Pacific Northwest Research Station

The Oregon white oak is a sun-loving tree. Without fire, conifers increasingly dominate the stand, and oak is eventually shaded out. Prior to Euro-America settlement, American Indians who gathered acorns for food frequently burned Oregon white oak woodlands to favor the tree. This killed the young conifers and led to grassy understories. These low-intensity grass fires often caused only minor canopy damage to oaks. Prescribed fire is once again being used in many Pacific Northwest woodlands to control competition from woody vegetation. Understory conditions have changed, however. Where shrubs have replaced grass, fire burns hotter and causes more tree damage.

Because white oak flowers and leaves form inside the buds one year prior to actual flowering, researchers found that buds are particularly vulnerable to summer fire damage. Researchers were able to relate fire intensity to acorn production. Even when tree buds are not killed, flowering and acorn production the following year may be reduced. This information will help land mangers effectively use prescribed burns to manage oak woodlands and wildlife that depend on acorns for food.

Contact: David Peter, dpeter@fs.fed.us, Threat Characterization and Management Program

Restoring burned forests on federal land in southwest Oregon may have multiple objectives, including protecting site productivity, maintaining diversity of species in the near and long term, and developing structurally complex mature stands. Station scientists have monitored various reforestation methods that were applied to burned plantations following the 2002 Timbered Rock Fire. They observed that removing competing vegetation had a greater effect on the composition of plant communities than did planting conifer seedlings.

Woody shrubs were cut back repeatedly during the first 3 years following conifer planning. This did not remove them from the sites, but the temporary reduction in their abundance resulted in the early development of different postfire vegetation communities, when compared to sites where woody shrubs were not cut back. Initial increases in the occurrence of nonnative invasive species were very small, suggesting that the reforestation treatments were not likely to directly lead to invasive species problems.

This information may be useful to land managers in southwest Oregon who are preparing postfire strategies for reforestation, vegetation, and fuel management. It can also be applied more broadly by modelers interested in predicting postfire vegetation and fuel dynamics at stand and landscape scales.

Contact: Paul D. Anderson, pdanderson@fs.fed.us, Land and Watershed Management Program

Rare, high-quality historical data are being used for model formulation, calibration, and testing to evaluate effects of a severe 1970 wildfire on flow regime and maintenance of water quantity and quality from streams draining forested headwaters. Following more than 25 years of inactivity at the Entiat Experimental Forest in eastern Washington, study watersheds were reinstrumented in 2004 with stream gauging, water quality, and meteorological stations to assess rate of recovery toward prefire conditions.

Preliminary results indicate that the wildfire resulted in initial peak flow increases in runoff of 100 percent or more from the burned area, and that these increases resulted from decreased evapotranspiration (the process through which plants draw up water), increased snow accumulation, and more rapid snowmelt. The effects of wildfire on watershed processes may be long-lived and depend, in part, on rehabilitation actions.

This ongoing research is providing predictive capability regarding effects of severe wildfire and hydrologic recovery processes after a major disturbance. These findings are informing management decisions regarding the need and strategies for postfire rehabilitation. They lay the foundation for future investigation of hydrologic effects of fuel-reduction treatments.

Contact: Richard D. Woodsmith, rwoodsmith@fs.fed.us, Threat Characterization and Management Program

For the first time, scientists were able to directly measure the effects of a hot wildfire on forest soils. The 2002 Biscuit Fire burned about half of twenty-seven 15-acre study plots east of Gold Beach, Oregon, established before the fire.

The fire burned at temperatures over 1,300 ºF, as evidenced by the melting of aluminum tags across the research plots—this is more than twice as hot as typical prescribed fires. Loss of topsoil and combustion of organic material were higher than most previous estimates. More than 10 tons per acre of carbon and 450 to 620 pounds per acre of nitrogen were lost, and nearly 60 percent of this came from the mineral topsoil below the organic layer.

The loss of topsoil and soil carbon can negatively affect a range of processes, including nutrient retention and water infiltration. To replace the documented amount of lost nitrogen would require nitrogen-fixing plants to dominate the forest for decades.

This study illustrates the dramatic effects of intense wildfire on soil nutrients and resulting site productivity. Understanding how fire can alter the available nutrients and thus change the productivity of a site is essential when undertaking reforestation efforts and projecting future stand development. Also, a forest that grows slower after an intense wildfire than it did before will have reduced rates of carbon sequestration.

