Biopower generation from mountain pine infested wood in Canada: An economical opportunity for greenhouse gas mitigation

ARTICLE IN PRESS

Renewable Energy 33 (2008) 1354–1363 www.elsevier.com/locate/renene

Biopower generation from mountain pine infested wood in Canada: An economical opportunity for greenhouse gas mitigation Amit Kumara,, Peter Flynna, Shahab Sokhansanjb a

Department of Mechanical Engineering, University of Alberta, Edmonton, Alberta, Canada T6G2G8 Department of Chemical and Biological Engineering, University of British Columbia, Vancouver, British Columbia, Canada V6T1Z3

b

Received 16 June 2006; accepted 10 July 2007 Available online 4 September 2007

Abstract Biomass is considered carbon neutral, and displacement of fossil fuel-based power by biomass-based power is one means to mitigate greenhouse gases. Large forest areas in British Columbia (BC), Canada, are infested by the mountain pine beetle (MPB). Dead wood from the infestation is expected to vastly exceed the ability of the pulp and lumber industry to utilize it; current estimates are that 200–600 million m3 of wood will remain unharvested over the next 20 years. Regions where the damaged wood is not harvested will experience loss of jobs in the forestry sector, increased risk of forest fire hazard, carbon emissions from burned or decaying wood, and uncertainty about timing of replanting since this usually occurs at harvest. This paper reports the results of a detailed preliminary techno-economic analysis of producing power from MPB killed wood. Power plant size and location are critical factors affecting overall power cost. Overall cost of power rises steeply at sizes below 300 MW net power output. By locating the power plant in an area of high infestation, transportation distances can be minimized. A 300 MW power plant would consume 64 million m3 of wood over a 20-year lifetime, and hence is a significant sink for otherwise unharvestable wood. Cost estimates are based on harvesting of whole dead trees with roadside chipping and transport to a central power plant located in either the Nazko or Quesnel regions of BC. A circulating fluidized bed boiler with a conventional steam cycle is a currently available technology demonstrated at 240 MW in Finland. The estimated power cost is $68 to $74 per MWh, which is competitive with other ‘‘green power’’ values in BC. Given recent values of export power in the Pacific Northwest, a 300 MW MPB power plant is viable with a carbon credit below $15 per tons of CO2. r 2007 Elsevier Ltd. All rights reserved. Keywords: Biomass power; Direct combustion; Power cost; Mountain pine beetle; Lodgepole pine

1. Background and overview Mountain pine beetle (MPB) infestation of lodgepole pine is creating a crisis for the forestry industry of British Columbia (BC), Canada. An estimated 10 million hectares is currently infested, and models predict the ultimate volume of otherwise merchantable timber lost to MPB will be 960 million m3 [1]. Some parts of Alberta have also been infected, and the potential exists for a trans-species jump to jackpine which would spread the infestation across Canada. Regions where the damaged wood is not harvested will experience loss of jobs in the forestry sector with an impact Corresponding author. Tel.: +1 780 492 7797; fax: +1 780 492 2200.

E-mail address: [email protected] (A. Kumar). 0960-1481/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.renene.2007.07.008

on the viability of communities. Unharvested areas may not be replanted in a timely manner since replanting occurs shortly after harvest. The unharvested biomass is a fire hazard to regenerating species and hence a risk of even more future economic damage. This unharvested wood, if left to decay in the stands, would release carbon into the atmosphere. Using surplus MPB killed trees to generate power would contribute to Canada’s efforts to reduce fossil carbon emissions while helping sustain the forestry industry in BC. Many plants around the world burn biomass to make heat, power or a combination of the two. Many of these plants are based on mill residues, for example bark, sawdust and trimmings, and hence are built at a small size that reflects the source of the biomass. An example of this is the 65 MW plant in Williams Lake, BC that uses about

ARTICLE IN PRESS A. Kumar et al. / Renewable Energy 33 (2008) 1354–1363

600,000 tons of sawmill residue per year. California has 28 direct combustion biomass power plants with a generation capacity of 558 MW and an additional 70 MW of generating capacity from cofiring of municipal waste; many other plants are located across the US. Europe has many biomass power plants, including several using straw as a fuel [2–5]. Several authors have noted that the cost of power from a biomass-based plant is dramatically lower for plants as size increases, even up to 500 MW (see, for example, [6–9]); capital efficiency outweighs increased transportation cost, with the biomass yield per gross area being a critical determinant of optimum size [8]. However, because many biomass projects to date are constrained by mill residue supply or by their demonstration nature, only one plant over 100 MW has been built, the 240 MW Alholmens power plant in Pietarsaari, Finland capable of running on 100% coal or 100% biomass [10]. The objective of this study is a preliminary technoeconomic assessment of using a portion of BC’s surplus MPB killed pine to generate power. Based on a preliminary study [11], two plant sizes in each of two locations in BC were analyzed: gross generation of 330 (Case 1) and 240 (Case 2) MW at West Road/Nazko River (N) and Quesnel (Q). The 240 MW size is identical to the Alholmens plant, and the 330 MW size was chosen as a reasonable scale up. The Nazko location is remote and in an area of very high concentration of MPB stands; the Quesnel site is an existing urban area closer to construction and operating labor but with a somewhat lower gross density of MPB trees in the surrounding area [12]. Internal power consumption reduces net output to 300 MW and 220 MW. All cost factors from wood harvesting with replanting through delivery of the power to the main BC transmission grid are included in the estimated power costs below [13].

