Remeasurement of Fuels and Stand Structure 13 Years After Logging of the Summit Fire, Malheur National Forest, Eastern Oregon

Project Description

The study was intended to examine the effects of logging after a severe wildfire burned through a set of 12 stands just north of the Middle Fork of the John Day River, Malheur County. The 16,000 hectare Summit Fire was caused by a lightning storm on August 13, 1996, on the North Fork John Day Ranger District (Umatilla National Forest), burned south onto the Long Creek Ranger District of the Malheur National Forest, and was declared officially controlled on September 16, 1996.

Within the Malheur Forest portion of the Summit Fire, an area of 8,103 hectare (72 percent) was judged to have burned at high severity (>80 percent large trees killed), including much of the area in the lower elevation ponderosa pine forests, which typically experience lower-severity fires. The short-term effects of postfire logging were evaluated by conducting a controlled, replicated experiment between 1997 and 1999. Logging occurred in the summers of 1998 and 1999, and consisted of the application of three treatments (unlogged control, commercial, and fuel reduction), applied to 12 experimental units (stands) in four replicate blocks. Data were taken from within about 258 measurement plots in the 12 units (stands) in 1997 (postfire, prelogging) and again in 1999 (postfire, postlogging). The longer-term effects of postfire logging on fuels and stand structure were evaluated by remeasuring fuel and stand structure variables in 2011, 15 years after the Summit Fire, and 13 years after postfire logging.

Purpose and Scope

We aimed to remeasure existing field studies to assess the effects of high-severity fire on vegetation succession, and to evaluate the effects of post-wildfire management.

Methods

All variables were measured pre- and postlogging from the same permanently established grid points. A total of 277 grid points were laid out in the 12 experimental units in August 1997, between 14 and 47 points per unit. Pre-treatment data were taken in the nine units of the Elk Creek, West Coyote, and Wray Creek blocks between August and October 1997. Post-treatment data were taken from July to September 1999, immediately following the termination of logging (post-treatment year 1), and from June to early August 2011 (post-treatment year 13). The same protocol was used on each of the three sampling occasions.

Key Findings

Short-Term Effects

1. Logging between June 1998 and April 1999 reduced mean basal areas to 46% in four stands experiencing a single entry to remove merchantable timber (commercial treatment), and down to 25% in four stands that experienced two entries, the first to remove merchantable timber, and the second to reduce fuels (fuel reduction treatment)
2. Stem density reduction followed prescription, and was consistent with the tendency for loggers to take the largest stems possible given commercial constraints and timber sale guidelines; smaller noncommercial stems were only taken in the fuel reduction treatment, in which the logger was paid through a separate service contract.
3. Logging reduced large snag densities far below those observed one year post-logging in unlogged stands, thereby reducing the habitat quality of logged stands for cavity-nesting bird species.  
4. Logging did not change ladder height or tree species composition
5. Logging increased total fuel mass, particularly in the slash fuel size class (1 to 100-hr fuels, < 7.62 cm diameter), with fuel reduction stands averaging 5.4 Mg/ha of slash fuel, commercial units 4.3 Mg/ha, and unlogged controls averaging 1.5 Mg/ha.
6. Logging activity did not cause short-term changes in the mass of duff or litter.
7. The Forest Vegetation Simulator Fuel and Fire Effects model (FVS-FFE) projections indicate that logging-induced differences in slash fuel would be sustained until about 15 years post-logging, but if a re-burn of moderate intensity were to occur during that time, all seedlings and saplings that began growth post-fire would be killed, even in un-logged stands, due to the influence of other components of the fuel bed (grasses, shrubs).
8. FVS-FFE model projections indicate that standing dead structure in all stands would collapse quickly, with 80% of snags falling within 15 years post-logging.
9. Stand collapse would be projected by FVS-FFE to result in accumulation of 1000-hr fuels (woody fuels > 7.62 cm diameter), especially in unlogged stands, such that unlogged stands would have 2 or 3-fold greater masses of heavy fuel at 25 and 50 years post-logging. 
10. Despite differences between unlogged and logged stands in heavy fuel accumulation for decades post-fire, FVS-FFE would not predict concomitant differences in young tree mortality should a re-burn occur, because coarse woody debris does not contribute to the kind of fire behavior that tends to kill trees

Intermediate-Term Effects

1. Mean basal area and density of trees that survived the 1996 Summit Fire remained unchanged from that measured in 1999, one year post-fire logging.
2. Mean basal area of dead trees in 2011 declined to about one-third that observed in 1999, and was proportional to logging treatment.
3. Ponderosa pine snags (dead trees > 30cm DBH) fell at a faster rate than Douglas-fir snags, declining by 2011 to 23% of the density measured one year after logging in 1999, compared to 34% for Douglas-fir
4. Snag density, especially Douglas-fir, remained higher in unlogged stands 13 years after post-fire logging
5. Mean forest floor mass declined for all treatments, but much more rapidly in logged stands. This is probably because of additional bark slough arising from higher mass of coarse woody debris in unlogged stands
6. Both sound and rotten 1000-hr woody fuel (> 7.62 cm diameter) increased markedly by 2011 in all stands regardless of treatment, as over 75% of trees killed by the 1996 Summit Fire fell down
7. Roughly half of coarse woody debris present in 2011 was decayed enough to be classified as 'rotten'
8. Unlogged stands contained about twice the mass of coarse woody debris greater than 15 cm diameter as compared with either commercial or fuel reduction stands
9. Logged stands that were entered twice (fuel reduction stands) had about 1/3 the density of sapling stems by 13 years post-fire logging, compared to unlogged stands
10. The 2008 Sunshine Fire, which burned through nearly one complete block within the Summit Study Area, erased most of the effects of the 1998 post-fire logging operation: snags were reduced to < 0.5 ha-1 for all treatments, forest floor and fuel mass were reduced to near 0, log mass declined to about 1/4 that observed in stands unburned by the Sunshine Fire, and shrubs and seedlings were nearly eliminated.