HomeMy WebLinkAboutGalena_biomass_assessment 2012 Assessment of Woody Biomass Energy Resources: Galena, Alaska: Presented to: State of Alaska Department of Natural Resources, Division of Forestery 3700 Airport way Fairbanks, AK 99709 By: Will Putman Tanana Chiefs Conference, Forestry Program 122 First Ave., Suite 600 Fairbanks, AK 99701 November, 2012
Assessment of Woody Biomass Energy Resources, Galena, Alaska i Executive Summary As part of an effort to assess the feasibility of proposed biomass energy projects at the community of Galena in Interior Alaska, an assessment of woody biomass resources was conducted for the vicinity of Galena. The assessment attempt s to leverage existing information as much as possible, including forest inventory information compiled by Tanana Chiefs Conference for previous projects and classified satellit e imagery. The assessment was completed as part of the work required for completion of assessments at 8 other communities in western Interior Alaska, including the other 3 communities that, along with Galena, are the villages that make up the holdings of Gana-A’Yoo Ltd., an ANCSA village corporation formed through the consolidation of the ANCSA entitlements for the villages of Galena, Koyukuk, Nulato, and Kaltag. The area considered for the assessment was defined by a 25-mile radius from Galena (~1.25 million acres), with the area additionally constrained to exclude areas closer to neighboring communities. A number of cost parameters were assumed and used to estimate costs of harvesting, transporting, and managing biomass resources across the landscape. The assessments result in woody biomass stocking and annual allowable cut estimates stored and maintained in a geodatabase, with the ability to query and report data by land cover type, ownership, biomass growth, biomass cost, distance from village, and ot her parameters. Highlights of the resulting data analysis include: • The percentage of land area determined to be associated with forested timber -bearing strata in the Galena project area was determined to be 31%. • As determined in the analysis, total wood y biomass in the Galena project area was determined to be about 5.5 million air-dry tons. • Using some simple growth modeling and estimates of existing stocking, estimates of Annual Allowable Cut (AAC) were generated. Total AAC for the entire Gale na project area was estimated at 166,741 tons. • Using the cost parameters assumed in the analysis, the cost of harvesting, transporting and managing the woody biomass was determined to range from $40 to $256 per ton. Not surprisingly, the most expensive biomass is farthest from the community because of the effect of the estimated transportation cost parameters. • There are extensive biomass stocks on Federal, State, and ANCSA corporation land holdings, but the areas closest to the villages are dominated by ANCSA corporation ownerships. • The data indicate the presence of significant amounts of recoverable woody biomass, particularly when viewed in terms of supporting relatively modest-sized thermal heating projects. Larger -scale projects, more demanding economic threshol ds, and information demands required by more detailed planning will require the collection and analysis of additional data.
Assessment of Woody Biomass Energy Resources, Galena, Alaska ii Table of Contents Land Cover ................................................................................................. 4 Forest Inventory data ................................................................................. 5 Woody Biomass Units ................................................................................. 7 Land Ownership ......................................................................................... 8 Site class 8 Estimating AAC and assigning rotation and growth parameters ................. 9 Cost modeling .......................................................................................... 11 List of Figures Figure 1: Location of Galena. ............................................................................ 2 Figure 2: Location of 25-mile radius Galena project area. ..................................... 3 Figure 3: Land ownership, Galena project area. ................................................ 20 Figure 4: Woody biomass dry ton stocking, Galena project area. ......................... 21 Figure 5: Woody biomass cost, Galena project area. .......................................... 22 List of Tables Table 1: Wood density of tree species in Interior Alaska. ...................................... 7 Table 2: Cost parameters used in the analyses. ................................................ 11 Table 3: Galena Biomass by Land Ownership .................................................... 17 Table 4: Galena Biomass by Village Proximity. .................................................. 17 Table 5: Galena Biomass by Estimated Cost. .................................................... 18 Table 6: Galena Biomass Dry Tons by Ownership and Village Proximity. ............... 18 Table 7: Galena Biomass by species. ............................................................... 19 INTRODUCTION............................................................................................. 1 DATA COMPONENTS ...................................................................................... 4 DATA PROCESSING AND ANALYSIS ............................................................. 13 RESULTS ...................................................................................................... 16 FUTURE STEPS............................................................................................. 23
Assessment of Woody Biomass Energy Resources, Galena, Alaska 1 INTRODUCTION Rapidly increasing fossil fuel costs have resulted in a heightened sense of urgency when considering the ability of small communities to absorb these costs and maintain so me sense of community sustainability. There are few places where this is more severe than rural communities in Interior Alaska, where fossil fuel dependence, energy costs, and remoteness are conspiring to produce an energy crisis that is becoming increasi ngly difficult for these small communities to deal with. These conditions have resulted in increased interest in any available form of alternative energy that may possibly be deployed. In Interior Alaska, the presence of apparently large amounts of woody biomass has increased the consideration of biomass energy systems to help address this crisis. Funding was provided by the Alaska Energy Authority to the State of Alaska Division of Forestry (DOF) to conduct biomass assessment work at several communities, including Galena , and a cooperative agreement was set up between DOF and the Tanana Chiefs Conference (TCC) Forestry Program to facilitate the work. At the same time, TCC Forestry was also contracted by the Interior Regional Housing Authority to conduct biomass assessment work at eight communities in western Interior Alaska, including the 3 communities that, along with Galena, make up the holdings of Gana-A’Yoo Ltd., the Alaska Native Claims Settlement Act (ANCSA) corporation formed by merging the ANCSA assets of the villages of Galena, Koyukuk, Nulato, and Kaltag. As a result, the data processing required to assemble information for Galena was included as part of an overall effort for the 4 Gana-A’Yoo villages. The assessment is designed to be prelimi nary in nature, and is intended to leverage existing information as much as possible. Additional work of a more detailed nature is currently being conducted at Galena; this assessment should be considered to augment that effort if at all possible, but is in any case not intended to replace it. Galena, with a population of 487 (2011 DCCED estimate), is located on the Yukon River in western Interior Alaska (Figure 1.) Although Galena serves as a central hub community for the region, it is not on a contiguous highway system, and is accessible only by air, water, or overland trails. The largest nearby urban centers providing goods and services are Fairbanks and Anchorage, aprox. 272 and 328 air miles away, respectively. With any proposed woody biomass energy project, a number of basic questions arise concerning the biomass supply, including: • How much biomass is there in the vicinity of the community? • What are the characteristics of the biomass (size, species, quality)? • Where is the resource located? • Who own s the resource? • What are the costs associated with getting the resource to an energy facility? • What management restrictions are there are on the resource? • Considering growth rates, cover type conversions, and other factors, what is the sustainability of the resource? • How large an array of biomass energy facilities could be economically supported on a sustainable basis by the local biomass resource?