1355

(MoFR) [12,17]. Actual amounts of MPB killed trees and the fraction that are surplus to existing forestry operations is under current re-evaluation within the MoFR, and yield figures may be adjusted in the future. In this study we assume that MPB killed trees are cut and skidded to the roadside. At the roadside the whole tree is chipped and chips are transported to the plant by large chip vans; hence limbs and tops are recovered as fuel. In this study we have assumed a value of 20% for the residues, and hence actual yield is 25% higher than merchantable volume. The final average standing yield per gross hectare for lodgepole pine is estimated at 64.1 m3 for the West Road/Nazko River location and 37.5 m3 for Quesnel location. Gross hectares include all other land uses such as other forest species and non-forestland use. 3. Fuel properties Moisture content of wood is one of the most important characteristics for its use as fuel. For a dead tree, water in the wood has a tendency to reach equilibrium with the surrounding air. The equilibrium moisture content (EMC) of wood stored outdoors is a function of the surrounding temperature and relative humidity of the air. In this study we estimated the EMC of the wood based on equations developed by W. Simpson of the United States Department of Agriculture Forest Service [18]. The temperature and relative humidity of Williams Lake, which is approximately in the center of the study area, are assumed to be representative of the study area. Based on 20 years of data [19], the estimated average daily temperature and relative humidity used in this study are 4.2 1C and 67.6%, respectively. The calculated value of EMC is 13% (dry basis); other assumed fuel properties are given in Table 1. The value of EMC has a critical impact on the available lower heating value (LHV) of the wood.

2. Biomass source and characteristics 4. Scope and cost The Province of British Columbia has a total land area of 94 million hectares. Timber productive forestland area is about 55% of the total land area. Timber productive volume for the province is about 10,595 million m3 [14]. As of August 2003, the annual allowable cut for the province was about 74.4 million m3/year of wood [15]. British Columbia’s forest consists of both coniferous and deciduous tree species; coniferous species include lodgepole pine, Douglas fir, spruce, hemlock, cedar, and true firs. Lodgepole pine is most susceptible to MPB attack, with mature trees being most vulnerable because their larger diameter and thicker bark protects the beetle from predators. The beetles have a symbiotic relationship with the blue stain fungus, which actually causes the death of the tree [16]. Infestation leaves relatively dense stands of dead trees. Fig. 1 shows the study area and the two study locations. The yield of surplus MPB killed trees for 60 year or older lodgepole pine stands is estimated from reports and discussions with the BC Ministry of Forests and Range

Note: All currency figures in this report are expressed in 2004 Canadian dollars unless otherwise noted. Costs from the literature have been adjusted to the year 2004 using historical inflation rates; an inflation rate of 2% is assumed for 2005 and beyond. MW refers to electrical megawatts unless otherwise noted. Cost factors are developed in detail for each element of the scope [13]. 4.1. Components of delivered cost of biomass This study is based on the existing practices in the forest industry of western Canada including clear cutting and skidding of whole trees to the edge of the logging road. Trees are drawn from throughout the long-term harvest area, giving a constant average transportation distance to the power plant over the life of the plant. Trees are cut by a feller-buncher, moved to the side of a logging road by a grapple skidder, chipped and loaded into 48 foot trailers

ARTICLE IN PRESS A. Kumar et al. / Renewable Energy 33 (2008) 1354–1363

1356

Fig. 1. Map of the study region in British Columbia, Canada (derived from Ref. [15]). Table 1 Fuel wood properties Items

Values

Average annual equilibrium moisture content (%, dry basis)

13

Higher heating values (MJ/ dry kg) Density of logs at given moisture content (kg/m3)

20

Ash in wood (%) Hydrogen content of wood (%, dry basis) Basic specific gravity for lodgepole pine Lower heating value of delivered wood chips (MJ/kg)

455.3

2.5 5.98 0.38

16.6

Comments/sources Based on the average temperature and relative humidity of Williams Lake. Calculated using equations given by Simpson [18] This is the average heating value of softwood [20] Calculated based on equations given in [21]. Density is for lodgepole pine logs at 13% EMC [22] [13] The density of logs at 13% EMC is estimated using equations in [23] [24]

for transport to the power plant. Table 2 summarizes the biomass delivery cost factors used in this study. (We draw heavily on cost studies by the Forest Engineering Research Institute of Canada (FERIC); for a detailed literature review of harvesting cost, see [13].) Harvesting for fuel wood is simpler and involves fewer steps than harvesting for lumber or pulp: trees are not bucked or delimbed, and thus residues are not left at the roadside, and trees are not loaded onto trucks but rather left at the roadside for chipping. Hence, costs in this study are at the lower end of the range of FERIC estimates [25]. MoFR and the Canadian Forest Service conduct ongoing resource and logging studies, including some specific to the Quesnel region (see, for example, [26]). As with FERIC figures, the MoFR figures reflect operations for pulp and lumber recovery that are not required for fuel wood, and hence some components of tree to roadside costs in this study are lower than MoFR figures. Similarly, recovery of fuel wood would not require waste and residue surveys for purposes of royalty calculation. Road

