|FEIS Home Page|
Abstract: Although little fire history information is available, the cool, wet climate and infrequent lightning in Alaskan mountain hemlock ecosystems suggest that fires were historically rare or absent. Even during years of extreme drought, moist fuels are likely to impede ignition and fire spread in these ecosystems, and presettlement fires are generally thought to have been small and of little ecological significance. Fires were more likely in mountain hemlock communities adjacent to or intergrading with south-central subboreal white spruce and Lutz spruce forests than in coastal, high-elevation, or subalpine mountain hemlock communities. LANDFIRE modelling estimates mean fire-return intervals ranging from 833 years in Alaskan subboreal mountain hemlock-white spruce forest to 1,112 years in the northern variant of Alaskan Pacific maritime mountain hemlock forest. However, some fire-return interval estimates used in those models are based on the time elapsed since fire last occurred in an area, and do not reflect a mean fire-return interval, but a fire-free period that resulted in current forest structure and composition. It is possible that fire-free intervals were even longer in most Alaskan mountain hemlock ecosystems. Contemporary changes in climate, fuel characteristics, and fire patterns in south-central Alaska suggest that fire activity is likely to increase and may lead to broadscale changes in plant community composition, structure, and distribution. Although climate change impacts are complex, interdependent, and largely uncertain, warming temperatures and lengthening frost-free and growing seasons suggest that mountain hemlock ecosystems may increase in productivity, shift upward in elevation, and possibly decrease in area. Changes in precipitation patterns are difficult to predict; however, warming temperatures alone indicate that growing and fire seasons will be longer and that fuels in Alaskan mountain hemlock ecosystems will be dry enough to carry fire over a longer period, even if precipitation increases substantially. Mountain hemlock is susceptible to fire injury and mortality and is slow to regenerate after fire; therefore, these ecosystems are particularly susceptible to changes in community structure and composition if fire activity increases.
|Figure 1—Treeline, Lower Paradise Lake, Chugach National Forest, Alaska. Photo ©2005 Steve Baskauf.|
Citation for this synthesis:
Zouhar, Kris. 2017. Fire regimes of Alaskan mountain hemlock ecosystems. In: Fire Effects Information System, (Online). U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Missoula Fire Sciences Laboratory (Producer). Available: https://www.fs.fed.us/database/feis/fire_regimes/AK_mountain_hemlock/all.html .
Very little information was available regarding fire regimes in Alaskan mountain hemlock communities as of 2017. Paleological fire history studies from mountain hemlock forests in southwestern British Columbia [22,40] are discussed in order to provide a regional context for climate-vegetation-fire interactions. These and related studies are covered in more detail in the Fire Regime Synthesis on Pacific Northwest mountain hemlock communities. This synthesis does not review the large body of primary literature on climate and vegetation trends throughout the Holocene (~12,000 years ago to present), but relies mostly on literature reviews of these topics (e.g., ) to provide context for discussing potential future changes in fire activity related to climatic and vegetation trends. Other reviews and syntheses of information on fire and other disturbance regimes are frequently cited in this synthesis (e.g., [2,4,8,25,31,37,42,72,78,86,100]). Information on climate change included in this synthesis comes primarily from the following reviews: [5,43,44,51,69,92,94,110].
Common names are used throughout this synthesis. See table A2 for a list of common and scientific names and links to FEIS Species Reviews of plant species that commonly occur in Alaskan mountain hemlock communities or are otherwise mentioned in this synthesis.
|Figure 2—Land cover distribution of Alaskan mountain hemlock (northern) communities based on the LANDFIRE Biophysical Settings (BpS) data layer . Numbers indicate LANDFIRE map zones. Click on the map for a larger image and zoom in to see details.|
Geography and climate: Wet coniferous forests and woodlands characterized by mountain hemlock occur at subalpine elevations along the Pacific Coast from southern Alaska to northern California . This synthesis covers Alaskan mountain hemlock ecosystems in south-central Alaska, from the eastern Kenai Peninsula south to around Yakutat (figure 2); an area that includes Prince William Sound and the Copper River Delta. Information on mountain hemlock communities in southeastern Alaska is included in the synthesis on Fire regimes of Alaskan Pacific maritime communities, and those in Washington, Oregon, and California are covered in Fire regimes of Pacific Northwest mountain hemlock communities.
Alaskan mountain hemlock communities occur throughout the Chugach National Forest in south-central Alaska, which is a transitional zone between boreal forests and the northernmost coastal temperate rainforest in North America. The area is dominated by heavily glaciated mountain ranges; over 30% of the area is covered by ice and snow, and 10% is rock. Forests make up about 16% of the area on the Kenai Peninsula, with mountain hemlock forests primarily on side slopes at low to mid elevations, and white, Sitka, and Lutz spruce dominating valley bottoms and lower side slopes. Forests cover about 20% of the area on Prince William Sound and are dominated by Sitka spruce, mountain hemlock, and western hemlock. About 10% of the land area of the Copper River Delta is forested; dominant species are Sitka spruce and western hemlock [27,99].
Alaskan mountain hemlock ecosystems occur from sea level to elevational treeline, in areas with short, cool summers, rainy falls, and long, cool, wet winters with heavy snow cover for 5 to 9 months . On the Kenai Peninsula, monthly precipitation declines from January through June, and abruptly increases from July through September. A brief period of relative drought occurs in June, making it the most favorable time for fires .
|Table 1—Climatic data from areas where mountain hemlock ecosystems occur in south-central Alaska .|
|Location||Mean annual temperature||Mean annual
(climate is transitional between maritime and continental)
|39 °F (3.9 °C)
at low elevations
20 °F (-6.7 °C)
at upper elevations
|Prince William Sound||40 °F (4.4 °C)
at sea level
32 °F (0 °C)
at upper elevations
|80 inches (200 cm)
at sea level
>300 inches (760 cm)
at upper elevations
|Copper River Delta||34 °F (1.1 °C) to 42 °F (5.6 °C)||80 inches (200 cm)
at sea level
200 inches (500 cm)
|Chugach and Saint Elias||cold summers and winters||precipitation occurs mainly as snow at elevations >8,000 feet (2,500 m)||up to 26.7 feet
Vegetation and site characteristics: At low elevations along the Gulf of Alaska coast, mountain hemlock is part of the temperate rainforest ecosystem, where it codominates with Sitka spruce or western hemlock. Yellow-cedar occurs in the overstory in isolated locations. At intermediate elevations, mountain hemlock dominates closed- to open-canopy forests. At higher elevations up to timberline, mountain hemlock dominates subalpine parkland composed of a mosaic of tree islands interspersed with meadows [21,56,77]. White spruce or Lutz spruce may be present on inland sites in the Kenai and Chugach mountains, but they typically have <15% canopy cover. Paper birch may be common in early-seral stages on inland sites .
Mountain hemlock ecosystems are among the most species-rich communities in southern coastal Alaska. They occur at the highest elevations (1,500 feet (460 m) or more, on average), and typically on steep slopes (35%-40%). Forest Inventory and Analysis (FIA) data from 2004 to 2008 showed that mountain hemlock forests were some of the oldest in southeastern and south-central Alaska (>40% of stands were >200 years old, ~10% were >300 years old), but they had lower live basal area than most other types. Alaska's coastal region contains slower growing, older forests than other U.S. coastal regions . Barrett and Christensen  provide biomass and volume estimates and stand distribution data for all forest types in southern coastal Alaska.
Table A1 lists the Biophysical Settings (BpSs) covered in this synthesis, provides links to their full descriptions, and summarizes data generated by LANDFIRE's successional modeling for those BpSs .
The following descriptions of Alaskan mountain hemlock communities are modified from NatureServe  unless otherwise cited. Corresponding BpS series are given in parentheses after NatureServe's ecological system name.
Alaskan Pacific Maritime mountain hemlock forest (16481) occurs along the Gulf of Alaska and Pacific Coast from Kenai Fjords through southeastern Alaska. This ecological system occurs primarily in the maritime region and, less frequently, in the subboreal transition on the inland side of the Kenai and Chugach mountains. In south-central Alaska it occurs from sea level to around 2,000 feet (600 m); in southeastern Alaska it occurs from about 1,000 to 3,300 feet (300-1,000 m). This system occurs on relatively stable slopes and benches where soils are generally well drained. It is common on north-facing slopes and uncommon on south-facing slopes. Associated canopy trees vary by region. Sitka spruce or western hemlock may codominate in the northern range. In the subboreal region, Lutz spruce may be present, but its cover is typically less than 15%. Throughout the entire range of the system, the dominant understory shrub is typically ovalleaf huckleberry . Plant communities in this system are further described by DeVelice et al. .
Fire regimes for this ecological system are described in two separate syntheses because fire occurrence varies geographically. This synthesis covers mountain hemlock forests in the subpolar rainforest region (west and north of Yakutat Bay ); these communities occur from the Kenai Mountains south to around Glacier Bay and upper Lynn Canal . Fire was historically rare in these communities (see Historical fire regimes). Mountain hemlock forests in southeastern Alaska are covered in the synthesis on Fire regimes of Alaskan Pacific maritime forests; fire was historically absent from those communities.
Alaskan Pacific Maritime subalpine mountain hemlock woodland (16481) occurs on upper slopes of mountain ranges along the Gulf of Alaska coast, including the Kenai, Chugach, St Elias, Fairweather, and Coast mountains, from the Kenai Fjords to southeastern Alaska and northern British Columbia. This system is characterized by patches of mountain hemlock forest and parkland interspersed with alpine heath or tall shrub communities. It is similar to Alaskan Pacific Maritime mountain hemlock forest, but it occurs at or near elevational treeline just below alpine tundra or forb meadows on upper mountain sideslopes, shoulders, and bedrock outcrops from about 1,600 to 2,600 feet (500-800 m) elevation. Mountain hemlock often has a stunted growth form (krummholz) in this system. Soils are typically shallow and are derived from glacial and colluvial deposits as well as residual bedrock. Treeline mountain hemlock is most common on north-facing slopes with late snowpack. It also occurs on east and west aspects but is uncommon on south-facing slopes . Treeline appears to be advancing upward in some areas (P. Hennon, personal observation cited in ).
Alaskan subboreal mountain hemlock-white spruce forest (16780) occurs on slopes and rolling terrain on the inland side of the Kenai and Chugach mountains. Soils are mesic and permafrost is rare. This forest type represents the transition from maritime to south-central boreal forest. Lutz spruce and mountain hemlock codominate the canopy; paper birch and balsam poplar may also be present. Ericaceous shrubs, Sitka alder, devil's-club, and feather mosses are common . Plant communities in this system are further described by DeVelice et al. .
Reference period: In order to understand how current vegetation and fire regimes developed, fire histories must be long enough to characterize the frequency and range of variability of fire occurrence in the system of interest . In ecosystems where fire is rare or infrequent and fire-free periods last for centuries to millennia, it is useful to understand past changes in vegetation and fire occurrence over similar time scales. Knowledge of the time since the last fire is not adequate for understanding historical fire frequency or assigning historical fire-return intervals in these ecosystems. Without long-term data, inferences about historical fire regimes are questionable and not likely to provide insights needed to manage for projected climate changes .
Charcoal records from lake sediments and soil profiles broaden the reference period and are well suited for reconstructing the incidence of past fire activity and its relationship to changing climate and vegetation, especially in ecosystems with fire-return intervals that exceed the age of the oldest trees . Fire reconstructions that span millennia show the historical range of variability in fire activity over the duration of a vegetation type and can offer insights into the role of large-scale climate changes, the influence of prehistoric human activity, and the causes and consequences of major reorganizations of vegetation [92,107,108]. Hundreds of charcoal records are available that span the Holocene (i.e., the past ~12,000 years) . These provide an opportunity to examine the links between fire, fuel, and climate on multiple temporal and spatial scales and through periods of substantial, and sometimes rapid, climate change [37,72,92,108]. In particular, times of rapid change or extreme climate anomalies (e.g., the last glacial-interglacial transition (~12,000-10,000) , the Medieval Warm Period (~950-1250), and the Little Ice Age (LIA, ~1650-1890)) may offer insights into potential future changes and provide perspectives on drivers of fire regimes and fire-climate-vegetation interactions [4,92].
Although long-term paleoecological records may include periods when the range of variability in climate, fire, or other ecosystem properties would have limited relevance to the present , the range of conditions during the Holocene, and especially those of the late Holocene (around 4,000 years ago to present), are arguably a better indicator of the true range of climate and vegetation variation possible on a particular landscape than are the last few centuries alone. Around 4,000 years ago, the large-scale controls of climate, and presumably the range of climatic variability, were approaching those of the last few centuries. This period is therefore highly relevant when considering the historical range of variability of ecosystems in this region (reviews by [37,46]). Understanding past climatic variability and its impacts on fuels and fire regimes may help managers anticipate potential effects of changing climates on disturbance regimes and forest succession [26,92] and plan management in that context , particularly when future climatic conditions are predicted to diverge substantially from those of the past few centuries.