Contact: Bernard Bormann, bbormann@fs.fed.us, Land and Watershed Management Program

How much stored carbon did the 2002 Biscuit Fire release? To answer this question, scientists estimated the amount of carbon in 24 separate fuel “pools” by using prefire data from areas later burned by the Biscuit Fire in southern Oregon. They then used postfire estimates to check combustion factors for the various pools of carbon and estimated that 2.9 to 3.5 terragrams of carbon were emitted during the fire. This is about 16 times the annual net ecosystem production of this landscape before the wildfire.

Few studies have attempted to empirically quantify wildfire-induced carbon exchange between terrestrial vegetation and the atmosphere. This study may lead to further use of pre- and postfire data to determine the carbon outputs to the atmosphere from large wildfires. As more pre- and postfire pairs of plots are measured, a better estimate of the carbon released may be modeled.

Contact David Azuma, dazuma@fs.fed.us, Resource Monitoring and Assessment Program

Station scientists, working with university collaborators, demonstrated that the area of wildfire burned in the American West was significantly controlled by climate (combinations of precipitation, temperature, and drought) during most of the 20th century (1916-2003). Climate-fire relationships differed by vegetation type, represented by 16 ecoprovinces (broad geographic regions with similar ecological properties) across the West.

Fire in most mountainous ecoprovinces, except in the arid Southwest, is driven by low precipitation, dry conditions, and high temperatures in the summer fire season. Fire in arid ecoprovinces, or those dominated by grasses or shrubs, is strongly associated with precipitation or drought in the previous year, which affects the amount of fuel for the current year. Despite the influences of fire exclusion, other landuse changes, or both, the amount of burned area is still substantially controlled by climate. In the future, this will likely depend on ecosystem-specific seasonal variation in climate and its effects on fuel conditions in northern mountainous ecosystems and fuel availability in arid ecosystems.

Despite the overarching influence of climate, it will be important to link this understanding of climate drivers to land use changes and broad-scale management strategies.

Contact: Don McKenzie, donaldmckenzie@fs.fed.us, Threat Characterization and Management Program

Slope stabilization treatments are applied after severe wildfires to increase soil cover, reduce soil erosion, protect water quality, and reduce risks to human life and property. The effectiveness of many common slope stabilization treatments, such as reseeding, remains in question, however.

Station scientists assessed different reseeding treatments after several recent fires on the Okanogan-Wenatchee National Forest. They found that seeding winter wheat over portions of the burned area produced little or no additional soil cover following high-severity wildfires in the dry coniferous forest. They also tried seeding yarrow, a perennial native forb, and found this often increased cover of yarrow significantly, particularly when applied in combination with nitrogen fertilizer. In general they found the increased cover of seeded species was often offset by reduced cover of native, nonseeded vegetation. These findings suggest that seeded species compete with recovering native vegetation for limited soil resources after wildfire in dry coniferous forests, and that efforts to monitor the effectiveness of slope stabilization treatments by measuring seeded species cover alone are likely to overestimate treatment effects on soil cover, runoff, and erosion.

Contact: Dave W. Peterson, davepeterson@fs.fed.us, Threat Characterization and Management Program

To effectively use fire as land management tool, it is helpful to know the typical severity of historical fires in the area. In this study, scientists found a broader suite of forest structural conditions and patch sizes supported native fire regimes of mixed-conifer forests than was formerly recognized.

After analyzing aerial photos from the early 1900s of unharvested forest on about 750,000 acres, scientists found that mixed-severity fires were most prevalent, regardless of forest type. The structure of mixed-conifer patches in the eastern Washington Cascades, in particular, was formed by a mix of fire severities. In moist mixed conifers, stand-replacement fire effects were more widespread than surface fire effects, whereas in dry mixed conifers, surface fire effects were more widespread by nearly 2 to 1. The relatively low abundance of old, parklike or similar forest patches in comparison with the abundance of young and intermediate-aged patches, and evidence of partial-stand and stand-replacing fires suggested that variable fire severity and nonequilibrium patch dynamics were primarily at work.

This suggests that some of the contemporary effects from wildfires may better meet management objectives than once thought, particularly when the objective is to create patch conditions achieved through low- and mixed-severity fires.

Contact: Paul Hessburg, phessburg@fs.fed.us, Ecological Process and Function Program

Logging activities can compact the soil, reducing its pore size and decreasing oxygen availability and movement of water and nutrients to tree roots. To alleviate compaction, land managers may fracture the subsoil, a practice known as subsoiling.

In this study, scientists examined the effects of compaction and subsoiling after postfire logging on the soil microbes in a mixed-conifer forest in central Oregon. These bacteria and fungi perform the complex biological and chemical processes that render essential nutrients available for the healthy growth of forests.

