INTRODUCTION

We are experiencing unprecedented global change and these changes will accelerate in the coming decades. Globally, forests are subject to many drivers of anthropogenic change (McIntyre et al. 2015, Franklin et al. 2016, Balch et al. 2017) that may reduce their capacity to deliver expected levels of ecosystem services (Asner et al. 2015). Because these anthropogenic drivers interact in surprising ways and their future magnitude is also uncertain, the future of forests are highly uncertain (Millar et al. 2007, Lindner et al. 2014, Luce et al. 2016, Wang et al. 2016, Boulanger et al. 2018). Natural resource managers must account for uncertainty when making decisions (Nichols et al. 2011, Lindner et al. 2014) but climate change magnifies uncertainty and can be a barrier to management action (Polasky et al. 2011, Adams 2013, Scheller and Parajuli 2018).

Despite decades of progress, there remain many sources of uncertainty that constrain our capacity to understand and forecast future forests, including parameter uncertainty (the data that feed into the models), model uncertainty (reflecting our overall understanding of how the system operates), and inherent uncertainty (unresolvable uncertainty; Higgins et al. 2003, Morin and Thuiller 2009, Reyer et al. 2016). Nevertheless, within the domains of forest ecology and management there are opportunities to quantify and evaluate the sources of uncertainty through the use of scenario planning. Scenarios in combination with forecasting models are a common approach to quantifying uncertainty by attempting to identify outcomes from a variety of inputs, states, and actions (Peterson et al. 2003). By identifying the plausible or potential bounds of the primary drivers of system change, uncertainty due to each (or neither) can be estimated (Polasky et al. 2011). In the case of forests and climate change, the climate forecasts themselves are a substantial source of uncertainty as they reflect a range of social, economic, and technologic variables themselves (Van Vuuren et al. 2011). Model uncertainty can also be substantial (Petter et al. 2020). Forest disturbances contribute considerable uncertainty to our understanding of forest futures (Hicke et al. 2006, Millar et al. 2007, Littell et al. 2010, Scheller et al. 2011, Anderegg et al. 2015, Seidl et al. 2016, Bognounou et al. 2017, Coen et al. 2018, Stephens et al. 2018). On the contrary, negative feedbacks among disturbances, in combination with ecological memory, may reduce uncertainty. Disturbances do not necessarily compound; there can be negative feedbacks among disturbances that tend to reduce the magnitude (i.e., tree mortality) of subsequent disturbances such as with insects and high severity fire where fire risk is reduced after needle drop (Meigs et al. 2016). Forested landscapes have long ecological memories (Sun et al. 2013, Johnstone et al. 2016)—consisting of the biotic elements, their age, and spatial distribution—that limit their future behavior (Rhemtulla et al. 2009, Loudermilk et al. 2013, Perring et al. 2016).

We provide an analytical approach that quantifies the key drivers of forest change, climate, disturbance, and forest management using scenarios paired with simulation modeling to forecast and quantify future uncertainties. We focused on uncertainty generated by climate change (including associated changes to wildfire and insect outbreaks), forest management (including thinning, prescribed fire, and fire suppression), and other sources using a fully factorial experimental design (similar to Seidl and Lexer 2013); we conducted an analysis of the variance generated by uncertainty (e.g., Seidl and Lexer 2013).

Our combination of scenarios and modeling enabled forecasting of an array of potential futures as dictated by climate and management. The information generated can subsequently inform long-term strategic management planning (Sturtevant et al. 2007), which can, in turn, answer this question: Can management continue to guide outcomes on this landscape in spite of changing climate?

We assessed three metrics that we forecasted through time that reflect present day management objectives, including the restoration of a more natural fire regime dominated by low-intensity fire; the reduction of high-risk, high-intensity wildfires; and the maintenance of potential C sequestration. We address this issue within the Lake Tahoe Basin (LTB), which is well-suited for landscape modeling because (1) the forests are mostly in public ownership, which allows for a unified approach to forest and fire management, (2) wildfires there have been historically confined within its steep basin boundaries, and (3) the climate is expected to warm but will remain characterized by winter snow and dry summers.

METHODS

Our analysis was a component of a larger effort to examine social and ecological resilience in the Lake Tahoe West (LTW) study area under alternative management strategies as part of a collaborative landscape restoration effort. This larger effort is the subject of various articles in this special feature, and more information about that project is available here: https://www.nationalforests.org/regional-programs/california-program/laketahoewest. This core of this effort involved modeling ecological change in the forests of LTW over time.

We forecast climate and management interactions using the LANDIS-II simulation modeling framework; LANDIS-II simulates management and climate forcings to quantify uncertainty (Scheller et al. 2007). We simulated five management strategies varying in overall intensity and specific management activities deployed (Table 1) and eight climate projections in a fully factorial design. We selected our 40 scenarios to reflect the full range of plausible climate and management projections that were then replicated three times, therefore encapsulating most of the uncertainty from both (while recognizing that there is potential uncertainty beyond what is currently regarded as plausible). The behavior of the dominant disturbances, wildfire and insects, were dependent upon both climate and management.

