Figure 1. Forest Service Acres Burned in Relation to Millions of Board Feet Cut

Some critics have argued that, because timber harvesting removes biomass from the forest, it also reduces the extent and severity of forest fires. A correlation analysis, relating acres burned to .timber harvest volume, was performed to test part of this hypothesis — that the extent of forest fires is related to the quantity of timber harvested. The coefficient of determination (r²) is the most frequently used statistic to assess the correlation, between two variables; an r² of 1.00 indicates an absolutely perfect correlation, while an r² of 0.00 indicates a perfectly random relationship. The coefficient of correlation (r) is also used sometimes, since it indicates the direction of the correlation (positively or negatively related) as well. The coefficients of determination and of correlation were calculated for 1980-1999, 1960-1999, and 1987-1999.1 The results are shown in table 2. The coefficients of determination (r²) are quite low, with the highest being an r² of 0.1362 for 1987-1999. The analysis finds that, for this period, less than 14% of the variation in acres burned is related to the variation in harvests; for other periods, the relationship is even weaker. The coefficients of correlation are also low. More surprising is that the they are positive for 1980-1999 and 1987- 1999, indicating fewer acres burned in association with lower timber harvests, contrary to the hypothesis.

In assessing this relationship — acres burned with timber harvests — qualitatively, the conclusion of the correlation analysis is not surprising. Timber harvesting removes the relatively large diameter wood that can be converted into wood products, but leaves behind the small material, especially twigs and needles. The concentration of these "fine fuels" on the forest floor increases the rate of spread of wildfires.2 Thus, one might expect acres burned to be positively correlated with timber harvest volume.

It should be noted that this discussion focuses on the extent of fire, but not on the severity. Areas with heavier fuel loadings almost certainly bum more intensely than areas with lesser fuel loadings. Timber harvesting does remove fuel, but it is unclear whether this fuel removal is significant, because the proportion of fuel removed is unknown and because the relative importance of large-diameter fuels in fire intensity is unknown. Furthermore, while it seems likely that more intense fires cause more resource damage, damage appraisal methods are relatively unsophistocated. Thus, timber harvesting might reduce the severity of forest fires, but given currently available information, a quantitative analysis of this benefit is infeasible.

Table 1. National Forest Timber Harvests and Acres Burned on Forest Service-Protected Lands
(in millions of board feet and total acres burned)

Table 2. Coefficients of Determination and of Correlation
for Harvest Volume-Acres Burned Comparisons

Footnotes

1 1987 was chosen as likely to maximize the correlation, since 1987 was the peak harvest year.

2 See: Robert E. Martin and Arthur P. Brackebusch, "Fire Hazard and Conflagaration Prevention" Environmental Effects of Forest Residues Management in the Pacific Northwest: A State-of-Knowledge Compendium (Owen P. Cramer, ed.), Gen. Tech. Rept. PNW-24 (Portland, OR: USDA Forest Service, 1974).