ARTICLE IN PRESS A. Kumar et al. / Renewable Energy 33 (2008) 1354–1363 Table 2 Delivered cost of biomass Components

This studya (based on FERIC’s estimates)

Felling ($ per m3) Skidding ($ per m3) Tree-to-roadside ($ per m3) Silviculture ($ per m3) Roads and infrastructure ($ per m3) Overheads ($ per m3) Chipping ($ per m3) Hauling ($ per m3) Total delivered cost ($ per m3)

6.00 3.00 9.00 3.15 3.90 5.00 5.00 6.76b 32.81

a All the costs have been estimated based on discussions with personnel from FERIC Western Division and are close to FERIC’s lower estimate of biomass delivered cost [25]. b Hauling cost for Quesnel location at 300 MW net power (Case 1Q) where average distance of transport is 62 km with a winding factor of 1.2. Hauling cost for West Road/Nazko location at the same capacity (Case 1N) is $5.85 per m3 for an average transport distance of 48 km with a winding factor of 1.2. For Cases 2Q and 2N, hauling costs are $6.20 per m3 and $5.41 per m3, respectively.

construction, infrastructure and silviculture costs are comparable to MoFR and Canadian Forestry Service (CFS) estimates and are a significant component of overall cost. As noted above, one key benefit to BC from recovery of fuel wood is replanting of MPB infested areas in a timely manner. 4.2. Capital cost, scale factor and boiler technology of power plant This study is based on a single cycle direct combustion circulating fluidized bed power plant similar to the Alholmens design. Circulating fluidized beds have an advantage of high carbon conversion, high tolerance for variable moisture in wood, minimal need for biomass size communition, and low emissions [27]. Both bubbling fluidized beds (BFB) and circulating fluidized beds (CFB) have been applied to biomass; for a discussion of the relative merits at large plant size, see [10]. The capital cost of a power plant depends on the size; most of the costs reported in the literature are for smaller scale plants, often built for demonstration purposes. A detailed literature review of biomass power plant capital costs was done by Kumar et al. [8]. The plant costs for the biomass boilers used in this study for the Quesnel location for Case 1Q (330 MW gross) and Case 2Q (240 MW gross) are $1875 per installed kW and $2024 per installed kW, respectively. At similar capacities, the cost for West Road/Nazko River location for Case 1N and Case 2N are $1960 per installed kW and $2120 per installed kW, respectively. A scale factor of 0.75 was used to adjust capital costs for plants of different size; values for cost and scale factor reflect both literature studies and discussions with boiler manufacturers and a power plant

1357

construction firm [8,27–30]. The scale factor value of 0.75 is conservative, i.e. high. The higher cost of Nazko vs. Quesnel arises from an allowance for a remote construction site that would require more camp-based construction labor [30]. Note that a comparable value for a new coalfired plant in Alberta is $1400 per kW at a size of 500 MW gross. Power plant cost estimates used in this study are higher than reported values for the Alholmens plant of 700–850 h per kW; they are based on both discussion with a boiler manufacturer, Kaverner Power Inc. [27] and on adjustments to reported values for coal-fired power plants based on an analysis of design differences between biomass and coal [31]. 4.3. Other cost factors We have assumed that power plant maintenance costs (parts plus labor) are 2% of the initial capital cost of the plant, which gives a maintenance cost in the range of $4.65 to $5.35 per MWh, depending on plant size. The biomass power plant is assumed to be a stand-alone company, and an administration staffing level of 8 is included for each case, with an additional two laboratory staff primarily for biomass quality monitoring. If a larger firm owned and operated the biomass power plant, savings in administration costs would be possible; however, these are not a significant cost factor in the overall cost of power. Average staff compensation is assumed to be $45 h 1. Kumar et al. [13] reported similar numbers and gives the details on distribution of staff for a 450 MW biomass power plant and we assume that there would not be a significant difference in the number of staff for a 300 MW biomass power plant. Ash disposal costs have been estimated based on an average ash haul distance of 50 km and spreading in the forest; for details see [8]. A reclamation and site recovery cost of 20% of original capital cost, escalated, is assumed in this study, spent in the 20th year of the project. Because the charge occurs only in the last year, it is an insignificant factor in the cost of power. Power cost is calculated to provide a 10% pre-tax return on total investment. The most likely mechanism for initiating this project would be a fixed price contract with a large distributor (BC Hydro) that would enable significant debt financing, leveraging the return on equity to an adequate level for a project developer. Note that purchase of green power is consistent with Policy Action 20 of the BC Energy Plan of 2002 that set a target of 50% of incremental generation in BC being supplied from green sources for a 10-year period. 4.4. Location of plant and transmission of generated power There are significant cost differences between a Quesnel and Nazko location. Quesnel has a lower density of trees, and hence transportation costs for biomass are higher per unit of power output. There are several offsetting cost factors, including being adjacent to a transmission line and

