Economics and Carbon Offset
Potential of Biomass Fuels

This work was awarded through Environment Canada’s Environmental Innovation Program and Supported by CANMET/NRCan

This work was funded by the Federal Panel on Energy Research and Development (PERD) Contract 23341-6-2010/00 1/SQ

Therefore, it appears that on a more nutrient enriched site (Ecomuseum), willow and switchgrass will sequester soil organic carbon, compared to corn. However, soil carbon levels may decrease at a given time, but for what length is unknown.

By using the 1998 data and comparing the perennial systems with the corn system, and averaging the values for both sites, switchgrass was calculated to decrease soil organic carbon levels, at 0-60 cm, by approximately 0.06 t/ha/yr, and SRF willow increased organic carbon, at 0-60 cm, by 1.45 t/ha/yr. In years to come, these figures may be expected to change, as many studies point to perennial cropping systems sequestering less soil carbon (even losing soil carbon) during the first years of establishment, compared with subsequent years.

With the above data, economic, energy and carbon budgets were developed (Chapter 4) for SRF willow and switchgrass cropping practices. It was estimated that soil organic carbon accumulation rates of 104 kg C/ha/yr and 121-144 kg C/ha/yr, for SRF willows and switchgrass, respectively, would be sufficient to offset any carbon emitted during the combustion of fossil fuels used to grow and transport these crops. At higher rates of soil organic carbon accumulations, the crops acted as carbon sinks. With proper site selection and cropping practices, the latter possibility is attainable. When no soil carbon was sequestered by the cropping systems, fossil fuel carbon emissions were computed to be approximately one-twentieth of those of liquid fuel combustion, such as diesel.

Assuming a soil organic carbon accumulation rate of 500 kg C/ha/yr for both SRF willows and switchgrass, along with the carbon being sequestered by switchgrass and SRF willow plant material, simulations were performed. The base case simulations indicated SRF willows to be, overall, more efficient in carbon sequestering than switchgrass.

The energy budget developed during this study also indicated SRF willows to be able to produce more energy per unit of fossil fuel input (output:input ratio of nearly 30) compared with switchgrass (output:input ratio of approximately 20). However, from an economic standpoint, switchgrass was estimated to be less expensive to grow.

The results generated in this study suggest that the production of switchgrass and SRF willows should be carbon neutral, with a possibility for these crops to act as carbon sinks under proper fertility management, compared with corn.

This study cautions that when growing short rotation forestry species and perennial grasses for biomass, while obtaining high yields is of primary concern, care must be taken to maintain soil fertility, lest the loss of soil carbon. The study also indicates that soil organic carbon accumulation is dependent on a number of variables, as well as interactions, such as the annual amount of biomass returned to the soil, as well as the amount of nutrients returned to the soil. It appears that soil carbon accumulation is furthermore dependent on time related factors. The longer the rotations, the better the chances of increasing the soil organic carbon pool.

Table of Contents

1. Objectives 2
2. Scope 2

                    Chapter 2. Field Productivity Assessments of Bioenergy Feedstocks 4

1. Site Background and Description 4
2. Cropping Regime 8
1. SRF Willows 8
2. Switchgrass 8
3. Corn 9
3. Growing Conditions 9
4. Harvestable Biomass Yields12
   1. SRF Willows12
   2. Switchgrass14
   3. Corn15
   4. Comparison between Willow, Switchgrass, and Corn16
5. Conclusions17

Chapter 3. Carbon Storage Study 18

                3.1. Methodology 19

                3.1.1. Sampling, Site Description, and History 19
                3.1.2. Modified Mebius Procedure to Determine Carbon Concentration in 1996 22
                3.1.3. LECO Combustion Process to Determine Carbon Concentration in 1998 22
                3.1.4. Calculations of Soil Profile Carbon on a Constant Mass Basis 22
                3.1.5. Statistical Analyses 22
                3.2. Results 24
                3.2.1. Aboveground Plant Yield and Carbon Content 24

1. Switchgrass24
2. Willow25
3. Corn26
4. Comparing the Ecosystems28
   1. Biomass Yields28
   2. Carbon Concentrations30

                3.2.2. Belowground Plant Yield and Carbon Content 32

1. Switchgrass32
2. Willow33
3. Corn33
4. Abandoned Field and Forest34
5. Comparing the Ecosystems35

1. Root Biomass35
2. Root Carbon Sequestered37

2. Soil Carbon40

1. Switchgrass40
2. Willow41
3. Corn42
4. Abandoned Field and Forest43
5. Comparing the Ecosystems44
6. Total Yield and Carbon in199649

2. Discussion51

1. Plant Carbon51
2. Soil Organic Carbon in 199653
   1. 1996 Accumulation Rates56
3. Soil Organic Carbon in 199856
   1. 1998 Accumulation Rates59

                3.4. General Conclusion 60

                Chapter 4. Economic, Energy, and Carbon Budgeting 62

1. Methodology62
   1. SRF Willows62
      1. Economics62
      2. Energy Budgeting65
      3. Carbon Emissions66
   2. Switchgrass69
      1. Economics69
      2. Energy Budgeting71
      3. Carbon Emissions71
2. Results71
   1. SRF Willows71
      1. Economics71
      2. Energy Budget72
      3. Carbon Analysis73
   2. Switchgrass75
      1. Economics75
      2. Energy Budget75
      3. Carbon Analysis77

                Chapter 5. Summary and Conclusions 83

1. Summary83
2. General Conclusions84
3. Directions for Further Research86

                Appendix 1: Cropping Practices for SRF Willows, Switchgrass, and Corn
                Appendix 2: Literature Review for Carbon Storage Study
                Appendix 3: Additional Tables for Chapter 4
                Appendix 4: Model Simulation Results for SRF Willow
                Appendix 5: Model Simulation Results for Switchgrass – Spring
                Appendix 6: Model Simulation Results for Switchgrass – Fall