Assessment of Woody Biomass Energy Resources, Galena, Alaska 2 Figure 1: Location of Galena. This report is an attempt to document an approach to answer these questions with available information, using information management tools such as a geographic information system (GIS) and relational databases. The process described here is meant to present a model for the handling of information to answer these questions, and in that regard does not constitute an end product. In those cases where information is lacking or unavailable, assumptions have been made and documented, with the idea that improved information in the future can be used to improve the model. It is intended that the model itself be a useful tool in the land management required to support proposed biomass energy projects. The geographic extent of the community’s assessment was defined as a radi us of 25 miles surrounding Galena. In addition, only those areas that were closer to Galena than an adjacent community were included in the assessment extent , so that some areas closer to Koyukuk or Nulato were excluded (Figure 2).
Assessment of Woody Biomass Energy Resources, Galena, Alaska 3 Figure 2: Location of 25-mile radius Galena project area .
Assessment of Woody Biomass Energy Resources, Galena, Alaska 4 DATA COMPONENTS The biomass assessment relied heavily on computerized geographic information system (GIS) and relational database technologies to store, process, query, and analyze data. The GIS software used was ArcGIS 10.0 from ESRI, Inc., and the relational database software used was Microsoft Access. The GIS was used to spatially define the location of various attributes of the landscape, the combination of those attributes for any giv en location on the landscape, and to produce acreages and biomass stocking estimates associated with any combination of attributes. A relational database was used to relate the attribute information stored in GIS data layers to tabular datasets such as biomass stocking information derived from existing forest inventory datasets, cost parameters, and lookup tables, allowing the generation of GIS layers of derived information such as biomass stocking, annual allowable cut, and biomass cost estimates. The b asic data input to the biomass assessment model consisted of land cover data, forest inventory data, and land ownership. Additional data components were derived from the basic input datasets, including raster datasets for site class, biomass stocking, biomass annual allowable cut, village proximity, and biomass cost estimates. Land Cover Typically, land cover is characterized from sources of remotely sensed image data such as aerial photography or satellite imagery. For Gal ena, high resolution imagery (spatial resolution <= 1m) was available, and there is the possibility of medium resolution (spatial resolution 2.5m) Spot 5 imagery becoming available through the Alaska Statewide Digital mapping Initiative (SDMI) for portions of the project area . However, the time and funding required to classify the imagery into classified land cover data layers was prohibitive given the scope of the project . As a result, it was decided instead to attempt to rely on classified image layers made available through the LandFire program, an interagency vegetation, fire and fuel characteristics mapping program sponsored by the U.S. Department of the Interior and the U.S. Forest Service (http://www.landfire.gov). LandFire data products consist of up to 50 data layers generated for all land areas within the United S tates, including Alaska. Within Alaska, the data layers are generated from classified LandSat satellite imagery at a spatial resolution of 30 meters. Existing vegetation is described in 3 layers ; existing vegetation type (evt), existing vegetation height (evh) and existing vegetation cover, or density (evc). In addition, a layer for biophysical settings (bps) showed potential for attempting to model potential productivity of a site. Th e advantages of the LandFire data include the comprehensive coverage of the data over the entire country, and the apparent detailed vegetation classification that appeared to be relatable to forest inventory stocking data from old forest inventory data on file at TCC. Potential disadvantages of use of the LandFire data include it’s relatively coars e spatial resolution (30m), and anecdotal and objective evidence that would lead one to question the accuracy of the LandFire classifications. In either case, t he landscape-level nature of this biomass assessment and the preliminary nature of the assessment led to the decision to utilize the LandFire datasets.
