Upland Forest Change

Select Findings from Upland Forest Analysis

From the valley floor east to the crest of the Cascade Range, air temperatures decrease and precipitation increases as elevation rises (Fig. 1). These gradients drive a change from maritime conifer to cool needleleaf forest and ultimately to subalpine conifer forest. This vegetation mix is expected to change as rising temperatures create a less-favorable climate for existing vegetation and as forest fires increase in frequency and intensity. Here we highlight some of the key findings from WW2100 forest modeling. For more detailed analysis, refer to Turner et al. (2015, 2016).

Wildfire

• Under the various climate scenarios, frequency of wildfires relative to the historical period increases as temperature increases.
• Under the Low Climate Change (LowClim) scenario, the area burned per year is only slightly below the historical rate.
• Under the Reference Case (Reference) scenario, the simulated forest area burned per decade in the 2010-2100 period is 0.6% per year (vs. 0.2% per year in recent decades). Fire tends to be concentrated in particular years, with as much as 25% of the forested area burning in a high fire year late in the 21st century (Fig. 2).
• The High Climate Change (HighClim) scenario (warmest temperatures) induces the largest areas of fire per year, with average area burned per year increasing by a factor of nine relative to the historical period.
• In the Reference and HighClim scenarios, the proportion of the Willamette River Basin that is recently burned and relatively open increases. (Note: WW2100’s simulations did not include pest and pathogen disturbances, which are also likely to increase with climate warming.) More relatively open areas result in lower mean leaf area. These decreases in leaf area lead to reduced growing season evapotranspiration, despite higher evaporative demand due to higher temperatures.
• Several of the WW2100 alternative scenarios influence the forest uplands. In the case of the Upland Wildfire Suppression (FireSuppress) scenario, the incidence of fire is maintained at the contemporary rate (0.2% of area per year). This assumption results in an increase in the proportion of the forest area with a difference (disequilibrium) between climate and potential vegetation type. In the Extreme scenario, the average area burned rises to 0.8% per year. Results in terms of vegetation change are similar to the Reference scenario. In the Managed scenario, the assumed fire rate is low relative to the Reference scenario and rotation age is reduced. These assumptions mean that the landscape is able to sustain contemporary rates of harvest on public and private lands.

Harvest

• Under climate change, the increased prevalence and power of forest fires is expected to affect how and if trees, such as the commercially important Douglas-fir, are harvested. The WW2100 forest team found that as temperatures increase from 2010-2100 (note: all three representative scenarios show some degree of warming), forest fires increase and mature forests available for harvest correspondingly decrease. (Worth noting in these results: Today’s forest age class distribution in the Willamette River Basin differs substantially between public and privately owned forestland. A significantly larger proportion of the forestland in older age classes are on public land, which also tends to be at higher elevations.)
• Under the LowClim and Reference scenarios, the harvest rate on private land is stable at about 1.5% of the area per year, and the harvest rate on public lands is stable at a rate of about 0.5% per year (Fig. 3).
• Under the more severe fire regimes in the HighClim scenario, harvestable forest area decreases, such that the harvest rate begins declining towards the end of the century on both private and public forestland (Fig. 3).

Vegetation Shifts

Plant species in the Northern Hemisphere are moving to higher latitudes and elevations in response to climate change. These climate-induced shifts in vegetation are due primarily to rising temperatures, as plant species migrate to areas with temperatures ranges they are adapted to. This shift in vegetation, already observed in the Northwest, is expected to continue under climate change. WW2100’s forest team produced the following findings concerning vegetation shifts and climate change for the Willamette River Basin:

• The climate-induced shifts in potential vegetation cover type for the Willamette River Basin under the three climate scenarios are proportional to the magnitude of the climate change (Fig. 4). Under the LowClim scenario (least warming), there is little change in potential vegetation type, whereas with the HighClim scenario (highest warming) the potential vegetation cover type changes over the entire Willamette River Basin by the end of the 21st century.
• Existing forest vegetation types tend to be replaced by other types of forest.
• Subalpine forests, now dominated by Subalpine fir, are replaced by cool conifer forests, such as the Pacific silver fir.
• At the mid-elevations, the maritime conifer forest, generally associated with Douglas-fir and western hemlock, shift to more mixed forest types with hardwood species such as the Big Leaf Maple, and conifer species such as the Grand fir increasing in dominance.
• Potential vegetation cover type in the relatively low elevation Willamette Valley periphery change from maritime conifer to subtropical mixed forest in the Reference and HighClim scenarios, here driven by the coldest month temperature. The most likely broadleaf species to replace the existing Willamette Valley maritime conifer species are the Pacific madrone and tanoak. Both species currently dominate at mid-elevations in the Sierra Nevada Mountains 800 km (500 miles) to the south. These evergreen broadleaf species would benefit from the projected warmer winters and can tolerate drier summers.
• The change in the actual vegetation type following a disturbance lags the change in potential vegetation type, meaning what vegetation type the climate could support based on the MC2 model runs, in all scenarios. By 2100 in the LowClim scenario, the area of upland vegetation in disequilibrium is 22% of the total forest area. However, that proportion rises in the two warmer climate scenarios (53% in the Reference and 56% in HighClim scenarios) by the end of the century. Much of the speckling in the actual vegetation cover by 2100 (Fig. 5b) is where stand replacing disturbances, either harvests or fire, induces a change in vegetation cover type to the underlying potential vegetation cover type (Fig. 5a).

Figure 5. Vegetation distribution in 2100 for the Reference scenario: a) Potential Vegetation Cover type (from MC2), a) Actual Vegetation Cover type (from Envision).  (Figure from Turner, 2015)

Conclusions

The recent climate in the western U.S. is already warmer than in previous decades, and increases in tree mortality have been linked to climate change. Spatially explicit landscape simulation of potential and actual vegetation could be particularly effective in adaptation efforts. Using climate observations, stands in different locations could be regularly assessed for the degree to which the vegetation type is out of equilibrium with the local climate, and hence at risk for attack by pests and pathogens. The most vulnerable stands could be prioritized for thinning or harvest.

Dynamic global vegetation models (DGVMs) driven by the latest downscaled climate data could provide resource managers with guidance on what type of vegetation to replant after a disturbance. Our results support the conclusion that climate change will become an increasing influence on forest management decisions throughout the 21st century. The projected increase in the risk of fire points to investments in fire management.