                Chapter 2: Field Productivity Assessments of Bioenergy Feedstocks

                    Chapter 3: Carbon Storage Study

                    Chapter 4: Economic, Energy, and Carbon Budgeting

A1.1.1 Cropping Practices for SRF Willows – Seedfarm A1-1

A1.1.2 Cropping Practices for SRF Willows – Ecomuseum A1-2

A1.2.1 Cropping Practices for Switchgrass – Seedfarm A1-3

A1.2.2 Cropping Practices for Switchgrass – Ecomuseum A1-4

A1.3.1 Cropping Practices for Corn – Seedfarm A1-5

A1.3.2 Cropping Practices for Corn – Ecomuseum A1-6

Appendix 3: Additional Tables for Chapter 4

Table A4.1: Summary of SRF Willow Results A4-1

Table A4.2: Economic Budget for SRF Willows A4-2

Table A4.3: Energy Budget for SRF Willows A4-3

Table A4.4: Carbon Budget for SRF Willows A4-4

Table A4.5: Simulation Results for SRF Willows A4-5

Appendix 5: Model Simulation Results for Switchgrass – Spring

Table A5.1: Summary of Switchgrass - Spring Results A5-1

Table A5.2: Economic Budget for Switchgrass – Spring A5-2

Table A5.3: Energy Budget for Switchgrass – Spring A5-3

Table A5.4: Carbon Budget for Switchgrass – Spring A5-4

Table A5.5: Simulation Results for Switchgrass – Spring A5-5

Appendix 6: Model Simulation Results for Switchgrass – Fall

Table A6.1: Summary of Switchgrass - Fall Results A6-1

Table A6.2: Economic Budget for Switchgrass - Fall A6-2

Table A6.3: Energy Budget for Switchgrass – Fall A6-3

Table A6.4: Carbon Budget for Switchgrass – Fall A6-4

Table A6.5: Simulation Results for Switchgrass – Fall A6-5

Figure 1.1: Organization of Study 3

Figure 2.1: Ecomuseum Site 6

Figure 2.2: Seedfarm Site 7

Chapter 3: Carbon Storage Study

Chapter 4: Economic, Energy, and Carbon Budgeting

Figure 4.1: Economic Cost Breakdown for SRF Willow 72

Figure 4.2: Carbon Budget Breakdown for SRF Willow 73

Figure 4.3: Carbon Requirements for SRF Willow 74

Figure 4.4: Economic Cost Breakdown for Spring and Fall Switchgrass 76

Figure 4.5: Carbon Budget Breakdown for Spring and Fall Harvested Switchgrass 77

Figure 4.6: Carbon Requirements for Spring and Fall Harvested Switchgrass 78

Figure 4.7: Comparison of Carbon Requirements for Cropping Systems 79

Figure 4.8: Energy Budget for Cropping Systems 81

BTU British Thermal Unit

C elemental carbon

GHG greenhouse gas(es)

GJ Gigajoule = 0.9478 MBTU

cm centimeter

cmol centi-mole = 0.01*1 mole

CO₂ carbon dioxide

g gram

ha hectare = 2.47 acres

HHV higher heating value

K elemental potassium

kg kilogram

Mg Megagram = one metric tonne = 2,200 lbs = 1.1 ton

m metre

mm millimetre

N elemental nitrogen

n.a. not available

odMg oven dried Megagram

odt oven dried tonne (metric)

P elemental phosphorous

ppm parts per million = 1 mg L^-1

SNK Student-Newman-Keuls statistical test for comparisons

SRF short rotation forestry

t tonne

tonne one metric tonne- one Megagram = 1.1 ton

yr year

$ Canadian dollar

^oC degrees Celsius

In Kyoto, Canada pledged to reduce its greenhouse gas (GHG) emissions by the years 2008-2012, to a level at least 6% below its 1990 greenhouse gas emission level (UNFCCC, 1997). The leading GHG is carbon dioxide (CO₂), followed by methane (CH₄) and nitrous oxides (NO_x) (IPCC, 1995). In order for Canada to achieve these reductions in emission rates, several strategies are available: 1. increase energy use efficiency; 2. develop renewable energy sources, and 3. develop new carbon sinks.

When energy is produced using plant biomass, the opportunity exists to address the greenhouse effect through the latter two strategies. During photosynthesis, plants transform CO₂, water, and solar energy into biomass and oxygen. The consequent carbon release into the atmosphere from the burning of biomass stems from the carbon incorporated into the plant tissue that was trapped from the atmosphere during the growing cycle. The carbon released is thus effectively recycled into the plant biomass through photosynthesis. The only net CO₂ loading in a biofuel system results from the use of fossil fuels during the production and processing of the biomass crop. Nonetheless, when perennial biomass crops are used to produce bioenergy feedstocks, the potential exists to sequester carbon in the soil, creating a carbon sink. This sink may even more than offset the release of carbon stemming from the cultivation and processing of the feedstocks.

Perennial biomass crops such as switchgrass (Panicum virgatum L.) and short rotation forestry (SRF) plantations are examples of crops that are able to achieve reductions in atmospheric CO₂ loading through fossil fuel substitution and at the same time possibly act as carbon sinks. Perennial crops also have the potential to reduce soil erosion and nutrient leaching due to their permanent ground cover and more extensive root systems than conventionally grown cash crops. The development of a biomass-to-biofuel industry would benefit the farm economy by introducing new markets for agricultural products, and create jobs in rural regions, as processing plants are likely to be located near the source of raw biomass material.

2. Objectives

The broad objective of this study was to develop economic, energy, and carbon emission budgets associated with the cultivation of switchgrass and SRF willow plantations grown as part of a CO₂ offset strategy in eastern Canada.

