Compounding Crises: Moving Beyond the "Single Stressor" View of Forest Health

1. Introduction: The New Reality of Forest Disturbance

The global forest estate is currently navigating a period of unprecedented environmental transformation. For the better part of the twentieth century, the discipline of forest ecology operated under a paradigm of compartmentalization. Disturbance agents—the discrete events that disrupt ecosystem structure and release resources—were largely studied in isolation. Fire ecologists meticulously characterized burn severity and return intervals; entomologists mapped the life cycles of bark beetles and defoliators; pathologists studied the epidemiology of fungal diseases; and climatologists modeled drought as a meteorological anomaly. This reductionist approach was necessary to build a foundational understanding of how individual stressors function. However, the emerging reality of the Anthropocene is that these disturbances no longer respect the boundaries of scientific subdisciplines. They are colliding, overlapping, and amplifying one another in ways that defy historical precedent.

Recent comprehensive reviews, most notably the 2025 synthesis by Dudney, Edwards, Harvey, and Seidl published in the Annual Review of Ecology, Evolution, and Systematics, argue that the defining feature of modern forest dynamics is the interaction of disturbances.¹ As climate change alters the thermal and hydrological baseline of the planet, it is dissolving the temporal and spatial buffers that once kept these events separate. A drought is no longer just a period of water stress; it is a physiological pre-conditioning event that alters a tree’s defense chemistry against beetles and lowers its ignition threshold for fire. A windstorm is no longer just a structural reset; it is a catalyst that reorganizes fuel loads and microclimates, setting the stage for subsequent insect outbreaks or high-severity fires.

This report serves as a deep-dive investigation into this "unifying framework" of interacting disturbances. By synthesizing data from over 160 recent studies, we explore how climate change is modifying these interactions through two primary pathways: by shifting the characteristics of the disturbances themselves (intensity, frequency, seasonality) and by altering the strength and direction of the interactions between them.¹ We will traverse from the microscopic scale of xylem cavitation and resin duct failure to the landscape scale of disturbance networks and biome tipping points. The analysis reveals a troubling trend: while nature historically contained many "dampening" or antagonistic feedbacks that stabilized ecosystems, the current climate trajectory is overwhelmingly favoring "amplifying" or synergistic feedbacks.² These positive feedback loops threaten to push forests past critical thresholds, leading to ecosystem transitions where forests fail to recover and are replaced by shrublands, grasslands, or novel non-analog communities.

The urgency of this research cannot be overstated. As the frequency of "megadisturbances" increases—from the massive pyrocumulonimbus fires of the North American West to the desiccation of the Amazon rainforest—managers and policymakers are finding that traditional models, built on the assumption of independent events, are failing. To steward forests through the twenty-first century, we must learn to see the forest not as a collection of trees, but as a complex network of potential interactions waiting to be triggered.

2. A Unifying Framework for Disturbance Interactions

One of the persistent challenges in disturbance ecology has been a lack of a common language. When a fire burns a drought-stressed forest, is that a "compound event," a "linked disturbance," or a "cascade"? Without precise terminology, it is impossible to compare studies or build predictive models. The framework proposed by Dudney et al. (2025) resolves this by categorizing interactions into seven distinct classes based on their structural relationships and causal flow.² Understanding these classes is the first step toward unraveling the complexity of modern forest change.

2.1 The Seven Classes of Interaction

The seven classes represent a hierarchy of complexity, moving from simple linear cause-and-effect relationships to complex, self-reinforcing webs.

Class 1: Linked Disturbances

This is the most fundamental unit of interaction. A linked disturbance occurs when one disturbance event (Disturbance A) alters the likelihood, extent, or severity of a subsequent disturbance (Disturbance B).⁴ The connection is typically unidirectional and sequential. The classic example is the relationship between windthrow and wildfire. When a storm topples trees, it rearranges the fuel complex, moving biomass from the canopy to the forest floor. This does not cause a fire immediately, but it alters the potential for fire severity if an ignition occurs later.

Critically, linked disturbances can be separated by significant time lags. A defoliation event by the western spruce budworm might reduce a tree's carbon reserves today, creating a linked vulnerability to a bark beetle attack five or ten years later.⁶ Climate change amplifies these links by shortening the recovery time between events, meaning the "memory" of the first disturbance is still fresh when the second one strikes.