Assessment of Woody Biomass Energy Resources, Galena, Alaska 5 The LandFire data layers are provided as raster datasets, with classifications provided for individual pixels, or cells in an image. This is in contrast to vector datasets, which define areas as polygons defined by line segments running between x-y coordinates. Previous analyses compiled by TCC have used land cover data in vector formats, and relied on standard vector overlay techniques to analyze biomass stocking with ownership, proximity to a village, etc. Using land cover data in a raster format dictated that the analyses for these assessments be based on raster techniques and processing. Forest Inventory data In I nterior Alaska, as in many places, woody biomass is a forest resource. The process of trying to assess the amount and location of forest resources falls under the purview of forest inventories, a traditional and essential component of forestry and forest management. This project is essentially a form of forest inventory, with particular interests and requirements that are driven by the land management required to support proposed biomass energy projects. The most prominent forest inventor y effort s to date in the vicinity of Galena are a number of inventories conducted by the Forestry Program at Tanana Chiefs Conference on village corporation lands and Native allotments. The village corporation inventories were conducted on individual village ANCSA corporation lands with a selected status at the time of the inventory. ANCSA village corporation inventories were conducted for all 4 Gana-A’Yoo villages in the 1980’s and 90’s, although field sampling was only conducted at Koyukuk and Kaltag. In addition, an inventory was conducted upriver on village corporation lands at Ruby in 1987. For the allotment inventories, the entire TCC region was subdivided into 8 subunits, and a separate inventor y project was conducted for each subunit, with the overall work occurring from 1987 to 1993. In the vicinity of Galena, allotment inventories were conducted for the Doyon-Melozitna and Doyon-Koyukuk subunits. The Native allotment inventories were conducted on Native allotment parcels with a status of pending or better at the time of the inventories. The protocols and processes used in the corporation and allotment inventories were very similar, and utilized a process that included the following steps: 1. The area included in the inventory was interpreted for land cover type using high -altitude color-infrared aerial photographs dating from the late 1970s. 2. Forested stands delineated on the aerial photographs were attributed with a cover type code that included a determination of primary tree species, primary tree size class (dwarf, reproduction, poletimber, sawtimber), secondary tree species, secondary species tree size class, and overall tree density (low, medium, and high crown closure). Non -forested areas were attributed for cover types such as water, tall shrub, bog, barren/cultural, etc.
Assessment of Woody Biomass Energy Resources, Galena, Alaska 6 3. Forested cover types covering the highest proportion of area were selected for field sampling by randomly selecting accessible stands within those types. 4. Field sampling was accomplished by visiting the selected stands on the ground and installing a series of variable radius plots and conducting tree measurements. Sample trees were measured for species, tree diameter, tree height, and percent defect, and a small number of white spruce trees were measured for radial growth and age 5. The collected field data were processed and compiled in the office with a computer to produce timber volume per acre figures by species and size class within strata defined as groupings of similar cover types. 6. The volume per acre figures were then extrapolated to all forested areas withi n the extent of the project. 7. Some years after the completion of the inventories in the early ‘90s, the spatial data represented by the cover type maps prepared in the invent ory were digitized into a GIS, and the processed timber volume data was incorporated into a digital relational database. The most important component provided to the biomass assessment as a result of the forest inventories are the tabular timber volume and stocking estimates. The stocking data generated from the field measurements are used to produce estimates of the amount of woody biomass present in each forested cover type. The tree data processing produced estimates of board -foot and cubic-foot t imber volume s per acre by tree species and size class. For the purposes of evaluating a forest resource as an energy source, it is most appropriate to focus on the cubic -foot estimates, since they represent the total woody biomass volume on the main stem of trees below a minimum top diameter (usually 4”), and not just the amount of recoverable wood when processing trees for lumber. There are a number of serious limitations in this available forest inventory data that need to be considered. The inventor ies a re quite “extensive”, that is, the geographic scope was relatively large and the intensity of the field sampling was relatively low, particularly for the allotment inventories . Forest cover types with relatively low acreages were not field sampled at all, but were lumped into similar types that were sampled, with resulting inaccuracies in the volume estimates. The photography used to produce the land cover typing was less than 15 years old at the time the inventories were conducted, but is now more than 30 years old, and does not take into account the changes that have no doubt occurred on the landscape. The data collection was focused on the standing stock, and what little growth information was collected is difficult to apply in any meaningful way with regards to estimates of site and forest growth. Only the biomass represented by the main boles of trees is included in the volume estimates, with no attention paid to whole tree biomass or non -timber species such as alder or willow. That being said, the data contained in th ese old inventory project still provide a useful starting point for evaluation of woody biomass energy resources.
Assessment of Woody Biomass Energy Resources, Galena, Alaska 7 Table 1: Wood density of tree species in Interior Al aska. White spruce, Paper birch, Aspen and Balsam poplar figures are from the State of Alaska, Department of Commerce (http://www.commerce.state.ak.us/ded/ dev/for est_products/forest_products5.htm); Black spruce figures are from a Canadian website maintained by Lakehead University in Ontario (http://www.borealforest.org/); Tamarack figures are from an engineering website (http://www.engineeringtoolbox.com/weigt-wood -d_821.html). Tree Species Green Density (lbs/cubic foot) Air -dry density (lbs/cubic foot) Air -dry tons/cord White spruce 36 31 1.31 Black spruce 32 28 1.19 Paper birch 48 38 1.62 Aspen 43 27 1.15 Balsam poplar 38 24 1.02 Tamarack 47 37 1.57 Woody Biomass Units As mentioned previously, the cubic-foot (CF) estimates of wood volume that are one of the products of a forest inventory analysis are appropriate when evaluating the volume of woody biomass as an energy source. However, the energy value of wood per unit volume varies somewhat by species because of varying wood densities, so it is common to report woody biomass in units of weight, commonly tons (1 ton=2,000 lbs). This matter is further complicated by the variability of wood weight per unit volume because of moisture levels in the wood. There are three units commonly used to report woody biomass by weight: Green tons, or the weight of the wood in tons at moisture levels found