Metrics

A variety of metrics were used to evaluate social and ecological resilience as part of the larger LTW research effort; these metrics were selected and constructed with input from a group of stakeholders as well as the research team (see Abelson et al. 2022). For this analysis, we examined three metrics that reflect important landscape dynamics relevant to forest management: (1) area burned at high intensity, (2) total area burned by wildfire; and (3) landscape carbon density. Note that our fire module was set up to represent fire intensity, specifically to approximate different classes of flame lengths and crown fire, but it serves also a measure of fire severity (Scheller et al. 2019). Although many dimensions of fire regime are important to consider, area burned at high-intensity may be more informative than percent area burned at high-intensity or severity. Area burned at high severity has been widely reported as a measure of departure in fire history studies (e.g., Safford and Stevens 2017), but that measure has potential to distort understandings of landscape change. For example, if a small area burns mostly at high-intensity in one year, followed by a large area mostly at low-intensity in another year, that outcome may be very consistent with a resilient landscape condition. However, if those relationships are switched, i.e., a small area at low-intensity followed by a large area at high-intensity, that outcome is indicative of departure from reference conditions. However, a metric based upon average percent high-severity does not effectively distinguish between those two outcomes.

Total area burned by wildfire is an important process variable, although in itself is not generally indicative of ecological resilience for fire-adaptive landscapes. Area burned by wildfire is important for understanding larger social and ecological processes, such as costs of suppression, smoke emissions, and restoration of functional fire, which were important indicators to stakeholders (Abelson et al. 2022). Carbon storage, both within the forested landscape and overall system (including wood products) represents another value important to some stakeholders (especially because funding programs have been established to support management that stores carbon). For this analysis, we considered carbon density, or mass per unit area (Mg ha^-1) within the landscape as a measure of this social-ecological value. We focus on these three indicators to consider resilience in fire-adapted ecosystems under climate change. Much recent research has suggested potential to both restore fire and secure carbon stores (e.g., Krofcheck et al. 2017, Loudermilk et al. 2017, Liang et al. 2018). However, some research has also suggested that carbon stocks in contemporary forests of the Sierra Nevada have exceeded historical references in some areas with a long-history of fire suppression (Harris et al. 2019), which suggests that reductions in carbon may be consistent with ecological restoration. Therefore, this analysis sheds light on important trade-offs when considering alternative management strategies in light of climate change.

Study area

The Lake Tahoe Basin (LTB) consists of 70,000 ha of predominantly forested land around Lake Tahoe in the Sierra Nevada of California and Nevada, USA (Fig. 1). The majority of the LTB forested area is under the management of the USDA Forest Service. The climate is a Mediterranean-influenced continental climate with warm to hot summers and most of the precipitation falling as snow in the winter. Annual precipitation averages a little over 1000 mm per year (ranging from 400 mm to 2000 mm; Fig. A1.1), with a mean minimum monthly temperature average around -7 °C and mean maximum monthly temperature around 24 °C (PRISM 30-year averages; Fig. A1.2). Most forests are mixed conifer, with the composition varying across topography and soils. At higher elevations, red fir (Abies magnifica A. Murr.) dominates, while in lower elevations Jeffrey pine (Pinus jeffreyi Grev. & Balf.) and white fir (Abies concolor Gord. & Glend.) dominate. Sugar pine (Pinus lambertiana Dougl.) and incense-cedar (Calocedrus decurrens (Torr.) Florin) are important components of the lower elevation forests. Shrub fields exist throughout elevation classes, featuring species primarily from the Ceanothus and Arctostaphylos genera.

Much of the basin was heavily logged beginning in the late 19th century to support mining operations in the greater area (Taylor 2004). Following the subsequent recovery of the forests and the institution of fire suppression policies, the present-day forests have become denser and feature more shade-tolerant tree species at the expense of less shade-tolerant pines (Barbour et al. 2002). Loudermilk et al. (2013) project that this trend will continue over the next 80 years. Wildfires were much more frequent prior to Euro-American settlement, with small fires happening nearly every year in some watersheds, while larger fires occurred once every 35 years (Taylor and Beaty 2005). Several species of bark beetles are also present in the Basin and have caused mortality across large areas of forest (Scheller et al. 2018).

Forest and disturbance modeling

We chose the LANDIS-II simulation framework because it simulates forest succession, disturbance, and management over long time periods and wide spatial extents (Scheller et al. 2007). In LANDIS-II trees and shrubs species are modeled individually as species-age cohorts, each species has its own life history attributes (e.g. shade tolerance, fire tolerance, dispersal ability, etc.), and multiple cohorts can occupy the same space. This allows species to respond uniquely to the multiple and interactive drivers (Scheller et al. 2007). Moreover, each species has its own range of temperature and water optimums, and so each responds to the future climate projections differently. Cohort establishment, likewise, was dependent on climate conditions, and it was assumed that there would only be natural regeneration on this landscape. Species parameters are detailed in Loudermilk et al. (2013), Kretchun et al. (2016), and Scheller et al. (2018). Initial aboveground biomass results were validated against Wilson et al. (2013; see Appendix 1 for supplemental methods and Fig. A1.4).