Class 2: Compound Disturbances

Compound disturbances are defined by their coincidence in space and time (or close succession) and, crucially, by their multiplicative effect on recovery.⁷ While linked disturbances focus on how one event changes the next event, compound disturbances focus on how the combination changes the ecosystem state.

The most potent examples of compound disturbances involve impacts on regeneration. For instance, if a wildfire burns a forest and is immediately followed by a severe drought during the critical seedling establishment phase, the forest may fail to regenerate entirely.⁹ Neither the fire (which the forest is adapted to) nor the drought (which seedlings might survive in unburned shade) would have caused a state shift on its own. It is the combination that exceeds the ecosystem's resilience. This phenomenon is increasingly responsible for "recruitment failure" in forests globally.

Class 3: Synergistic Effects

Synergy refers to the outcome of an interaction where the combined effect is greater than the sum of the individual effects.² This is often referred to in the literature as an "amplifying" effect. If Disturbance A causes 10% mortality and Disturbance B causes 10% mortality, a synergistic interaction might result in 50% mortality.

Climate change is overwhelmingly increasing the frequency of synergistic interactions.¹⁰ A prime mechanism for this is the interaction between heat and drought. Hotter droughts (global change-type droughts) deplete soil moisture while simultaneously increasing the atmosphere's demand for water (vapor pressure deficit), leading to hydraulic failure in trees that might have survived a cooler drought of the same precipitation deficit.

Class 4: Antagonistic Effects

Antagonistic interactions occur when one disturbance dampens or buffers the effects of another.² While less discussed in the context of climate catastrophe, these negative feedbacks are essential stabilizing forces in natural systems.

For example, a high-severity wildfire consumes available fuel. For a period of decades following the fire, the burned area may act as a fuel break, limiting the spread of subsequent fires. Similarly, a blowdown event might remove the specific large-diameter host trees required by a bark beetle, thereby halting the expansion of an outbreak in that immediate area.¹¹ However, researchers note that climate change may be eroding these buffers. Rapid vegetation regrowth (like invasive grasses) can shorten the fuel-break effect of fires, converting an antagonistic interaction into a synergistic one.

Class 5: Self-Feedbacks

This class describes a single disturbance type interacting with itself over time. In a stable system, self-feedbacks are often negative (stabilizing). A fire burns fuel, reducing the likelihood of another fire. An insect outbreak kills susceptible hosts, causing the population to crash due to starvation.¹²

However, climate change is increasingly inverting these loops, creating positive (destabilizing) self-feedbacks. The "grass-fire cycle" is the archetype of this inversion. A fire in a dry forest opens the canopy, allowing invasive annual grasses (like cheatgrass, Bromus tectorum) to invade. These grasses cure early in the season and ignite easily, increasing the fire frequency. More fire leads to more grass, and the cycle accelerates, locking the system into a non-forest state.¹³

Class 6: Coupled Feedbacks

Coupled feedbacks involve two different disturbance agents interacting in a bidirectional loop.¹⁴ This is distinct from a "linked" disturbance because the causality flows both ways.

The relationship between drought and fire in the Amazon is a critical coupled feedback. Drought increases fire frequency and intensity. Fire, in turn, kills trees and opens the canopy, which reduces local transpiration and increases the amount of solar radiation reaching the understory. This microclimatic change intensifies the local drought conditions, further increasing fire risk.¹⁵ These feedback loops are the engines of "critical transitions," capable of driving massive biomes toward alternative stable states.

Class 7: Disturbance Networks

The most complex class is the disturbance network. This framework recognizes that in many landscapes, disturbances do not occur in isolated pairs or chains but in a web of causality.³ A "cascade" implies a linear progression (A -> B -> C), but a network allows for non-linear connections, feedback loops, and multiple starting points.