ARTICLE IN PRESS 1358

A. Kumar et al. / Renewable Energy 33 (2008) 1354–1363

being better located relative to existing population and transportation. A dedicated 100 km transmission line is included in the Nazko cases, and line loss of 1% of power is factored into net output. Construction cost is affected by the relative remoteness of the project from a skilled labor supply [30]. Relative to a tidewater location, a construction premium of 5% is assessed for Quesnel, to reflect the need to house a portion of the construction workforce in a camp or transport them from a location such as Prince George, BC, but a premium of 10% is assessed for Nazko because virtually the entire labor force would be camp based. 4.5. Power plant characteristics In this study we have assumed a capacity or overall availability factor of 0.85 for the first year, 0.88 for the second, and 0.90 thereafter. The plant is assumed to be base loaded, i.e. operating at full capacity 7  24 h when available, which is a reasonable assumption in BC where plants in the region (Alberta/BC/US Northwest) with a higher net marginal cost (fired by natural gas) provide nonbase load power and is also consistent with a firm power purchase agreement. The 240 MW Pietarsaari, Finland plant reports an availability of 93.5%, with 1.5% unplanned outages and 5% planned shutdowns [32]. This study assumes an LHV efficiency of 39% [8,10,29]. Power

costs are calculated for a plant life of 20 years; the actual physical life of the assets would exceed 20 years if alternate sources of fuel such as forest harvest residues or mill residues are identified. 5. Results 5.1. Resource requirement and power cost Table 3 gives the amount of wood required over 20 years to support the biomass power plant, the geographical collection area and the power cost for all four cases. Note that if all of the minimum assumed available 200 million m3 of otherwise unharvested MPB wood were to be used for power production, it would support three 300 MW power plants producing, over their life, 164 TWh of electricity. Ten such plants could be built at the higher estimate of 600 million m3 of surplus merchantable volume. Costs are for the first year of operation at full capacity (year 3), but are deflated back to the base year 2004. Delivered cost of biomass is in the range of 43–49% of the total power cost, followed by capital cost (39–42%) and operation and maintenance cost (12–15%). Transportation cost is in the range of 22–26% of the biomass delivered cost, which is close to the figures reported in other studies [8,33,34]. Biomass storage cost is not a significant component of total

Table 3 Resource requirement and power cost composition for a MPB killed tree biomass based power plant life of 20 years West Road/Nazko River, BC

Quesnel, BC

Case 1N 300 MW output

Case 2N 219 MW output

Case 1Q 300 MW output

Case 2Q 221 MW output

62,670,780 50,136,620 112  112

45,717,290 36,573,830 95  95

62,099,310 49,679,450 145  145

45,717,290 36,573,830 125  125

13.30 7.62 2.93 3.63 4.93

13.30 7.05 2.93 3.63 4.93

13.17 8.71 2.90 3.60 4.88

13.17 8.00 2.90 3.60 4.88

Total delivered biomass cost ($ per MWh) Capital cost recovery ($ per MWh)

32.42 28.58

31.85 30.93

33.26 27.05

32.55 29.21

Operation and maintenance cost components Storage cost at plant ($ per MWh) Operating cost for plant ($ per MWh) Maintenance cost for plant ($ per MWh) Administration cost for plant ($ per MWh) Ash disposal cost ($ per MWh) Transmission ($ per MWh)

0.62 4.94 1.44 0.55 0.49 1.49

0.62 5.35 1.97 0.76 0.49 1.75

0.61 4.68 1.43 0.55 0.48 0.00

0.61 5.05 1.95 0.75 0.48 0.00

9.53 70.53

10.93 73.71

7.76 68.08

8.84 70.60

Amount of biomass required over 20 years (actual m3) Amount of biomass required over 20 years (merchantable m3) Project draw area (km  km). Note: only the surplus MPB killed trees within this area are used for fuel Cost elements Delivered biomass cost components Harvesting cost ($ per MWh) Transportation cost ($ per MWh) Silviculture cost ($ per MWh) Road construction cost ($ per MWh) Chipping cost ($ per MWh)

Total operation and maintenance cost ($ per MWh) Total power cost from MPB killed wood ($ per MWh)

ARTICLE IN PRESS A. Kumar et al. / Renewable Energy 33 (2008) 1354–1363

1359

105 Power Cost ($/MWh)

90 75 60 45

Power cost at Quesnel (no transmission line)

30

Power cost at West Road/Nazko River (including transmission line)

15 0 0

100

200 300 Power Plant Capacity (net MW)