The ignition, spread, and intensity of fires (both wild and prescribed) were modeled using the Social-Climate Related Pyrogenic Processes (SCRPPLE v. 2.1) extension (Scheller et al. 2019). Simulated fire regimes are sensitive to climate; recent wildfires (2000–2016) were used to parameterize fire spread and size. Five fire experts working in the region provided their estimates of the mortality of three fire intensities for varying species and age combinations.

Three beetle species—Jeffrey pine beetle (Dendroctonus jeffrey), mountain pine beetle (Dendroctonus ponderosae), and fir engraver beetle (Scolytus ventralis)—that cause the majority of insect mortality within the LTB, as well as white pine blister rust (Cronartium ribicola), were simulated using a modified version of the Biological Disturbance Agent (BDA v.2.0.1) extension (Sturtevant et al. 2004); the modification triggers outbreaks following climate water deficit (CWD) and minimum winter temperature thresholds. The extension requires insect-specific resource requirements and assigns a species-specific vulnerability that varies by age. Mortality at an outbreak site is determined by tree species’ age and host susceptibility probabilities based from empirical field studies (Egan et al. 2010, 2016) and expert opinion. The parameters for insect spread and their resultant mortality are outlined in Kretchun et al. (2016). Additionally, results from the Insect and Disease Detection Survey (1993–2017) were used to validate the model results under historical climate conditions (see Fig. A1.5). However, there were challenges associated with using a climate threshold as a trigger approach as it ignores the brood mechanics and so does not capture the epidemic “wave” pattern of Egan et al. (2016). As such, the model underestimates peaks and overestimates troughs; instances where population dynamics override climate controls. All model parameters, and the model and extension versions used, are available on GitHub at: https://github.com/LANDIS-II-Foundation/Project-Lake-Tahoe-2017/.

Management modeling

We developed five scenarios that represent unique approaches to achieving multiple management objectives: restore a low-intensity fire regime; reduce the risk of high-intensity fires; and maintain carbon sequestration. These scenarios were co-developed with managers representing multiple agencies operating within LTB along with input from stakeholder groups operating in the region. For details of area treated annually and treatment frequency for each scenario, see Table 1. Scenario 1 features no fuels management paired with a high fire suppression. Scenario 2 focuses on reducing wildfire hazard in wildland-urban interface (WUI) area (1.5 miles from urban development) through hand or mechanical thinning (based upon accessibility) along with high effort fire suppression; it was closest to the current, business-as-usual strategy because understory prescribed burning has been rather limited. Scenario 3 increases the intensity and extent of vegetation thinning treatments. This scenario focuses on hand and mechanical treatments in the WUI and general forest, with hand treatments occurring in the wilderness as well. Scenario 4, the fire-focused strategy, uses prescribed and managed natural ignitions, along with some limited thinning in the WUI (akin to Scenario 2) to reduce fuels and restore forest structure. Prescribed fire was constrained to be low-intensity fire only, based upon guidance from managers regarding their intent. Scenario 5 was similar to Scenario 4, but with higher levels of prescribed burning. In Scenario 4 and Scenario 5, natural ignitions were not suppressed in management zones outside of the WUI. The amount of area treated under the five scenarios ranges from 0% to 11% of the landscape annually. The amount removed by thinning treatments were based on recent treatments within the Basin, and moreover followed the same approach of a thin-from-below up to a set diameter size class (27 cm dbh for hand thinning, 61 cm dbh for mechanical thinning) and slope restrictions (< 30%) for mechanical operations.

Climate modeling

In keeping with the 4th California Climate Assessment, future climate projections were derived from four Global Change Models (GCM; CanESM2, CNRM5, HADGEM2, and MIROC5) under two different relative concentration pathways (RCP) (4.5: which is an “optimistic” scenario of emissions; and 8.5: which represents a “business-as-usual” uncontrolled emissions scenario) using the localized constructed analogs downscaling methodology (Pierce et al. 2014). Climate futures ranged substantially with respect to precipitation: some projecting an increase of around 30% more annual precipitation (CanESM2 8.5), others projecting an almost equivalent decrease (MIROC5 8.5). With the CanESM2 8.5 projection, summers were projected to see an increase in summertime precipitation. Under the MIROC5 GCM, the area is expected to see increasing frequency and persistence of summertime droughts (Fig. A1.3).

Analysis

In order to differentiate between management and climate sources of uncertainty, an analysis of variance was performed using climate and management scenario as group factor variables for every time step of the model run and for our three metrics (landscape carbon density; area burned at high-intensity, and total area burned by wildfire). This analysis was repeated using decadal averages of each metric to reduce temporal autocorrelation associated with persistent climate events like multi-year droughts. We also examined a climate by management interaction effect although doing so produced too few degrees of freedom at the annual or decadal scale. The analysis was performed using the “car” package (3.0) in R (3.5.2). The reported explained variance is in terms of the sum of squares (SS), which can be apportioned into treatment effect and error. Error represents other sources of variation not explained by climate or management and results from stochastic model behavior not specifically related to climate or management, e.g., randomized ignition locations and seed dispersal.

RESULTS

Total area burned by wildfire

Climate was the main driver for total area burned. Despite the different approaches taken with respect to management practices among the different scenarios, there was little difference in the total area of wildfire (Fig. 2B).