For instance, a disturbance network might involve a tropical cyclone that causes windthrow (Node A). This windthrow facilitates a bark beetle outbreak (Node B) and simultaneously increases fuel loads for fire (Node C). The fire might then trigger a landslide (Node D) by removing root cohesion. Network analysis allows ecologists to identify "hub" disturbances—those that have the highest connectivity and potential to trigger system-wide collapse.¹⁴

2.2 Table 1: Summary of Disturbance Interaction Classes

Interaction Class	Definition	Ecological Example	Climate Change Implication
Linked	Disturbance A alters likelihood/severity of Disturbance B (sequential).	Windthrow -> increased fuel -> increased Fire severity.	Shorter intervals between events reduce recovery, amplifying linked effects.
Compound	A + B occur close in time; combined effect > sum of parts (recovery failure).	Fire + Post-fire Drought -> Seedling mortality -> Forest loss.	Increased frequency of both heatwaves and fires maximizes overlap probability.
Synergistic	Interaction amplifies the net impact or extent (Outcome-focused).	Drought + Bark Beetles -> Exponential tree mortality.	Nonlinear responses to warming drive "megadisturbances."
Antagonistic	Interaction dampens the net impact (Buffer).	Severe Fire -> Fuel removal -> Reduced Fire spread.	Climate-driven invasive species (grasses) may erode these natural buffers.
Self-Feedback	Disturbance alters its own future likelihood.	Fire -> Invasive Grass -> More Fire (Positive Feedback).	Shifts from negative (stabilizing) to positive (runaway) feedbacks.
Coupled Feedback	Bidirectional interaction between two agents.	Vegetation Loss <-> Regional Drying (Amazon).	Drives critical transitions (tipping points) to new biomes (e.g., Savanna).
Disturbance Network	Complex, multi-node web of interactions.	Drought -> Pathogen -> Fire -> Landslide.	High complexity makes prediction difficult; requires network modeling.

3. Physiological Mechanisms: The Engine of Vulnerability

To truly understand why disturbance interactions are shifting from dampening to amplifying, we must descend from the landscape scale to the physiological scale. The vulnerability of a forest stand is ultimately determined by the biological limits of individual trees—specifically, their hydraulic architecture and their carbon-based defense systems. Climate change is exploiting the trade-offs inherent in these systems.

3.1 Hydraulic Failure and the Live Fuel Moisture Paradox

One of the most critical mechanisms linking drought and fire is the status of Live Fuel Moisture Content (LFMC). For decades, fire behavior models primarily focused on "dead fuel moisture"—the dryness of sticks and logs on the ground. Live vegetation in the canopy was often treated as a dampening factor, a heat sink that slowed fire down. However, recent research utilizing process-based models like SurEau-Ecos-FMC has revolutionized this understanding.¹⁹

LFMC is not a static variable; it is dynamically linked to the plant's water potential (Psi, Ψ). Water potential is a measure of the tension or "suction" required to pull water from the soil up to the leaves. As soil moisture depletes during a drought, trees must exert greater tension (more negative water potential) to access water.

Research in Mediterranean forests and Western US conifers has identified a critical threshold known as the turgor loss point. When a tree's water potential drops below this species-specific limit, the cells in the leaves lose their rigidity (turgor). Stomata close to prevent desiccation, but if the drought deepens, the leaves begin to dehydrate rapidly. This leads to a precipitous drop in LFMC.¹⁹

Crucially, the relationship between drought intensity and flammability is nonlinear. A forest can maintain relatively high LFMC during moderate drought. But once the turgor loss point is crossed, the canopy transitions rapidly from a fire-resistant barrier to a highly flammable fuel source. This physiological "switch" explains why modern fires, driven by "hotter droughts," are exhibiting extreme fire behavior (such as crown runs) that were historically rare. The drought acts as a linked disturbance, pre-conditioning the canopy for catastrophic combustion.

3.2 Carbon Starvation and the Failure of Resin Defenses

The interaction between drought and biotic agents (insects and pathogens) is governed by a different physiological currency: Non-Structural Carbohydrates (NSC), or sugars and starches.

Conifers, such as pines and spruces, have evolved sophisticated chemical defenses against bark beetles. When a beetle bores into the bark, the tree attempts to "pitch it out" by flushing the wound with toxic, sticky resin. This resin is composed of terpenes and requires significant carbon resources to produce.²¹

Under normal conditions, trees generate enough carbon through photosynthesis to maintain growth, respiration, and defense. However, under drought stress, trees close their stomata to save water. This creates a dilemma: closing stomata halts photosynthesis. The tree is now starving for carbon. It must rely on stored NSCs.