when the material is freshly cut, often in the neighborhood of 50% moisture by weight; air dry tons, or the weight of the wood when enough moisture has been removed from the wood to make it feasible to efficiently recover energy from the wood through combustion, commonly in the neighborhood of 20% moisture by weight; and bone-dry tons, the weight of the wood with all moisture removed. For the purposes of this analysis, the unit of air-dry tons (also referred to in this document as “d ry tons”) is used, the weight of the wood in the form most likely to be used in a heating project. The literature is inconsistent in terms of wood density values for the species found in Interior Alaska, but representative values (and their sources) are p resented in Table 1. Another unit used to measure wood is the “cord”, traditionally used to measure fuelwood. A cord is defined as the amount of minimally processed wood (bucked, split) that can be stacked in a space measuring 4’x4’x8’. Because of the airspace and inconsistency inherent in stacking cordwood, the cord is a relatively imprecise measure, but is nonetheless in common use in fuelwood transactions. The volume space of a cord, 128 cubic feet, is
Assessment of Woody Biomass Energy Resources, Galena, Alaska 8 sometimes thought to contain roughly 100 cubic feet of wood (a “cunit”) when the air space between wood chinks in the stacked wood is considered. Other estimates put the conversion at 85 cubic-feet of roundwood per cord. Using the conversion factors presented in Table 3 at 85 CF/cord, the number of ai r-dry tons in a cord varies from approximately 1.0 tons for balsam poplar to 1.6 tons for paper birch. Land Ownership A key component of the analysis is the determination of which individual or organization owns or has management responsibilities for the lands on which the biomass resource is found. In th is analysis, this is accomplished through the use of a GIS layer that defines land ownership in the vicinity of the project . Spatial data of land ownership were acquired fro m several sources and combined into the ownership layer: • Generalized land status, from the Bureau of Land Management (BLM) • Native Allotments, from BLM • ANCSA Corporation conveyed lands, from Doyon, Ltd. The data from the various sources vary in quality and precision; specifically, the generalized land status data is available statewide, but only shows categories of land ownership to the nearest section (square mile). Because of that, allotment lands and ANCSA corporation lands as defined in the other sources were given priority over the generalized land status when combining the land ownership data. The data acquired from Doyon for conveyed ANCSA lands allows the land status to be defined for the regional corporation (Doyon Ltd.) and the village corporations, but the ANCSA land as coded in the BLM generalized land status made no distinction between regional and village corporations. As a result, any listing of individual ANCSA corporations in the results in this report refers to the location of conveyed ANC SA corporation lands as defined in the Doyon data, and any reference to “ANCSA misc.” refers to ANCSA selected or patented lands as defined in the BLM generalized land status data with no distinction between individual corporations. Site class It was assumed that site productivity is a critical factor when attempting to determine the growth of biomass on the landscape, a key factor when evaluating biomass sustainability. For the purposes of this analysis, four broad site classes were defined to describe the location of site class areas in the project area. The four site classes defined were: • Site Class 0 – areas incapable of producing woody biomass such as rivers, lakes, seasonally submerged sandbars, wetland bogs, etc. • Site Class 1 – areas of relatively poor site in terms of woody biomass production, such as poorly drained areas and north-facing slopes with underlying continuous permafrost. These sites may have cover types such as tall shrubs, dwarf shrubs (dwarf birch, etc .), black spruce or other slow -growing unproductive cover types. • Site Class 2 – areas of intermediate productivity such as lower slopes adjacent to wetlands, areas underlain by permafrost but with some productive tree cover, etc. • Site Class 3 – Areas of relatively high productivity such as south-facing slopes, well-drained benchlands, and productive riparian sites.
Assessment of Woody Biomass Energy Resources, Galena, Alaska 9 In the GIS, all parts of the project area w as classified into one of the four site classes, using the LandFire biophysical settings (bps) layer and a lookup table in the database assigning a site class to each bps classification, creating site class raster datasets for covering the project area. Estimating AAC and assigning rotation and growth parameters In order to assess sustainability, the traditional forestry concept of Annual Allowable Cut (AAC) was applied. AAC is deemed to be the maximum level of annual harvest that is possible in perpetuity without diminishment of the level of harvest or the amount and quality of the resource. There are a variety of techniques used to calculate AAC, including the “Hanzlik formula”, which was designed to attempt to deal with areas still in an unmanaged “old-growth” state. The Hanzlik formula uses mature standing volume, rotation length, and growth (increment) as parameters required to calculate AAC: Allowable cut (AAC) = (Mature Standing Volume / Rotation ) + Growth Standing volume is determined from the inventory data as described above, but figures for rotation length and growth are more difficult to determine or estimate. “Rotation”, or “rotation length” refers to the hypothetical length of time required for a forest stand to reforest, grow, and replace itself after harvest. At first glance this appears quite simple, but there are a number of complicating factors, including: • What species the stand regenerates to – different species will grow at different rates and mature at different time intervals. • Site potential may vary over time; in fact, in Interior Alaska, the act of harvesting (or other disturbances, such as fire) may change the growth potential of a site, and as a result, the anticipated rotation length. • Anything other than even-aged management may complicate the determination of rotation length, particularly if it involves multiple tree species and multiple stand entries in a rotation. • Differing economic conditions or other factors may dictate a different array of forest products requiring material to reach different sizes or ages to be marketable. Similarly, “growth” can be a concept that may be simple to visualize, but involves a number of factors that make it difficult to determine with any precision. The ability to gauge the capacity of woody biomass to grow and replace itself after harvest is a critical component of any assessment that would attempt to evaluate the sustainability of the resource. Unfortunately, this is one area where hard data to drive the analysis is in short supply. It is an exce edingly complex situation that is being modeled – growth rates of individual trees and the stands they grow in vary by site, species, tree age, stand age, stand density, reproductive capacity, disturbance regime, and other factors, and all in cumulative and interactive ways. Growth models for the boreal forest are in development at the University of Alaska Fairbanks and with the U.S. Forest Service and may prove to be useful. In the meantime, this effort applies some broad and exceedingly gross assumption s in an attempt to get a handle on growth and sustainability.