Specific objectives were to:

1. evaluate the potential for soil carbon storage, and below- and aboveground plant carbon storage, in commercial switchgrass and SRF willows plantations;
2. conduct an on-site comparison of the carbon sequestering potential of perennial biomass crops to a commercial row crop;
3. develop energy, carbon and economic budgets associated with the production of switchgrass and SRF willows;
4. assess the final carbon offset potential for switchgrass and SRF willow biomass feedstocks using carbon requirement ratios.

Figure 1.1 shows the organization of the study (see next page).

The field experiments for this three-year study were conducted on two 5-hectare sites using three crops: switchgrass (Panicum virgatum L.), short rotation forestry willows(Salix alba x glatfelteri L.), and corn (Zea mays L.). Short rotation forestry willows and switchgrass had been grown on the sites for three years when this project was initiated in 1996. Corn was introduced on the sites in 1996. All three crops were grown on areas of St. Bernard and Chicot fine sandy loam, with some pockets of Chateauguay clay loam. The field productivity assessments presented in Chapter 2 were performed randomly across the plots, regardless of soil type. All sampling in Chapter 3 was restricted to areas of Chicot fine sandy loam. Both experimental sites were located in southwestern Quebec. Therefore all results and recommendations of this study pertain only to similar climatic and soil conditions.

The model used to investigate the economics, energy requirements and carbon emissions associated with the cultivation of SRF willows and switchgrass was developed in accordance with the most recent research information available. Assumptions used and limitations of the analysis are stated in Chapter 4.

Switchgrass and short rotation forestry (SRF) willows were grown on quasi-commercial size plantations, monitored by REAP-Canada, since 1993 in Ste-Anne-de-Bellevue, Quebec. The establishment methods, cropping and maintenance practices, and biomass yields up until, and including, 1995 are described in an earlier report (Samson et al., 1995). This chapter contains some of the background information on the maintenance and harvestable biomass yields attained from 1996 to 1998 for switchgrass, SRF willows and corn.

Two sites were established in 1993 by REAP-Canada to evaluate the adaptability and potential productivity of different varieties of SRF willow and switchgrass, a warm-season perennial grass. Both sites (Ecomuseum and Seedfarm), of approximately 5 hectares each, were located on the Emile Lods Agronomy Research Centre of McGill University in Ste-Anne-de-Bellevue, Quebec (42° 25’N lat., 75° 56’W long.).

The sites were identified as "Ecomuseum" and "Seedfarm", and represented two different levels of productivity at the onset of the experiment, with the Ecomuseum being inherently more fertile, and the Seedfarm less fertile. The Ecomuseum site received regular dairy manure applications of approximately 20 Mg/ha/yr. The field was considered of good quality for row crop production and hence was mostly under corn production for the past twenty years. The Seedfarm site had a hard pan layer at approximately 30 cm, and had some areas which were prone to summer water deficiencies, and therefore had limitations for row crop production, such as corn. Prior to 1993, the field was mostly in a corn-alfalfa rotation.

Both sites were composed mainly of a Chicot fine sandy loam, with some areas of St-Bernard loam, as well as pockets of Chateauguay clay loam. The assessment of soil texture, in 1993, at the two sites revealed sand, silt and clay contents of 48.6%, 35.0% and 16.4%, respectively for the Seedfarm, and 64.0%, 19.5% and 16.6%, respectively for the Ecomuseum (Samson et al., 1995). From these soil compositions, the soil at the Seedfarm was primarily deemed to be a loam, while the Ecomuseum was a fine sandy loam.

Both sites had the same experimental design, which consisted of 6 blocks (replications) at each site, with 3 treatments (willow, switchgrass, and corn) in each block, arranged in a randomized complete block design (RCBD).

The plots used for the corn treatments were former plots of SRF willows that were planted in 1994. These SRF willows were mowed and plowed under at both sites, in the summer of 1995 (as the clones suffered from recurring pest infestations from the willow sawfly (Nematus sp.)). The mowed material, approximately 2-4 od (oven dried) Mg/ha, was left on the surface and incorporated into the soil during the plowing process. In the spring of 1996, these plots were planted to corn, so as to measure carbon storage in an annual row crop (Chapter 3). The layout of the Ecomuseum and the Seedfarm are shown in Figures 2.1 and 2.2.

The SRF willows used in this study were planted in the spring of 1993 and coppiced during the winter of 1996 (January 1996), after 3 years of growth. The coppicing was undertaken to allow the study to begin with a new harvest cycle (cycle 2 in Table 2.4), which is more representative of the potential crop growth the plots have to offer. The stands were considered to be fully established after coppicing. Cropping practices for 1993 to 1995 are presented in detail in Samson et al. (1995) and summarized in Appendix 1.

The spacing between each tree was 0.92 m, giving rise to a planting density of approximately 11,000 trees/ha, with a survival rate of approximately 9,500 trees/ha at the Seedfarm and 10,000 trees/ha at the Ecomuseum site. The coppiced trees at both sites were fertilized in June 1996 with 77 kg N/ha and 84 kg K/ha. Based on soil tests, it was determined that phosphorus application was not necessary. The trees at the Seedfarm site were sprayed with quizalofop ethyl (Assureâ ) in July 1996, for grass weed control. No fertilizer or herbicide applications were performed during the 1997 growing season. In 1998 the trees were fertilized with 125 kg N/ha, 31 kg P/ha and 75 kg K/ha.

Short Rotation Forestry willow yields were assessed in December 1996, 1997 and 1998 by harvesting two randomly selected areas (each consisting of two rows x 5 meters long) in every plot, at a height of 10 cm. A representative tree from each plot was chopped and oven-dried to determine the dry weight for the yield calculations.

Switchgrass plots were also seeded in 1993 (for details see Samson et al., 1995). The stands were considered to be fully established in their third growing season, in 1995. A full description of the establishment and cropping practices between 1993 and 1995 are outlined in Samson et al. (1995) and summarized in Appendix 1.