Historical range of variability: Mountain hemlock communities are stable over long periods and rarely experience large-scale disturbances . Before European settlement, most of the landscape in the mountain hemlock zone in Alaska was old-growth forest, showing no signs of large-scale disturbance for many centuries (, Langille 1924 and Holbrook 1924 cited by ). Conifers, including mountain hemlock, likely dominated these sites for several thousand years (e.g., [22,50]). Studies of similar mountain hemlock communities in coastal southwestern British Columbia reveal ancient stands and estimates of 1,000  to more than 1,500 [64,81] years without stand-level disturbances. In a stand of Pacific silver fir, yellow-cedar, mountain hemlock, and western hemlock on Mt Cain on northern Vancouver Island, several trees of each species exceeded 900 years old, and the oldest trees (yellow-cedars) exceeded 1,400 years . Paludification may lead to conversion from old-growth mountain hemlock forest to peatland ecosystems on poorly drained sites .
The most common disturbances in Alaskan Pacific maritime mountain hemlock forests and woodlands include windthrow, fungal pathogens, snow avalanches, and soil creep [17,77]. These typically cause mortality of individual or small groups of trees, resulting in small-patch (or gap-phase) succession as regeneration occurs under canopy gaps (e.g., [17,63,64,81]). At high elevations and in areas with severe growing conditions, regeneration occurs only on microsites where snow melts earliest (e.g., small mounds or patches near trees), resulting in a patchy distribution of tree islands typical of subalpine parkland communities . In the mountain hemlock zone in the Coast Ranges of southwestern British Columbia, the estimated forest turnover time via gap-phase processes varied from 280 to 1,000 years . Climate and weather, especially extreme events, may have an important role in small-scale mortality of canopy and subcanopy trees (review by ).
Because large-scale disturbances are rare, successional patterns have not been studied in Alaskan Pacific maritime mountain hemlock forest and woodlands. Because of the dense shrub layer and poor growing conditions on Alaskan mountain hemlock forest and woodland sites, mountain hemlock communities probably require substantial time to return to old-growth conditions after large-scale disturbance .
Large-scale disturbances were historically more common in Alaskan subboreal mountain hemlock-white spruce forests than in Alaskan Pacific maritime mountain hemlock forests. Mountain hemlock-white spruce forests are more likely to experience large, albeit very infrequent, fires (see Historical fire regimes, below) . They are also susceptible to spruce beetle infestations, which can reach epidemic proportions and impact large areas. The most recent infestation, which began in 1987 , impacted over 1 million acres (~429,000 ha) on the Kenai Peninsula over about 15 years . See Postsettlement fuels for additional details.
Historical fire regimes: Fire has occurred in mountain hemlock forests in the Kenai Mountains during the past century (see Postsettlement fires), but whether fire was an important disturbance process within this ecosystem prior to European settlement is unclear. Very little is known about historical fire regime characteristics in Alaskan mountain hemlock ecosystems because fire is rare, and because tree-ring-based fire history methods are not effective in these communities due to a scarcity of even-aged stands and fire-scarred trees . Mountain hemlock and other subalpine mixed-conifer stands are typically multiaged, and tree ages do not indicate stand age because the time between stand-initiating events is often longer than the life span of individual trees [63,64].
Presettlement fires were probably rare in Alaskan mountain hemlock forests, particularly in the maritime region , and they were likely absent from mountain hemlock communities in southeastern Alaska . Presettlement fires in maritime ecosystems were thought to have been small and of little ecological significance (e.g., ). Fires were more likely in mountain hemlock communities that intergrade with white spruce and Lutz spruce forests in the transitional zone from maritime to south-central subboreal forests . On the Chugach National Forest, cool, moist mountain hemlock forests occur on north-facing mountain sideslopes and ridges, and show some evidence of rare fire occurrence (charred bowls resembling shallow fire scars) at the edges of stands near the lower elevational boundary of mountain hemlock forests . Additional anecdotal evidence from the forest zone in the Kenai Mountains (charcoal observed in soil pits) (USDA Forest Service 2002, cited in [55,56]) and radiocarbon dates on soil charcoal ranging from 4,500 to 570 years ago (Reiger 1995, cited by [85,99]) suggest a history of infrequent fire, but at unknown intervals. It is also unclear which plant communities these charcoal samples were taken from, although it is more likely that fires occurred in the western portion of the Kenai Mountains, where spruce forests dominate the landscape, lightning is more likely, and the climate is drier due to a rain shadow effect . Fire is included in the LANDFIRE model for the northern variant of Alaskan Pacific maritime mountain hemlock forest, although fire may not occur (or is extremely rare) in some areas to which this model applies .
Table A1 summarizes data generated by LANDFIRE succession modeling for the Biophysical Settings (BpSs) covered in this synthesis. The range of values generated for fire regime characteristics in Alaskan mountain hemlock communities is shown in table 2.
|Table 2—Modeled fire intervals and severities in Alaskan mountain hemlock communities .|
in each fire regime group
| ¹Average historical fire-return interval derived from LANDFIRE succession modeling (labeled "MFRI" in LANDFIRE).
²Percentage of fires in 3 fire severity classes, derived from LANDFIRE succession modeling. Replacement-severity fires cause >75% kill or top-kill of the upper canopy layer; mixed-severity fires cause 26%-75%; low-severity fires cause <26% [11,59].
³NA (not applicable) refers to BpS models that did not include fire in simulations.
Fire history and related studies in adjacent vegetation on the Kenai Peninsula [15,85] and in similar subalpine communities in coastal British Columbia  are referenced in this synthesis because they are sometimes cited as estimates for mean fire-return intervals in Alaskan mountain hemlock communities (e.g., [55,56,77]). Paleoecological studies from mountain hemlock forests in southwestern British Columbia [22,40] are discussed to provide a regional context for climate-vegetation-fire interactions during the Holocene.
Fuel characteristics: Descriptions of fuel characteristics in Alaskan mountain hemlock communities were rare in the available literature (as of 2017), with the exception of a fuel model for "Dwarf Tree Mountain Hemlock Scrub" described by Cella et al. , which would apply to subalpine mountain hemlock woodlands. Fuel characteristics may also be inferred from climate data, information on forest structure and composition, and disturbance history.
Fuels in Alaskan mountain hemlock communities are usually too wet to burn. Fires are less common in coastal Alaska than in the interior because of lower incidence of lightning, lower temperatures, higher amounts of rainfall, and wet conditions year-round [2,7,8,36,53,70,79,80,86,96,104]. Even during years of extreme drought, high fuel moisture content may impede ignition and fire spread in these ecosystems. Historically, years of high fire activity in the Pacific Northwest typically occurred when persistent high pressure ridges formed along the Pacific Coast, which blocked westerly winds, reduced precipitation, and allowed fuels to dry for extended periods . The same is true in coastal forests of British Columbia, especially the northern coast [10,31,83]. Exceptions may be found where rainforest transitions into subboreal forest on the Kenai Peninsula [7,85] and in areas of severe rain shadow such as Lynn Canal north of Juneau [80,100], where abundant fuels may become dry enough to carry fire during drought. This regional-scale rainshadow pattern is also apparent on the mainland coast of British Columbia and on Vancouver Island .
Topography influences fuel moisture and fire spread at local scales by directly affecting fuel exposure to irradiance and wind, which can dry fuels and make some parts of the landscape more susceptible to fire . In mountain hemlock ecosystems, southerly aspects tend to be drier and therefore may be more susceptible to burning .
Forest structure influences fuel loading and arrangement. Alaskan mountain hemlock stand structures are described in vegetation classifications (e.g., [17,77,101]). Mountain hemlock ecosystems range from closed-canopy forests (60%-100% overstory cover), to open-canopy woodlands (25%-59% cover), to subalpine krummholz (10%-24% cover) characterized by patches of trees interspersed with alpine heath or tall shrubs grading into alpine tundra and forb meadow systems [17,77,101]. Closed mountain hemlock forests typically have a well-developed shrub layer (about 3 feet (1 m) tall with about 65% cover) and mature mountain hemlock trees ranging from 55 to 75 feet (18-25 m) tall and 15 to 20 inches (38-50 cm) DBH. Open mountain hemlock communities have a well-developed shrub layer, with overstory trees averaging 30 to 70 feet (9-21 m) tall and 10 to 20 inches (25-50 cm) DBH . Shrubs may act as ladder fuels in these systems. In the mountain hemlock zone of the Pacific Northwest, closed or parkland forests have the highest probability of burning, due to dead fuel loads that can desiccate during east wind events and flammable lichens in lower branches [2,3]. Those mountain hemlock ecosystems typically have thick layers of humus (>4 inches (10 cm)) overlying mineral soils (e.g., ). Prolonged drought can dry these fuels and increase opportunities for ground fires [2,40]. The same is likely true for Alaskan mountain hemlock forests.
Where Alaskan mountain hemlock forests intergrade with spruce forests in south-central Alaska, spruce beetles can cause substantial mortality of spruce, altering fuel characteristics and raising concerns about increased fire hazard (e.g., [15,19,35,86,89]). However, mixed forests dominated or codominated by mountain hemlock showed fewer changes in fuel characteristics, stand structure, and species composition than spruce-dominated stands following spruce beetle outbreaks on the Kenai Peninsula [19,89]. See Contemporary Changes in Fuels and Fire Regimes for more information on these studies and related topics.
Fire season and ignition sources: Fires can occur in south-central Alaska from about April through October, although most occur during May, June, and July . The fire season is more restricted in coastal areas. On the Kenai Peninsula, a brief dry period in June can cause low fuel moisture, making it the most favorable time for fires . In western Canada, large areas burned when persistent ridges of high pressure created conditions necessary for large fires (i.e., dry convective storms, lightning, and high winds) [2,40].
Drought conditions and weather anomalies must coincide with ignitions for fire to occur in these wet to mesic forest types, and ignitions are rare . Lightning is much less common in coastal than in interior Alaska [36,80]. It is generally infrequent in cool, rainy coastal climates and typically occurs only during anomalously warm, dry weather . Based on reports of fires suppressed between 1957 and 1979, south-central Alaska had the fewest lightning fires in the state (with the exception of Southeast Alaska, the Aleutian Islands, and the Alaska Peninsula, which were excluded from the study because fires are so rare). Only three lightning fires per million acres (>405,000 ha) were recorded in south-central Alaska during those 23 years. A total of 13 lightning fires burned 0.01% of the area in the Pacific Border Ranges of Alaska during that period (0.6 fire/million acres/23 years) . In south-central Alaska, lightning is most common on the inland side of the Kenai and Chugach mountains and in upper Lynn Canal . It is very rare in southeastern Alaska, with strikes recorded in major towns only once every 1 to 2 decades . The probability of lightning increases from southeastern Alaska southward along the British Columbia coast [8,42]. Fire records from 1940 to 1982 in British Columbia indicate that lightning fires were least common on the northern coast, and incidence of lightning-caused fires and average fire size were greater on the inner mainland coast near Prince Rupert than on Haida Gwaii. Mountain hemlock forests at higher elevations had more lightning-caused fires than low-elevation forests on the coast .
Ignitions in Alaskan mountain hemlock forests may come from adjacent white and Lutz spruce-dominated forests, which burn more frequently. Observations in the Kenai Mountains suggest that fires may spread upslope from valley-bottom Lutz spruce stands, but often stop at the lower boundary of mountain hemlock-dominated forests [56,85,99].
Humans have inhabited the northwestern Pacific Coast for the last ~8,000 years , and Native Alaskans used fire for a variety of purposes such as cooking, warming, making canoes, felling large trees, clearing burial grounds, and drying or smoking fish and meat. These fires may have occasionally escaped and burned surrounding forests during infrequent periods of dry weather, but the effects were probably local [42,100]. Fire may have been used to maintain open spaces and promote growth of desirable plants; this has been documented on the northern coast of British Columbia  and in the coastal wet forests of the Pacific Northwest . However, because coastal tribes got much of their subsistence food from the ocean and rivers, they may have used fire less frequently than tribes whose settlements were in rainshadow climates  or farther inland . Some coastal peoples burned brush to facilitate hunting , particularly on dry south-facing slopes and valley bottoms . The seasonality and frequency of fire use by Native Americans varied with the ecosystem and the desired outcome .
Fire frequency: Although fire history studies for Alaskan mountain hemlock ecosystems are not available, the cool, wet climate and low incidence of lightning suggest long fire-free intervals. Long fire-return intervals are common in forests on the northwestern coast of North America, where exceptional droughts or successive years of drought are needed to create conditions conducive fire . Estimated fire frequencies for adjacent ecosystems on the Kenai Peninsula [15,85], and in similar forest types in coastal British Columbia (e.g., [40,64,81]) indicate that fire is rare in those forests as well. This is fairly consistent with LANDFIRE estimates of mean fire-return intervals ranging from 833 years in Alaska subboreal mountain hemlock-white spruce forest  to 1,112 years in the northern variant of Alaskan Pacific maritime mountain hemlock forest (table 2) . However, fire-free periods may be even longer on some sites (e.g., [40,64,81,82]), and the concept of a "mean" fire-return interval may not be meaningful in ecosystems where fire is so rare.