Assessment of Woody Biomass Energy Resources, Galena, Alaska 10 For both growth and rotation, the approach taken was to establish an optimal value for each, then adjust the values based on other conditions. Based on TCC inventory data, maximum biomass stocking in high-volume spruce stands, presumably on good sites, is in the neighborhood of 60 tons/acre. Employing the concept of mean annual increment (MAI), and assuming a stand age of 120 years to produce this volume, this would indicate a maximum mean annual increment of 0.5 tons/acre/year on the best sites. Interestingly, roughly similar rates can be arrived at with productive hardwood stands; TCC’s inventory data indicates total biomass tons of well -stocked cottonwood, birch, or aspen stands to be in a somewhat lower range (~20-50 tons/acre), with lower stand ages to be expected to produce those volumes (~50-80 years). Based on this, a value of 0.5 tons/acre/year is assumed as an optimum mean annual growth rate. Optimal rotation length is assumed to be 50 years, based on a hypothetical rotation length for the deciduous broadleaf tree species (birch, aspen, and balsam poplar). Although white spruce has traditionally been the favored species for timber management in Interior Alaska, it is assumed that managing for hardwoods is desirable from a woody biomass perspective because of faster juvenile growth rates, shorter rotations, ease in regenerating, importance in wildlife habitat, and desirability from a community wildfire protection perspective. Several key assumptions were made to facilitate adjusting the optimum growth and rotation figures based on the availability of existing information. The assumptions used in this analysis to estimate growth and rotation include: 1. Fully stocked stands will show best realization of potential growth . 2. Lower site quality will result in longer rotations and slower growth. The first assumption of stand stocking levels influencing relative growth can be dealt with most directly using the stand density component of the cover type calls coming from the LandFire evc layer. Each of the evc codes related to density of a tree canopy were assigned a relative growth rate expressed as a proportion of optimum growth: LandFire evc Class Growth Proportion 151 (Tree Canopy >= 10 and < 25%) 0.3 152 (Tree Canopy >= 25 and < 60%) 0.6 153 (Tree Canopy >= 60 and <= 100%) 1.0 Similarly, the second assumption of relative growth varying by site quality was handled by taking the site class codes as assigned to areas on the lan dscape and adjusting the optimal rotation of 50 years upwards for poorer site classes, as well as assigning degraded growth proportions for lower sites: Site Class Growth proportion Rotation (years) 0 0 none 1 0.3 90 2 0.6 70 3 1.0 50 Using this ap proach, annual allowable cut was seriously degraded for those areas interpreted to be of poor site quality, by calculating a lower current growth and by using a longer rotation in the AAC formula.
Assessment of Woody Biomass Energy Resources, Galena, Alaska 11 Table 2: Cost parameters used in the analyses. Cost Type Cost Stumpage (payments to owner), cost per ton $ 5 Harve st Costs Costs per acre $300 Costs per ton of woody biomass $ 10 Transportation costs Cost/ton/mile off -road $ 6 Reforestation – cost per acre $100 Misc. Admin – cost per acre $ 20 By applying a series of update queries in the database, al lowable cut was determined for the project areas; since this was a raster analysis, this was done on a pixel -by-pixel basis based on the LandFire datasets. Growth for each pixel was determined by multiplying the optimum growth rate (0.5 tons/acre/year) by the growth proportion number assigned to the stand density of the pixel , and multiplied again by the growth proportion assigned to the site class of the pixel . Rotation length for each pixel was determined by applying the rotation length assigned to the site class of the area . The resulting figures for growth and rotation were used with the overall stocking of each pixel in the Hanzlik formula to generate an AAC for each pixel . The resulting AAC figures for each pixel are not meant to mean that some calculated portion of every pixel is a portion of the volume cut in any given time frame, but refers to the contribution that the resource represented by the area of that pixel contributes to the harvestable volume of biomass over the project as a whole. Through the other attributes assigned to each pixel through the creation of overlaid raster datasets, both standing stock and AAC figures can be broken out by ownership, proximity to the village, or other area attributes. Cost modeling In addition to estimates of the amount and growth of the woody biomass resource, it is also useful to estimate the costs involved in making the biomass available to an energy facility. This estimation could include the modeling of costs associated with harvesting, transport, reforestation, stumpage, and other costs. At this stage of the project, much is unclear in terms of type of harvest and equipment to be used, the nature and extent of the transportation network to be established and other cost factors, but all of these factors can be modeled in the GIS and reported back from the database. Table 2 presents a list of cost factors used in this analysis as an example of how these costs could be modeled. Per acre costs were converted into costs per ton. Per acre cost parameters such as harvest costs per acre and reforestation costs per acre have the effect of driving up relative costs per ton of woody biomass for low volume areas. Es timated transportation costs were driven solely by distances from the village, with the off-road transportation cost parameter of $6/ton/mile being applied. Harvest costs are broken into two components, cost per ton and cost per acre (Table 4). This is an attempt to recognize that some costs associated with
Assessment of Woody Biomass Energy Resources, Galena, Alaska 12 harvesting will remain relatively fixed per ton, while other costs associated with mobilization, equipment movement, etc. may remain relatively fixed per unit area. Other costs associated with biomass supply could include reforestation costs and other management costs, and stumpage payments made to a landowner. The reforestation costs initially used in this analysis are based on a lowering of known planting costs, assuming that some level of natural regeneration or other techniques may be used. This cost modeling can be modified in the future with changes to the cost parameters, modification of the modeling used to assign costs, etc. to create updated cost scenarios. Since the cost per ton is determined by area, as is the annual allowable cut, one interesting ramification of this is that it is possible to evaluate AAC based on different cost thresholds.