400

500

Fig. 2. Power cost as a function of capacity for MPB killed wood based plant.

cost; it reflects 3 months storage at the plant to provide a buffer for road restrictions. Transmission cost for the West Road/Nazko River location is about 2% of the total cost of power. Fig. 2 shows the power cost as a function of plant capacity. In theory, the optimum power plant size would be 450 MW of power generation (an assumed upper size limit based on maintaining grid stability in the event of sudden unit outage), but in practical terms a unit of 300 MW would reduce the risk to the project developer because it is a reasonable design extension from the Alholmens plant. Note that the incremental reduction in power cost with increasing size is very substantial at small power plant sizes but relatively small above 300 MW. The general shape of Fig. 2 sharply dropping power cost that flattens at larger sizes, arises from the competition between capital efficiency and longer transportation distances, and is consistent with previous studies of optimum size of biomass plants [6,8,9,35,36]. This pattern is not observed for fossil fuel plants where delivered fuel cost is either independent of or drops with increasing plant size. Note that Fig. 2 likely understates the power cost for small plant sizes, since efficiency (power out per unit of heat in) is lower for small plants, an effect not factored into this study. 5.2. Carbon credit from MPB killed tree biomass-based power An MPB wood-based power plant is likely to displace a base loaded fossil fuel power plant, i.e. because a biomassbased plant is constructed the need for an incremental fossil fuel plant somewhere in the Alberta/BC/Pacific Northwest region is postponed. In Alberta and portions of the US incremental base load plants burn coal, and that assumption is used in this study, i.e. that the available carbon credit from the MBP wood plant is the assumed displacement of the equivalent amount of coal to generate 300 or 220 MW. Life cycle emissions from a biomass power plant and an Alberta coal power plant are used to estimate the carbon reduction. For both types of power plant, emissions were

estimated for production, transportation, plant construction and decommissioning and energy conversion. For an MPB biomass-based plant, total emission is 42.3 g of CO2 per kWh (derived from [8]). This value includes emissions during transport of biomass for a distance of 48 km; carbon emitted from combustion of wood is not a net contributor of CO2 because forest regrowth takes up this carbon. For a coal power plant the comparable value is 984.6 g of CO2 per kWh [8]. Although the transportation distances for biomass power plants in West Road/Nazko River and Quesnel locations are different, the emissions are not significantly different. In this study the emissions for both the locations are assumed to be the same. A market for carbon credits may emerge in Canada; the value of credits in the future is unknown. Fig. 3 shows the relationship between a future carbon credit in Canadian dollars per tonne of CO2 and the effective reduction in power price from this project. As a first approximation, every dollar per tonne of CO2 reduces the power cost by $1 per MWh. The disposition of carbon credits would be a matter of negotiation between BC Hydro and the project developer. There is also a potential subsidy for biomass power: in the 2005 budget the Canadian Federal Government announced its intention to apply a support payment to biomass power of $0.01 per kWh ($10 per MWh). We do not know if this support payment will be implemented and would be available to a project of the size and scope in this study. However, if it is available it would have a significant impact on the economics of the project, reducing the effective cost of power from biomass power plant to $60 to $63 per MWh. Power costs in Alberta and the US Pacific Northwest run $60 to $70 per MWh; power from surplus MPB killed trees is attractive both relative to other sources of green power and potentially in its own right. 6. Discussion Power from surplus MPB killed trees in BC is a large source of base load green power available at an attractive price. It can help to meet the objective of BC to draw half of its incremental power from green sources. Low-cost

ARTICLE IN PRESS A. Kumar et al. / Renewable Energy 33 (2008) 1354–1363

1360

power from MPB killed trees is only available from a large power plant; the extent of the devastation caused by MPB will support 3–10 such plants. Power cost from each project will depend on transportation cost, but the transportation component of total power cost is small, less than 15% for all cases in this study. Generating power from MPB killed trees conveys many other benefits that are not quantified in this study. These include:



  

Earlier replanting of devastated forests. In BC, forests are replanted by the firm that cuts them, and the earlier dead MPB trees are harvested the sooner new forests will be replanted. Reduction of forest fire hazard. If surplus MPB killed trees are not harvested for power they will constitute an increased forest fire risk. Sustained employment in the forestry sector in BC. The MPB devastation threatens both individual employment and community survival. A significant contribution to meeting GHG emission reduction targets. A 300 MW renewable power plant is



substantial compared to many alternate GHG mitigation alternatives. The development of engineering and operating expertise in biomass power plants. Given the vast forest and agricultural resources in BC and Canada, development of this capability in house is of merit.