Some "fire frequency" estimates (e.g., those cited in [56,77]) are based on an estimate of how much time has elapsed since the last fire occurred in an area and do not reflect a mean fire-return interval, but a fire-free period that resulted in the current forest structure and composition. For example, based on a preliminary analyses of charcoal from soil profiles and on estimated ages (ranging from 200 to >1,000 years old) of large canopy trees growing on equally large, uncharred stumps, Lertzman and Krebs [63,64,65] estimate that mountain hemlock stands at Cypress Provincial Park in the North Shore Mountains of southwestern British Columbia have not experienced a major fire for over 1,500 years. Fire was also likely absent for 1,500 years or more from mountain hemlock stands at Mt Cain on northern Vancouver Island, where a reconstruction of stand history showed no signs of stand-level disturbance .
Evidence of past fires in Alaskan mountain hemlock communities is sparse. An unpublished report documenting "fire history for the mountain hemlock/Lutz spruce vegetation type of the Kenai Peninsula's mountainous portion of the Chugach National Forest" provides evidence of both historical fires (see Postsettlement fires) and prehistorical fires in those forests, although some sites had no evidence of past fires. Charcoal samples taken from subsurface soil layers at 4 sites dominated by mountain hemlock and 1 site dominated by Lutz spruce provide evidence of presettlement fires (table 3) . Anecdotal evidence indicated that charcoal was present in "most" soil pits examined in the forest zone of the Kenai Mountains, suggesting "widespread fire occurrence" but at unknown dates or intervals (USDA Forest Service 2002, cited in [55,56]). It is not clear whether those were mountain hemlock, spruce, or mixed forests. On the southwestern Kenai Peninsula, radiocarbon-dated soil charcoal from 22 sites in white spruce-Lutz spruce forests around Paradox Lake ranged from about 90 to 1,500 years old . Although mountain hemlock does not occur in the immediate study area around Paradox Lake, these data are sometimes cited as estimates of fire frequency in Alaskan mountain hemlock-white spruce forests (e.g., [55,77]). While fires in white and Lutz spruce-dominated forests may burn into adjacent mountain hemlock forests under extreme fire weather, observations in the Kenai Mountains suggest that fires often stop at the lower boundary of mountain hemlock-dominated forests [56,85]. On the Chugach National Forest, mountain hemlock stands on mountain sideslopes and ridges from the coastal zone to the interior and to timberline have evidence of rare fire occurrence at stand edges, including "charred bowls resembling shallow fire scars" that were found in small remnant gaps and at the edge of burns at the lower boundary of mountain hemlock stands. On mid and lower portions of mountain sideslopes, mountain hemlock mixes with Lutz spruce and broadleaf species, such as paper birch, balsam poplar, and quaking aspen. These species indicate relatively recent fire and are widespread on areas burned during the settlement period—predominately in valley bottoms and lower sideslopes .
|Table 3—Evidence of past fires from mountain hemlock forest sites in the Kenai Mountains . Blank cells indicate that no data were available.|
|Study site||Vegetation type||Charcoal evidence (radiocarbon date, where measured)||Fire severity||Age structure|
|Forest sites >200 years old, possibly burned during the presettlement period|
|North Kenai Lake
|Lutz spruce/menziesia-moss||surface and subsurface soil charcoal
(~1,540 BP, subsurface)
|Palmer Creek||mountain hemlock-Sitka spruce/menziesia-moss||no charcoal||no fire evidence||uneven|
|Hunter||mountain hemlock-Lutz spruce/devil's-club-spreading woodfern||subsurface soil charcoal
|North Kenai Lake
|mountain hemlock/menziesia||subsurface soil charcoal
|Schillter Creek||mountain hemlock/menziesia||subsurface soil charcoal (~2,470 BP)|
|Black Mountain||mountain hemlock/menziesia||subsurface soil charcoal
|Forest sites possibly burned during the settlement period, up to ~1924|
|West Swan Lake||mountain hemlock-white spruce/arctic dwarf birch-black crowberry||charred stumps||stand replacing||uneven|
|Tenderfoot Hillslope||mountain hemlock/menziesia||no charcoal||no fire evidence||uneven|
|Crescent Lake Trail||mountain hemlock/menziesia||charred stumps||mixed severity||uneven|
|Manitoba East||mountain hemlock/moss||charred stumps||stand replacing||uneven|
|Manitoba West||mountain hemlock/moss||no charcoal||no fire evidence||uneven|
|Pass Creek||mountain hemlock/menziesia||charred stumps||stand replacing||uneven|
|Rimrock Creek||mountain hemlock-Lutz spruce-Kenai birch/black crowberry||charred stumps||mixed severity||uneven|
|Upper Palmer Creek||mountain hemlock-Lutz spruce-Kenai birch/menziesia||charred stumps||mixed severity||uneven|
Fire type and severity: Little is known about fires in Alaskan mountain hemlock forests, although Agee  suggests that fires tended to be infrequent and stand-replacing in Pacific Northwest mountain hemlock forests. Due to its shallow roots, low-hanging branches, and flammable foliage, mountain hemlock is easily killed by fire [2,73], so fires in these forests may be of high severity with regard to canopy mortality. However, observations by Potkin  suggest that mixed-severity fires may have been historically important in mountain hemlock forests in the Kenai Mountains (table 3). This is supported by fire severity mapping of the 2005 Irish Channel Fire that burned in mountain hemlock forest on the Kenai National Wildlife Refuge, which showed approximately 22% of the area within the burn perimeter was unburned, ~38% of the area was burned at low severity, ~38% at moderate severity, and ~1% at high severity . See Postsettlement fires for more information.
Fire pattern and size: No data are available regarding historical fire pattern and size in Alaskan mountain hemlock communities. Some reports suggest that fires were large [2,55,56,77]. However, fire size in these systems is driven, in part, by forest distribution and structure, which is often patchy and grades into shrub, tundra, rock, and ice . The fragmented nature of these forests in the Kenai Mountains, for example, likely limited fire spread . The 2005 Irish Channel Fire burned ~925 acres (374 ha) of mountain hemlock forest in the Kenai Mountains .
Drought resulting from the 1912 Novarupta eruption (near Mt Katmai) may have contributed to large fires on the Kenai Peninsula from 1913 to 1915 . In the mountain hemlock zone in the Olympic Mountains, Washington, past fires occurred mainly on steep, south-facing slopes . In mountain hemlock forests in Oregon, some historical fires were >7,900 acres (3,200 ha); however, most fires tended to be <1,300 acres (500 ha) .
Disturbances that kill mountain hemlock may have long-term impacts, because mountain hemlock forests grow very slowly [63,64,65,81] and mountain hemlock growth is sensitive to several climate parameters [38,48].
Postsettlement fuels and stand structure: Contemporary fuel conditions in Alaskan mountain hemlock ecosystems may reflect changes brought about by human activities, disturbance from insects, and changing climate. These changes may affect the likelihood, size, and severity of wildfire. Forest management activities in the Chugach National Forest (e.g., vegetation management for wildlife habitat, fuel reduction, and land development) have likely had less impact on mountain hemlock than other communities because mountain hemlock typically occurs in more remote areas (e.g., ). Climate changes and insect outbreaks have likely had a greater impact on fuel characteristics in mountain hemlock ecosystems.
Climate change alters landscape composition and structure over time and appears to be reducing the alpine zone and snow and ice fields while expanding forested areas and subalpine communities (e.g., ), and impacts vary among sites. In the western Kenai Mountains of south-central Alaska, timberline (the altitudinal boundary between forest and woodland) rose very little between 1951 and 1996, but 29% of the forest-alpine tundra ecotone area increased in woodiness, closed-canopy forest expanded 14%, shrub cover expanded 4%, unvegetated areas decreased 17.4%, and tundra decreased 5% per decade. Area of open woodland remained constant but changed location. Cool aspects changed more dramatically than warm aspects. In general, the forest-tundra ecotone shifted upward on cooler, presumably more mesic aspects near seed sources; on warm aspects the density of shrubs and trees increased, but the ecotone did not rise. Treeline (altitudinal boundary between woodland and shrubland) rose ~160 feet (~50 m) on cool, northerly aspects, but not on other aspects .
The relationship between temperature and mountain hemlock growth in Alaskan forests has changed over time [29,48] and may vary with latitude  and, in Alaska, with altitude . Consequently, its growth and distribution may be affected unevenly as climate warms . For example, warming temperatures may increase growth rates of mountain hemlock at high and mid elevations. At low elevations, however, warming temperatures that result in earlier loss of insulating snowpack may increase susceptibility of mountain hemlock to root damage from late frosts [48,61]. Tree-ring analyses of mountain hemlocks at several sites in Glacier Bay National Park and near Juneau reveal changes in the relationship between mountain hemlock growth and climate when comparing growth patterns during the LIA (late 1800s) to those that followed. In high-elevation stands, mountain hemlock growth rates showed no significant relationship with temperature during the LIA and an increasingly positive relationship with temperature during the 20th century, suggesting increased growth at high elevations with further warming. In mid-elevation stands, growth rates were significantly correlated with temperature during both periods. In low-elevation stands, growth rates were correlated with temperature during the LIA, but showed a steadily weakening response to temperature during the 20th century (P < 0.05). This negative trend in correlation between growth and temperature at low-elevation sites is similar to that observed in the declining yellow-cedar, raising concerns for a similar decline in mountain hemlock on low-elevation sites with further climate warming .
Mountain hemlock trees can persist in low light conditions in the understory of mature forests and then grow rapidly when canopy trees die or are removed [19,73], so mountain hemlock-dominated forests may increase when associated spruce is killed. Following frequent and severe spruce beetle outbreaks in south-central Alaska during the late 20th and early 21st centuries that caused substantial spruce mortality, mountain hemlock-white spruce forests in the Kenai Mountains showed moderate white spruce mortality (46% reduction in basal area), increased cover of mountain hemlock, and a 28% decrease in area, while mountain hemlock forest showed a 22% increase in area .
Spruce beetle outbreaks may also increase the likelihood and severity of fire due to the rapid growth of grasses and the abundance of beetle-killed trees [35,86,89]. However, stands dominated or codominated by mountain hemlock had fewer changes in fuel characteristics, stand structure, and species composition than spruce-dominated stands following outbreaks [19,89]. Between 1987 and 2000 fuel heights, fine fuels, and sound large fuels increased in spruce beetle-impacted forests on the Kenai Peninsula, while organic layer depths and rotten large fuels decreased in all forest types analyzed (white spruce, paper birch, and mountain hemlock). The only changes that were significant in mountain hemlock forest plots, however, were a decrease in organic layer depth and an increase in large, sound (1,000-hour) fuels (P<0.05) . Bluejoint reedgrass cover did not increase in the Kenai Mountains or along the Gulf of Alaska where mountain hemlock codominates .
Although mountain hemlock forests show few changes in fuel characteristics following spruce beetle outbreaks, it is widely believed by land managers and the public that changes in adjacent white and Lutz spruce-dominated forests set the stage for large and severe wildfires (e.g., [35,78,86,97,98,99]). Wildfires in these spruce forests could spread into adjacent mountain hemlock forests (see Fire season and ignition sources). However, few studies have been published on the relationship between fire activity and spruce beetle outbreaks, and those available lack consensus. Studies on the Kenai Peninsula showed no relationship between fire activity and spruce beetle outbreaks during the 1800s or during the past ~2,500 years [15,86]. In contrast, Hansen  found a strong positive relationship between spruce beetle outbreak and the subsequent probability of large wildfire (>1,240 acres (>500 ha)) in spruce forests on the western Kenai Peninsula between 2001 and 2009. Eleven large wildfires and 324 small wildfires (<500 ha) occurred in that area during that time. The large fires burned about 3.8% of the study area, and approximately 97% of the small fires were within ~6 miles (10 km) of a road. The occurrence (P < 0.01) and duration (P < 0.001) of the spruce beetle outbreak increased the probability of large wildfires, but neither affected the probability of small fires (P > 0.05) . Wildfire risk may depend on time since tree death and specific beetle-caused changes in stand structure .
Evidence from the Kenai Peninsula shows that Alaskan wetlands that may have served as firebreaks in the past are drying and succeeding to upland habitat as a result of climate changes [16,52]. These wetlands may become "fuel bridges" as they convert to shrublands and forests. With a warmer and drier climate, fires can spread more effectively across these large areas of continuous fuels , perhaps increasing the possibility of spread into mountain hemlock forests.
