides and other pesticides sold. A 1991 Pesticide Registrant Survey of 1990 sales, conducted by Agriculture Canada and Environment Canada, reported total Canadian sales of pesticides of 34,000 tonnes of active ingredients. This included all users of pesticides (not just agriculture) and included herbicides, insecticides, fungicides and fumigants. Herbicides accounted for about 80` by weight of sales of active ingredients.

For Saskatchewan, the survey reported sales of herbicides of 9,246 kg active ingredients (A.I.), 184 kg A.I. of insectides, 108 kg A.I. fungicides and an undisclosed amount of fumigants (probably low, since most sales were reported to be in other provinces). In another study (Coxworth et al., 1995), it was calculated that agricultural use of herbicides in 1990 was about 8,800 tonnes of active ingredients. Estimated energy for manufacture was 1.721 PJ, including ERD factors for the fossil fuels and electricity used in production. It is believed that a reasonable estimate of total pesticides used by agriculture in Saskatchewan in 1990 would be 9,000 tonnes A.I. As described in the materials and methods section, the estimated energy costs would be 1.760 PJ or about 1.467 PJ without ERD factors. Since pesticide expenses were $194,086,000 in 1990, the energy costs per $100 million would be 0.756 PJ. Total energy expended on pesticides in Canada are shown in Table 12.

Table 12. Energy expended (PJ) for pesticides in Canadian agriculture. 1990 - 1996.

Year Energy expended (PJ)

1990 5.5

1991 4.8

1992 5.0

1993 5.2

1994 6.4

1995 6.8

1996 6.8


The total amount of energy spent on pesticides was low, compared to other energy related costs. There appearded to be a tendency for pesticide energy costs to be rising since about 1991. This seems consistent with the increase in cropped land in Saskatchewan and an increase in land under some form of reduced or zero tillage in western Canada.

4.7. Total Energy Costs for Agriculture. 1990 to 1996. Tables 13a and 13b summarizes all energy costs, direct and indirect, for agriculture for the period under study. Figures 8a,8b show trends in total energy use, and the trends in the major energy consumption sources.

Table 13a. Energy consumption (PJ) in Canadian agriculture. 1990 to 1996. Farm business and personal use combined.

Energy 1990 1991 1992 1993 1994 1995 1996


Natural gas 23.2 23.2 25.1 31.2 23.6 22.9 26.3

NGLs 7.1 5.2 6.8 6.2 4.9 4.4 4.9

Prim. Elec. 34.7 34.3 34.1 34.1 34.8 34.0 38.5

Pet'1 Prod. 139.8 132.5 157.5 126.8 131.1 145.3 155.5

Fertilizers 91.2 87.6 93.9 98.1 104.5 107.7 116.3

Machinery 69.1 66.6 66.2 66.1 66.8 67.0 67.4

Building 35.9 37.2 36.0 34.2 33.2 33.9 35.7

Pesticides 5.5 4.8 5.0 5.2 6.4 6.8 6.8

Total 406.5 391.4 424.6 401.9 405.3 422.0 451.4

NGLs = natural gas liquids (propane,etc.)

Prim. Elec. = primary electricity (GWh X 3.6 = TJ)

Pet'1 Prod. = refined petoleum products, all types.

Table lab. Energy consumption (PJ) in Canadian agriculture. 1990 to 1996. Farm business only.

Energy 1990 1991 1992 1993 1994 1995 1996


Natural gas 14.6 14.4 15.3 18.8 14.0 13.4 15.1

NGLs 4.2 3.0 3.8 3.3 2.5 2.2 2.4

Prim. Elec. 23.0 22.7 22.6 22.6 23.0 22.5 25.4

Pet'1 Prod. 114.4 105.1 106.8 101.4 105.5 113.4 121.1

Fertilizers 91.2 87.6 93.9 98.1 104.5 107.7 116.3

Machinery 69.1 66.6 66.2 66.1 66.8 67.0 67.4

Buildings 35.9 37.2 36.0 34.2 33.2 33.9 35.7

Pesticides 5.5 4.8 5.0 5.2 6.4 6.8 6.8

Total 357.9 341.4 349.6 349.7 355.9 366.9 390.2


4.8. General Comments. Energy consumption have risen since 1991, with the major increases coming from increased consumption of petroleum products and fertilizers. This is the case whether one measures farm business only or farm business and personal use of energy.