Area burned at high intensity

The area of high intensity (> 8’ flame length) fire was most limited under the intensive third scenario (Fig. 2C). The influence of climate on high-intensity area burned was closely tied to the precipitation values for each climate projection (Fig. A1.1). As a result of increasing summertime precipitation, the CanESM2 8.5 climate projection resulted in the least of amount of high-intensity fire. The persistent droughts toward the end of the century forecast under the MIROC5 RCP 4.5 and 8.5 climate projections resulted in the largest high-intensity area burned by the end of the century (approximately 2-3 times higher than the CanESM2 RCP 8.5 projection; Fig. 2C).

Landscape carbon density

Our simulations project that forest carbon would increase through time as this forest recovers from historical logging. There was not a substantial amount of variation among the climate projections until the end of the century, when there was the greatest divergence among the different climates (Fig. 2A). The intensive mechanical treatment scenario (Scenario 3, subjected to the most thinning, produced an initial decline in C density (Fig. 2). However, this was offset over the century as the thinned stands were less likely to experience high-severity fire, resulting in a higher rate of carbon sequestration. The high fire use scenario (Scenario 5) had the lowest C density at the end of the century, in part due to greater removal of C in surface fuels, dead wood, and standing biomass from long-term use of prescribed fire (as compared to hand or mechanical thinning treatments used in Scenario 3).

Variance decomposition

Group predictors (climate and management scenario) explained a range of variation in landscape C density, total area burned, and area burned at high intensity, ranging from 11 to 98% of variance (adjusted R-squared) in a given year (Figs. 3, 4, and 5). Management strategy and climate projection explained substantially more of the variation in landscape C density (and were always significant; Fig. 5) as compared to total area burned or high-intensity area burned (Figs. 3 and 4, respectively). This difference highlights the uncertainty of fire generally and indicates that there are other sources of uncertainty, including the stochastic numbers of fires and their locations.

For total area burned, management explained less variance than with high-intensity fire at annual and decadal time scales. Climate was the main driver of total area burned, while management was the main driver of area burned at high-intensity (Fig. 3). For high-intensity fire, climate explained more variance than management although the error term—functionally the stochasticity of disturbance within the model—accounted for much of the total variance (Fig. 4A). When aggregated to a decadal time step, management explained substantially more variance in the area of high severity fire (see Fig. 4B). At the decadal scale, large fluctuations in climate are averaged out (wet years can follow dry years, except during periods with multi-year droughts forecast), while the area treated was deterministic (i.e., determined by the management scenarios).

DISCUSSION

Uncertainty regarding future forest conditions has important implications for derived social values, including ecosystem services (Hou et al. 2013, Hamel and Bryant 2017). From carbon storage to wildlife habitat, forest benefits are dependent on the interactions of climate, disturbance, and management. When those benefits drive the local economy, such as recreation in the Lake Tahoe Basin, quantifying the contribution of individual drivers improves decision making regarding the forest and the benefits in question, which is the focus on another manuscript in this special feature (Abelson et al. 2022). We focused on the primary sources of uncertainty—climate and management and interactions with fire—in our simulations of forest landscape change in order to assess the ability of management generally to shape future forest conditions. Our results suggest areas where management can have the greatest influence (high-intensity fire ~ landscape C density > total area burned). Although this landscape is unique in its centralized ownership, which may limit the broader applicability of this study, there is now greater movement toward whole landscape planning for the State of California.

The amount of area treated was more important in reducing area of high-intensity fire than the type of treatment used: as the area treated increased, the area burned by high-intensity fire declined because a fire would have a larger likelihood of intersecting a treated area. This is evident from the order of the results in Figure 2C and how they align with the number of hectares treated in each management scenario (i.e., 1, 2, 4, 3, 5). However, the variance explained by management waned as climate change uncertainty increased over the century, implying that the effectiveness of management may decline after 40 years. Management efforts within the LTB may need to increase substantially through time, more than managers considered when co-designing the management strategies tested. In the near term, more aggressive initial treatment may delay widespread mortality, and in the long term, promote the transition to a more drought-tolerant species mix (Elkin et al. 2015). In general, our results suggest that to increase the capacity of LTB forests to remain forests, management would need to cover a greater proportion of the landscape in a shorter period of time (Drever et al. 2006). Without this increased level of investment and activity, the rate of change brought by management actions may not keep the forest within a desirable condition (Johnstone et al. 2016).

Fuel treatments (hand and mechanical thinning) locally reduce fire intensity and rate of spread. The extent to which this holds true at the landscape-scale is debated (Campbell et al. 2012, Restaino and Peterson 2013), because in low-ignition environments there is only a small chance that fire will intersect with a treatment. At the same time, climate change and associated higher temperatures will reduce fuel moisture and will generate larger and more intense fires (Westerling and Bryant 2008). Our results suggest that management reduces fire intensity at landscape scales when the accumulation of treated area is large enough that there is a high chance a wildfire will intersect a treatment. This finding is consistent with other studies in the region that have found that both area treated and incidence of wildfire would need to increase from recent historical levels for that intersection to take place (Chiono et al. 2017, Krofcheck et al. 2017), and that there may be thresholds of area treated above which that have even greater effectiveness in reducing the risk of high severity fire (Stevens et al. 2016). The variance in area of high-intensity fire and total area burned explained by management declined over the century, which reflects the warming and drying trend of climate change.