Postsettlement fires: While information specific to mountain hemlock communities is lacking, reviews of historical literature and early accounts by explorers and settlers suggest that prior to European settlement, which began in the late 1800s, most of the coniferous forests on the Kenai Peninsula were in late successional stages (e.g., [18,85]). Fire frequency increased in south-central Alaska in the late 1800s and early 1900s with the arrival of miners and the construction of railroads. According to an unpublished report by Potkin in 1997 , much of the landscape mosaic on the peninsula at that time was comprised of early-successional vegetation, including grass, shrub, and broadleaf types that were the result of human-caused fires that occurred during and after European settlement. These large patches of relatively homogeneous, early-seral species may last 100 years or more. For example, large areas of mature birch near the community of Hope, Alaska, are the result of human-caused fires that occurred in the early 1900s . Holbrook (1924, cited in [85,95]) mapped approximately 30,000 acres (>12,000 ha) of burned area on the Chugach National Forest. The largest fires on the Forest occurred during periods of mining and mineral exploration from 1849 to 1902, and during railroad development between 1903 and 1953 [85,99]. From 1914 to 1953 on the Kenai Peninsula portion of the Chugach National Forest, an average of 22.5 fires occurred each year, about 73% of which were related to the railroad. Fires on the Kenai Peninsula were particularly frequent and large between 1913 and 1915 and are attributed to drought brought on by the 1912 eruption of Novarupta (near Mt Katmai) and ignitions from railroad activity. After the end of the steam engine era around 1954, fires decreased in both size and number on the Kenai Peninsula; however, recreational use since then has also resulted in a high incidence of human-caused fires [85,95].
|Figure 3—Aerial view of the 2014 Funny River Fire. Photo by John Morton, Kenai National Wildlife Refuge.|
Since the 1950s, the major causes of fires on the Kenai Peninsula have been campfires and debris burning. By the end of the 20th century, over 99% of fires were human-caused [85,86]. From 1914 to 1997 a total of 1,364 fires burned about 65,000 acres (26,300 ha) on the peninsula [85,95]. Presumably, these fires burned mostly in subboreal spruce forests because areas with limited human access, such as mountain hemlock forests, had few fires during settlement and postsettlement periods (e.g., ). In July 2005, a year with record high temperatures in Alaska and numerous lightning-caused fires on the Kenai Peninsula, lightning ignited the Irish Channel Fire in mountain hemlock forest on the Kenai National Wildlife Refuge. The fire was not suppressed, and it slowly burned about 925 acres (374 ha)  at mostly low and moderate severity . Several other fires larger than 100 acres (40 ha) occurred on the Kenai Peninsula in the early 2000s. The largest of these was the Funny River Fire that was ignited by humans on 19 May 2014, and burned almost 200,000 acres (~81,000 ha) by early June . The fire was mixed severity and burned in several forest types, including about 500 acres (~200 ha) of mountain hemlock forest . Lack of snow in late winter and very little rain (0.39 inch (10 mm)) in the preceding month resulted in dry fuels that led to rapid spread under windy conditions . See the synthesis on Fire Regimes of Alaskan white spruce communities for more information on that fire. For more information on contemporary and historical fires in Alaska, see the Alaska Interagency Coordination Center, Incident Information maps.
Climate change and fire regimes: Global climate warming during the past century is unequivocal, and scientific evidence shows major and widespread ecosystem changes throughout the globe that are associated with increasing air and ocean temperatures . Projections generally indicate that a warmer world will have more fire, although the potential impacts vary geographically and among ecosystems .
Climate change is caused, in part, by alterations in atmospheric concentrations of greenhouse gases and aerosols, and anthropogenic activities are modifying both the average state and variability of climate by adding greenhouse gasses—particularly carbon dioxide and methane—to the atmosphere . Current atmospheric concentrations of these gasses far exceed the natural range over the last 650,000 years and have increased markedly (35% and 148%, respectively) since the beginning of the industrial era in 1750. Both past and future anthropogenic emissions will continue to contribute to warming temperatures for more than a millennium, due to the time scales required for the removal of these gasses from the atmosphere .
Fire regimes in Alaskan mountain hemlock communities are strongly climate-driven, and are therefore potentially susceptible to climate changes. Fire is rare in these ecosystems because lightning is rare and fuels are typically too wet to burn. Increasing temperatures, changing moisture relationships, and changing storm patterns have the potential to increase fire frequency by drying fuels; changing the composition, structure, and distribution of fuels; lengthening the fire season; and altering lightning patterns . An increase in fire activity at many high-elevation coastal sites during the Medieval Warm Period (~950-1250)  suggests that more frequent and severe fires might be expected with further warming and decreased spring snowpack, as has been shown in the western United States (e.g., [105,106]). However, the effects of climate changes are complex, and complexity increases at finer spatial scales due to local variability in factors that affect fuel characteristics such as local weather patterns, topography, dominant vegetation, disturbance history, and management history . The following sections describe recorded and projected climate changes in south-central Alaska, how these changes are expected to alter fuels and possibly fire regimes, and some climate change considerations for land managers:
Observed and projected climate changes: Climate datasets indicate substantial changes in southern coastal Alaska during the past century [93,110] including increased temperatures, changing precipitation patterns, reduced depth and duration of snowpack, and increasing storm intensities. These changes have the potential to affect the probability and characteristics of wildfire [43,94] by changing fuel characteristics, ignition patterns, fire season length, and fire weather. It is very likely that continued greenhouse gas emissions at or above the current rate will cause further warming and result in changes that exceed those observed during the 20th century . Projections of future effects of climate change are based on global circulation models, assuming continued increases in greenhouse gases and, because the rate of increase is unknown, most models present a range of future predictions .
During the late 20th and early 21st centuries, Alaska has warmed at more than twice the rate of the rest of the United States , and mean annual temperature in the coastal region of south-central and southeastern Alaska increased 2 to 3 °F (1.1-1.6 °C) , with especially large increases in winter temperatures [43,51,94]. Temperature increases in south-central Alaska were driven, in part, by a warm-phase of the Pacific Decadal Oscillation (PDO) that began in 1976 (Hartmann and Wendler 2005, cited by ).
|Table 4—Increases in mean annual air temperature (MAAT) in southern coastal Alaska between 1971 and 2008 by location .|
|Valdez||3.76 °F (2.09 °C)|
|Yakutat||2.75 °F (1.53 °C)|
|Kodiak||0.87 °F (0.48 °C)|
Warmer temperatures, especially in winter, have led to longer snow-free seasons, more precipitation falling as rain instead of snow, changes in vegetation, an accelerated rate of glacier recession, and loss of ice and permafrost [5,110], all of which can contribute to longer or more active fire seasons [5,86]. The number of snow-free days across Alaska increased an average of 10 days during the latter part of the 20th century , and the length of the growing season increased almost 7 days per decade in south-central and southeastern Alaska between 1949 and 1997. The first snow-free week in Alaska occurred 3 to 5 days earlier per decade from 1972 to 2000, and the duration of the snow-free period extended 3 to 6 days longer per decade . Glacial recession has been documented in several areas. Exit Glacier, for example, retreated approximately 1.5 miles between 1815 and 2010 .
Records show a pattern of generally decreasing precipitation in south-central Alaska since the mid-20th century. At the Kenai airport, precipitation declined by 10% (from 20 inches (513 mm) to 18 inches (463 mm)) between 1944-1967 and 1968-2007 . Along with decreased precipitation, increased evapotranspiration associated with higher temperatures and longer growing seasons is likely to reduce available moisture and result in drier conditions [32,43]. Local meteorological records show increasing temperatures and decreasing water balance in the Kenai Lowlands , including a 55% decline in available water since 1968. About one-third of this decline was due to higher summer temperatures and increased evapotranspiration, and two-thirds was due to lower annual precipitation .
Regardless of the models or greenhouse gas-emission scenarios used, it is universally expected that warming will continue throughout the globe, with larger temperature increases at higher latitudes. It is very likely that hot extremes, heat waves, heavy precipitation events, and more intense storms will become more frequent throughout the globe [43,69,94].
Air temperatures in Alaska are expected to continue to increase [90,110], with winter temperatures increasing at a higher rate than summer temperatures [43,94]. Projections indicate that southern coastal Alaska will have the largest increase in frost-free days of any area in North America during the 21st century because current winter mean temperatures hover close to the 32 °F (0 °C) freezing threshold. For example, projected temperatures for the Kenai Peninsula show that temperatures in March and November are expected to shift from below freezing to above freezing. Warmer winters would lead to longer growing seasons, altered distribution of snow cover, considerably reduced snow accumulation in some areas, and earlier snowmelt and peak runoff [23,43,46,94]. With a 1.8 °F (1.0 °C) increase in temperature, the growing season is projected to increase by 20 to 40 days in Alaska by 2100 (vs. 1961-1990), with coastal areas seeing the greatest increases .
Precipitation is projected to increase in most areas of coastal Alaska, particularly in fall and winter [51,90,110], but more will fall as rain instead of snow. Nonetheless, many locations are expected to continue to have decreased water availability, increased drought stress, and overall drier conditions during summer due to increased evapotranspiration with higher temperatures [32,43] and changes in atmospheric circulation patterns . Weather extremes are generally expected to be more common with climate change, so that occasional droughts are more likely . Projections frequently show greater extremes of the El Niño Southern Oscillation (ENSO) and PDO indices in the future, indicating the potential for long periods without precipitation, more severe droughts in some locations, and more precipitation, flooding, and stormy weather at other times and locations [32,44]. Above-average temperature and precipitation are typical along the Alaskan coast during and just after moderate to strong El Niño episodes .
Projected changes in fuels and fire regimes: Both anthropogenic climate changes (especially increased temperatures and aridity) and natural climate variability (overlapping patterns of ENSO, PDO, etc.) are implicated in increased wildfire activity observed in the late 20th and early 21st century in interior Alaska  and the western United States [1,92,106]. Modeling by Krawchuk et al.  and Bachelet et al.  predicted increases in fire probability and area burned along the southern coast of Alaska . Because a smaller spring snow pack would have a compounding effect on summer fuel moisture, high-elevation sites, where mountain hemlock forests are most common, could become disproportionately more susceptible to fire under a warmer climate [39,40,62].
As of 2017, no information was available specifically addressing the potential effects of climate change on fire regimes in Alaskan mountain hemlock communities. Contemporary changes in fire activity observed in south-central Alaska seem to be mostly anthropogenic, and not directly caused by climate changes (see Postsettlement fires). Nonetheless, widespread changes in vegetation and fuel characteristics have been observed that are at least partly attributed to climate warming (see Postsettlement fuels). These types of changes are expected to continue throughout the northern Pacific coastal region and possibly become more widespread and severe with further warming [33,43,90,94]. These changes may result in more frequent, severe, and larger fires in Alaskan mountain hemlock ecosystems by reducing fuel moisture and altering fuel structure, composition, and distribution through species range shifts and altered disturbance regimes.
Climate change may have direct and indirect effects on fire regimes by creating conditions more or less favorable for burning, by changing the frequency and severity of other climate-mediated disturbances, and by changing vegetation and fuel characteristics across the landscape [26,110]. Fire occurrence over long periods reflects direct as well as indirect (i.e., vegetation and human land use) climatic controls, with a strong potential for interactions and feedbacks between fire and its controls . Increased fire in Alaskan mountain hemlock forests could accelerate or exacerbate changes brought about by other factors, which could, in turn, increase fire potential.
Climate change and fuel structure and distribution: Plant species' responses to climate change and other disturbances can include changes in vigor, phenology, growth rates, mortality, and occupancy of particular sites , leading to changes in the distribution of vegetation formations (forest, shrubland, grassland) across the landscape, and changes in species composition and structure within those types . Alaskan coastal  and high-elevation ecosystems [39,40,62,90,103] may be particularly sensitive to these effects of climate change.
Large vegetation shifts, such as those from woodland to forest or alpine tundra to woodland, are expected to alter fire regimes . Forests are projected to remain the dominant formation in northern Pacific coastal ecosystems, but their distribution and composition may change substantially due to range shifts, expansions, and contractions of tree species. Coniferous forests are expected to expand in south-central Alaska and may serve as biome refugia. However, increased incidence of fire, insect and disease outbreaks, mortality, and aridity will also affect tree establishment and survival .
Direct and indirect effects of climate warming are altering vegetation production, composition, and distribution in Alaskan coastal ecosystems with corresponding changes in the amount, composition, structure, and distribution of fuels on the landscape. Many of the changes in vegetation already observed in Alaskan ecosystems may be attributed to higher temperatures, reduced depth and duration of snowpack, elevated snowline, longer growing seasons, decreasing water balance, glacial recession, and altered disturbance regimes (see Postsettlement fuels and Fire regimes of Alaskan Pacific maritime ecosystems), all of which are expected to continue along with associated changes in productivity, species compositions, and competitive relationships; increased mortality; moisture and nutrient limitations; range shifts; increases in populations of damaging agents (insects, diseases, herbivores); and changing storm patterns and associated disturbances [12,43,94,110]. These changes will continue to alter fuel characteristics, which will impact the likelihood and characteristics of fire. However, there is a great deal of uncertainty regarding the potential impacts of these changes in specific ecosystems.
Changes in productivity, range shifts, and mortality: High-elevation Alaskan mountain hemlock ecosystems range from subalpine mountain hemlock forest to mountain hemlock woodlands in the forest-tundra ecotone. It is generally thought that with warming temperatures and lengthening frost-free and growing seasons, mountain hemlock ecosystems will increase in productivity, shift upward in elevation, and decrease in area [43,73,87,94]. Given the complexity of the landscapes that Alaskan mountain hemlock ecosystems inhabit and the interactions of other processes affecting their productivity and distribution, these patterns will vary among sites. In high-elevation areas in the northern Pacific coastal region, tree establishment in subalpine meadows is increasing, some treelines are advancing upslope, and some treelines are retreating [28,29,94]. The suite of processes that limit tree growth at treeline are complex and not fully understood, and forest ecotone change depends on biophysical processes that vary both by region and by landscape, such that rapid rise in treeline can occur on some slopes while nearby slopes have no treeline rise [28,29].