If the drought is prolonged, the tree enters a state of carbon starvation. It no longer has the excess energy to synthesize resin or maintain turgor pressure in the resin ducts. The defense system collapses. This creates a window of massive vulnerability. A beetle population that would normally be repelled by healthy trees can now successfully mass-attack and kill the drought-stressed stand.²²

This mechanism creates a strong synergistic interaction. The drought does not necessarily kill the trees directly, but it removes their shield. Furthermore, research on Pinus lambertiana (sugar pine) indicates that this relationship is also threshold-dependent. There is a tipping point in water stress where resin duct pressure drops to zero. This explains the "eruptive" nature of recent beetle outbreaks: the forest appears healthy until the drought intensity crosses that invisible physiological line, triggers defense failure, and allows the beetle population to explode.²³

3.3 The "Time Lag" Vulnerability

Physiological stress leaves a legacy. Interactions do not always happen simultaneously. Defoliating insects, such as the western spruce budworm, consume the needles of trees, which are the tree's photosynthetic factories. Studies have shown that heavy defoliation can reduce a tree's carbon uptake for years or even a decade after the insect has gone.⁶

During this recovery period, the tree is operating on a carbon deficit. It prioritizes regrowing needles over producing defense chemicals. This creates a "linked disturbance" vulnerability with a significant time lag. A tree defoliated in 2015 might be uniquely susceptible to a bark beetle attack or a pathogen infection in 2020, even if the climate in 2020 is favorable. This "memory" of past stress complicates management, as a forest that looks green and recovered may effectively be immunocompromised.

4. Biome Case Study: The Arid West and the Beetle-Fire Nexus

Nowhere is the complexity of interacting disturbances more visible—or more contentious—than in the conifer forests of the Western United States. Here, the interplay between drought, bark beetles, and wildfire has generated a massive body of research and intense debate.

4.1 The Myth of the "Beetle Bomb"

For years, the conventional wisdom (and public perception) was that beetle-killed trees, being dead and dry, would inevitably lead to more severe wildfires. This was known as the "beetle bomb" hypothesis. However, the scientific reality, revealed through rigorous studies of linked disturbances, is far more nuanced and time-dependent.²¹

The interaction changes depending on the stage of the beetle attack:

1. Green Attack (Weeks): The trees are infested but still retain green needles. Flammability is largely unchanged.

2. Red Stage (1-3 Years): The trees die, and needles turn red and dry out. The canopy LFMC drops to near zero. In this specific window, flammability can increase, particularly the potential for crowning (fire moving through tree tops), because the fuel is incredibly dry and the chemical changes in the needles (terpene content) may boost combustibility.²¹

3. Gray Stage (Decades): The needles fall off, leaving standing gray snags. Surprisingly, fire severity often decreases in this stage regarding crown fire, simply because there is no fuel in the canopy to carry the flames. The interaction becomes antagonistic. The fire stays on the ground.

However, climate change is altering this trajectory. While the "Gray Stage" might reduce crown fire risk, the heavy accumulation of logs on the ground (as snags fall) increases the duration of burning and soil heating. This can lead to soil sterilization and root damage, a different kind of severity. Furthermore, under extreme fire weather (high wind, high heat), the dampening effect of the Gray Stage can be overwhelmed, leading to mass fires that consume everything regardless of fuel structure.²⁴

4.2 Synergies with Drought

The real driver in the Western US is not just Beetle -> Fire, but Drought -> Beetle + Fire. This is a disturbance network where drought acts as the master driver. Recent analyses suggest that the correlation between beetle outbreaks and large fires is often driven by the fact that both are caused by the same underlying drought conditions, rather than one causing the other directly.²⁵

However, when they do interact, the results can be transformative. The loss of canopy from beetles changes the microclimate, increasing wind speed and solar radiation hitting the forest floor. If a fire does occur, these altered conditions can increase the rate of spread. This is a classic example of how a biotic disturbance (beetles) alters the physical environment to facilitate an abiotic disturbance (fire).¹

5. Biome Case Study: Tropical Tipping Points and the Amazon

In the humid tropics, the interaction of disturbances is creating an existential threat to the world's largest rainforest. The Amazon basin is currently teetering on the edge of a Coupled Feedback loop that threatens to flip the ecosystem from rainforest to savanna.

5.1 The Moisture Recycling Feedback

The Amazon rainforest essentially creates its own weather. Trees pump vast amounts of water from the soil into the atmosphere through transpiration. This moisture forms "flying rivers" that drift west and fall as rain, sustaining the forest further inland.²⁶ This is a positive feedback loop that maintains the rainforest state.