In addition to management and climate uncertainty, we estimated the error term or unexplained uncertainty, which is the variance unexplained by either management or climate alone. This variance is not the same as error propagation, which is the combination of all the uncertainty of all the variables used in the development of these results (Morgan et al. 1990). In our modeling framework, unexplained variance of landscape C density may include a fire-by-climate effect: as the climate warms, the number of fires increases, but probabilistically. A given warm and dry day may or may not produce multiple wildfires. Therefore, the “inherent” variance (Higgins et al. 2003) increases over time in parallel with climate for our estimates of landscape C density. In contrast, the inherent variance remained relatively constant for total and high-intensity area burned because maximum area burned for each fire was a linear function of climate (Scheller et al. 2019).

Although there is and always will be uncertainty about the future, management has an important role to play in shaping the future forest conditions. Management actions can shape landscape conditions in spite of climate uncertainty up to a certain point. Areas, or metrics, where management is less effective suggests the need for new thinking about the kind (planned vs. reactionary), intensity, rate, or placement of treatments. Given the recent extremity of the climate conditions across the western U.S. only highlights the need for treatment and the need for new thinking.

LITERATURE CITED

Abelson, E. S., K. M. Reynolds, A. M. White, J. W. Long, C. Maxwell and P. N. Manley. 2022. Evaluating pathways to social and ecological landscape resilience. Ecology and Society, in press.

Adams, M. A. 2013. Mega-fires, tipping points and ecosystem services: managing forests and woodlands in an uncertain future. Forest Ecology and Management 294:250-261. https://doi.org/10.1016/j.foreco.2012.11.039

Anderegg, W. R. L., J. A. Hicke, R. A. Fisher, C. D. Allen, J. Aukema, B. Bentz, S. Hood, J. Lichstein, A. K. Macalady, N. McDowell, Y. Pan, K. Raffa, A. Sala, J. D. Shaw, N. L. Stephenson, C. Tague, and M. Zeppel. 2015. Tree mortality from drought, insects, and their interactions in a changing climate. New Phytologist 208(3):674-683. https://doi.org/10.1111/nph.13477

Asner, G. P., P. Brodrick, C. B. Anderson, N. Vaughn, D. E. Knapp, and R. E. Martin. 2015. Progressive forest canopy water loss during the 2012-2015 California drought. Proceedings of the National Academy of Sciences 113(2):E249-E255. https://doi.org/10.1073/pnas.1523397113

Balch, J. K., B. A. Bradley, J. T. Abatzoglou, R. Nagy, E. J. Fusco, and A. L. Mahood. 2017. Human-started wildfires expand the fire niche across the United States. Proceedings of the National Academy of Sciences 114(11):2946-2951. https://doi.org/10.1073/pnas.1617394114

Barbour, M., E. Kelley, P. Maloney, D. Rizzo, E. Royce, and J. Fites‐Kaufmann. 2002. Present and past old‐growth forests of the Lake Tahoe Basin, Sierra Nevada, US. Journal of Vegetation Science 13(4):461-472. https://doi.org/10.1111/j.1654-1103.2002.tb02073.x

Bognounou, F., L. De Grandpré, D. Pureswaran, and D. Kneeshaw. 2017. Temporal variation in plant neighborhood effects on the defoliation of primary and secondary hosts by an insect pest. Ecosphere 8(3):e01759. https://doi.org/10.1002/ecs2.1759

Boulanger, Y., M.-A. Parisien, and X. Wang. 2018. Model-specification uncertainty in future area burned by wildfires in Canada. International Journal of Wildland Fire 27(3):164-175. https://doi.org/10.1071/WF17123

Campbell, J. L., M. E. Harmon, and S. Mitchell. 2012. Can fuel‐reduction treatments really increase forest carbon storage in the western US by reducing future fire emissions? Frontiers in Ecology and the Environment 10(2):83-90. https://doi.org/10.1890/110057

Chiono, L. A., D. L. Fry, B. M. Collins, A. H. Chatfield, and S. L. Stephens. 2017. Landscape-scale fuel treatment and wildfire impacts on carbon stocks and fire hazard in California spotted owl habitat. Ecosphere 8(1):e01648. https://doi.org/10.1002/ecs2.1648

Coen, J. L., E. N. Stavros, and J. A. Fites-Kaufman. 2018. Deconstructing the King megafire. Ecological Applications 28(6):1565-1580. https://doi.org/10.1002/eap.1752

Development and structure of the Canadian Forest Fire Behavior Prediction System. 1992. Forestry Canada. Information Report ST-X-3. Forestry Canada, Headquarters, Fire Danger Group and Science and Sustainable Development Directorate, Ottawa, Ontario, Canada.