Mean annual temperatures in alpine ecosystems are projected to exceed those of current subalpine ecosystems by the 2080s in much of the northern Pacific coastal region [90,103]. Rising winter temperatures, reduced depth and duration of snowpack, and increased elevation of snowpack may lead to altitudinal and latitudinal expansion of subalpine forests, shifting altitudinal boundaries between forest and woodland (timberline) and between woodland and shrubland (treeline), and losses of alpine tundra ecosystems [28,43,90,94,103]. Modeling by Wang et al.  predicts that by the 2080s, British Columbia's coastal alpine ecosystems will shift upward in elevation more than 600 feet (200 m), subalpine mountain hemlock will shift upward almost 1,500 feet (455 m) and northward almost 50 miles (75 km), and coastal western hemlock will shift upward more than 1,000 feet (323 m) and northward about 43 miles (69 km).
With less snowpack to limit tree establishment, the forest-tundra ecotone in the mountains of south-central Alaska will likely change to a woodier landscape with less tundra and more closed-canopy forest [28,29], resulting in a more continuous distribution of fuels at high elevations [40,62] as trees and shrubs replace alpine tundra vegetation. A 75% to 90% loss of tundra to boreal and temperate forest is projected for Alaska, statewide [28,94]. Examination of the forest-tundra ecotone in the western Kenai Mountains showed increased woodiness in 29% of the area between 1951 and 1996—a period of warming and drying across the Kenai Peninsula . Ecosystem boundaries in both the Kenai (1950-1996) and Chugach (1972-2012) mountains shifted upwards in elevation, as did growing season isotherms. Timberline advanced more rapidly than treeline, but models suggest a close match between the upward advance of tall woody ecosystems and the expectation due to warming .
A review by Haufler et al.  suggests a rise in treeline of 2,000 to 3,000 feet (600-900 m) in southern coastal Alaska with climate warming. However, treeline advance was less than expected in the mountains of south-central Alaska from 1950 to 1996, given the degree of climate warming, and was more likely on mesic slopes near seed sources [28,29]. Using FIA data from south-central and southeastern Alaskan coastal forests and modeled distribution and productivity of western hemlock, Sitka spruce, yellow-cedar, western redcedar, mountain hemlock, and shore pine, Caouette et al.  found indications of changing relationships between distribution and elevation for mountain hemlock, yellow-cedar, and Sitka spruce that may be indicative of a changing climate. Mountain hemlock and yellow-cedar appeared to be shifting upslope in some areas, and Sitka spruce shifting downslope. This study suggests that western hemlock, Sitka spruce, western redcedar, and shore pine may be favored with warmer conditions, and that mountain hemlock may have diminished suitable habitat due to reduced snow accumulation and subsequent root damage from late frosts . Upward movements of western hemlock and other midelevation trees are expected as higher elevations become more suitable habitat, and these species may replace mountain hemlock on some sites . However, some insects (e.g., spruce beetle, spruce aphid), parasites (e.g., hemlock dwarf mistletoe), and herbivores (e.g., mule deer) that impact these tree species may also be favored by warmer temperatures, and may therefore offset gains in their productivity  and possibly favor mountain hemlock on some sites. A review by Tillman and Glick  suggests that mountain hemlock in western Washington is vulnerable to warmer summers, reduced snowpack, and frost damage if earlier snowmelt triggers shoot growth before the last frost, but that microhabitat variability created by complex mountain terrains may provide refugia. Warmer temperatures and reduced snowpack may increase mountain hemlock growth and productivity on some sites (e.g., Washington and northern Oregon) and reduce productivity in areas subject to drought stress (e.g., southern Oregon and low-elevation distribution limits) [84,94].
Insects and disease: Warming temperatures and stress caused by drought, heat, and other disturbances are generally expected to contribute to increases in frequency and severity insect and disease outbreaks throughout western North America (e.g., [14,49]) and northern Pacific coastal forests [31,43,90,94,110], largely due to the direct effects of temperature on development and survival of over-wintering life stages. Forest insects and fungal pathogens are also expected to more fully occupy the current range of their host tree species and expand ranges northward and to higher elevations along with their hosts as climate warms . These changes seem less likely to affect mountain hemlock in Alaska and more likely to impact its associated tree species. The relationships between climate and biotic disturbance agents are complex and interdependent, and readers are encouraged to see the primary literature cited in these reviews for additional details: [31,43,44,94,110].
Mountain hemlock seems to have fewer pests and damaging agents than associated tree species in Alaska, although this could change if climate warming allows these pests to expand their ranges to higher latitudes and elevations. With reduced snow cover and longer growing seasons, for example, hemlock dwarf mistletoe, a potentially damaging parasite that causes reduced vigor and occasionally death , could migrate upslope and have increased disease effects on mountain hemlock . Hemlock dwarf mistletoe occurs throughout the range of mountain hemlock, but rates of infection decrease from Washington north, and it rarely infects mountain hemlock in Alaska . Analysis of forested plots in southeastern Alaska indicates that climate limits both abundance and distribution of hemlock dwarf mistletoe to a subset of the range of its primary tree host, western hemlock, with infection varying from a high of 20% of trees at sea level, to only 5% at about 700 feet (200 m) . Cold temperatures can impact dwarf mistletoes at several stages of their life cycle , and thus limit their rate of spread, effects on individual trees, and overall range. Predictions from three climate envelope models indicated a modest potential increase in abundance and range expansion for western hemlock, and extremely large potential increases in abundance of hemlock dwarf mistletoe, mostly within its current range but including a westward expansion in potential habitat .
Several defoliating insects, bark beetles, and wood-boring insects attack mountain hemlock but do not cause extensive damage, and in the southern part of its range (in California), mountain hemlock is susceptible to laminated root rot and several other fungal pests . The range and impacts of these pests may increase with climate warming; however, damage and mortality of associated tree species from other agents is more likely. For example, climate change may increase spruce beetle outbreaks [14,19,31,49,78,110], which have the potential to cause rapid and broad-scale mortality of spruce. Changes in fuel characteristics and possibly increased fire hazard are expected in conjunction with these outbreaks, from both direct mortality and because trees weakened by infestation and infection are less tolerant of drought and heat stress; more susceptible to other insects, pathogens, and disturbances; and more likely to break or fall during wind storms [31,43,94,110]. Increased fire hazard in white and Lutz spruce and mixed white spruce-mountain hemlock forests may increase fire hazard in adjacent mountain hemlock ecosystems. See Postsettlement fuels for details on the interactions between spruce beetle outbreaks, fuel characteristics, and fire regimes.
Altered disturbance regimes: Projected climate changes are expected to impact the frequency and severity of disturbances other than fire, which will impact vegetation structure and species composition at landscape scales, changing fuel characteristics and having potential feedbacks on fire regimes. Altered weather patterns including a northern shift in storm tracks and greater storm intensities are one consequence of increasing ocean temperatures, and increased storm intensities are expected in southern coastal Alaska . Wind damage, floods, and landslides can be expected to increase on terrain where they are already risk factors . Wind damage is common in mountain hemlock forests in the coastal strip of British Columbia and Alaska , and climate change may affect wind storm frequency and severity [14,31,49,110]. Resulting damage may indirectly affect patterns and severity of wildfire. The prediction of future forest disturbance regimes is in its infancy, but managers may wish to adjust plans accordingly where there is consensus among projections .
Climate change and fuel moisture: Fuels in Alaskan mountain hemlock ecosystems are usually too wet to burn, except in areas where droughts sometimes occur, such as Lynn Canal north of Juneau [80,100], and areas where adjacent ecosystems are more likely to burn and carry fire into mountain hemlock stands, such as the Kenai Mountains [7,85]. In areas where droughts and fires occasionally occurred historically, they are likely to increase in frequency and severity with further warming . Increased frequency and severity of water limitation and drought have already been observed on the Kenai Peninsula (see Postsettlement fuels) and are expected to continue [32,43]. June water availability is projected to decrease 10% to 75% in south-central Alaska (June 2090-2099 vs. June 1961-1990) .
Warmer and drier conditions in systems with ample fuel loads and ignition sources will increase the incidence and likely the size and severity of wildfires. The effect of warmer and wetter conditions is less straightforward , and projections of future precipitation patterns have a higher degree of uncertainty than those for temperature [31,74]. A significant decrease in area burned throughout British Columbia from 1920 to 2000 was strongly related to significant increases in precipitation ranging from 22.7% to 45.9% (P < 0.05). However, in the mountain hemlock zone, summer precipitation, drought indices, and temperature were equally strong predictors of area burned (P < 0.001). Significant increases in both mean summer temperature (+0.92 °F (0.51 °C); P < 0.10) and summer precipitation (+3.2 inches (82 mm); P < 0.01) were evident from 1920 to 2000 in the mountain hemlock zone . Nonetheless, projected temperature increases alone indicate that growing and fire seasons will be longer, and that fuels in Alaskan mountain hemlock ecosystems will be dry enough to carry fire over a longer period of time  even if precipitation increases substantially . Flannigan et al.  calculated that in order to maintain fuel moisture levels as temperatures increase, precipitation would have to increase 5% to 15% for every degree of warming. A 15% increase in precipitation would be required to maintain moisture levels in fine surface fuels, 10% for upper forest floor (duff) layers, and about 5% for deep organic soils . Increased precipitation is projected in some parts of the northern Pacific coastal region, but rarely of this magnitude . Furthermore, precipitation has tended to decline and water limitation to increase in south-central Alaska as climate warms .
Water limitation not only increases the potential for drier fuels during a longer portion of the fire season, it also constrains the growth and distribution of many tree species, and makes some more susceptible to attack from insects and disease [6,87,94]. These impacts have the potential to alter fuel structure and distribution at landscape scales.
Climate change and fire characteristics: While temperatures are sure to increase in the future, changes in precipitation patterns are more difficult to predict, making projections for future fire regimes less certain in ecosystems where fire activity is driven by precipitation (e.g., ). A substantial increase in summer precipitation could keep fuels moist and maintain low levels of fire activity in Alaskan mountain hemlock forests, as was observed in British Columbia from 1920 to 2000 . However, projected temperature increases alone indicate that growing and fire seasons will be longer and suggest that fuels in Alaskan mountain hemlock ecosystems will be dry enough to carry fire over a longer period of time [62,99] even if precipitation increases substantially . An increase in annual area burned is likely with longer fire seasons .
Drier fuels alone can increase the likelihood of fire, but fire activity is also driven by extreme fire weather , which is also likely to increase with climate warming. Extreme fire weather could render a greater portion of the landscape susceptible to fire . A warmer planet will have a moister atmosphere, a larger number of extreme convective storms, and a greater density of lightning discharges [39,43]. Increased incidence of lightning, especially during extended drought, would certainly increase the likelihood of wildfires in coastal Alaska . Increasing human populations and associated sources of ignitions may also create a greater fire risk .
Simulations for California over the next century predict increased area burned in subalpine forests (review by ). It is unclear whether such changes in fire frequency can be expected in more northern Pacific subalpine communities [22,40], although modelling by Bachelet et al.  projects increases in future area burned along the southern and western coasts of Alaska . Fires in coastal ecosystems are likely to be severe for dominant conifers and other plants that are not well adapted to survive fire.
Management considerations under a changing climate: A warmer climate and altered disturbance regimes are likely to change fuel characteristics and may increase the likelihood of fire in Alaskan mountain hemlock ecosystems. Alaskan mountain hemlock ecosystems have historically experienced little or no fire, and are therefore particularly susceptible to changes in community structure and composition with increases in fire activity. Mountain hemlock is susceptible to fire injury and mortality and is slow to regenerate after fire , especially if conditions are dry. Tree regeneration at higher elevations is largely unpredictable and only partially correlated with time-since-disturbance, because it is substantially affected by browsing, snow creep, and the duration of the snowpack . Increased fire activity, especially in conjunction with changes in other disturbances, can therefore exacerbate or accelerate vegetation changes caused by climate warming and may catalyze relatively rapid changes [37,66,87,94]. Fire activity could change faster than many terrestrial species may be able to accommodate . Understanding thresholds for systems beyond which changes are irreversible may be an important component of forest management in a changing climate .
Continued glacier recession and more frequent and severe disturbances will lead to large areas undergoing both primary and secondary succession, and the novel climatic conditions present as succession begins will impact long-term trajectories  such that novel communities are likely to develop . The probability of nonnative, invasive plant establishment and spread is also likely to increase with climate warming and increased disturbance [43,110], although this is less likely in remote mountain hemlock ecosystems.
Consideration of intermediate and long-term climatic trends is essential for anticipating and planning forest management . Given that the atmospheric concentrations of greenhouse gasses will likely continue to increase and exacerbate climate change effects for the foreseeable future, adaptive actions to reduce ecosystem vulnerability, increase the capacity to withstand or be resilient to change, and/or accept their transformation to novel systems compatible with likely future conditions are warranted . Land managers are tasked with devising regionally-specific but flexible strategies to adapt to a warmer climate, while maintaining the essential values of ecosystem productivity, integrity, and biodiversity . Although uncertainty and gaps in knowledge exist, resources are available to help plan for and address climate change impacts (e.g., ). Implementing strategic adaptation actions early may reduce severe impacts and prevent the need for more costly actions in the future . Haufler et al.  list and describe a large number of initiatives, programs, and organizations focused on management adaptations for climate change, and they outline a strategic plan for adapting management with regard to climate change in south-central and southeastern Alaska. They describe several landscape conservation strategies, including (but not limited to) monitoring key indicators of climate change, establishing model watersheds to support monitoring and adaptive management, assessing risks to identify changes that can be mitigated, and mitigating and adapting to anticipated and observed effects . Examples of planned or ongoing adaptation efforts in the northern Pacific coastal region are described by Tillmann and Glick . Sommers et al.  provide some specific suggestions so managers can use information about fire history and climate change to better understand potential fire regimes in the face of climate change, and use this information to help shape fire and fuel management decisions in the 21st Century. Haughian et al.  stress the importance of a greater emphasis on risk analysis over productivity, as well as managing for flexibility and resilience.
