COAL RESOURCE EVALUATION METHODOLOGY
Coal resource evaluation employs digital computer technology (for example, software), similar to manual methods, for isopaching and planimetering coal beds and zones, and overburden to categorize and calculate quantities of coal. These manual and computer methods determine the thickness and volume of coal, overburden thickness of rocks above the coal, and reliability categories; however, the computer method is deemed more efficient, more readily repeatable, more detailed, and faster than the manual method. Detailed methods for data preparation and computer calculation of coal resources for the assessed Fort Union and equivalent coal in the Powder River Basin is discussed by Ellis and others (1999; in press). See specific chapters in this CD-ROM for a general discussion of the methodology used to assess the coal resources of other basins in this study (Williston Basin WN; Greater Green River Basin GN; Hanna and Carbon Basins HN). The coal resource evaluations in these basins consist of estimating the volumetrics of 18 coal beds and zones (Wyodak-Anderson, Rosebud, Knobloch, Harmon, Hansen, Hagel, Beulah-Zap, Ferris 23, 25, 31, 50, and 65, Hanna 77-79 and 81, Johnson-107, and Deadman coal beds or zones). These selected Tertiary coal beds and zones, except the Knobloch and Johnson-107, were mined and produced more than 30 percent of the total U.S. coal production in 1997. Thus, these selected coal deposits are important to development in the next few decades.

VOLUMETRICS
In order to calculate the volume for each of the 18 coal beds and zones, it is necessary to determine the thickness and areal distribution of the coal. The volume of coal is calculated using these parameters and is then converted from acre-ft to short tons by factoring in the density (number of short tons per acre-ft for the apparent coal rank) of the coal. Thus the formula used for resource calculation is: short tons = acres x coal thickness x coal rank or density conversion factor (for example, 1,750 for lignite, 1,770 for subbituminous, 1,800 for bituminous, and 2,000 for semianthracite and anthracite) (Wood and others, 1983).

As described under the section on construction of the database, coal beds existing in vertical proximity were assigned to and correlated over large areas as a coal zone throughout their depositional extent. Throughout this extent an individual coal zone may contain one or more coal beds recorded in each drill hole. Thus, given the number of coal beds and their continuity per coal zone, a method was devised to measure and combine the net thickness of the coal beds of one zone at each data-point location. This method was applied after the coal beds in the coal zone were correlated, taking into account all the geological factors discussed under the section on database construction. Correlations defined the coal beds included in the coal zone and the elevations of the top and base of the coal zone. The amount of overburden (thickness of rocks above the coal zone) was calculated by subtracting the grid of the surface elevations from digital elevation models (DEMs) and the grid of the elevations of the top of the coal zone. As a part of this method, the rock in the coal zone from each drill hole was also divided into partings and splits as defined by Wood and others (1983). Rock partings exist when the thickness of the rock between coal beds in the zone is less than the thickness of the underlying and overlying coal beds. Rock splits exist when the thickness of the rock between the coal beds in the zone is more than the thickness of either the underlying or overlying coal beds. These criteria are tailored to the estimation of coal resources where a coal bed bifurcates into two or more beds, as in the coal zone. This classification is required in order to accurately calculate the net coal thickness of coal within a coal zone.

COAL RESOURCE CATEGORIES
A standardized and universally accepted method for reporting reliability categories of coal resources has been established for the U.S. Geological Survey (USGS) by Wood and others (1983). This coal resource classification system is an expansion of the system adopted by the U.S. Geological Survey (1976) and Averitt (1975) in reporting the 1974 U.S. coal resources. In 1976, the USGS and the U.S. Bureau of Mines modified the system used by Averitt (1975), which served as the standard reference for coal resource/reserve assessment by many Federal and State agencies. Our coal resource reporting categories generally follow the 1983 USGS methodology (Wood and others, 1983), but with the addition of coal quality.

The basic concept in the USGS coal resource classification system is based on the geologic assurance that it is directly related to the distance from drill holes (control points) where coal thickness and overburden are measured. The specified distances from drill holes (control points) are reliability circles: 0–0.25 mi radius for measured coal, 0.25–0.75 mi radius for indicated coal, 0.75–3.0 mi radius for inferred coal, and beyond 3 mi radius for hypothetical coal. Thus, the classification system is designed to quantify the amounts of coal (1) that are known or identified resources (measured, indicated, and inferred reliability categories), and (2) that remain to be identified (hypothetical reliability category). These distance reliability categories, as well as the net coal thickness category (2.5–5 ft, 5–10 ft, 10–20 ft, 20–40 ft, and >40 ft) and the overburden thickness category (0–100 ft, 100–200 ft, 200–500 ft, and 500–1,000 ft), are required for reporting resources for the Fort Union lignite and subbituminous coal (>500 ft is used because no or little overburden exceeds 1,000 ft). In addition, coal resources were reported by county, state, 7.5-minute quadrangle, Federal versus non-Federal surface and coal ownership, coal quality (using ash and sulfur contents, and pounds of SO 2 /million Btu), and apparent coal rank (as indicated by moist, mineral-matter free Btu; American Society for Testing and Materials, 1997) categories. The coal resources reported by these categories do not include mined out and leased areas, and areas of burned coal and associated rocks or clinker. Also, coal resources of Indian tribal lands were not calculated.

COMPUTERIZED METHODS
Several computer software programs were utilized to create digital information for the calculation of the coal resources. ARC/INFO and ArcView software were used to create layers of spatial digital information or coverages (for example, state boundaries, counties, geological boundaries, mine and lease boundaries, quadrangle maps, clinker, point locations, etc.). These coverages are in Lambert Conformal Conic projection, Clarke Spheroid 1866, with parameters of first standard parallel of 33°, second standard parallel of 45°, and central meridian of 106°.

The computer method for calculating coal resources involved compensating for the irregular distribution (x and y values) of drill-hole data (dense versus sparse distribution) within the study basins. In order to compensate for this distribution, a rectangular grid was superimposed over the data area and a Z value (for example, top and bottom of coal beds, net thickness, etc.) was interpolated by computer. This gridding procedure was performed by using the earthVision software. Different grid sizes and algorithms were used to test either the “coarseness” or “fineness” of the grids in order to generate the appropriate isopach maps. Thus, resource calculations were accomplished by using grid nodes and sub-nodes values in the earthVision program (Ellis and others, 1999, in press; Roberts, 1998). Coal thickness isopach and overburden maps generated from this procedure were utilized to produce unioned coverages using ARC/INFO. These coverages are composed of combined layers or polygons of spatial information containing attributes (for example, counties, State, Federal versus nonFederal ownerships). After unioning all the coverages, the resulting coverage was clipped to the coal zone areal extent. The ARC/INFO union coverage polygon files were imported into earthVision for volumetric calculations (assigned in short tons). This method, which used the net coal thickness grid node and sub-grid node values of the coal thickness, was determined to be the most accurate method for estimation of coal resources by computer (Ellis and others, 1999; in press).

The coal resources are reported in tables, which contain rows such as overburden, coal thickness, and distance reliability categories. The tables also include columns such as coal quality (percent sulfur and ash), lbs SO 2 /million Btu, and apparent coal rank. Coal resources within each of these categories are reported as short tons (in millions to billions) in two significant figures. Schuenemeyer and Power (in press) developed a procedure for the estimation of uncertainty or measurement of error in the volume of coal resources as a part of the USGS National Coal Resource Assessment.