Assessment of Woody Biomass Energy Resources, Galena, Alaska 13 DATA PROCESSING AND ANALYSIS Starting with the basic datasets described above, there were several data processing steps that were conducted to prepare and analyze the data and prepare for the generation of tables and maps showing the analysis results. The spatial data raster processing steps described below used geoprocessing tools in the GIS software, with the use of the tools being automated somewhat through the creation of script tools written in Python, a scripting language used with ArcGIS software. The data processing steps implemented for each project area w ere: 1. Data were downloaded from the LandFire website for the LandFire data layers to be used in the analysis. Since the Gana-A’Yoo villages are close enough to each other that a 25-mile radius around any one of the villages overlaps that of adjacent villages, a spatial extent defining an area including 25-mile radii around all 4 villages was used to define the LandFire data download, and all 4 villages were processed together. Data were downloaded for the LandFire evt, evc, evh, and bps layers in ArcInfo GRID format. 2. A geodatabase was created, and the downloaded data were imported into it as raster datasets. All resulting datasets, both raster and vector, were also stored in the geodatabase, which was created as an ArcGIS personal geodatabase in MS Access format, and which also served as the repository for the other data base structures in the analysis; lookup tables, strata stock tables, queries, data entry forms, reports, etc. 3. The evt, evc, and evh layers were combined into a new raster layer (called lf_tch) containing the combined attributes of vegetation type, vegetation cover (density) and vegetation height. This produced a VAT (value attribute table) describing all possible combinations of the attributes from the combined raster layers. In the case of the 3 Gana-A’Yoo villages being processed together, this produced a table of 374 combinations of vegetation type, coverage, and height classes. 4. The VAT was exported into a database table, (called tch_classes), and a column was added to the table to hold information on strata ID. 5. Each row in the tch_classes table was assigned a strata ID from the TCC forest inventories . Non-forested vegetation types (shrubland, wetland, water, barren, etc.) were assigned to non -forested strata not associated with any timber volume. Forested vegetation types were subjectively assigned to the most appropriate strata from nearby TCC forest inventory projects. To aid in this rather complex, manual, and very subjective process, a form was developed in MS Access. 6. The database contained a table called strata_biomass that had been processed to contain biomass stocking values (in tons and cords) for all strata defined in the TCC inventories. In ArcGIS, the tch_classes table and the strata_biomass table were joined and the tch_classes table and the lf_tch VAT were joined to associate each cell in the combined vegetation raster with strata biomass stocking values. This joined raster is used to create a series of raster datasets of biomass stocking with the ArcGIS Spatial Analyst “lookup” command. Raster datasets were created for overall dry ton stocking, dry tons by species, and cords by species. 7. Similarly, a site class raster was created for the project area. The LandFire biophysical settings raster (lf_bps) VAT was exported to a database table
Assessment of Woody Biomass Energy Resources, Galena, Alaska 14 (bps_classes) and a column for site class code was added to the bps_classes table, and each row of the bps_classes table was coded for site class using the codes 0 through 3 described above. In ArcGIS, the bps_classes table was joined to the lf_bps raster layer, and the “lookup” command was used to create a site class raster for the project area. 8. A raster of annual allowable cut (AAC) was created by first creating rasters of growth adjustment by density values, growth adjustment by site values, and rotation adjustment by site values, and then executing a m ap algebra raster calculation for AAC using an application of the Hanzlik formula with the rasters for biomass stocking, growth as determined from the growth adjustment rasters, and rotation as determined from the rotation adjustment raster. The growth by density a djustment raster was created in a process similar to that used to create the stocking and site class rasters by joining the LandFire vegetation density raster (lf_evc) to a growth_by_density table in the database to relate the evc codes to a dens ity adjustment factor and creating a growth by density adjustment raster with a lookup command. Similarly, the growth adjustment by site and rotation adjustment by site rasters were created by joining the site class raster to lookup tables in the database (growth_by_site, rotation_by_site) and creating the adjustment rasters with lookup commands. 9. A raster dataset was created defining the proximity to the nearest village in miles up to a 25 mile radius using ArcGIS spatial analyst commands. In addition, a raster dataset defining which village was closest to each pixel in the dataset was created , to account for those villages whose 25-mile radii overlap. 10. A raster dataset of biomass costs per ton was created by applying the cost parameters described above to previously created raster datasets. A harvest cost raster was created by dividing the harvest per acre parameter by the biomass stocking per acre raster, and adding the result to the harvest cost per ton parameter. A transportation cost raster was created by multiplying the village proximity raster and multiplying it by the off-road transportation cost parameter. A total cost per ton raster was created by adding the harvest cost raster, the transportation cost raster, the reforestation parameter and the administration cost parameters divided by the biomass stocking raster (to convert those parameters to per -ton units), and the stumpage parameter. 11. A vector layer of land ownership was created for the project by overlaying generalize d land status (from BLM) with conveyed ANCSA land data (from Doyon, Ltd.), and Native allotment locations (from BLM). These are overlapping datasets, but a unique ownership was identified for all areas through the overlay commands applied, with a priority given to the location of Native allotment parcels, the next lowest priority given to the conveyed ANCSA data, and the least priority given to the generalized land status. The resulting polygons were attributed for owner and owner class. Native allotments were coded with the B LM serial number as the owner and “Native allotment” as the owner class. ANCSA conveyed lands were coded with the name of the ANCSA corporation as the owner (usually either the local village corporation or Doyon, Ltd.) and an owner class of “ANCSA corp”. The remaining lands were identified from the generalized land status data with some level of agency
Assessment of Woody Biomass Energy Resources, Galena, Alaska 15 ownership; State lands were identified as “State patented” or “State selected” as the owner and “State of Alaska” as the owner Class; federal lands identif y the agency (USFWS, NPS, BLM) as the owner and “Federal” as the owner class. To be compatible with the raster analysis used in these analyses, the tools used to query the data convert the vector ownership layer to a raster dataset for processing. 12. The lay ers described above for ownership, village, village proximity, and biomass cost were combined together into a single raster layer, called the “combined parameters layer”, attributed for all parameters. To do this, vector layers such as ownership were conv erted to rasters, and to keep the number of parameter combinations to a reasonable number, layers containing continuous data (biomass cost and distance to village) were converted into class categories; for example, instead of using the calculated biomass c ost numbers directly, the biomass costs were grouped into increments of $20/ton ($20 -40/ton, $40-60/ton, etc.), and the distances in the village proximity raster were converted to 1-mile classes (1-2 miles, 2-3 miles etc.). 13. Using spatial analyst commands i n ArcGIS, tables of statistics were generated by analyzing the stocking rasters with the combined parameters layer. Each table generated summarized one component of biomass stocking with all combinations of the parameters. Tables were generated for summa ry statistics for overall dry tons, dry ton annual allowable cut, dry tons by species, and cords by species. Once the statistics table were generated, it was possible to produce summary tables of biomass stocking by various attributes using standard datab ase reporting tools. The datasets resulting from the process described above allow querying and displaying the data with multiple combinations of attributes. For example, one can query the data to show those areas and the biomass stocking amounts for a pa rticular ownership and under a particular cost threshold. Or, perhaps one would want to query the data show the estimated annual allowable cut on a particular ownership within a specified distance of the village. Two tools were prepared as ArcGIS Python script tools to facilitate querying the data: 1. A GIS interactive query tool allows a user to interactively specify query parameters for village, ownership, owner class, and maximum biomass cost per ton, view the calculated values for total biomass and annual allowable cut in a brief tabular display, and have the areas in question highlighted on the map in ArcGIS. 2. A GIS statistics generation tool generates a table of statistics that is stored in the database and can be used to drive reports showing biomass stocking and annual allowable cut by distance class, cost class, owner, owner class, and village.