Detailed knowledge of historical fire regime characteristics in Alaskan mountain hemlock ecosystems is lacking, likely due to the historical rarity of fire. Some "fire frequency" estimates (e.g., those cited in [56,77]) are based on an estimate of how much time has elapsed since the last fire occurred in an area, and do not reflect a mean fire-return interval, but a fire-free period that resulted in current forest structure and composition. LANDFIRE models calculated mean fire-return intervals of 833 to 1,112 years for Alaskan mountain hemlock ecosystems in south-central Alaska [55,56] based on similar, and often anecdotal, information. It is possible that fire-free intervals were longer in montane and subalpine mountain hemlock ecosystems in south-central Alaska (see Fire frequency), and the concept of a "mean" fire-return interval may not be meaningful in ecosystems where fire is so rare.
Contemporary changes in fire patterns, fuel characteristics, and climate suggest that fire activity is likely to increase in these ecosystems and may lead to broadscale changes in plant community composition, structure, and distribution. While it is reasonable to expect that vegetation composition, structure, and productivity will change with changing climate and disturbances, future trends in climate, weather events, and tree mortality are difficult to project with confidence. Regional differences in topography and vegetation types drive disturbance regimes and will continue to do so under changing climates . A better characterization of climate variability, seasonal patterns, and frequency of severe events—rather than long-term trends in climate means—is needed to better understand potential ecological responses. Because global climate models do not account for local variability, fine-scale monitoring is important for evaluating assumptions based on larger-scale models .
The direct and causal relationships between climate, weather, fuel moisture content, ignition probability, and fire behavior make plausible estimates of climate change impacts on wildfire possible  and, since the early 1990s, researchers have been using projected changes in temperature and precipitation derived from global climate models to explore the potential influence of climate change on disturbance regimes, especially fire regimes (e.g., [34,54,67,68]). However, the nature of these changes and their effects on individual ecosystems are difficult to discern with precision, due to the complex and non-linear interactions between weather, vegetation, and people , as well as fire-climate-vegetation feedbacks that could have a further warming effect on global climate (e.g., through fire-related emissions) and affect other natural disturbances that kill or defoliate trees. The result of interactions between these phenomena and fire activity is very difficult to predict. For models projecting the future of species distributions, especially those of plants, methods are needed that explicitly integrate the effects of fire activity on vegetation, in addition to species range changes based on plant-climate relationships alone . To better understand the implications of potential increases in the frequency and scale of fire in Alaskan mountain hemlock ecosystems, more information is needed regarding the consequences of fire and disease on successional trajectories of contemporary forest types .
Long-term paleoecological data might provide insights needed to understand forest response to projected climate changes  and can help to clarify the links between fire, fuel, and climate through periods of substantial climate change, revealing which processes are most important in controlling fire occurrence . Applications of paleoecological data to future forecasts are complicated because climate-vegetation (fuel) combinations in the future may not resemble those of any time in the past. Nonetheless, an understanding of the processes that produced the variability observed in paleoecological records may be used to test the mechanistic models used for predicting future variations in fire  and help managers to anticipate potential effects of future climates on trajectories of forest change, particularly when conditions are predicted to exceed the historical ranges of variation .
Table A1: Summary of modeled fire regime information for Biophysical Settings covered in this synthesis
|Table A2—Plants that commonly occur in Alaskan mountain hemlock communities or were mentioned in this synthesis. Follow links to FEIS Species Reviews for additional information on those species.|
|Common name||Scientific name|
|balsam poplar||Populus balsamifera subsp. balsamifera|
|Kenai birch||Betula papyrifera var. kenaica|
|Lutz spruce||Picea × lutzii|
|mountain hemlock||Tsuga mertensiana|
|paper birch||Betula papyrifera|
|Pacific silver fir||Abies amabilis|
|quaking aspen||Populus tremuloides|
|shore pine||Pinus contorta var. contorta|
|Sitka spruce||Picea sitchensis|
|western hemlock||Tsuga heterophylla|
|western redcedar||Thuja plicata|
|white spruce||Picea glauca|
|Alaskan bellheather||Harrimanella stelleriana|
|Aleutian mountainheath||Phyllodoce aleutica|
|arctic dwarf birch||Betula nana|
|black crowberry||Empetrum nigrum|
|bog blueberry||Vaccinium uliginosum|
|dwarf bilberry||Vaccinium caespitosum|
|hemlock dwarf mistletoe||Arceuthobium tsugense|
|mountain cranberry||Vaccinium vitis-idaea|
|ovalleaf huckleberry||Vaccinium ovalifolium|
|Sitka alder||Alnus viridis subsp. sinuata|
|yellow mountainheath||Phyllodoce glanduliflora|
|American skunkcabbage||Lysichiton americanus|
|calthaleaf avens||Geum calthifolium|
|claspleaf twistedstalk||Streptopus amplexifolius|
|common camas||Camassia quamash|
|darkthroat shootingstar||Dodecatheon pulchellum|
|heartleaf twayblade||Listera cordata|
|strawberryleaf raspberry||Rubus pedatus|
|threeleaf foamflower||Tiarella trifoliata|
|bluejoint reedgrass||Calamagrostis canadensis|
|fewflower sedge||Carex pauciflora|
|grassyslope arctic sedge||Carex anthoxanthea|
|manyflower sedge||Carex pluriflora|
|sedge||Carex, Eriophorum spp.|
|tall cottongrass||Eriophorum angustifolium|
|tufted bulrush||Trichophorum caespitosum|
|variegated sedge||Carex stylosa|
|Ferns and fern allies|
|field horsetail||Equisetum arvense|
|deer fern||Blechnum spicant|
|long beechfern||Phegopteris connectilis|
|queen's-veil maiden fern||Thelypteris quelpaertensis|
|spreading woodfern||Dryopteris expansa|
|western oakfern||Gymnocarpium dryopteris|
|feather moss||Hylocomium spp.|
|Schreber's moss||Pleurozium schreberi|
|splendid feather moss||Hylocomium splendens|
1. Abatzoglou, John T.; Williams, A. Park. 2016. Impact of anthropogenic climate change on wildfire across western US forests. Proceedings of the National Academy of Sciences. 113(42): 11770-11775. 
2. Agee, James K. 1993. Fire ecology of Pacific Northwest forests. Washington, DC: Island Press. 493 p. 
3. Agee, James K.; Smith, Larry. 1984. Subalpine tree reestablishment after fire in the Olympic Mountains, Washington. Ecology. 65(3): 810-819. 
4. Alaback, P.; Veblen, T. T.; Whitlock, C.; Lara, A.; Kitzberger, T.; Villalba, R. 2003. Climatic and human influences on fire regimes in temperate forest ecosystems in North and South America. In: Bradshaw, G. A.; Marquet, P. A., eds. How landscapes change: Human disturbance and ecosystem fragmentation in the Americas. Berlin: Springer-Verlag: 49-87. 
5. Alaska Fire Science Consortium. . Research summary: Alaska climate change adaptation series--wildfires, [Online]. In: Library--Newsletters, fact sheets and summaries. Fairbanks, AK: Alaska Fire Science Consortium (Producer). Available: http://www.frames.gov/files/7913/4764/4448/CES_Wildfire_and_Climate_Summary.pdf [2017, March 6]. 
6. Allen, Craig D.; Macalady, Alison K.; Chenchouni, Haroun; Bachelet, Dominique; McDowell, Nate; Vennetier, Michel; Kitzberger, Thomas; Rigling, Andreas; Breshears, David D.; Hogg, E. H. (Ted); Gonzalez, Patrick; Fensham, Rod; Zhang, Zhen; Castro, Jorge; Demidova, Natalia; Lim, Jong-Hwan; Allard, Gillian; Running, Steven W.; Semerci, Akkin; Cobb, Neil. 2010. A global overview of drought and heat-induced tree mortality reveals emerging climate change risks for forests. Forest Ecology and Management. 259(4): 660-684. 
7. Anzinger, Dawn; Radosevich, Steven R. 2008. Fire and nonnative invasive plants in the Northwest Coastal bioregion. In: Zouhar, Kristin; Smith, Jane Kapler; Sutherland, Steve; Brooks, Matthew L., eds. Wildland fire in ecosystems: Fire and nonnative invasive plants. Gen. Tech. Rep. RMRS-GTR-42-vol. 6. Ogden, UT: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station: 197-224. 
8. Arno, Stephen F. 2000. Fire in western forest ecosystems. In: Brown, James K.; Smith, Jane Kapler, eds. Wildland fire in ecosystems: Effects of fire on flora. Gen. Tech. Rep. RMRS-GTR-42-vol. 2. Ogden, UT: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station: 97-120. 
9. Bachelet, D.; Lenihan, J.; Neilson, R.; Drapek, R.; Kittel, T. 2005. Simulating the response of natural ecosystems and their fire regimes to climatic variability in Alaska. Canadian Journal of Forest Research. 35(9): 2244-2257. 
10. Banner, A.; MacKenzie, W.; Haeussler, S.; Thomson, S.; Pojar, J.; Trowbridge, R. 1993. A field guide to site identification and interpretation for the Prince Rupert Forest Region, Parts 1 and 2. Land Management Handbook No. 26. Victoria, BC: Ministry of Forests Research. Variously paginated. 
11. Barrett, S.; Havlina, D.; Jones, J.; Hann, W.; Frame, C.; Hamilton, D.; Schon, K.; Demeo, T.; Hutter, L.; Menakis, J. 2010. Interagency fire regime condition class guidebook (FRCC), [Online], (Version 3.0). In: Interagency fire regime condition class website. U.S. Department of Agriculture, Forest Service; U.S. Department of the Interior; The Nature Conservancy (Producers). Available: https://www.frames.gov/files/7313/8388/1679/FRCC_Guidebook_2010_final.pdf [2017, March 1]. 
12. Barrett, Tara M.; Christensen, Glenn A., tech eds. 2011. Forests of southeast and south-central Alaska, 2004-2008: Five-year forest inventory and analysis report. Gen. Tech. Rep. PNW-GTR-835. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station. 156 p. 
13. Barrett, Tara M.; Latta, Greg; Hennon, Paul E.; Eskelson, Bianca N. I.; Temesgen, Hailemariam. 2012. Host-parasite distributions under changing climate: Tsuga heterophylla and Arceuthobium tsugense in Alaska. Canadian Journal of Forest Research. 42(4): 642-656. 
14. Bentz, Barbara J.; Regniere, Jacques; Fettig, Christopher J.; Hansen, E. Matthew; Hayes, Jane L.; Hicke, Jeffrey A.; Kelsey, Rick G.; Negron, Jose F.; Seybold, Steven J. 2010. Climate change and bark beetles of the western United States and Canada: Direct and indirect effects. BioScience. 60(8): 602-613. 
15. Berg, Edward E.; Anderson, R. Scott. 2006. Fire history of white and Lutz spruce forests on the Kenai Peninsula, Alaska, over the last two millennia as determined from soil charcoal. Forest Ecology and Management. 227(3): 275-283. 
16. Berg, Edward E.; Hillman, Kacy McDonell; Dial, Roman; DeRuwe, Allana. 2009. Recent woody invasion of wetlands on the Kenai Peninsula lowlands, south-central Alaska: A major regime shift after 18000 years of wet Sphagnum-sedge peat recruitment. Canadian Journal of Forest Research. 39(11): 2033-2046. 
17. Boggs, Keith; Klein, Susan C.; Flagstad, Lindsey; Boucher, Tina.; Grunblatt, Jess; Koltun, Beth. 2008. Landcover classes, ecosystems and plant associations: Kenai Fjords National Park. Natural Resource Technical Report NPS/KEFJ/NRTR-2008/136. Fort Collins, CO: U.S. Department of the Interior, National Park Service, Natural Resource Program Center. 222 p. 
18. Boucher, Tina V. 2003. Vegetation response to prescribed fire in the Kenai Mountains, Alaska. Res. Pap. PNW-RP-554. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station. 59 p. 
19. Boucher, Tina V.; Mead, Bert R. 2006. Vegetation change and forest regeneration on the Kenai Peninsula, Alaska following a spruce beetle outbreak, 1987-2000. Forest Ecology and Management. 227(3): 233-246. 
20. Boyd, Robert. 1999. Introduction. In: Boyd, Robert, ed. Indians, fire, and the land in the Pacific Northwest. Corvallis, OR: Oregon State University: 1-30. 
21. Brett, Robert B.; Klinka, Karel. 1998. A transition from gap to tree-island regeneration patterns in the subalpine forest of south-coastal British Columbia. Canadian Journal of Forest Research. 28(12): 1825-1831. 