The biomass produced on these sites was left to overwinter after each growing season and was harvested the following May, prior to spring re-growth. Following harvesting, starting in 1994, the stands at both sites were fertilized each year with 50 kg N/ha as urea. In 1997, an additional 25 kg N/ha was applied at the Seedfarm site, 4 weeks after the first application, due to a slight yellowing of the switchgrass. A period of hot dry weather followed the first urea application in 1997, which may have caused some urea volatilization. No herbicide was sprayed following establishment, and no insect or disease problems were experienced.

Beginning in 1994, fall switchgrass yields were assessed, at both sites, by hand-sickling four 1 m² quadrats, at a cutting height of 10 cm in each plot. A 200 g subsample from each sample was obtained and oven-dried for 24 hours to determine the dry weight.

Grain corn was sown in May from 1996 to 1998, initially on the plowed down SRF willow plots, described in section 2.1. Secondary spring tillage, prior to planting, consisted of disking in 1996 and 1998, and harrowing in 1997. Sowing was performed using a four-row corn seeder. Corn hybrid Pioneer^® 3921 was sown in 1996, Pride^® K206 in 1997, and Pioneer 3893 in 1998. An average density of 80,000 plants per hectare was achieved in all years.

The average rate of fertilizer application over the three years was 170 kg N/ha, 77 kg P/ha, and 65 kg K/ha. In 1996, the weed control consisted of a combination of mechanical (inter-row cultivation) and chemical methods (metolachlor (Dual^®) and dicamba/atrazine (Marksman^®) applied as post-emergent herbicides. The same chemical control was used in 1997, but no inter-row cultivation was performed. In 1998, dimethenamid and a mixture of dicamba/atrazine, as well as bentazon were applied. Harvesting of the grain corn took place each year in October, using a conventional corn combine with an eight-row header. The stalks and leaves remained on site and were incorporated into the soil during plowing. Cropping practices for each year, and site, are summarized in Appendix 1.

Grain corn yields were evaluated in October from 1996 to 1998. Samples were harvested from two randomly selected areas (each consisting of two rows x 5 meters long) in the center of each plot. The samples were subsequently shelled and dried, to determine moisture content and dry yield.

The experimental sites were situated on the western side of the island of Montreal. On the southeastern side of the island of Montreal, the St. Lawrence River flows from east to west. On the north side of the island, Riviere des Prairies flows from east to west. The Lac des Deux Montagnes and the Lac St-Louis are situated on the west side of the island. Due to the region’s abundance of rivers and lakes, the West Island is noted for its microclimate. It receives sufficient precipitation throughout the year; total average precipitation from May to September is 423 mm (Table 2.1), and has hot, humid summers (Table 2.2) with corn heat units (CHU) reaching >3,000 for the growing season (Table 2.3). The Montreal/Dorval International Airport is located approximately 20 km east of the research station, while the Les Cedres meteorological station is approximately 15 km west of the research station, both at similar latitudes to the site. Throughout the duration of the study, monthly precipitation was relatively normal during the growing season, with the exception of higher than normal precipitation in April and May of 1996 and July 1997, and lower than normal in August 1996, and April and May of 1998. Temperatures were near the normal range in 1996 and 1997 with the exception of a cool May in 1997. In 1998 the temperature was higher than normal, especially during the early spring (in May 1998, the average temperature was 4.5^oC above the normal) and fall period, which extended the growing season. The CHU for the region followed the mean, with a minimum amount of deviations.

^‡ Total increment for 1993 and 1994.

*Sampled in April 1999, due to technical difficulties with samples in November 1998

^† Does not include block 4

Means within a column followed by the same letter, or no letter, are not significantly different

at P>0.05 level according to Duncan’s Multiple Range test.

Means within a column followed by the same letter, or no letter, within a site are not significantly different at the 0.05 level according to Duncan’s Multiple Range test.

2.4.3. Corn

In 1996, grain corn yields were high on both sites considering the relatively late sowing date of May 29 (Table 2.6). Yields at the Seedfarm and at the Ecomuseum were 8.6 odMg/ha and 8.3 odMg/ha, respectively. In 1997, poor germination, poor weed control, and herbivore damage contributed to low yields, resulting in no data being collected for some plots, at both sites. In 1998, an early growing season and mild weather contributed to a productive growing season for corn. Grain yields attained an average of 8.0 odMg/ha, on both sites, with no pest problems, and little to no weed problems experienced. Corn stalk and leaf biomass data are presented in Chapter 3.

2.4.4. Comparisons Between Willow, Switchgrass, and Corn

The mean yields, from 1996 to 1998, on the two fully established plantations of switchgrass and SRF willows were very similar, at 12.0 and 12.3 odMg/ha, respectively (Table 2.7). Therefore, both of these biomass crops can obtain relatively productive yields once fully established. However, SRF willow yields were more variable within each year, while the switchgrass yields were more stable.

Switchgrass had consistent yields across the plantations, in contrast to the SRF willow, where the most variable results were found at the Seedfarm. The switchgrass yields were almost identical at both sites (12.2 and 12.5 odMg/ha, at the Seedfarm and the Ecomuseum, respectively), while overall, the SRF willow at the Seedfarm had 21% higher yields than at the Ecomuseum (Table 2.7).

The corn plots had productive grain yields in 1996 and 1998. The microclimate of the region, as well as the relatively high CHU are usually beneficial factors for the production of corn. Nevertheless, the average grain yield of the corn was approximately one third lower than that of the two perennial crops, with two-year average grain yields for both sites of 8.2 odMg/ha. A three year corn yield average for the sites would be approximately 7.5 odMg/ha, assuming a harvestable yield of 5.9 odMg/ha (same as for the Seedfarm) was obtained for the Ecomuseum in 1997.

At the onset of the experiment, the Ecomuseum site was of inherently higher soil fertility, due to the many years of dairy manure application. However, after 6 years without manure application at either site, the Seedfarm site is beginning to emerge as the more productive site. The soil texture analysis indicated the Seedfarm to have, from its composition, a more productive soil texture (loam classification), as opposed to the fine sandy loam classification of the Ecomuseum.