Drever, C. R., G. Peterson, C. Messier, Y. Bergeron, and M. Flannigan. 2006. Can forest management based on natural disturbances maintain ecological resilience? Canadian Journal of Forest Research 36(9):2285-2299. https://doi.org/10.1139/x06-132

Egan, J. M., W. R. Jacobi, J. F. Negron, S. L. Smith, and D. Cluck. 2010. Forest thinning and subsequent bark beetle-caused mortality in Northeastern California. Forest Ecology and Management 260(10):1832-1842. https://doi.org/10.1016/j.foreco.2010.08.030

Egan, J. M., J. M. Sloughter, T. Cardoso, P. Trainor, K. Wu, H. Safford, and D. Fournier. 2016. Multi-temporal ecological analysis of Jeffrey pine beetle outbreak dynamics within the Lake Tahoe Basin. Population Ecology 58(3):441-462. https://doi.org/10.1007/s10144-016-0545-2

Elkin, C., A. Giuggiola, A. Rigling, and H. Bugmann. 2015. Short- and long-term efficacy of forest thinning to mitigate drought impacts in mountain forests in the European Alps. Ecological Applications 25(4):1083-1098. https://doi.org/10.1890/14-0690.1

Franklin, J., J. M. Serra-Diaz, A. D. Syphard, and H. Regan. 2016. Global change and terrestrial plant community dynamics. Proceedings of the National Academy of Sciences 113(14):3725-3734. https://doi.org/10.1073/pnas.1519911113

Hamel, P., and B. P. Bryant. 2017. Uncertainty assessment in ecosystem services analyses: seven challenges and practical responses. Ecosystem Services 24:1-15. https://doi.org/10.1016/j.ecoser.2016.12.008

Harris, L. B. A. E. Scholl, A. B. Young, B. L. Estes, and A. Taylor. 2019. Spatial and temporal dynamics of 20th century carbon storage and emissions after wildfire in an old-growth forest landscape. Forest Ecology and Management 449:117461. https://doi.org/10.1016/j.foreco.2019.117461

Hicke, J. A., J. Logan, J. Powell, and D. Ojima. 2006. Changing temperatures influence suitability for modeled mountain pine beetle (Dendroctonus ponderosae) outbreaks in the western United States: climate change and insect outbreaks. Journal of Geophysical Research: Biogeosciences 111(G2). https://doi.org/10.1029/2005JG000101

Higgins, S. I., J. S. Clark, R. Nathan, T. Hovestadt, F. Schurr, J. Fragoso, M. R. Aguiar, E. Ribbens, and S. Lavorel. 2003. Forecasting plant migration rates: managing uncertainty for risk assessment. Journal of Ecology 91(3):341-347. https://doi.org/10.1046/j.1365-2745.2003.00781.x

Hou, Y., B. Burkhard, and F. Müller. 2013. Uncertainties in landscape analysis and ecosystem service assessment. Journal of Environmental Management 127:S117-S131. https://doi.org/10.1016/j.jenvman.2012.12.002

Johnstone, J. F., C. D. Allen, J. Franklin, L. E. Frelich, B. J. Harvey, P. E. Higuera, M. C. Mack, R. K. Meentemeyer, M. R. Metz, G. L. Perry, T. Schoennagel, and M. G. Turner. 2016. Changing disturbance regimes, ecological memory, and forest resilience. Frontiers in Ecology and Environment 14:369-378. https://doi.org/10.1002/fee.1311

Kretchun, A. M., E. L. Loudermilk, R. M. Scheller, M. D. Hurteau, and S. Belmecheri. 2016. Climate and bark beetle effects on forest productivity — linking dendroecology with forest landscape modeling. Canadian Journal Forestry Research 46:1026-1034. https://doi.org/10.1139/cjfr-2016-0103

Krofcheck, D. J., M. D. Hurteau, R. M. Scheller, and E. Loudermilk. 2017. Restoring surface fire stabilizes forest carbon under extreme fire weather in the Sierra Nevada. Ecosphere 8(1):e01663. https://doi.org/10.1002/ecs2.1663

Liang, S., M. D. Hurteau, and A. L. Westerling. 2018. Large‐scale restoration increases carbon stability under projected climate and wildfire regimes. Frontiers in Ecology and the Environment 16(4):207-212. https://doi.org/10.1002/fee.1791

Lindner, M., J. Fitzgerald, N. E. Zimmermann, C. Reyer, S. Delzon, E. van der Maaten, M. Schelhaas, P. Lasch, J. Eggers, M. van der Maaten-Theunissen, F. Suckow, A. Psomas, B. Poulter, and M. Hanewinkel. 2014. Climate change and European forests: what do we know, what are the uncertainties, and what are the implications for forest management? Journal of Environmental Management 146:69-83. https://doi.org/10.1016/j.jenvman.2014.07.030

Littell, J. S., E. Oneil, D. McKenzie, J. Hicke, J. Lutz, R. Norheim, and M. Elsner. 2010. Forest ecosystems, disturbance, and climatic change in Washington State, USA. Climatic Change 102(1-2):129-158. https://doi.org/10.1007/s10584-010-9858-x

Loudermilk, E. L., R. M. Scheller, P. J. Weisberg, and A. Kretchun. 2017. Bending the carbon curve: fire management for carbon resilience under climate change. Landscape Ecology 32:1461-1472. https://doi.org/10.1007/s10980-016-0447-x