22. Brown, K. J.; Hebda, R. J. 2003. Coastal rainforest connections disclosed through a late-quaternary vegetation, climate, and fire history investigation from the mountain hemlock zone on southern Vancouver Island, British Columbia, Canada. Review of Palaeobotany and Palynology. 123(3-4): 247-269. 
23. Caouette, J. P.; Steel, E. A.; Hennon, P. E.; Cunningham, P. G.; Pohl, C. A.; Schrader, B. A. 2016. Influence of elevation and site productivity on conifer distributions across Alaskan temperate rainforests. Canadian Journal of Forest Research. 46(2): 249-261. 
24. Cella, Brad; Allen, Jennifer; Butteri, Peter; Cole, Frank; Fleming, Michael; Howard, Kato; Koltun, John; Long, Gene; Martyn, Parker; Murphy, Karen; Noonan-Wright, Erin; Passek, Jan; Scott, Joe; See, John; Slaughter, Kent; Sorbel, Brian; Theisen, Skip. 2008. Fuel model guide to Alaska vegetation. Unpublished report on file with: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory, Missoula, MT. 62 p. [+ appendices]. 
25. Daniels, Lori D.; Gray, Robert W. 2006. Disturbance regimes in coastal British Columbia. Journal of Ecosystems and Management. 7(2): 44-56. 
26. Daniels, Lori D.; Maertens, Thomas B.; Stan, Amanda B.; McCloskey, Shane P. J.; Cochrane, Jed D.; Gray, Robert W. 2011. Direct and indirect impacts of climate change on forests: Three case studies from British Columbia. Canadian Journal of Plant Pathology. 33(2): 108-116. 
27. DeVelice, R. L.; Hubbard, C. J.; Boggs, K.; Boudreau, S.; Potkin, M.; Boucher, T.; Wertheim, C. 1999. Plant community types of the Chugach National Forest: Southcentral Alaska. Technical Publication R10-TP-76. Anchorage, AK: U.S. Department of Agriculture, Forest Service, Chugach National Forest. 375 p. 
28. Dial, Roman J.; Berg, Edward E.; Timm, Katriina; McMahon, Alissa; Geck, Jason. 2007. Changes in the alpine forest-tundra ecotone commensurate with recent warming in southcentral Alaska: Evidence from orthophotos and field plots. Journal of Geophysical Research. 112(G4): 1-15. 
29. Dial, Roman J.; Smeltz, T. Scott; Sullivan, Patrick F.; Rinas, Christina L.; Timm, Katriina; Geck, Jason E.; Tobin, S. Carl; Golden, Trevor S.; Berg, Edward C. 2016. Shrubline but not treeline advance matches climate velocity in montane ecosystems of south-central Alaska. Global Change Biology. 22(5): 1841-1856. 
30. Dickman, Alan; Cook, Stanton. 1989. Fire and fungus in a mountain hemlock forest. Canadian Journal of Botany. 67(7): 2005-2016. 
31. Dorner, Brigitte; Wong, Carmen. 2003. Natural disturbance dynamics in coastal British Columbia. St. Victoria, BC: Scientific Background Report for the Coast Information Team. 68 p. 
32. Flannigan, M. D.; Wotton, B. M.; Marshall, G. A.; de Groot, W. J.; Johnston, J.; Jurko, N.; Cantin, A. S. 2016. Fuel moisture sensitivity to temperature and precipitation: Climate change implications. Climatic Change. 134(1): 59-71. 
33. Flannigan, Mike D.; Krawchuk, Meg A.; de Groot, William J.; Wotton, B. Mike; Gowman, Lynn M. 2009. Implications of changing climate for global wildland fire. International Journal of Wildland Fire. 18(5): 483-507. 
34. Flannigan, Mike; Cantin, Alan S.; de Groot, William J.; Wotton, Mike; Newbery, Alison; Gowman, Lynn M. 2013. Global wildland fire season severity in the 21st century. Forest Ecology and Management. 294: 54-61. 
35. Flint, Courtney G.; Haynes, Richard. 2006. Managing forest disturbances and community responses: Lessons from the Kenai Peninsula, Alaska. Journal of Forestry. July/August: 269-275. 
36. Gabriel, Herman W.; Tande, Gerald F. 1983. A regional approach to fire history in Alaska. BLM-Alaska Tech. Rep. 9. Anchorage, AK: U.S. Department of the Interior, Bureau of Land Management. 34 p. 
37. Gavin, Daniel G.; Hallett, Douglas J.; Hu, Feng Sheng; Lertzman, Kenneth P.; Prichard, Susan J.; Brown, Kendrick J.; Lynch, Jason A.; Bartlein, Patrick; Peterson, David L. 2007. Forest fire and climate change in western North America: Insights from sediment charcoal records. Frontiers in Ecology and the Environment. 5(9): 499-506. 
38. Gedalof, Ze'ev; Smith, Dan J. 2001. Dendroclimatic response of mountain hemlock (Tsuga mertensiana) in Pacific North America. Canadian Journal of Forest Research. 31: 322-332. 
39. Hallett, Douglas J.; Anderson, R. Scott. 2010. Paleofire reconstruction for high-elevation forests in the Sierra Nevada, California, with implications for wildfire synchrony and climate variability in the late Holocene. Quaternary Research. 73(2): 180-190. 
40. Hallett, Douglas J.; Lepofsky, Dana S.; Mathewes, Rolf W.; Lertzman, Ken P. 2003. 11,000 years of fire history and climate in the mountain hemlock rain forests of southwestern British Columbia based on sedimentary charcoal. Canadian Journal of Forest Research. 33(2): 292-312. 
41. Hansen, Winslow D. 2013. Linked disturbance interactions in south-central Alaska: Implications for ecosystems and people. University of Alaska, Fairbanks. 133 p. [+ tables & figures]. Thesis. 
42. Harris, Arland S.; Farr, Wilbur A. 1974. The forest ecosystem of southeast Alaska. Chapter 7: Forest ecology and timber management. Gen. Tech. Rep. PNW-25. Portland. OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Forest and Range Experiment Station. 109 p. 
43. Haufler, J. B.; Mehl, C. A.; Yeats, S. 2010. Climate change: Anticipated effects on ecosystem services and potential actions by the Alaska Region, U.S. Forest Service. Seeley Lake, MT: Ecosystem Management Research Institute. 40 p. [+ appendices]. 
44. Haughian, Sean R.; Burton, Philip J.; Taylor, Steve W.; Curry, Charles L. 2012. Expected effects of climate change on forest disturbance regimes in British Columbia. BC Journal of Ecosystems and Management. 13(1): 1-24. 
45. Hawksworth, Frank G.; Wiens, Delbert. 1996. Dwarf mistletoes: Biology, pathology, and systematics. Agriculture Handbook 709. Washington, DC: U.S. Department of Agriculture, Forest Service. 401 p. 
46. Hennon, Paul E.; McKenzie, Carol M.; D'Amore, David V.; Wittwer, Dustin T.; Mulvey, Robin L.; Lamb, Melinda S.; Biles, Frances E.; Cronn, Rich C. 2016. A climate adaptation strategy for conservation and management of yellow-cedar in Alaska. Gen. Tech. Rep. PNW-GTR-917. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station. 368 p. [+ appendices]. 
47. Hess, Jason C.; Scott, Carven A.; Hufford, Gary L.; Fleming, Michael D. 2001. El Nino and its impact on fire weather conditions in Alaska. International Journal of Wildland Fire. 10(1): 1-13. 
48. Jarvis, Stephanie K.; Wiles, Gregory C.; Appleton, Sarah N.; D'Arrigo, Rosanne D.; Lawson, Daniel E. 2013. A warming-induced biome shift detected in tree growth of mountain hemlock [Tsuga mertensiana (Bong.) Carriere] along the Gulf of Alaska. Arctic, Antarctic, and Alpine Research. 45(2): 211-218. 
49. Jenkins, Michael J.; Hebertson, Elizabeth; Page, Wesley; Jorgensen, C. Arik. 2008. Bark beetles, fuels, fires and implications for forest management in the Intermountain West. Forest Ecology and Management. 254(1): 16-34. 
50. Jones, Miriam C.; Peteet, Dorothy M.; Kurdyla, Dorothy; Guilderson, Thomas. 2009. Climate and vegetation history from a 14,000-year peatland record, Kenai Peninsula, Alaska. Quaternary Research. 72: 207-217. 
51. Kelly, Brendan P.; Ainsworth, Thomas; Boyce, Douglas A. Jr.; Hood, Eran; Murphy, Peggy; Powell, Jim. 2007. Climate change: Predicted impacts on Juneau. Report to: Mayor Bruce Botelho and the City and Borough of Juneau Assembly. Juneau, AK: Scientific Panel on Climate Change, City and Borough of Juneau. 86 p. 
52. Klein, Eric; Berg, Edward E.; Dial, Roman. 2005. Wetland drying and succession across the Kenai Peninsula Lowlands, south-central Alaska. Canadian Journal of Forest Research. 35(8): 1931-1941. 
53. Kramer, Mark G.; Hansen, Andrew J.; Taper, Mark L.; Kissinger, Everett J. 2001. Abiotic controls on long-term windthrow disturbance and temperate rain forest dynamics in southeast Alaska. Ecology. 82(10): 2749-2768. 
54. Krawchuk, Meg A.; Moritz, Max A.; Parisien, Marc-Andre; Van Dorn, Jeff; Hayhoe, Katharine. 2009. Global pyrogeography: The current and future distribution of wildfire. PLoS ONE. 4(4): e5102. 
55. LANDFIRE Biophysical Settings. 2009. Biophysical Setting 7516780: Alaska sub-boreal mountain hemlock-white spruce forest. In: LANDFIRE Biophysical Setting Model: Map zone 75, [Online]. In: Vegetation Dynamics Models. In: LANDFIRE. Washington, DC: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory; U.S. Geological Survey; Arlington, VA: The Nature Conservancy (Producers). Available: https://www.landfire.gov/national_veg_models_op2.php. 
56. LANDFIRE Biophysical Settings. 2009. Biophysical setting 7716481: Alaskan Pacific Maritime Mountain Hemlock Forest - Northern. In: LANDFIRE Biophysical Setting Model: Map zone 77, [Online]. In: Vegetation Dynamics Models. In: LANDFIRE. Washington, DC: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory; U.S. Geological Survey; Arlington, VA: The Nature Conservancy (Producers). Available: https://www.landfire.gov/national_veg_models_op2.php. 
57. LANDFIRE Biophysical Settings. 2009. Biophysical setting 7816482: Alaskan Pacific Maritime Mountain Hemlock Forest - Southeast. In: LANDFIRE Biophysical Setting Model: Map zone 78, [Online]. In: Vegetation Dynamics Models. In: LANDFIRE. Washington, DC: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory; U.S. Geological Survey; Arlington, VA: The Nature Conservancy (Producers). Available: https://www.landfire.gov/national_veg_models_op2.php. 
58. LANDFIRE Biophysical Settings. 2009. LANDFIRE Vegetation Product Descriptions, Biophysical Settings, [Online]. In: Vegetation Dynamics Models. In: LANDFIRE. Washington, DC: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory; U.S. Geological Survey; Arlington, VA: The Nature Conservancy (Producers). Available: https://www.landfire.gov/NationalProductDescriptions20.php [2017, January 10]. 
59. LANDFIRE Rapid Assessment. 2005. Reference condition modeling manual (Version 2.1). Cooperative Agreement 04-CA-11132543-189. Boulder, CO: The Nature Conservancy; U.S. Department of Agriculture, Forest Service; U.S. Department of the Interior. 72 p. On file at: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory, Missoula, MT. 
60. LANDFIRE. 2008. Alaska refresh (LANDFIRE 1.1.0). Biophysical settings layer. In: LANDFIRE data distribution site, [Online]. In: LANDFIRE. U.S. Department of the Interior, Geological Survey (Producer). Available: https://landfire.cr.usgs.gov/viewer/ [2017, January 10]. 
61. Laroque, Colin P.; Smith, Dan J. 2003. Radial-growth forecasts for five high-elevation conifer species on Vancouver Island, British Columbia. Forest Ecology and Management. 183: 313-325. 
62. Lertzman, Ken; Gavin, Daniel; Hallett, Douglas; Brubaker, Linda; Lepofsky, Dana; Mathewes, Rolf. 2002. Long-term fire regime estimated from soil charcoal in coastal temperate rainforests. Conservation Ecology. 6(2): 5 p. 
63. Lertzman, Kenneth P. 1992. Patterns of gap-phase replacement in a subalpine, old-growth forest. Ecology. 73(2): 657-669. 
64. Lertzman, Kenneth P.; Krebs, Charles J. 1991. Gap-phase structure of a subalpine old-growth forest. Canadian Journal of Forest Research. 21(12): 1730-1741. 
65. Lertzman, Kenneth Peter. 1989. Gap-phase community dynamics in a sub-alpine old growth forest. Vancouver, British Columbia: The University of British Columbia. 168 p. Dissertation. 