There are two critical steps to realize the potential of MPB killed trees as a power plant fuel: identify trees that are surplus to existing forestry operations (MoFR) and confirm that transmission capacity is available and that the power would be purchased (BC Hydro). When these two steps are complete, the existing Request for Proposal process in BC would enable development. This study illustrates that density of biomass, i.e. yield per gross hectare, is offset by other site factors in a comparison of West River/Nazko and Quesnel. The cost of transmission and remote construction more than offset lower transportation and silviculture costs at the Nazko site. In addition, there are some intangible benefits to the Quesnel location: Quesnel has rail service that might enable the future delivery of woody biomass from more distant

Reduction in Power Cost ($ / MWh)

60 50 40 30 20 10 0 0

5

10

15

20

25

30

35

40

45

50

55

Carbon Credit ($ / tonne CO2) Fig. 3. Impact of carbon credit on power cost based on displacement of base load coal generation in Western Canada or the North Western US.

90

Power cost ($/MWh)

80 70 60 50 40

Power cost for Case 1N - West Road/Nazko River 300 MW plant Power cost for Case 1Q - Quesnel 300 MW plant

30

Power cost for Case 2N - WestRoad/Nazko River 219 MW plant

20

Power cost for Case 2Q - Quesnel 221 MW plant

10 0 0.40

0.50

0.60 0.70 0.80 Merchantable tree/Fuel tree

0.90

1.00

Fig. 4. Impact of ratio of merchantable volume to total volume of a tree on power cost for West Road/Nazko River and Quesnel locations for a biomass power plant.

ARTICLE IN PRESS A. Kumar et al. / Renewable Energy 33 (2008) 1354–1363

1361

90 Power cost ($/MWh)

80 70 60 50

Power cost for Case 1N - West Road/Nazko River 300 MW plant

40

Power cost for Case 1Q - Quesnel 300 MW plant

30

Power cost for Case 2N - West Road/Nazko River 220 MW plant Power cost for Case 2Q - Quesnel 220 MW plant

20 10 0 10

20

30

40

50

60

Moisture content (%, dry basis) Fig. 5. Impact of moisture content on power cost for West Road/Nazko River and Quesnel locations for a biomass power plant.

Table 4 Sensitivities for a MPB killed tree based biomass power plant for West Road/Nazko River and Quesnel locations Cost element

Case 1N ($ per MWh)

Case 1Q ($ per MWh)

Case 2N ($ per MWh)

Case 2Q ($ per MWh)

Base case Biomass production and delivery related sensitivities Biomass yield is 25% higher per gross hectare Biomass yield is 25% lower per gross hectare Biomass felling and skidding cost is 50% higher Biomass felling and skidding cost is 50% lower Biomass transportation cost is 25% higher Biomass transportation cost is 25% lower Biomass chipping cost is 25% higher Biomass chipping cost is 25% lower

70.53

68.08

73.71

70.60

70.11 71.13 74.82 66.24 72.42 68.62 71.76 69.29

67.55 68.86 72.33 63.83 70.26 65.90 69.30 66.86

73.35 74.22 78.00 69.42 75.47 71.94 74.94 72.47

70.14 71.27 74.85 66.35 72.60 68.60 71.82 69.38

73.87 67.19 69.64

71.24 64.92 67.16

77.32 70.09 72.84

74.01 67.19 69.70

70.02 69.27 70.03 75.16

67.57 66.89 67.59 72.47

73.01 72.35 73.21 78.73

69.92 69.32 70.11 75.34

Biomass power plant related sensitivities Capital cost of plant 10% higher Capital cost of plant 10% lower Efficiency of power plant is increased from 39% to 40% (LHV) Staffing cost is reduced by 25% Maintenance cost is reduced by 25% Ash disposal has zero cost Pretax return on capital is 12% rather than 10%

locations, and Quesnel has more industry and hence the potential for a purchaser of low-quality heat (steam or hot water), improving the economics of the power plant. Several sensitivities were explored to determine the robustness of the estimate of power cost. Three factors explored in detail are the ratio of total tree to merchantable volume, moisture content, and combustion technology. The first two arise from uncertainty about the characteristics of the properties of MPB killed trees: will the branches detach in the field during harvesting, and what is the final level of EMC. Figs. 4 and 5 illustrate that neither of these factors will overwhelm the power cost estimate; the impact over a wide range is moderate. We performed a detailed analysis of gasification based on estimates of the capital cost of gasification that are less certain than those for direct combustion [31,37]. Gasification would achieve a higher power conversion efficiency at

a higher capital cost, and appears to achieve a power cost of 10% less than for direct combustion. Despite this we believe that a developer would select direct combustion based on it being a commercially proven technology, while gasification of wood and integrated power production has not been demonstrated above 10 MW. Table 4 shows other sensitivity values. Again, no factor is found to overwhelm the power cost from an MPB biomass power plant. 7. Conclusions The cost of generating power using MPB wood in a 300 MW net (330 gross power) direct combustion power plant is $70.53 per MWh for plant located in West Road/ Nazko River with a new dedicated transmission line and $68.08 per MWh for plant located in Quesnel without