Climate change and deforestation are disrupting this cycle through a disturbance network involving three key agents:

1. Vapor Pressure Deficit (VPD): As the atmosphere warms, its capacity to hold water increases. This raises the VPD, which pulls moisture out of the vegetation and soil at an accelerated rate.²⁷

2. Fire Leakage: Historically, the Amazon was too wet to burn. Natural fires were rare and small. However, fragmentation (from logging) and drought are drying out the edges of forest patches. Fire from agricultural lands "leaks" into the standing forest.

3. The Dieback Loop: A surface fire in a tropical forest does not need to be intense to be destructive. Most Amazonian trees have thin bark (having evolved without fire resistance). A slow, low-intensity fire can kill 50% of the trees. When these trees die, the canopy opens. The "pump" of transpiration weakens. Less moisture goes into the atmosphere. The local climate gets drier. The next fire burns hotter and deeper.¹⁵

5.2 The Specter of Savannization

This coupled feedback—where vegetation loss begets climate drying, which begets more vegetation loss—is the mechanism of the dreaded "tipping point." Models suggest that if the Amazon loses a certain percentage of its cover (estimates range from 20-25%), or if global warming exceeds a certain threshold, the moisture recycling system could collapse.³⁰

The result would be a critical transition to a "derived savanna" or scrubland. This new state would be maintained by a new set of feedbacks: frequent fires (promoted by grasses) would kill any tree seedlings attempting to recolonize the area. This hysteresis means that even if deforestation stops or the climate cools slightly, the rainforest cannot easily return. The system has fallen into a "basin of attraction" defined by fire and grass.⁹

6. Biome Case Study: Boreal Forests and the Immaturity Risk

In the high-latitude boreal forests, the interaction of climate change and fire frequency is creating a "Compound Disturbance" that threatens the persistence of coniferous forests.

6.1 Serotiny and the Fire Cycle

Many boreal trees, such as black spruce (Picea mariana) and jack pine (Pinus banksiana), are serotinous. They store their seeds in resin-sealed cones that stay on the tree for years. These trees depend on fire to melt the resin and release the seeds onto a nutrient-rich ash bed. This is an ancient adaptation to a specific disturbance regime: high-intensity crown fires occurring every 80 to 150 years.³³

The resilience of this system depends on the interval between fires. The trees need enough time to grow, reach sexual maturity, and produce a new "bank" of cones. This usually takes 15 to 25 years.

6.2 The Immaturity Risk Mechanism

Climate change is drastically shortening the fire return interval in the boreal zone. We are now seeing areas re-burning in less than 15 years—sometimes in less than 5. This creates a phenomenon known as Immaturity Risk.⁷

If a second fire strikes before the regenerating stand has produced cones, the seed source is obliterated. There are no seeds left on the trees, and the soil seed bank is exhausted. This compound disturbance breaks the cycle of recovery.

The result is often a transition to deciduous species like aspen or birch, whose light, wind-blown seeds can travel long distances to colonize the burned site. In worse scenarios, where soil organic layers are consumed, the site may transition to open lichen woodlands or heathlands. This "regime shift" alters the carbon balance of the boreal zone, potentially turning a carbon sink into a carbon source.³⁴

6.3 The Wind-Beetle-Fire Cascade

Boreal and temperate forests are also subject to Disturbance Cascades initiated by wind. Severe storms (derechos or cyclones) cause extensive blowdown. This provides a massive pulse of breeding material for bark beetles (like the spruce beetle), which thrive in fresh downed logs.

The beetle population builds up in the windthrow and then "spills over" into the surrounding healthy forest, triggering an epidemic.¹⁰ The dead trees and the slash from the windstorm then cure, creating a fuel matrix that is susceptible to high-severity fire. This linear cascade (Wind -> Beetle -> Fire) is becoming more frequent as climate change increases the intensity of storms and creates longer fire seasons for the final step of the chain to occur.⁴

7. Methodological Frontiers: Untangling the Web

As the complexity of these interactions grows, the traditional tools of ecology—simple correlations and ANOVA tests—are becoming insufficient. How do we prove that drought caused the fire, if beetles were also present? How do we measure the strength of a feedback loop? To address this, disturbance ecology is adopting advanced methodologies from data science and complexity theory.