Assessment of Woody Biomass Energy Resources, Galena, Alaska 16 RESULTS Following are selected results of the analysis , with tabul ar results produced from the statistical summaries generated by the statistics generation tool described above, and sample maps of the generated spatial data. As indicated above, these results as displayed constitute only a portion of the possible combinations and ways to view the data, both in tabular form or on maps. “Forested area” refers to those portions of the project area that have been associated with a forest inventory stratum that have woody biomass estimates. It does not include those areas tha t have a LandFire classification not associated with any woody biomass stocking estimates, including low-volume types such as dwarf black spruce or shrubland types. As determined in this analysis, forested area for Galena is 312,233 acres, or 31% of the project area. The amount of biomass found on ANCSA corporation lands (both regional and village) was about 36% of the total (Table 3). Perhaps more importantly, 78% of the biomass within 10 miles of the village was found on ANCSA lands (Table 6), highlighting the importance of the ANCSA corporations, parti cularly the village corporation, Gana -A’Yoo, in the ownership of the most accessible, least expensive biomass resources. Over half of the estimated biomass stocking was found to be white spruce , with the bulk of the remainder, about 33%, determined to be comprised of birch (Table 7). One observation is the relatively low level of cottonwood stocking despite the known presence of extensive productive riparian cottonwood stands; this may be due to the subjective assignment of inventory strata to the LandFire land classifications. The nature of the LandFire classifications did not readily associate themselves with forest inventory cottonwood strata, and as a result, may be underestimated in the analysis.
Assessment of Woody Biomass Energy Resources, Galena, Alaska 17 Table 3: Galena Biomass by Land Ownership Annual Allowable Cut Forested Ownership Air -dry Tons Cords (AAC, tons/year) Acres Gana-A'Yoo Ltd. 798,384 575,741 23,104 39,918 Doyon Ltd. 1,169,666 852,559 33,722 62,951 Native allotment 53,436 38,358 1,617 2,633 BLM 1,102,514 782,841 35,903 70,923 USFWS 398,370 287,754 13,097 24,696 Military 25,517 18,707 731 1,379 State of Alaska 1,964,464 1,429,884 58,566 109,732 All ownerships: 5,512,350 3,985,844 166,741 312,233 Table 4: Galena Biomass by Village Proximity. Proximity to Annual Allowable Cut Forested village (miles) Air-dry Tons Cords (AAC, tons/year) Acres 0 - 1 6,514 4,521 187 391 1 - 2 34,004 24,718 847 1,636 2 - 3 45,148 33,206 1,121 2,055 3 - 4 59,715 44,020 1,501 2,563 4 - 5 81,242 60,216 1,942 3,410 5 - 6 99,967 74,083 2,278 4,273 6 - 7 107,509 79,018 2,683 4,911 7 - 8 138,166 100,738 3,665 6,429 8 - 9 164,996 119,536 4,652 7,855 9 - 10 174,777 127,990 4,742 8,362 10 - 11 228,961 165,665 6,946 12,739 11 - 12 237,166 170,966 7,406 13,673 12 - 13 232,511 166,982 7,158 12,635 13 - 14 245,528 175,988 7,560 13,720 14 - 15 220,909 158,405 6,921 12,772 15 - 16 237,234 170,920 7,384 13,717 16 - 17 281,332 201,119 8,886 16,184 17 - 18 322,256 231,207 10,153 18,721 18 - 19 353,743 253,602 11,229 20,412 19 - 20 360,884 260,704 11,180 21,006 20 - 21 361,259 260,370 11,327 22,305 21 - 22 360,091 259,069 11,295 22,190 22 - 23 370,126 270,211 11,219 22,512 23 - 24 359,006 260,881 11,118 21,937 24 - 25 429,306 311,708 13,341 25,825 Totals: 5,512,350 3,985,844 166,741 312,233
Assessment of Woody Biomass Energy Resources, Galena, Alaska 18 Table 5: Galena Biomass by Estimated Cost. Biomass Cost Annual Allowable Cut Forested ($/ton) Air-dry Tons Cords (AAC, tons/year) Acres 40 - 60 46,566 34,539 1,103 1,654 60 - 80 229,462 169,567 5,431 9,047 80 - 100 393,962 289,188 10,012 16,745 100 - 120 580,739 420,300 16,963 26,281 120 - 140 711,054 508,138 22,169 37,794 140 - 160 940,759 675,437 29,421 51,052 160 - 180 1,115,388 802,122 34,361 61,145 180 - 200 1,052,554 759,954 32,736 62,809 200 - 220 353,021 260,494 11,196 31,707 220 - 240 65,888 49,054 2,472 10,046 240 - 260 22,957 17,051 877 3,954 Totals: 5,512,350 3,985,844 166,741 312,233 Table 6: Galena Biomass Dry Tons by Ownership and Village Proximity. Land Ownership : Proximity to Native State of village (miles) ANCSA Corp. Allotments Federal Alaska Total 0 - 1 6,514 6,514 1 - 2 34,004 34,004 2 - 3 45,037 110 45,148 3 - 4 59,679 35 59,715 4 - 5 67,786 852 12,603 81,242 5 - 6 68,893 3,656 39 27,379 99,967 6 - 7 70,320 2,502 1,109 33,578 107,509 7 - 8 104,670 1,771 7,096 24,629 138,166 8 - 9 132,654 3,973 11,923 16,446 164,996 9 - 10 120,878 3,748 21,449 28,702 174,777 10 - 11 167,809 3,631 23,826 33,695 228,961 11 - 12 170,881 3,021 24,210 39,054 237,166 12 - 13 130,014 7,868 23,903 70,727 232,511 13 - 14 155,560 313 42,076 47,579 245,528 14 - 15 126,176 47,910 46,822 220,909 15 - 16 142,370 2,852 58,222 33,790 237,234 16 - 17 140,960 5,390 90,419 44,563 281,332 17 - 18 119,397 4,204 125,204 73,450 322,256 18 - 19 70,188 1,497 141,698 140,360 353,743 19 - 20 29,536 3,437 136,130 191,781 360,884 20 - 21 4,582 343 167,192 189,142 361,259 21 - 22 142 155,695 204,254 360,091 22 - 23 161,035 209,091 370,126 23 - 24 2,726 146,387 209,894 359,006 24 - 25 1,541 140,878 286,888 429,306 Totals: 1,968,050 53,436 1,526,401 1,964,464 5,512,350