Scientists found differences among stands with respect to relative bacterial species abundance, but no difference among treatments. The cumulative number of bacterial species and fungal species did not differ significantly among the compacted soil, subsoiled, or fire-only treatments. Over multiple sampling seasons, however, slightly more bacterial and fungal species were found in the compacted soil treatment. These findings suggest that microbial communities in the sandy loam soils in this study were generally tolerant of postfire harvest disturbance.

Contact: Jane Smith, jsmith01@fs.fed.us, Land and Watershed Management Program

A changing climate means trees and other plant species may be suited to different areas than they are currently. This has profound implications for biodiversity and economic concerns. To help anticipate this change, scientists developed the MC1 Dynamic General Vegetation Model which projects what might grow where under different temperature and precipitation regimes. The model also projects fire frequency under different climate scenarios. These climate scenarios are linked to high, moderate, and low trajectories for future greenhouse emissions. Simulations showed that areas of tropical woodland, temperate mixed forest, and tundra and alpine vegetation are potentially the most vulnerable to impacts of future climate change, and that approximately one-third of the global land surface could experience increased fire frequency.

Numerous federal, state, and private land management entities have asked Station scientists about these findings. The Nature Conservancy is considering these research results as it incorporates climate change into conservation action plans.

Contacts: James Lenihan, jlenihan@fs.fed.us; Ecological Process and Function Program

Following the massive Biscuit Fire in southern Oregon in 2002, Station scientists had the unique opportunity to study the effects a past fire and subsequent management can have on future fire severity. Parts of the area that burned in 2002 also were burned in 1987 by the Silver Fire and then alternately salvage-logged, planted, and managed to establish Douglas-fir plantations or left to naturally regenerate.

Using satellite and aerial imagery and management records, scientists found that areas burned severely in the Silver Fire tended to burn again at high severity in the 2002 Biscuit Fire, regardless of management treatment. Areas that were salvaged-logged and planted after the initial fire, however, burned more severely than comparable unmanaged areas, suggesting that slash left over from logging and fuel conditions in conifer plantations can increase fire severity for some time despite removal of large woody fuels.

This study is the first to look at how postfire management can affect vegetation burn severity in a subsequent fire. Continued research is needed to further understand the causes of these results and to better understand the tradeoffs associated with post-fire forest management.

Contact: Thomas A. Spies, tspies@fs.fed.us, Ecological Process and Function Program

In this study, scientists used rare historical data from the Entiat Experimental Forest to create models to evaluate the effects of a severe 1970 wildfire on streams draining forested headwaters. Preliminary results indicate that runoff from the burned area more than doubled. These increases in peak flows resulted from less vegetation present to draw ground water, greater snow accumulation, and more rapid snowmelt.

After more than 25 years of inactivity, the study watersheds were re-instrumented in 2004 with stream gauging, water quality, and meteorological stations to assess the rate of recovery toward prefire conditions. This data will also be used in future investigations of the effects of prescribed burning and other fuel control treatments on hydrologic processes.

The information gained from this research will enable scientists and managers to better predict the effects of severe wildfire on hydrology and to better understand the hydrologic recovery process after major disturbances. It will also inform management decisions regarding the need and strategies for postfire rehabilitation.

Contact: Richard D. Woodsmith, rwoodsmith@fs.fed.us, Threat Characterization and Management Program

Slope stabilization treatments often are applied after high-severity wildfires to reduce soil erosion, protect water quality, and reduce risks to human life and property resulting from landslides. The effectiveness of many common slope stabilization treatments remains in question, however, and high costs limit the extent to which some popular treatments, like aerial mulching, can be used. In a study of seeding and fertilization treatment effects following the 2004 Pot Peak Fire near Chelan, Washington, scientists found that fertilizing with nitrogen and sulfur increased plant cover significantly during the first two growing seasons following the severe wildfire. Fertilization alone was more effective than seeding alone; however, spreading fertilizer and yarrow seeds, a native forb, produced the greatest increase in plant cover.

If further studies can establish the consistency of these results, fertilization can provide a viable alternative to seeding or mulching as a slope stabilization treatment. Fertilizer is widely available and relatively inexpensive to purchase and apply, and carries low risk of introducing exotic plant species into severely disturbed landscapes. Ongoing studies will help us better understand possible secondary effects (good or bad) of fertilization on vegetation composition and succession.

Contact: Dave W. Peterson, davepeterson@fs.fed.us, Threat Characterization and Management Program