7.1 Causal Inference and Directed Acyclic Graphs (DAGs)

One of the most promising tools for untangling "spaghetti" networks of causality is the Directed Acyclic Graph (DAG). A DAG is a visual and mathematical representation of assumptions about causal relationships.³⁷

In a DAG, variables are nodes connected by arrows (edges) representing causal influence. For example, a researcher might draw:

• Drought -> Fuel Moisture -> Fire Intensity

• Drought -> Beetle Outbreak -> Fire Intensity

By explicitly mapping these paths, researchers can identify "confounders" (variables that influence both the cause and the effect, creating spurious correlations) and "colliders" (variables influenced by two others). This allows for more rigorous statistical testing. For instance, using DAGs, researchers can determine if the correlation between beetles and fire is real or if it is merely because both are driven by the common confounder of drought.³⁹ This moves the field from observing "linked patterns" to understanding "linked mechanisms."

7.2 Network Analysis

Originally developed for social sciences and computer networks, Network Analysis is now being applied to forest landscapes. In this approach, the landscape is modeled as a network where forest stands are nodes and disturbances are the connections between them.¹⁴

This approach is crucial for understanding connectivity. For example, fire spreads through a landscape based on the connectivity of fuels. Network metrics like "centrality" or "betweenness" can identify which patches of forest are critical for spreading a disturbance. If a specific stand acts as a "hub" for beetle dispersal, targeting management there could break the network.

Climate change is essentially "rewiring" these networks, often increasing connectivity (e.g., by drying out wetlands that used to act as firebreaks). Modeling these changes allows researchers to predict how a disturbance introduced in one corner of a landscape might propagate through the entire system.¹⁶

7.3 Process-Based Modeling

Statistical models based on historical data are failing because the future has no historical analog. "Novel" climates produce "novel" disturbance interactions. To bridge this gap, ecologists are turning to Process-Based Models (PBMs) like LANDIS-II, FireBGCv2, and iLand.¹⁴

These models do not rely on past correlations. Instead, they simulate the fundamental physiological processes: photosynthesis, respiration, water uptake, and decomposition. They simulate the "rules" of life. By running these models under future climate scenarios, researchers can see how these rules play out in new environments. For example, a PBM can simulate whether a tree will die from carbon starvation before it dies from hydraulic failure, and how that timing affects its flammability. These simulations are the primary tool for forecasting "immaturity risk" and biome shifts.⁴⁰

8. Management Implications: From Resistance to Resilience

The "Unifying Framework" of disturbance interactions forces a paradigm shift in forest management. The traditional goal of management was often Resistance: preventing change, suppressing fire, stopping beetles. The prevalence of synergistic, amplifying interactions suggests that resistance may effectively be futile in many ecosystems.

8.1 Adaptive Resilience and Transition

Managers are increasingly adopting the concept of Adaptive Resilience. This acknowledges that the forest of the future will not look like the forest of the past. If a "Compound Disturbance" (Fire + Drought) makes it impossible for the native species to regenerate, management might shift to facilitating transition.³²

This could involve "assisted migration"—planting tree species (or genotypes) from warmer, drier regions that can survive the new disturbance regime. It might mean accepting that a forest will become a shrubland and managing that shrubland to preserve soil stability and biodiversity, rather than fighting a losing battle to replant trees that will die in the next drought.

8.2 Breaking the Chain

Understanding the specific biological mechanism of an interaction allows for surgical intervention. If we know that the link between windthrow and fire is the connectivity of surface fuels, we can use targeted thinning or prescribed burning to "decouple" that interaction.⁴¹

Similarly, understanding the "time lag" of linked disturbances offers a window of opportunity. If a forest is defoliated by budworms, managers know it is now hyper-vulnerable to bark beetles for the next decade. This allows for proactive thinning to reduce competition and boost the vigor of the remaining trees before the beetles arrive, effectively dampening the synergistic potential of the interaction.⁶

9. Conclusion: The Scientific Frontier

The study of interacting disturbances represents the "frontier" of modern ecology.¹ We are moving away from a view of nature as a static backdrop interrupted by occasional, independent events. Instead, we are seeing the forest as a dynamic, interconnected system where energy and stress flow through complex networks of feedback loops.

The evidence synthesized in this report is stark. Climate change is not merely turning up the dial on individual hazards; it is fundamentally rewiring the connections between them. Through Linked, Compound, and Synergistic mechanisms, the interactions between drought, fire, wind, and pests are pushing ecosystems toward Critical Transitions. The "tipping points" of the Amazon and the Boreal forest are not theoretical worst-case scenarios; they are the logical endpoints of the disturbance networks we are currently observing.