Loudermilk, E., R. M. Scheller, P. J. Weisberg, J. Yang, T. E. Dilts, S. L. Karam, and C. Skinner. 2013. Carbon dynamics in the future forest: the importance of long-term successional legacy and climate-fire interactions. Global Change Biology 19(11):3502-3515. https://doi.org/10.1111/gcb.12310

Luce, C. H., J. Vose, N. Pederson, J. Campbell, C. Millar, P. Kormos, and R. Woods. 2016. Contributing factors for drought in United States forest ecosystems under projected future climates and their uncertainty. Forest Ecology and Management 380:299-308. https://doi.org/10.1016/j.foreco.2016.05.020

McIntyre, P. J., J. Thorne, C. Dolanc, A. L. Flint, L. E. Flint, M. Kelly, and D. Ackerly. 2015. Twentieth-century shifts in forest structure in California: denser forests, smaller trees, and increased dominance of oaks. Proceedings of the National Academy of Sciences 112(5):1458-1463. https://doi.org/10.1073/pnas.1410186112

Meigs, G. W., H. Zald, J. L. Campbell, W. S. Keeton, and R. E. Kennedy. 2016. Do insect outbreaks reduce the severity of subsequent forest fires? Environmental Research Letters 11:045008. https://doi.org/10.1088/1748-9326/11/4/045008

Millar, C. I., N. Stephenson, and S. Stephens. 2007. Climate change and forests of the future: managing in the face of uncertainty. Ecological Applications 17(8):2145-2151. https://doi.org/10.1890/06-1715.1

Morgan, M. G., M. Henrion, and M. Small. 1990. Uncertainty: a guide to dealing with uncertainty in quantitative risk and policy analysis. Cambridge University Press, Cambridge, UK.

Morin, X., and W. Thuiller. 2009. Comparing niche- and process-based models to reduce prediction uncertainty in species range shifts under climate change. Ecology 90(5):1301-1313. https://doi.org/10.1890/08-0134.1

Nichols, J. D., M. Koneff, P. J. Heglund, M. G. Knutson, M. E. Seamans, J. E. Lyons, J. Morton, M. Jones, G. S. Boomer, and B. K. Williams. 2011. Climate change, uncertainty, and natural resource management. Journal of Wildlife Management 75(1):6-18. https://doi.org/10.1002/jwmg.33

Perring, M. P., P. De Frenne, L. Baeten, S. L. Maes, L. Depauw, H. Blondeel, M. M. Carón, and K. Verheyen. 2016. Global environmental change effects on ecosystems: the importance of land‐use legacies. Global Change Biology 22(4):1361-1371. https://doi.org/10.1111/gcb.13146

Peterson, G. D., G. S. Cumming, and S. R. Carpenter. 2003. Scenario planning: a tool for conservation in an uncertain world. Conservation Biology 17(2):358-366. https://doi.org/10.1046/j.1523-1739.2003.01491.x

Petter, G., P. Mairota, K. Albrich, P. Bebi, J. Brůna, H. Bugmann, A. Haffenden, R. M. Scheller, D. R. Schmatz, R. Seidl, et al. 2020. How robust are future projections of forest landscape dynamics? Insights from a systematic comparison of four forest landscape models. Environmental Modelling & Software 134:104844. https://doi.org/10.1016/j.envsoft.2020.104844

Pierce, D. W., D. R. Cayan, and B. L. Thrasher. 2014. Statistical downscaling using localized constructed analogs (LOCA). Journal of Hydrometeorology 15:2558-2585. https://doi.org/10.1175/JHM-D-14-0082.1

Polasky, S., S. Carpenter, C. Folke, and B. Keeler. 2011. Decision-making under great uncertainty: environmental management in an era of global change. Trends in Ecology & Evolution 26(8):398-404. https://doi.org/10.1016/j.tree.2011.04.007

Restaino, J. C., and D. Peterson. 2013. Wildfire and fuel treatment effects on forest carbon dynamics in the western United States. Forest Ecology and Management 303:46-60. https://doi.org/10.1016/j.foreco.2013.03.043

Reyer, C. P. O., M. Flechsig, P. Lasch-Born, and M. van Oijen. 2016. Integrating parameter uncertainty of a process-based model in assessments of climate change effects on forest productivity. Climatic Change 137(3-4):395-409. https://doi.org/10.1007/s10584-016-1694-1

Rhemtulla, J. M., D. J. Mladenoff, and M. K. Clayton. 2009. Legacies of historical land use on regional forest composition and structure in Wisconsin, USA (mid-1800s—1930s—2000s). Ecological Applications 19(4):1061-1078. https://doi.org/10.1890/08-1453.1

Safford, H. D., and J. T. Stevens. 2017. Natural range of variation for yellow pine and mixed-conifer forests in the Sierra Nevada, southern Cascades, and Modoc and Inyo National Forests, California, USA. Gen. Tech. Rep. PSW-GTR- 256. U.S. Forest Service, Pacific Southwest Research Station, Albany, California, USA. https://www.srs.fs.usda.gov/pubs/55393