This study has not examined energy efficiency, i.e., the energy required per unit of food produced. In 1996, for example, Saskatchewan produced 28,768,200 tonnes of all annual crops (Saskatchewan Agricultural Statistics, 1997), which was 21.5t above the average of the 1991 - 1995 period. This high production was partly due to favorable weather, but was also probably related to more land in annual crops (less summerfallow) and greater use of fertilizers and fuels to achieve this higher production. It would be desirable to calculate trends in energy costs per tonne (or megajoule of food energy) of crop or animal product. This has been done for Saskatchewan by Stirling and Kun (1995) for the period up to 1990. This type of study should be continued into the future.

Analysis of energy use did not include data on what country the energy was expended in. Thus, for example, some of the costs of phosphate fertilizer were probably spent in other countries for the mining, upgrading and transport of phosphate rock to Canadian fertilizer plants. In studies of GHG emissions by different countries, the country of origin of GHG emissions associated with various stages of production of products used in their final form in Canada would need to be determined.

This study has not examined methods to improve the efficiency of use of inputs to produce food in Canada. Possibilities appear to exist. For example, a study for The Netherlands (very intensive agriculture) showed that in the short-term, nitrogen fertilizer use could be reduced cost-effectively (e.g., by reducing losses due to leaching and evaporation) nearly 40` without yield losses (Worrell et al., 1995).

This study has also not examined the relationship between economic costs and prices and energy costs and energy use. For example, pesticides are not a large energy cost item, compared to fuel and fertilizer, for example (see Table 13). But they are a significant economic cost item. Another example: the rise in the price of a tonne of wheat in 1995 and 1996 was probably a factor in the greater use of fertilizers in the period 1995 and 1996.

Other factors may also be important in assessing the environmental role of energy inputs. Greater use of fertilizer, coupled with less land in summerfallow and more land in crop, might be expected to increase soil organic matter, at least until a new equilibrium is established at a higher soil organic matter content. This sequestering of carbon dioxide as increased soil organic matter would offset some of the GHG emissions from the production and use of inputs into agriculture for some period of years, until a new soil organic matter equilibrium was reached.

References and Source of Data.

Agriculture Canada and Environment Canada. 1991. Agriculture Canada - Environment Canada 1991 Pesticide Registrant Survey. Data obtained from the Farm Inputs Markets Unit, Agriculture and Agri-Food Canada, Ottawa, Ontario K1A 0C5.

Bhat, M.G., English, B.C., Turhollow, A.F. and Nyangito, H. 1994. Energy in Synthetic Agricultural Inputs: Revisited. Oak Ridge National Laboratory Report ORNL/Sub/90-99732/2. Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831-6285, U.S.A.

Campbell, C.A., Lafond, G.P., Selles, F., McConkey, B.G., Hahn, D. and Wen, G. 1997. Effect of fertilizer, legumes and cropping frequency on soil organic matter in a long-term rotation -changes after 6 yeats of zero tillage. Proceedings of the Workshop an Greenhouse Gas Research in Agriculture, Quebec City, Quebec, March 12-14. Contact the authors c/o Semiarid Prairie Agricultural Research Centre, Swift Current, SK S9H 3X2.

Coxworth, E., Entz, M.H., Henry, S., Bamford, K.C., Schoofs, A., Ominski, P.D., Leduc, P. and Burton, G. 1995. Studies of the Effects of Cropping and Tillage Systems on the Carbon Dioxide Released by Manufactured Inputs to Western Canadian Agriculture: Identification of Methods to Reduce Carbon Dioxide Emissions. Report to Agriculture and Agri-Food Canada, Lethbridge Research Station, Lethbridge, AB T1J 4B1.

Coxworth, E., Hultgreen, G. and Leduc, P. 1994. Net Carbon Balance Effects of Low Disturbance Seeding Systems on Fuel, Fertilizer, Herbicide and Machinery Usage in Western Canadian Agriculture. Report to TransAlta Utilities, Calgary AB. Available from B. Coxworth, 1332 10th St. E., Saskatoon, SK S7H 0J3.

Doyle, P.J. and Cowell, L.E. 1993. Balance of nutrient inputs (fertilizers) and exports (grains) in Alberta, Manitoba and Saskatchewan. Chapter 1 in Impact of Macronutrients on Crop Responses and Environmental Sustainability on the Canadian Prairies (Editors D.A. Rennie, C.A. Campbell and T L. Roberts) Can. Soc. Agron., Ottawa, Ont. K1P SH4.

Environment Canada. 1997. Canada's Second National Report on Climate Change. Catalogue No. En 21-125/1997E. Available from Environment Canada, Ottawa, Ontario, 1-800-668-6767.

FEUS. 1983. Farm Energy Use 1981. Tables and data comparing the 1981 and the 1997 surveys (crop years 1981 and 1996) were obtained from M. Korol, Policy Branch, Agriculture and Agri-Food Canada, Ottawa, Ontario. Also Statistics Canada Catalogue 21-519.