66. Littell, Jeremy S.; Oneil, Elaine E.; McKenzie, Donald; Hicke, Jeffrey A.; Lutz, James A.; Norheim, Robert A.; Elsner, Marketa M. 2010. Forest ecosystems, disturbance, and climatic change in Washington State, USA. Climate Change. 102(1-2): 129-158. 
67. Liu, Yongqiang; Goodrick, Scott L.; Stanturf, John A. 2013. Future U.S. wildfire potential trends projected using a dynamically downscaled climate change scenario. Forest Ecology and Management. 294: 120-135. 
68. Liu, Yongqiang; Stanturf, John; Goodrick, Scott. 2010. Trends in global wildfire potential in a changing climate. Forest Ecology and Management. 259(4): 685-697. 
69. Loehman, Rachel; Anderson, Greer. 2009. Understanding the science of climate change: Talking points: Impacts to western mountains and forests. Natural Resource Report NPS/NRPC/NRR --2009/090. Fort Collins, CO: U.S. Department of the Interior, National Park Service, Natural Resource Program Center. 37 p. 
70. Lotan, James E.; Alexander, Martin E.; Arno, Stephen F.; French, Richard E.; Langdon, O. Gordon; Loomis, Robert M.; Norum, Rodney A.; Rothermel, Richard C.; Schmidt, Wyman C.; van Wagtendonk, Jan. 1981. Effects of fire on flora: A state-of-knowledge review: National fire effects workshop: Proceedings. 1978 April 10-14; Denver, CO. Gen. Tech. Rep. WO-16. Washington, DC: U.S. Department of Agriculture, Forest Service. 71 p. 
71. Marlon, J. R.; Bartlein, P. J.; Walsh, M. K.; Harrison, S. P.; Brown, K. J.; Edwards, M. E.; Higuera, P. E.; Power, M. J.; Anderson, R. S.; Briles, C.; Brunelle, A.; Carcaillet, C.; Daniels, M.; Hu, F. S.; Lavoie, M.; Minckley, T.; Richard, P. J. H.; Scott, A. C.; Shafer, D. S.; Tinner, W; Umbanhowar, C. E., Jr.; Whitlock, C. 2009. Wildfire responses to abrupt climate change in North America. Proceedings of the National Academy of Sciences. 106(8): 2519-2524. 
72. Marlon, Jennifer R.; Bartlein, Patrick J.; Daniau, Anne-Laure; Harrison, Sandy P.; Maezumi, Shira Y.; Power, Mitchell J.; Tinner, Willy; Vanniere, Boris. 2013. Global biomass burning: A synthesis and review of Holocene paleofire records and their controls. Quaternary Science Reviews. 65: 5-25. 
73. Means, Joseph E. 1990. Tsuga mertensiana (Bong.) Carr. mountain hemlock. In: Burns, Russell M.; Honkala, Barbara H., technical coordinators. Silvics of North America. Volume 1, Conifers. Agric. Handb. 654. Washington, DC: U.S. Department of Agriculture, Forest Service: 623-634. 
74. Meyn, Andrea; Schmidtlein, Sebastian; Taylor, Stephen W.; Girardin, Martin P.; Thonicke, Kirsten; Cramer, Wolfgang. 2013. Precipitation-driven decrease in wildfires in British Columbia. Regional Environmental Change. 13(1): 165-177. 
75. Morton, John M.; Berg, Edward; Newbould, Doug; MacLean, Dianne; O'Brien, Lee. 2006. Wilderness fire stewardship on the Kenai National Wildlife Refuge, Alaska. International Journal of Wilderness. 12(1): 14-17. 
76. Morton, John. 2014. Fire footprint - ghosts of fires past, [Online]. Augusta, GA: Morris Communications LLC, Peninsula Clarion (Producer). Available: http://peninsulaclarion.com/outdoors/2014-06-12-0 [2017, May 17]. 
77. NatureServe. 2013. International Ecological Classification Standard: Terrestrial Ecological Classifications of the United States and Canada. In: NatureServe Central Databases. Arlington, VA, (Producer). 1530 p. 
78. Negron, Jose F.; Bentz, Barbara J.; Fettig, Christopher J.; Gillette, Nancy; Hansen, E. Matthew; Hayes, Jane L.; Kelsey, Rick G.; Lundquist, John E.; Lynch, Ann M.; Progar, Robert A.; Seybold, Steven J. 2008. US Forest Service bark beetle research in the western United States: Looking toward the future. Journal of Forestry. 106(6): 325-331. 
79. Neiland, Bonita J. 1971. The forest-bog complex of southeast Alaska. Vegetatio. 22(1-3): 1-64. 
80. Noste, Nonan V. 1969. Analysis and summary of forest fires in coastal Alaska. Juneau, AK: U.S. Department of Agriculture, Forest Service, Pacific Northwest Forest and Range Experiment Station, Institute of Northern Forestry. 12 p. 
81. Parish, Roberta; Antos, Joseph A. 2004. Structure and dynamics of an ancient montane forest in coastal British Columbia. Oecologia. 141(4): 562-576. 
82. Parish, Roberta; Antos, Joseph A. 2006. Slow growth, long-lived trees, and minimal disturbance characterize the dynamics of an ancient, montane forest in coastal British Columbia. Canadian Journal of Forest Research. 36: 2826-2838. 
83. Parminter, John. 1983. Fire history and fire ecology in the Prince Rupert Forest Region. In: Trowbridge, R. L.; Macadam, A., eds. Prescribed fire--forest soils: Symposium proceedings; 1982 March 2-3; Smithers, BC. Land Management Report, Number 16. Victoria, BC: Province of British Columbia, Ministry of Forests: 1-35. 
84. Peterson, David W.; Peterson, David L. 2001. Mountain hemlock growth responds to climatic variability at annual and decadal time scales. Ecology. 82(12): 3330-3345. 
85. Potkin, Michele. 1997. Fire history disturbance study of the Kenai Peninsula mountainous portion of the Chugach National Forest. Unpublished report on file with: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory, Missoula, MT. 75 p. 
86. Ross, Darrell W.; Daterman, Gary E.; Boughton, Jerry L.; Quigley, Thomas M. 2001. Forest health restoration in south-central Alaska: A problem analysis. Gen. Tech. Rep. PNW-GTR-523. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station. 38 p. 
87. Ryan, Michael G.; Vose, James M. 2012. Effects of climatic variability and change. In: Vose, James M.; Peterson, David L.; Patel-Weynand, Toral, eds. Effects of climatic variability and change on forest ecosystems: A comprehensive science synthesis for the U.S. forest sector. PNW-GTR-870. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station: 7-95. 
88. Saperstein, Lisa; Fay, Brett; O'Connor, Josh; Reed, Brad. 2014. Use and effectiveness of fuel treatments during the 2014 Funny River Fire, Alaska. Anchorage, AK: U.S. Department of the Interior, Fish and Wildlife Service, Branch of Fire Management. 39 p. [+ appendices]. 
89. Schulz, Bethany. 2003. Changes in downed and dead woody material following a spruce beetle outbreak on the Kenai Peninsula, Alaska. Res. Pap. PNW-RP-559. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station. 9 p. 
90. Shanley, Colin S.; Pyare, Sanjay; Goldstein, Michael I.; Alaback, Paul B.; Albert, David M.; Beier, Colin M.; Brinkman, Todd J.; Edwards, Rick T.; Hood, Eran; MacKinnon, Andy; McPhee, Megan V.; Patterson, Trista M.; Suring, Lowell H.; Tallmon, David A.; Wipfli, Mark S. 2015. Climate change implications in the northern coastal temperate rainforest of North America. Climatic Change. 130(2): 155-170. 
91. Sherriff, R. L.; Berg, E. E.; Miller, A. E. 2011. Climate variability and spruce beetle (Dendroctonus rufipennis) outbreaks in south-central and southwest Alaska. Ecology. 92(7): 1459-1470. 
92. Sommers, William T.; Coloff, Stanley G.; Conard, Susan G. 2011. Synthesis of knowledge: Fire history and climate change. Joint Fire Science Project No. 09-2-01-09. Boise, ID: Joint Fire Science Program. 190 p. [+ appendices]. 
93. Sullivan, Patrick F.; Mulvey, Robin L.; Brownlee, Annalis H.; Barrett, Tara M.; Pattison, Robert R. 2015. Warm summer nights and the growth decline of shore pine in southeast Alaska. Environmental Research Letters. 10(12): 1-10. 
94. Tillmann, Patricia; Glick, Patty. 2013. Climate change effects and adaptation approaches for terrestrial ecosystems, habitats, and species: A compilation of the scientific literature for the North Pacific Landscape Conservation Cooperative Region. Seattle, WA: National Wildlife Federation, Pacific Regional Center. 30 p. 
95. USDA Forest Service. 2010. Resurrection River landscape assessment. Anchorage, AK: U.S. Department of Agriculture, Forest Service, Chugach National Forest, Seward Ranger District. 142 p. [+ appendices]. 
96. USDA Forest Service. [n.d.]. Preliminary forest plant associations of the Stikine Area, Tongass National Forest. R10-TP-72. Portland, OR: U.S. Department of Agriculture, Forest Service, Alaska Region. 126 p. 
97. USDA. 2008. Trail River Landscape Assessment. Anchorage, AK: U.S. Department of Agriculture, Forest Service, Chugach National Forest, Seward Ranger District. 156 p. 
98. USDA. 2009. Quartz Creek Landscape Assessment. Anchorage, AK: U.S. Department of Agriculture, Forest Service, Chugach National Forest, Seward Ranger District. 154 p. 
99. USDA. 2014. Chugach National Forest: Fire Management Plan. Anchorage, AK: U.S. Department of Agriculture, Forest Service, Chugach National Forest. 88 p. 
100. Veblen, Thomas T.; Alaback, Paul B. 1996. A comparative review of forest dynamics and disturbance in the temperate rainforests of North and South America. In: Lawford, Richard G.; Alaback, Paul B.; Fuentes, Eduardo, eds. High latitude rainforests of the west coast of the Americas: Climate, hydrology, ecology and conservation. Ecological Studies, Vol. 116. New York: Springer: 173-213. 
101. Viereck, L. A.; Dyrness, C. T.; Batten, A. R.; Wenzlick, K. J. 1992. The Alaska vegetation classification. Gen. Tech. Rep. PNW-GTR-286. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station. 278 p. 
102. Vose, James M.; Peterson, David L.; Patel-Weynand, Toral, eds. 2012. Effects of climatic variability and change on forest ecosystems: A comprehensive science synthesis for the U.S. forest sector. PNW-GTR-870. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station. 265 p. 
103. Wang, Tongli; Campbell, Elizabeth M.; O'Neill, Gregory A.; Aitken, Sally N. 2012. Projecting future distributions of ecosystem climate niches: Uncertainties and management applications. Forest Ecology and Management. 279: 128-140. 
104. Weaver, Harold. 1974. Effects of fire on temperate forests: Western United States. In: Kozlowski, T. T.; Ahlgren, C. E., eds. Fire and ecosystems. New York: Academic Press: 279-319. 
105. Westerling, A. L.; Hidalgo, H. G.; Cayan, D. R.; Swetnam, T. W. 2006. Warming and earlier spring increase western U.S. forest wildfire activity. Science. 313(5789): 940-943. 
106. Westerling, Anthony LeRoy. 2016. Increasing western US forest wildfire activity: Sensitivity to changes in the timing of spring. Philosophical Transactions of the Royal Society B. 371: 20150178. 
107. Whitlock, Cathy; Higuera, Philip E.; McWethy, David B.; Briles, Christy E. 2010. Paleoecological perspectives on fire ecology: Revisiting the fire-regime concept. The Open Ecology Journal. 3: 6-23. 
108. Whitlock, Cathy; Marlon, Jennifer; Briles, Christy; Brunelle, Andrea; Long, Colin; Bartlein, Patrick. 2008. Long-term relations among fire, fuel, and climate in the north-western US based on lake-sediment studies. International Journal of Wildland Fire. 17(1): 72-83. 
109. Wildland Fire Leadership Council (WFLC). 2017. 2005 Alaska: Irish Channel. In: Monitoring Trends in Burn Severity (MTBS), [Online]. Sioux Falls, SD: U.S. Department of the Interior, U.S. Geological Survey, National Center for Earth Resources Observation and Science; Salt Lake City, UT: U.S. Department of Agriculture, Forest Service, Remote Sensing Applications Center (Producers). map, colored. Available: http://fsgeodata.net/MTBS_Uploads/data/2005/maps/ak6040215016520050706_map.pdf [2017, May 17]. 
110. Wolken, Jane M.; Hollingsworth, Teresa N.; Rupp, T. Scott; Chapin, F. Stuart, III; Trainor, Sarah F.; Barrett, Tara M.; Sullivan, Patrick F.; McGuire, A. David; Euskirchen, Eugenie S.; Hennon, Paul E.; Beever, Erik A.; Conn, Jeff S.; Crone, Lisa K.; D'Amore, David; Fresco, Nancy; Hanley, Thomas A.; Kielland, Knut; Kruse, James J.; Patterson, Trista; Schuur, Edward A. G.; Verbyla, David L.; Yarie, John. 2011. Evidence and implications of recent and projected climate change in Alaska's forest ecosystems. Ecosphere. 2(11): 1-35.