*Corn yield is an average of 1996 and 1998

2.5. Conclusions

The main observations from the plantations, demonstrate that in southwestern Quebec, both SRF willows and switchgrass can have average productive yields of approximately 12 odMg/ha/yr, once they reach full establishment. Switchgrass yields have less variability across years, and sites, while willow yields can vary considerably. Seasons such as those experienced in 1998, with extended growing seasons and timely rainfalls, demonstrate the high yield capacity of SRF willows. Similarly, the trees can have low productivity years when dry summers or heavy pest infestations occur. Both of the perennial biomass crops, however, have demonstrated a high yield potential. Particularly considering the limited plant improvement which has occurred in this region. The Salix alba x glatfelteri willow was an imported clone, developed in the former Yugoslavia, and the Cave-In-Rock switchgrass was an unimproved variety, originating from seed gathered from a Southern Illinois prairie. With plant breeding programs specifically targeted to these crops, in this region, yields will undoubtedly be improved.

Chapter 3

CARBON STORAGE STUDY

The primary purpose of this component of the project was to quantify, for the first time, the impact switchgrass and SRF willows have on carbon cycling and carbon sequestering within an agricultural setting, in eastern Canada. Two main objectives were pursued within the scope of this study;

• compare the accumulation of above and belowground plant carbon and soil carbon in SRF willow and switchgrass systems;

• collect soil carbon data for comparison purposes from an intact forest ecosystem and a long-term annual cropping ecosystem, to enable the estimation of long-term soil carbon storage potential under SRF willows and switchgrass systems.

Claudia Zan, a graduate student who worked under the supervision of Dr. James Fyles, in the Department of Natural Resource Sciences at McGill University, performed these components of the study. Chapter 3 presents a summary of the methodology implemented and results obtained. A detailed literature review is presented in Appendix 2. Further details are available in her thesis (Zan, 1998).

2. Methodology

The following section describes the methodology (Zan, 1998) used to assess carbon storage on the sites.

3.1.1. Sampling, Site Description and History

Unlike the field productivity assessments carried out by REAP (see Chapter 2), samples referred to in this chapter were obtained exclusively from areas of Chicot fine sandy loam, a soil series belonging to the Grey Brown Podzolic group, to maintain consistency and to reduce the number of statistical variables.

The land under study was undulating to gently rolling, and the soil had a moderate moisture-holding capacity. At 0-20 cm, this soil was characterized as being a dark brown, fine sandy loam with a fine crumb structure and very friable. At 20-40 cm, the soil consisted of a yellow-brown to brown, fine light very friable sandy loam, with a uniform colour and some faint mottling, and a pH of 5.8 to 6.2. At 40-60 cm the soil was characterized as being a brown sandy loam or sandy clay loam with clay loam patches, and moderately friable to slightly firm, with occasional faint mottling, and a pH ranging from 6.0 to 6.4. At deeper layers, the soil was dark brown and characterized as a brown sandy loam with streaks and pockets of sandy clay or clay loam, mottled, with a pH of 6.5 to 6.9. Underlying this soil was calcareous till (Lajoie, 1960). The Seedfarm site was noted as having a plow pan at 30 cm, comprised of very compacted sand, giving rise to an almost gravel-like texture.

Biomass sampling began in October 1996 and consisted of obtaining aboveground and belowground samples of switchgrass, willow, and corn (including litter and weeds). Soil samples, to an average depth of 60 cm were obtained to determine the carbon content within each cropping system. Sampling was carried out in 4 blocks at two sites: Seedfarm (Figure 3.1) and Ecomuseum (Figure 3.2), both situated at the Emile A. Lods Agronomy Research Center, at the Macdonald Campus of McGill University, in southwestern Quebec (42° 25’N latitude, 75° 56’W longitude). The Seedfarm site was situated in an open field having a south-facing slope, whereas the Ecomuseum site was located on relatively flat, gently sloping land, protected by windbreaks to the east and west.

Figure 3.1. The Seedfarm Site. SWG = switchgrass; SRF = willow The rectangles represent the sampling area (sub-plot) with number of replicates (circles).

Figure 3.2. The Ecomuseum Site. SWG = switchgrass; SRF = willow. The rectangles represent the sampling area (sub-plot) with number of replicates (circles).

A second sampling was carried out in October 1997 in which only aboveground samples (including litter and weeds) were obtained from the already established switchgrass, willow, and corn plots. In addition, a mature, hardwood forest was sampled in 4 locations to determine the soil carbon content to a 60 cm depth. Two long term abandoned agricultural fields were also sampled for soil carbon and root production (at 2 locations in each). Once again, the soil sampling was restricted to areas of Chicot fine sandy loam.

The forest site was located at the Morgan Arboretum, at the Macdonald Campus of McGill University, and was situated 0.5 km west of the Ecomuseum site. The mature forest consisted of a beech (Fagus grandifolia) and maple (Acer saccharum) association, aged between 100 and 130 years old. It remained relatively untouched with the exception of salvage work that was being carried out periodically.

The first abandoned agricultural field was located at Stoneycroft Pond, approximately 0.2 km from the Ecomuseum site. The field was cultivated until the late 1970’s, and has been abandoned since (Titman R., personal communication). The dominant plant species today is Bromus tectorum (junegrass), which makes up roughly 90% of the plant cover. Other species include Rubus idaeus (wild raspberry), Cirsium arvense (Canada thistle), Asclepias syriaca (common milkweed), Crategus sp. (hawthorn bush), Malus sp. (apple trees less than 5 years old), and Solidago canadensis (goldenrod).

