SCIENCEXPRESS

Brevia

Post-Wildfire Logging Hinders Regeneration and Increases Fire Risk

D. C. Donato,1* J. B. Fontaine,2 J. L. Campbell,1 W. D. Robinson,2 J. B. Kauffman,3 B. E. Law1

1Department of Forest Science, Oregon State University, Corvallis, OR 97331, USA. 2Department of Fisheries and Wildlife,

Oregon State University, Corvallis, OR 97331, USA. 3Institute of Pacific Islands Forestry, USDA Forest Service, Pacific

Southwest Research Station, 60 Nowelo Street, Hilo, HI 96720, USA.

*To whom correspondence should be addressed. E-mail: dan.donato@oregonstate.edu

Recent increases in wildfire activity in the United States have

intensified controversies surrounding the management of

public forests after large fires (1). The view that postfire

(salvage) logging diminishes fire risk via fuel reduction, and

that forests will not adequately regenerate without

intervention that includes logging and planting, is widely held

and commonly cited (2, 3). An alternative view maintains that

postfire logging is detrimental to long-term forest

development, wildlife habitat and other ecosystem functions

(1). Scientific data directly informing this debate are lacking.

Here we present data from a study of early conifer

regeneration and fuel loads following the 2002 Biscuit Fire,

Oregon, USA, with and without postfire logging. Because of

the fire's size (~200,000 hectares), historic reforestation

difficulties in the region (4), and an ambitious postfire

logging proposal, the Biscuit Fire has become a national icon

of postfire management issues. We used a spatially nested

design of logged and unlogged plots replicated across the fire

area and sampled before (2004) and after (2005) logging (5).

Natural conifer regeneration on sites that experienced

high-severity fire was variable but generally abundant, with a

median stocking density of 767 seedlings per hectare,

primarily of Douglas-fir (Pseudotsuga menziesii) (Fig. 1A).

Such density exceeds regional standards for fully stocked

sites, suggesting that active reforestation efforts may be

unnecessary. Postfire logging subsequently reduced

regeneration by 71%, to 224 seedlings per hectare (Fig. 1A),

due to soil disturbance and physical burial by woody material

during logging operations. Thus, if postfire logging is

conducted in part to facilitate reforestation, replanting could

result in no net gain in early conifer establishment.

Postfire logging significantly increased both fine and

coarse downed woody fuel loads (Fig. 1B). This pulse was

comprised of unmerchantable material (e.g., branches), and

far exceeded expectations for postfire logging-generated fuel

loads (5, 6). In terms of short-term fire risk, a reburn in

logged stands would likely exhibit elevated rates of fire

spread, fireline intensity and soil heating impacts (7).

Postfire logging alone was notably incongruent with fuel

reduction goals. Fuel reduction treatments (prescribed

burning or mechanical removal) are frequently intended

following postfire logging, including in the Biscuit plan, but

resources are often not allocated to complete them (8). Our

study underscores that, after logging, mitigation of short-term

fire risk is not possible without subsequent fuel reduction

treatments. However, implementing these treatments is also

problematic. Mechanical removal is generally precluded by

its expense, leaving prescribed burning as the most feasible

method. This will result in additional seedling mortality and

potentially severe soil impacts due to long duration

combustion of logging-generated fuel loads. Therefore, the

lowest fire risk strategy may be to leave dead trees standing

as long as possible (where they are less available to surface

flames), allowing for aerial decay and slow, episodic input to

surface fuel loads over decades.

Our data show that postfire logging, by removing naturally

seeded conifers and increasing surface fuel loads, can be

counterproductive to goals of forest regeneration and fuel

reduction. In addition, forest regeneration is not necessarily in

crisis across all burned forest landscapes. The results

presented here suggest that postfire logging may conflict with

ecosystem recovery goals.

References and Notes

1. D. B. Lindenmayer et al., Science 303, 1303 (2004).

2. U.S. House Committee on Resources, Forest Recovery Bill

Hearing Press Release (November 9, 2005).

3. J. Sessions, P. Bettinger, R. Buckman, M. Newton, J.

Hamann, J. For. 102, 38 (2004).

4. S. D. Tesch, S. D. Hobbs, W. J. Appl. For. 4, 89 (1989).

5. Materials and methods are available as supporting material

on Science Online.

6. Timber decay associated with delays in postfire logging

was anticipated to exacerbate the observed pulse of surface

fuel. However, our data indicate that delay was responsible

for ~10% of woody fuel present after logging.

/ www.sciencexpress.org / 5 January 2006 / Page 1 / 10.1126/science.1122855

7. J. K. Agee, Fire Ecology of Pacific Northwest Forests

(Island Press, Washington, DC, 1993).

8. R. W. Gorte, “Forest Fires and Forest Health”

Congressional Research Service (Publication 95-511,

1996).

9. This work was supported by the Joint Fire Science

Program and DOE grant DE-FG02-04ER63917. We thank

field technicians and the Siskiyou National Forest.

Supporting Online Material

www.sciencemag.org/cgi/content/full/1122855/DC1

Materials and Methods

SOM text

References and Notes

21 November 2005; accepted 21 December 2005

Published online 5 January 2006; 10.1126/science.1122855

Include this information when citing this paper.

Fig. 1. Natural conifer regeneration (A) and surface woody

fuel loads (B) before and after postfire logging of the Biscuit

Fire, Oregon, USA. (A) Regeneration was abundant

following fire. Postfire logging significantly reduced seedling

densities (P<0.01, Wilcoxon signed rank test) from 767

seedlings ha–1 to 224 seedlings ha–1. (B) Postfire logging

significantly increased downed fine (P<0.01) and coarse

(P<0.05) woody fuel loads (Mg ha-1) relative to burn-only by

Wilcoxon signed rank test. To provide context, fuel data from

unburned stands are shown as reference for pre-fire

conditions (fuel loads in burn-logged stands were at or well

above pre-fire levels). Graphs of seedling densities and fine

(<7.62) and coarse (>7.62) surface woody fuels are medians +

SE; n = 8 stands for no burn, n = 9 for burn-only and burnlogged

(5).

/ www.sciencexpress.org / 5 January 2006 / Page 2 / 10.1126/science.1122855