ARTICLE IN PRESS 1362

A. Kumar et al. / Renewable Energy 33 (2008) 1354–1363

a new dedicated transmission line. Similar figures for two locations from a 240 gross capacity power plant are $73.71 per MWh and $70.60 per MWh, respectively. Delivered cost of biomass is in the range of 43–49% of the total power cost, followed by capital cost (39–42%) and operation and maintenance cost (12–15%). The cost of power from Quesnel location is lower than the power cost at West Road/Nazko River location. Two main reasons for this are transmission line cost and a higher capital cost premium for West Road/Nazko River location plants. The potential for a cogeneration project is higher for a Quesnel location. Hence, a Quesnel location is recommended based on current estimates of available surplus MPB killed trees. Total estimated MPB killed wood that would otherwise remain unharvested is 200–600 million m3. A 330 gross MW direct combustion MPB killed tree based power plant would require about 63 million m3 (50 million merchantable m3) and a 240 gross MW power plant would require about 46 million m3 (37 million merchantable m3) of wood over 20 years. The total projected area for a 330 gross MW biomass power plant from which biomass would be drawn is about 112 km  112 km (average transportation distance including winding factor is 48 km) for the West Road/ Nazko River location, and 145 km  145 km (average transportation distance 62 km) for the Quesnel location. Similar figures for a 240 gross MW power plant would be 95 km  95 km and 125 km  125 km for two locations, respectively. MPB killed wood provides a unique opportunity to convert otherwise wasted biomass in BC to useful electrical power, a project that would sustain jobs, contribute to a clean environment, potentially help Canada meet its obligations under the Kyoto accord, and put Canada at the forefront of biomass utilization. Acknowledgments This paper was presented at the Climate Change Technology Conference, 10–12 May 2006, Ottawa, Canada. The authors are grateful to BIOCAP Canada Foundation and the BC Government for providing the financial support to carry out this project. The authors thank Mr. Alex Sinclair, Vice President, Western Division, Forest Engineering Research Institute of Canada (FERIC) for his valuable comments on the harvesting, transportation and storage of biomass. We thank David Layzell (BIOCAP), Jamie Stephen (BIOCAP), Jack McDonald (FERIC), Tony Sauder (FERIC), Henry Benskin (Ministry of Forests and Range, BC) for their input and discussion. The authors are grateful to Hank Sherrod (Kvaerner Power Inc., USA), Pekka Saarivirta (Kvaerner Power Oy, Finland), Walt Sanders (Kvaerner Power Inc., USA) and Matti Jarvinen (Electrowatt-Ekono, Finland) for their input on biomass power plant, technology, efficiency and capital cost. The authors are also thankful to Mr. Marvin Eng (Research Branch, Ministry of Forests and Range, BC) for his help in determining the availability of the MPB infested wood.

Many others in the forestry and engineering community have provided valuable input. However, all the conclusions, recommendations and opinions are solely the authors, and have not been endorsed by any other party.

References [1] Clark B. Pine beetle crosses rockies. Vancouver Sun, 8 November 2005. p. A1–A2. [2] Wiltsee G. Lessons learned from existing biomass power plants. National Renewable Energy Laboratory Report no. NREL/SR-57026946, 2000. [3] Organisation for the Promotion of Energy Technologies. The world’s largest biofuel CHP plant Alholmens Kraft, Pietarsaari. Available from /http://www.tekes.fi/opet/pdf/Alholma_2002.pdfS. [4] Caddet Renewable Energy. Straw-fired CHP plant in Rudkobing— providing environmentally. Technical brochure no. 95, IEA, 1998. [5] Caddet Renewable Energy. The Sabro straw-fired district heating plant. IEA, 1988, Technical brochure no. 88, 1998. [6] Jenkins BM. A comment on the optimal sizing of a biomass utilization facility under constant and variable cost scaling. Biomass Bioenergy 1997;13(1/2):1–9. [7] Jenkins BM. Biomass resources in California. Quantities, availability and costs of supply. Presented at the AETC-ASAE annual meeting, Louisville, KY, 14–16 February 2005. [8] Kumar A, Cameron JB, Flynn PC. Biomass power cost and optimum plant size in western Canada. Biomass Bioenergy 2003;24(6):445–64. [9] Larson ED, Marrison CI. Economic scales for first-generation biomass-gasifier/gas turbine combined cycles fueled from energy plantations. J Eng Gas Turbines Power 1997;119:285–90. [10] Flynn PC, Kumar A. Trip report: site visit to the 240 MW Alholmens power plant, Pietarsaari, Finland. Report prepared for BIOCAP Canada Foundation, Queen’s University, Kingston, Ont. and the Province of British Columbia, November 2005, 13pp. Available from /http://www.biocap.caS. [11] Kumar A, Sokhansanj S, Flynn PC. British Columbia’s beetle infested pine: biomass feedstocks for producing power. Final report prepared for BIOCAP Canada Foundation, Queen’s University, Kingston, Ont. and the Province of British Columbia, April 2005, 56pp. Available from /http://www.biocap.caS. [12] Eng M. Initial analysis of the availability of a wood supply for power generation as a result of as a result of the current mountain pine beetle outbreak. Internal report, Research Branch, Ministry of Forests and Range, BC, 2005. [13] Kumar A, Flynn PC, Sokhansanj S. Feedstock availability and power costs associated with using BC’s beetle-infested pine. Final report prepared for BIOCAP Canada Foundation, Queen’s University, Kingston, Ont. and the Province of British Columbia, November 2005, 49pp. Available from /http://www.biocap.caS. [14] Wood SM, Layzell DB. A Canadian biomass inventory: feedstocks for a bio-based economy. Final report. Prepared for Industry Canada Contract #5006125. BIOCAP Canada Foundation, Queen’s University, Kingston, Ont., Canada, 2003. [15] Ministry of Forests and Range (MoFR). Timber supply and the mountain pine beetle infestation in British Columbia. Forest Analysis Branch, British Columbia, October 2003. [16] Pacific Forestry Centre. Mountain pine beetle. Insect–host interactions, outbreak origin. Canadian Forest Service, Natural Resources Canada, 2005. Available from /http://www.pfc.forestry.ca/entomology/mpb/ outbreak/interactions_e.htmlS. [17] Eng M. Personal communication. Research Branch, Ministry of Forests and Range, 2005. [18] Simpson WT. Equilibrium moisture content of wood in outdoor locations in the United States and worldwide. Research note FPLRN-0268, Forest Products Laboratory, Forest Service, United States Department of Agriculture, 1998.