The second agricultural field, also abandoned in the late 1970’s, was located at the Morgan Arboretum, close to the mature forest sampling site. The dominant plant species is Festuca sp. (fescue), with 99% of the plant cover. Other species include Asclepias syriaca, Cirsium arvense, Valeriana officinalis (common valerian), Ramnus sp. (buckthorn), and Vitis sp. (wild grape).

A third, and final sampling, was initiated on the switchgrass, willow and corn plots in October 1998. Soil sampling from 0 to 60 cm was conducted, adhering to the same methodology as used in 1996, except that instead of 6 subsamples obtained from the willow and corn plots, only 4 were obtained, with no loss in the resulting accuracy of soil carbon determination. Soil samples were not obtained from the abandoned fields, or from the forest site.

Aboveground biomass sampling was also carried out in 1998 to determine productivity (4 samples in each plot) using the same methodology as in 1996 and 1997, however no litter, weed biomass, or belowground biomass was sampled. Tissue samples were not analyzed for carbon content either.

3.1.2. Modified Mebius Procedure to Determine Carbon Concentration in 1996

The plant and soil samples were ground to pass a 100 mesh sieve, and analyzed for organic carbon content using a modified Mebius procedure, involving the digestion of each sample with an acidified dichromate solution (Yeomans and Bremner, 1988). A full description of this procedure is outlined in Zan (1998).

3. LECO Combustion Process to Determine Soil Carbon Concentration in 1998

In 1998, the soil samples were sent to the University of Guelph for analysis. A LECO 1400 C carbon analyzer was used to determine total soil carbon and soil organic carbon. Since this methodology varied from the Modified Mebius processes deployed in 1996, 60 of the stored and dried samples from 1996 (equally representing the 3 cropping systems and all 4 depths), were re-analyzed using a LECO method. A regression equation was derived (r²=0.99) from the 60 re-analyzed samples, and their corresponding 1998 samples. From this equation, 1996 carbon values were adjusted to the 1998 values, to enable comparisons between the years.

All soil samples were stored in air-tight plastic bags immediately after sampling, and stored at approximately 10^oC. Four to six weeks later, a subsample from each bag was placed in plastic, air tight vials and shipped to the University of Guelph, where they were analyzed for total and organic carbon by a LECO 1400 C carbon analyzer.

3.1.4. Calculation of Soil Profile Carbon on a Constant Mass Basis

When comparing different cropping systems (corn, switchgrass, and SRF willows), expressing soil carbon storage in terms of the concentration of carbon (% carbon) obtained from laboratory analyses, does not take into account the bulk density nor the thickness of the soil layers. Hence, the percent carbon does not account for differences in soil masses under different management systems. Therefore, all results were standardized to an equivalent soil mass of 7090 od Mg/ha in 1996, and 6910 odMg/ha in 1998, as outlined by Ellert and Bettany (1995).

3.1.5. Statistical Analyses

Data obtained for aboveground biomass, roots, and soil carbon were expressed on an oven dried mass per area basis (odMg/ha). Corn, switchgrass, and willow data were analyzed as a randomized complete block design, with the 3 crops as fixed treatments, in each of 4 randomized blocks, and 2 sites as main factors. Corn and willow were sampled 6 times per plot in 1996 and 1997, and 4 times in 1998, whereas switchgrass was sampled 4 times per plot every year.

Statistical analyses were performed using SAS (SAS Institute, 1994). A probability level of P £ 0.05 was considered significant. Data obtained for aboveground biomass and carbon were analyzed as analyses of variance (GLM procedure). Data obtained for root biomass and soil carbon were further analyzed as repeated measures analyses of variance, using depth as a repeated factor. Following significant ANOVA results, means were compared using the Student-Newman-Keuls (SNK) multiple range test. All analyses used plot means rather than subsample means.

3. Results

3.2.1. Aboveground Plant Yield and Carbon Content

3.2.1.1 . Switchgrass

The average total aboveground yields for 1996 and 1997 were 15,879 kg/ha at the Seedfarm, and 16,314 kg/ha at the Ecomuseum (Table 3.1). In 1998, the standing biomass of switchgrass was 12,657 kg/ha at the Seedfarm and 12,931 kg/ha at the Ecomuseum. For individual years, there were no significant yield differences between the sites. In 1997, switchgrass aboveground yields increased by 1,157 kg/ha at the Seedfarm site, and 811 kg/ha at the Ecomuseum site, compared to 1996. The standing yields obtained in 1998 represent an average decrease of 917 kg/ha from 1996, and an average decrease of 1,309 kg/ha compared to 1997. The 1998 yields were found to be significantly lower than the 1996 and the 1997 yields. One contributing factor to lower yields in 1998 may have been the relatively low rates of N fertilizer applied (50 kg/ha/yr), which may have been insufficient to sustain higher yields than those obtained, after 5 years of growth.

Due to greater corn aboveground yields in the first field season (1996), organic carbon analyses of aboveground plant material was significantly higher for corn in 1996 than in 1997. On average for both years, corn at the Seedfarm and Ecomuseum sites stored 6,226 kg/ha and 6,678 kg/ha of carbon, respectively (Table 3.3).

In 1996, at the Seedfarm site, the stalk/leaf and grain components contributed an average of 45% and 46%, respectively, to aboveground organic carbon storage. Similar results were obtained at the Ecomuseum, with percent contributions from stalk/leaf and grain of 45% and 47%, respectively. In 1997, however, the stalk/leaf component contributed 55% and 52% to the aboveground carbon storage at the Seedfarm and Ecomuseum sites, respectively. The grain portion contributed to 38% and 22% of total aboveground organic carbon for the Seedfarm and Ecomuseum sites, respectively. At the Ecomuseum site, there was a fairly large carbon contribution from weeds (22%).