FEUS. 1997. Farm Energy Use Survey. Tables and data were obtained from M. Korol, Policy Branch, Agriculture and Agri-Food Canada, Ottawa, Ontario.

Jaques, A.P. 1992. Canada's Greenhouse Gas Emissions: Estimates for 1990. Report BPS 5/AP/4. Environment Canada, Ottawa, Ont. Available from Publications, Environment Protection, Conservation and Protection, Environment Canada, Ottawa, Ontario K1A OH3.

Jaques, A., Neitzert, F. and Boileau, P. 1997. Trends in Canada's Greenhouse Gas Emissions (1990-1995). Environment Canada, Ottawa, Ontario K1A OH3.

Johnston, A. 1997. Comments on increasing consumption of fertilizers and increasing frequency of recropping in southern Saskatchewan. Telephone conversation with E. Coxworth, August, 1997. A. Johnston is with Agriculture and Agri-Food Canada, Melfort Research Station, Melfort, SK.

Korol, M. 1997. Personal communication to E Coxworth, November 13, 1997. Economist, Policy Branch, Agriculture and Agri-Food Canada, Ottawa, Ontario.

Lafond, G.P., Losppky, H. and Derksen, D.A. 1992. The effects of tillage systems and crop rotations on soil water conservation, seedling establishment and crop yield. Can. J. Plant Sci. 72: 103-115.

Lafond, G.P., Zentner, R.P., Geremia, R. and Derksen, D.A. 1993. The effects of tillage systems on the economic performance of spring wheat, winter wheat, flax and field pea production in east-central Saskatchewan. Can. J. Plant Sci. 73: 47-54.

Mosier, A.R., Duxbury, J.M., Freney, J.R., Heinemeyer, O. and Minami, K. 1996. Nitrous oxide emissions from agricultural fields: Assessment, measurement and mitigation. Plant and Soil 181: 95-108.

Reddy, A.K.N., Williams, R.H. and Johansson, T.B. 1997. Energy After Rio: Prospects and Challenges. United Nations Publications, United Nations Development Program, United Nations, New York, N.Y. 10017, U.S.A.

Rutherford, A. and Gimby, M. 1987. Farm Fuel Requirements, Enerdemo Results. Saskatchewan Research Council Report No. 199, Publication No. R-812-6-B-87. Saskatchewan Research Council, Saskatoon, SK.

Saskatchewan Agricultural Statistics. 1997. Agricultural Statistics Fact Sheet. Statistics Branch, Saskatchewan Agriculture and Food, Regina, SK S4S OB1.

Smith, W.N., Desjardins, R.L., Pattey, E. and Jaques, A. 1997. Estimated rates of carbon change in agricultural soils in Canada from 1970 to 2010. Proceedings, Workshop on Greenhouse Gas Research in Agriculture, Quebec City, Quebec, March 12-14. W.N.S. is at the Eastern Cereal and Oilseed Research Centre, Central Experimental Farm, Agriculture and Agri-Food Canada, Ottawa, Ontario K1A 0C6.

Southwell, P.H. and Rothwell, T.M. 1977. Report on Analysis of Output/Input Ratios of Food Production in Ontario. Available from the Engineering and Statistical Research Institute, Research Branch, Agriculture and Agri-Food Canada, Ottawa, Ontario K1A 0C6.

Statistics Canada Catalogue Number 21-603. Agriculture Economic Statistics. Statistics Canada, Ottawa, Ontario K1A OT6.

Statistics Canada Catalogue Number 57-003. Quarterly Report on Energy Supply-Demand in Canada. Statistics Canada, Ottawa, Ontario K1A OT6.

Statistics Canada Catalogue Number 62-004. Farm Input Price Index. Statistics Canada, Ottawa, Ontario K1A OT6.

Stirling, R. 1979. Use of Non-Renewable Energy on Saskatchewan Farms. Prepared for the Saskatchewan Science Council (now defunct). Energy Research Unit, Univ. of Regina, Regina, SK S4S OA2.

Stirling, R. and Kun, H. 1995. An Energy Inventory for Saskatchewan Agriculture 1976-1990. Final Report. Energy Research Unit Contribution No. 254. Energy Research Unit, Univ. of Regina, Regina SK S4S OA2.

Worrell, E., Meuleman, B. and Blok, K. 1995. Energy savings by efficient application of fertilizers. Resources, Conservation and Recycling 13: 233-250. (Quoted in Reddy et al., 1997, q.v.).