Subsurface fires in coal beds consume coal resources and contribute to the global emissions of CO2 and air pollutants. Many of these fires are found in China, India, Indonesia, and the United States. Combustion product gases at these coal fires exit through surface fissures that form over fires. These fissures are created when subsurface subsidence causes preexisting fractures in the area to widen. Fissures act as both inlets for air and exhaust for combustion gases. While remote sensing approaches have been used to quantify the rate of coal consumption and CO2 emissions at large scale fires that extend over large distances, methods for estimating the coal consumption and CO2 emissions values based on surface observations are less well established. In this paper, a coal fire near Durango, CO, is described. A combination of fissure mapping, thermocouple temperatures, and a cesium-vapor magnetometer survey was used to delineate the aerial extent of the current combustion zone and previously burned zones. Three methods were then used to estimate combustion rates at an active region at the site. In the first method, time-lapse, high-resolution topographic surveys were used to relate surface volumetric losses over the active region to coal consumption and rates of CO2 emission. In the second method, measured temperatures, gas compositions, and dimensions of an exhaust fissure were used in a simple natural convection chimney model to estimate rate values. The third method estimated coal consumption and CO2 emission rates by measuring the velocity of exhaust gases, gas compositions and exhaust fissure dimensions. For the second and third methods, C-13 isotope signatures were used to determine the fractions of CO2 that were emitted from coal and CH4 combustion or from CO2 in the native gas in the coal seam. A flux accumulation chamber was also used to quantify CO2 leakage rates from non-fissured regions over an active fire region. The three methods produced roughly consistent estimates of coal combustion and CO2 emission rates. (C) 2010 Elsevier B.V. All rights reserved.