ARTICLE IN PRESS A. Kumar et al. / Renewable Energy 33 (2008) 1354–1363 [19] Environment Canada. Canadian climate normals or averages 1971–2000, 2005. Available from /http://www.climate.weatheroffice. ec.gc.ca/climate_normals/index_e.htmlS. [20] Demirbas A. Calculation of higher heating values of biomass fuels. Fuel 1997;79(5):431–4. [21] Simpson WT. Specific gravity, moisture content, and density relationship for wood. General technical report FPL-GTR-76, Forest Products Laboratory, Forest Service, United States Department of Agriculture, 1993. [22] MacDonald J, Sauder T. Personal communication. Western Division, Forest Engineering Research Institute of Canada, April 2005. [23] Simpson WT, TenWolde A. Chapter 3—Physical properties and moisture relations of wood. Wood handbook—wood as an engineering material. General technical report FPL-GTR-113, Forest Products Laboratory, Forest Service, United States Department of Agriculture, 1999. [24] Broek RVD, Faaij A, Wijk AV. Biomass combustion power generation technologies. Background report 4.1 for the EU Joule II+ project: energy from biomass: an assessment of two promising systems for energy production. Report 95029, Department of Science, Technology and Society, Utrecht University, 1995. [25] MacDonald J. Personal communication. Western Division, Forest Engineering Research Institute of Canada, May 2005. [26] Stennes B, McBeath A. Bioenergy options for woody feedstock. Final draft report. Pacific Forestry Centre, Canadian Forestry Service, Victoria, BC, Canada, July 2005. Available from /http://mpb.cfs. nrcan.gc.ca/research/workshops/8-37/Stennes-McBeath_e.pdfS. [27] Sherrod H, Saarivirta P. Personal communication. Regional Manager, Sales and Service, Kvaerner Power North America, Irving, TX, USA, 75062 and Production Sales Manager, Power Generation

[28]

[29]

[30] [31]

[32] [33]

[34]

[35]

[36] [37]

1363

Systems, Kvaerner Power Oy, Kelloportinkatu 1D, Tampere, respectively, 2005. Marrison CI, Larson ED. Cost versus scale for advanced plantationbased biomass energy systems. In: The US Proceedings of the second biomass conference of the Americas: Energy, Environment, Agriculture, and Industry, Portland, Oregon, August 1995. p. 1272–90. Sherrod H, Saarivirta P. Personal communication. Regional Manager, Sales and Service, Kvaerner Power North America, Irving, TX, USA, 2006. Williams D. Personal communication. Chief estimator for Bantrel Corporation (an affiliate of Bechtel), Alberta, Canada, 2002. Cameron JB, Kumar A, Flynn PC. Does gasification improve the economics of biomass utilization? Presented at the second world conference and technology exhibition on biomass for energy, industry and climate protection, 10–14 May 2004, Palazzo del Congressi, Rome, Italy. Setala M. Personal communication. Production supervisor, Alholmens Kraft power plant, Pietarsaari, Finland, 2005. Perlack RD, Turhollow AF. Assessment of options for the collection, handling, and transport of corn stover. Report no. ORNL/TM-2002/ 44, 2002. Kumar A, Cameron JB, Flynn PC. Pipeline transport and simultaneous saccharification of corn stover. Biores Technol 2005; 96(7):819–29. Nguyen MH, Prince RGH. A simple rule for bioenergy conversion plant size optimization: bioethanol from sugar cane and sweet sorghum. Biomass Bioenergy 1996;10(5/6):361–5. Overend RP. The average haul distance and transportation work factors for biomass delivered to a central plant. Biomass 1982;2:75–9. Shilling NZ. Personal communication. Leader—process power plants, energy products. General Electric, USA, 2004.