3.2.1.4. Comparing the Ecosystems

3.2.1.4.1. Biomass Yields

At the Seedfarm, in 1996, corn and switchgrass produced significantly higher aboveground yields than willow (16,144 kg/ha, 15,300 kg/ha, and 10,664 kg/ha, respectively) (Figure 3.3). On the other hand, there were no significant differences found between any of the crops at the Ecomuseum site, with corn, switchgrass, and willow yields averaging 16,256 kg/ha, 15,909 kg/ha, and 16,290 kg/ha, respectively (Figure 3.3).

In 1997, at the Seedfarm site, switchgrass and willow had significantly greater aboveground annual yields than corn (16,457 kg/ha, 16,062 kg/ha, and 12,309 kg/ha, respectively). At the Ecomuseum site, willow had significantly greater (22,028 kg/ha) yields than corn (13,598 kg/ha). Switchgrass yields (16,720 kg/ha) were not significantly different from either willow or corn (Figure 3.3).

In 1998 at the Seedfarm, switchgrass, willow, and corn did not produce significantly different total aboveground yields (12,657 kg/ha, 22,950 kg/ha, and 20,653 kg/ha, respectively). At the Ecomuseum, no significant difference in aboveground biomass was observed either between switchgrass (12,931 kg/ha), willow (13,385 kg/ha) or corn (18,965 kg/ha). Corn had the highest aboveground yields in 1998 at both sites, however they were not significantly different from the willow or the switchgrass yields (Figure 3.4). This is partly attributed to the different corn varieties planted each year, as well as to the exceptional early growing season experienced in 1998. As well, willow total biomass was high and was very variable in 1998, with standard errors of ±6,688 kg/ha at the Seedfarm, and ±4,874 kg/ha at the Ecomuseum (Figure 3.4).

By drawing comparisons between the three individual cropping system yields (combining the data from both sites) between 1996, 1997 and 1998, it was found that switchgrass in 1998 produced significantly lower yields than in the two previous years. This decrease in yield after the fifth year is not normally expected for a warm season perennial grass system, and therefore may have been attributed to relatively low rates of fertilizer application. The willow aboveground shoot biomass in 1998 was not significantly different than in 1996 or in 1997. However, the corn total aboveground biomass (stalk/leaves, grain, and cobs) was significantly higher in 1998, compared with 1996 and with 1997.

3.2.1.4.2. Carbon Concentrations

At the Seedfarm site, at the end of the 1996 growing season, corn and switchgrass accumulated significantly more organic carbon in their aboveground biomass than willow (7,046 kg/ha, 6,473 kg/ha, and 4,994 kg/ha, respectively). On the other hand, there were no significant differences found between any of the crops at the Ecomuseum site, with corn, switchgrass, and willow carbon contents averaging 7,285 kg/ha, 6,999 kg/ha, and 7,679 kg/ha, respectively (Figure 3.5).

In 1997, at the Seedfarm site carbon in the aboveground biomass produced by willow and switchgrass was significantly higher than in corn (7,520 kg/ha, 6,983 kg/ha, and 5,406 kg/ha, respectively). At the Ecomuseum site during the same year, willow aboveground carbon content (10,433 kg/ha) was significantly greater than the carbon content of switchgrass (7,438 kg/ha) and corn (6,071 kg/ha) (Figure 3.4).

In terms of the plant percent carbon content, willow had a significantly higher carbon concentration in its aboveground biomass than switchgrass at the Seedfarm site. No differences were found between any of the crops at the Ecomuseum (Table 3.4).

3.2.2. Belowground Plant Yields and Carbon Content

3.2.2.1. Switchgrass

Since switchgrass lacks a root crown, all the roots were considered to be fine roots. The total root production as measured in 1996 was significantly greater at the Seedfarm site (8,068 kg/ha) than at the Ecomuseum site (5,965 kg/ha) (Table 3.5).

The total root carbon was significantly greater at the Seedfarm site (3,723 kg/ha) than at the Ecomuseum site (2,574 kg/ha). At the Seedfarm, the contribution to total root carbon storage at the 0-15, 15-30, 30-45, and 45-60 cm depth intervals were 53%, 24%, 11%, and 11%, respectively. At the Ecomuseum, the distribution was 45%, 27%, 15%, and 14%, respectively. It appears the more impervious layers of soils at the Seedfarm site confined most of the switchgrass roots to the surface horizons.

3.2.2.2. Willow

The roots of willow, as well as those of corn were comprised of the root crown, and the fine roots. In 1996, there were no significant differences in terms of fine and total root production in the 2 sites (Table 3.6). Fine root distribution up to 60 cm, at intervals of 15 cm, followed a ratio of approximately 8:9:1:1, at the Seedfarm site and of 28:18:4:1 at the Ecomuseum site.

There was no significant difference in fine root carbon between the Seedfarm (2,270 kg/ha) and Ecomuseum site (2,137 kg/ha). The percent contribution to root carbon storage by fine roots was 60% at the Seedfarm, and 58% at the Ecomuseum. There was also no significant difference between root crown carbon at the Ecomuseum (1,562 kg/ha) and at the Seedfarm (1,516 kg/ha). Root crown carbon storage was 40% and 42% of total root carbon at the Seedfarm and Ecomuseum sites, respectively. In total, there were no significant differences between total root carbon at the Seedfarm (3,786 kg/ha) and the Ecomuseum sites (3,699 kg/ha).

3.2.2.3. Corn

Corn fine root yields (Table 3.7) were significantly higher at the Seedfarm site (522 kg/ha) than at the Ecomuseum site (337 kg/ha). However, root crown yields were not significantly different at the Seedfarm site than at the Ecomuseum site. There were also no significant differences between total yields at both sites. The fine root distribution up to 60 cm, at intervals of 15 cm, followed a ratio of approximately 4:4:1:1 at the Seedfarm site, and of 12:5:1:1 at the Ecomuseum site.