Assessment of Woody Biomass Energy Resources, Galena, Alaska 19 Table 7: Galena Biomass by species. Tree Species Air -dry Tons Cords % of Total White Spruce 3,091,238 2,346,291 56.1% Black Spruce 380,214 319,508 6.9% Birch 1,842,779 1,141,039 33.4% Aspen 139,804 121,834 2.5% Cottonwood 58,316 57,172 1.1% All Species 5,512,350 3,985,844 100.0%
Assessment of Woody Biomass Energy Resources, Galena, Alaska 20 Figure 3: Land ownership, Galena project area .
Assessment of Woody Biomass Energy Resources, Galena, Alaska 21 Figure 4: Woody biomass dry ton stocking, Galena project area.
Assessment of Woody Biomass Energy Resources, Galena, Alaska 22 Figure 5: Woody biomass cost, Galena project area.
Assessment of Woody Biomass Energy Resources, Galena, Alaska 23 FUTURE STEPS As plans for proposed biomass heating projects move forward, what steps need to be taken to implement effective and sustained use of forest resources as a woody biomass supply at villages in Interior Alaska? This report constitutes a first-look assessment designed to assist in determining if the potential supply of woody biomass warrants pursuing the development of proposed biomass energy projects. Work underway by the Louden Council and Gana-A’Yoo Ltd. are currently addressing much of what is required to move forward, but additional steps that will need to be considered as proposed projects move f orward include: • Develop agreements with major landowners. As owners of the resource required to fuel a biomass energy project, any proposed project needs to have the commitment and participation of the landowners involved. In many cases, this means the required participation of the local ANCSA village corporation (Gana -A’Yoo) as the owner of the bulk of the lands in the immediate vicinity of a community. • With the involved landowners, develop forest management plans. The forest stewardship program, administered by the State of Alaska with federal funds, is one option for a landowner to receive planning assistance. A project involving multiple landowners would require coordinated planning among the landowners to best serve the project and the affected comm unity. Included in the issues to be addressed by these plans would be: Managing the biomass resources in a sustainable manner through reforestation and other forestry Best management Practices (BMPs), and ensuring compliance with the Alaska Forest Resource Practices Act (FRPA); Preparation of a transportation and access plan; Detailed harvest plans; Ensuring that the harvest of biomass for energy does not interfere with normal subsistence wood gathering and other forest products utilization by community residents; Work to avoid the natural tendency to harvest the most available resource first, with the resultant effect of making fuel costs prohibitively more expensive in the future; Coordinate biomass harvesting with other land management activities such as hazardous fuel mitigation, wildlife habitat enhancement, etc. • Work to develop local capacity for technical land management tasks, biomass harvesting and transportation, and other contractable services and small businesses required to make a biomass energy project functional. • Attempt to develop better biomass supply and growth data. This can include the development of more precise and accurate land cover mapping using higher-resolution imagery or aerial photography, and the installation of ground plots to determine more accurate estimates of biomass stocking. This work can be quite expensive, but can be scaled to fit the demands of a proposed project. For example, a combined heat and power project (CHP) projected to consume relatively large amounts of woody biomass would require tighter biomass stocking and sustainability estimates and more detailed planning than would a relatively small cordwood thermal heating projec t. The Alaska Energy Authority has recently worked to
Assessment of Woody Biomass Energy Resources, Galena, Alaska 24 develop standards for required information for projects of varying size, complexity, resource demands, and stage of development. • As projects come on line, develop monitoring programs to collect information on harvest and transportation costs to better inform decisions made for current and future projects.