Fuel appraisal photoseries plots were installed by Yellowstone's fire effects monitoring crew to represent the range of target fuels within the park. Our methods generally follow Ottmar et al. (2000) although we use metric system units and did not employ binocular photography.
Plot locations. YELL-2 LP0 Series (olive dots), YELL-3 Andesitic Forest Series (light blue dots) and Miscellaneous stands (also light blue dots). Ottmar et al. (2000) photoseries plot locations are shown in rust colored dots.
We targeted three specific conditions in priority order: 1.) opportunities for plot installation and observation ahead of wildfires involving LP0 in order to obtain pre- and post-fire fuel information in the context of the fire environment, 2.) sites adjacent to and best representing target areas that either recently burned (e.g., the Little Fire 2001) or had the opportunity to burn but did not (e.g., the Stone Fire 2001), and 3.) other sites that represent the environmental range of the target type. Of these the last comprises the majority. These sites were chosen to represent broad ranges in elevation, slope-aspect, habitat type, and other environmental factors.
Each photoseries sample plot represents the view of a camera with a 50 mm lens and is therefore shaped like a pie wedge that spans 32°. The plot is established by 5 lines 8° apart and marked every 10 m for a total of 25 points. The points are identified by the meter mark from the origin followed by the transect line (e.g., 10 meters on line 3 is 10.3). The origin and the end of each of the five lines were permanently marked with iron rebar. Late in the season after the vegetation reached its cured and most burnable state, the plot was photographed. Where available, an orange and white survey stake divided into 30 cm (1 ft) divisions was placed 10 m from the camera.
Plot coordinates are given in UTM NAD83/WGS84.
';
Plot layout.
Seedlings and overstory trees were recorded within 12 subplots of 2.0 m radius at 15.3, 20.2, 20.4, 25.3, 30.2, 30.4, 35.3, 40.2, 40.4, 45.1, 45.3, 45.5. Seedlings are defined as trees <1.4 m (breast height) tall for YELL-2. For YELL-3 the split was made at >2.5 cm DBH because that is the bottom limit for calculating canopy fuel features in Crown Mass module of the Fuels Management Analyst Suite (FPS 2005). Seedlings were tallied by live/dead and species. Five random seedling heights were measured as an indicator of ladder fuel height. Overstory trees were sampled for live/dead, species, diameter at breast height (DBH), and height. For YELL-3, additional tree characteristics were sampled to estimate canopy features in a 7 m radius (154 m2) subplot at 30.3. If additional plots were needed to get an adequate sample size, plots were sampled at 15.3, then 45.3. Overstory trees were sampled for live/dead, species, crown class, diameter at breast height (DBH), height, and crown base height. Crown class was subjectively judged based on crowding and amount of sunlight received.
Height and crown base height (CrBH) were measured directly if the tree was ~<4 m. Taller tree heights and crown base heights were calculated from distances and angles. The observer stands a measured distance from the tree and using a clinometer records the bottom angle, crown base angle, and top angle in percent. Tree height was calculated as (Top angle+Bottom angle) × Distance × 0.01. Crown base height was similarly calculated substituting crown base angle for top angle. Distance from the tree was measured with a measuring tape or an infrared-sonic distance measuring device, and slope-corrected by multiplying by Cos(Atan(Bottom angle × 0.01)). Measurements were converted to canopy fuel loading, volumetric canopy bulk density, segmental canopy bulk density, canopy base height, and canopy ceiling height using the FMA suite (FPS 2005).
Crown characteristics.
Crown classes.
Canopy measurement.
Herbaceous, shrub, and 1-hour woody fuels were collected in twelve 50 X 50 cm clip plots (0.25 m2). All above-ground biomass was cut and collected in paper bags by fuel type: 1-hour woody fuels, grasses and forbs, shrubs, and litter. Duff was not collected. Models regard 1-hour fuels as both woody twigs and dead, cured vegetation <6 mm (1/4"). The clip plots were sampled after mid-August to allow full growth and curing. The paper bags were placed in drying ovens at 103°C for several days and weighed and tared. Fuel load was calculated as the total mass divided by the area of the twelve sub-plots sampled.
Fuel transects were run at 23 points to measure 10-, 100-, and 1000-hour fuels. At each point an azimuth was randomly selected. Counts of fuel intersections with the transect were recorded for 10-hour fuels from 0-3 m, and for 100-hour fuels from 0-10 m. All intersecting 1000-hour logs were measured for diameter in rotten and solid classes. Litter and duff depths were measured at 3, 6, and 10 m. Calculations were made according to Brown et al. (1982).
Average fuel bed depth was recorded at all 25 points in the plot.
The percent cover of the two most common shrubs, forbs and graminoids was recorded within 6 of the subplots (20.2, 20.4, 30.2, 30.4, 40.2, and 40.4).
Often it was logistically not possible to install a time-intensive full photoseries plot ahead of an active wildfire. At Union and Phlox 2 a randomly selected 30-m fuel transect was substituted within the LP0 cover type area. Each end of the transect was permanently marked with iron rebar. Ten and 100-hour fuel intersections were sampled from 0 to 5 m and 0 to 10 m respectively. Thousand hour fuels were measured for diameter along the full transect in rotten and sound classes. One-hour woody fuels, litter, live herbaceous, and live shrub fuels were collected in two 50 × 50 cm sub-plots at 10 and 20 m. Litter and duff depth were measured at 5, 10, 20, and 25 m.
At Cub Creek the East Fire had already burned through the 19 ha (48 acre) 1955 Cub Creek burn. It was not possible to fit an entire plot in the burn area. Six clip plots, fuel transects, and tree density plots, and two canopy fuel plots were selected in areas that were unburned by the fire. In this respect, caution must be used because they may or may not represent pre-fire conditions.
Baker's Hole, Chickadee, and Chalcedony are included although different overall protocols were used (e.g., NPS fire effects monitoring plots). Canopy measurements are similar if not identical.
None of the alternative plots were used in custom fuel model development.
Although our subset of photoseries plots that are associated with some form of fire is small, we made some attempt to place fire behavior in our fuel models in the context of the fire environment in order to verify model outputs. Weather and fire behavior observations came from several sources. Fire Monitors recorded time, location, flame lengths, rates of spread, carrier fuels, temperature, relative humidity, cloud cover, and eye-level wind speed and direction, as possible and available. Fire monitors also collected samples of fuels for fuel moisture analysis. Most samples were collected in standard aluminum tins. Cookies from logs were collected in sealed plastic bags. Sedges and other herbaceous fuels represent "grab" samples of above-ground live and dead biomass without any effort to separate the two. All samples were weighed, oven-dried at 103°C for 48 hours, re-weighed and tared. Moistures are (wet wt - dry wt)/(dry wt) × 100%. Weather and modelled fuel moisture information were also obtained from remote, automated weather stations (RAWS), manual fire weather or climatological stations, or on-site weather data loggers. Some fire behavior observations were obtained from aerial reconnaissance flights.
Fuel models were initialized off TU1 for surface area to volume ratio (SAV), heat content, and other basic fuel model factors.