Fine root carbon (Table 3.7) was significantly higher at the Seedfarm site (170 kg/ha) than at the Ecomuseum site (105 kg/ha). Percent fine root contribution to carbon storage was 24% and 17% at the Seedfarm and Ecomuseum sites, respectively. Root crown carbon was not significantly different between the Ecomuseum site (529 kg/ha) and the Seedfarm site (525 kg/ha), which represented 76% and 83% of belowground carbon storage for the Seedfarm and Ecomuseum site, respectively.

3.2.2.4. Abandoned Field and Forest

At 0-60 cm, using a soil mass of 7,090 od t/ha, an average root yield of 13,058 kg/ha was obtained for both abandoned fields. The distribution of fine roots was concentrated in the top 15 cm of the soil profile, with 81% of the roots located in this area. The lower layers (15-30, 30-45, and 45-60 cm) contained 12%, 5%, and 2% of the total roots in the depths, respectively. From the belowground biomass production, a root carbon content of 45% was assumed, to calculate root carbon sequestering.

For the forest ecosystem, an average fine root yield of 12,315 kg/ha was sampled. Total root production was nonetheless estimated based on studies conducted in Ohio under similar late secondary growth deciduous forest.

3.2.2.5. Comparing the Ecosystems

3.2.2.5.1. Root Biomass

At both sites, switchgrass and willow had significantly higher root yields than corn (Figure 3.6). When comparing switchgrass with willow, there was no significant difference in root production between the systems, except for switchgrass at the Ecomuseum site, which had a significantly lower root yield. Detailed comparisons including the root production measured in the abandoned fields, and the estimates for the hardwood forest, are presented in Zan (1998).

Statistical analyses were performed on mean fine root production, for each of 4 sampling depths: 0-15, 15-30, 30-45, and 45-60 cm (Figure 3.7). At 0-30 cm, there were no significant differences between switchgrass and willow at either site, and corn had significantly lower fine root biomass than all other systems. At 30-45 cm, switchgrass at both sites produced significantly more roots than either the willow or the corn. At 45-60 cm, switchgrass at both sites had a significantly greater root yield than all other systems (Figure 3.7).

Overall, switchgrass had relatively low root biomass levels of 6-8 odMg/ha in these southwestern Quebec sites compared to other studies. In Virginia, Parish et al. (1990) found switchgrass root yields of 10.5 odMg/ha on average on three soils. As well, 14.5 –19.0 Mg/ha (Dahlman and Kucera, 1965) and 15.1 – 17.9 Mg/ha (Weaver and Zink, 1946) have been reported from the US Midwest. The relatively low root yields in the current study may be a result of several factors. One may be that switchgrass needs several more years to establish its root system fully in a more northerly location. Another may be that the favorable water conditions in the study region result in a relatively low investment in root biomass production compared with regions such as the US Midwest, which have lower rainfall to evaporation ratios. Finally, the soils were relatively shallow with compacted layers which inhibit rooting compared to deep prairie soils.

3.2.2.5.2. Root Carbon Sequestered

Comparisons of total root carbon at the Seedfarm site demonstrated that to an average sampling depth of 60 cm, both willow and switchgrass had significantly greater carbon accumulations than corn (3,786 kg/ha, 3,723 kg/ha, and 695 kg/ha, respectively). At the Ecomuseum site, willow root carbon was significantly greater than that of switchgrass, which was significantly higher than that of corn (3,699 kg/ha, 2,574 kg/ha, and 634 kg/ha, respectively) (Figure 3.8).

Figure 3.9 shows fine root carbon accumulations for each of the 4 sampling depths (0-15, 15-30, 30-45, and 45-60 cm). At the Seedfarm site, at a depth of 0-15 cm, switchgrass and SRF willow did not differ significantly in their fine root carbon content (1,991 kg/ha and 969 kg/ha, respectively). Both had significantly greater root carbon than corn (58 kg/ha). At the Ecomuseum site, both switchgrass and willow accumulated significantly more carbon than corn (1,157 kg/ha, 1,152 kg/ha, and 61 kg/ha, respectively).

At the 15-30 cm depth, a similar trend was observed at both sites. Switchgrass (Seedfarm: 905 kg/ha; Ecomuseum: 691 kg/ha) and willow (Seedfarm: 1,021 kg/ha; Ecomuseum: 779 kg/ha) had significantly more stored carbon than corn (Seedfarm: 75 kg/ha; Ecomuseum: 28 kg/ha).

At 30-45 cm, results differed from the 2 previous depth intervals, with switchgrass having significantly higher carbon accumulations (Seedfarm: 426 kg/ha; Ecomuseum: 378 kg/ha) than willow (Seedfarm: 152 kg/ha; Ecomuseum: 160 kg/ha) and corn (Seedfarm: 19 kg/ha; Ecomuseum: 9 kg/ha) at both sites. The same applied at 45-60 cm, switchgrass (Seedfarm: 347 kg/ha; Ecomuseum: 337 kg/ha) had significantly greater carbon than willow (Seedfarm: 100 kg/ha; Ecomuseum: 44 kg/ha) and corn (Seedfarm: 21 kg/ha; Ecomuseum: 5 kg/ha).

Table 3.8 summarizes the concentration of root percent carbon content for the three cropping systems, which shows root carbon content to be significantly lower for corn at 0-15 cm at both sites. At 15-30 cm, switchgrass had the highest percent root C at the Seedfarm site. Willow had significantly lower root carbon than switchgrass, but more than corn. No differences were detected at the 30-45 cm level, at either site. At the Ecomuseum, at 45-60 cm, corn had significantly lower carbon concentration than switchgrass or willow.

3. Soil Carbon

Soil carbon measurements were made in 1996 and in 1998. A different soil organic carbon analysis was used was used for both years, hence the values in 1996 were adjusted, by using a regression equation, to the values in 1998. The values presented for 1996 (text and tables) have not been adjusted using a regression equation. However, when comparisons are drawn between 1996 and 1998, the 1996 soil organic carbon values were transformed. It is always stated when transformations on the 1996 data were conducted.