Scheller, R. M., J. B. Domingo, B. R. Sturtevant, J. S. Williams, A. Rudy, E. J. Gustafson, and D. J. Mladenoff. 2007. Design, development, and application of LANDIS-II, a spatial landscape simulation model with flexible temporal and spatial resolution. Ecological Modelling 201:409-419. https://doi.org/10.1016/j.ecolmodel.2006.10.009

Scheller, R., A. Kretchun, T. J. Hawbaker, and P. Henne. 2019. A landscape model of variable social-ecological fire regimes. Ecological Modelling 401:85-93. https://doi.org/10.1016/j.ecolmodel.2019.03.022

Scheller, R. M., A. Kretchun, E. L. Loudermilk, M. D. Hurteau, P. J. Weisberg, and C. Skinner. 2018. Interactions among fuel management, species composition, bark beetles, and climate change and the potential effects on forests of the Lake Tahoe Basin. Ecosystems 21:643-656. https://doi.org/10.1007/s10021-017-0175-3

Scheller, R., and R. Parajuli. 2018. Forest management for climate change in New England and the Klamath ecoregions: motivations, practices, and barriers. Forests 9(10):626. https://doi.org/10.3390/f9100626

Scheller, R. M., W. D. Spencer, H. Rustigian-Romsos, A. Syphard, B. Ward, and J. Strittholt. 2011. Using stochastic simulation to evaluate competing risks of wildfires and fuels management on an isolated forest carnivore. Landscape Ecology 26(10):1491-1504. https://doi.org/10.1007/s10980-011-9663-6

Seidl, R. and M. J. Lexer, 2013. Forest management under climatic and social uncertainty: trade-offs between reducing climate change impacts and fostering adaptive capacity. Journal of Environmental Management 114:461-469. https://doi.org/10.1016/j.jenvman.2012.09.028

Seidl, R., T. A. Spies, D. L. Peterson, S. L. Stephens, and J. A. Hicke. 2016. Searching for resilience: addressing the impacts of changing disturbance regimes on forest ecosystem services. Journal of Applied Ecology 53(1):120-129. https://doi.org/10.1111/1365-2664.12511

Stephens, S. L., B. Collins, C. Fettig, M. Finney, C. Hoffman, E. Knapp, M. North, H. Safford, and R. B. Wayman. 2018. Drought, tree mortality, and wildfire in forests adapted to frequent fire. BioScience 68(2):77-88. https://doi.org/10.1093/biosci/bix146

Stevens, J. T., B. Collins, J. Long, M. North, S. Prichard, L. Tarnay, and A. White. 2016. Evaluating potential trade‐offs among fuel treatment strategies in mixed‐conifer forests of the Sierra Nevada. Ecosphere 7(9):e01445. https://doi.org/10.1002/ecs2.1445

Sturtevant, B. R., A. Fall, D. D. Kneeshaw, N. P. P. Simon, M. J. Papaik, K. Berninger, F. Doyon, D. G. Morgan, and C. Messier. 2007. A toolkit modeling approach for sustainable forest management planning: achieving balance between science and local needs. Ecology and Society 12(2):7. http://dx.doi.org/10.5751/ES-02102-120207

Sturtevant, B. R., E. Gustafson, W. Li, and H. He. 2004. Modeling biological disturbances in LANDIS: a module description and demonstration using spruce budworm. Ecological Modelling 180(1):153-174. https://doi.org/10.1016/j.ecolmodel.2004.01.021

Sun, Z., H. Ren, V. Schaefer, H. Lu, J. Wang, L. Li, and N. Liu. 2013. Quantifying ecological memory during forest succession: a case study from lower subtropical forest ecosystems in South China. Ecological Indicators 34:192-203. https://doi.org/10.1016/j.ecolind.2013.05.010

Taylor, A. H. 2004. Identifying forest reference conditions on early cut-over lands, Lake Tahoe Basin, USA. Ecological Applications 14:1903-1920. https://doi.org/10.1890/02-5257

Taylor, A. H. and R. M. Beaty. 2005. Climatic influences on fire regimes in the northern Sierra Nevada mountains, Lake Tahoe Basin, Nevada, USA. Journal of Biogeography 32(3):425-438. https://doi.org/10.1111/j.1365-2699.2004.01208.x

Van Vuuren, D. P., and J. Edmonds, M. Kainuma, K. Riahi, A. Thomson, K. Hibbard, G. Hurtt, T. Kram, V. Krey, J. F. Lamarque, et al. 2011. The representative concentration pathways: an overview. Climatic Change 109:5. https://doi.org/10.1007/s10584-011-0148-z

Wang, T., G. Wang, J. Innes, C. Nitschke, and H. Kang. 2016. Climatic niche models and their consensus projections for future climates for four major forest tree species in the Asia-Pacific region. Forest Ecology and Management 360:357-366. https://doi.org/10.1016/j.foreco.2015.08.004

Westerling, A. L., and B. P. Bryant. 2008. Climate change and wildfire in California. Climatic Change 87:231-249. https://doi.org/10.1007/s10584-007-9363-z

Wilson, B. T., C. W. Woodall, and D. M. Griffith. 2013. Imputing forest carbon stock estimates from inventory plots to a nationally continuous coverage. Carbon Balance Management 8:1. https://doi.org/10.1186/1750-0680-8-1