To navigate this turbulent future, science and management must embrace complexity. We need the "Unifying Framework" to organize our observations. We need advanced physiological models to understand the hidden vulnerabilities of trees. And we need the courage to manage for change, rather than against it. Only by understanding the web of interactions can we hope to sustain the function and biodiversity of the world's forests in the face of an uncertain climate.

10. Table 2: Physiological Traits Determining Vulnerability to Interactions

Physiological Trait	Description	Role in Disturbance Interaction	Climate Sensitivity
Water Potential (Ψ)	The tension of water in the xylem.	Determines the "tipping point" for hydraulic failure and flammability changes.	Directly driven by soil moisture and Vapor Pressure Deficit (VPD).
Turgor Loss Point (ΨTLP)	The water potential at which leaves lose rigidity.	The threshold where LFMC drops precipitously; triggers fire susceptibility.	Species-specific; lower (more negative) values indicate drought tolerance.
Resin Duct Density	Number/size of resin canals in bark.	Primary physical defense against bark beetles.	Higher density = better defense. Can be reduced by past stress (defoliation).
Non-Structural Carbohydrates (NSC)	Stored sugars and starches.	Energy reserve for defense and recovery.	Depleted by drought (stomatal closure) and defoliation, leading to carbon starvation.
Serotiny	Cone sealing requiring fire to open.	Adaptation for regeneration after fire.	Becomes a liability ("immaturity risk") if fire intervals become shorter than maturation time.
Bark Thickness	Physical insulation of cambium.	Resistance to surface fire heat.	Critical for survival in "fire leakage" scenarios (e.g., Amazon trees lack this).

11. Table 3: Comparative Analysis of Interaction Dynamics by Biome

Biome	Key Disturbance Agents	Dominant Interaction Class	Critical Transition Risk
Western US Conifer	Drought, Bark Beetles, Fire	Synergistic: Drought primes beetles; Beetles/Drought prime fire.	Moderate: Shift from forest to shrubland/grassland in dry margins.
Amazon Rainforest	Drought, Fire, Deforestation	Coupled Feedback: Vegetation loss drives regional drying.	High: "Savannization" tipping point driven by moisture recycling collapse.
Boreal Forest	Fire, Insects, Wind	Compound: Short fire intervals (Immaturity Risk).	High: Conversion from Conifer to Deciduous or Lichen Woodland.
Mediterranean	Fire, Drought	Linked: Drought lowers LFMC; Fire promotes invasive grasses.	Moderate/High: "Grass-Fire Cycle" creating positive feedback loops.
Temperate Deciduous	Wind, Pathogens, Insects	Disturbance Network: Windthrow cascades to insects/pathogens.	Low/Moderate: Compositional shifts (species replacement) rather than biome loss.

Works cited

1. Climate Change Effects on Interacting Disturbances in Forest Ecosystems - Annual Reviews, accessed January 11, 2026, https://www.annualreviews.org/content/journals/10.1146/annurev-ecolsys-102723-052437

2. Climate Change Effects on Interacting Disturbances in ... - mediaTUM, accessed January 11, 2026, https://mediatum.ub.tum.de/doc/1838354/m2qtaxuhtqdqm9enr6xwpv4qr.annurev-ecolsys-102723-052437.pdf

3. Climate Change Effects on Interacting Disturbances in Forest Ecosystems - University of Washington, accessed January 11, 2026, https://depts.washington.edu/bjhlab/wordpress/wp-content/uploads/2025/11/Dudney_etal_2025_AnnuRevEcolEvolSyst_ClimateChangeEffectsInteractingDisturbancesForestEcosystems.pdf

4. (PDF) The Ecology of Disturbance Interactions - ResearchGate, accessed January 11, 2026, https://www.researchgate.net/publication/345760490_The_Ecology_of_Disturbance_Interactions

5. Novel Disturbance Regimes and Ecological Responses - Annual Reviews, accessed January 11, 2026, https://www.annualreviews.org/doi/pdf/10.1146/annurev-ecolsys-110421-101120

6. Outbreaks of Douglas-Fir Beetle Follow Western Spruce Budworm Defoliation in the Southern Rocky Mountains, USA - MDPI, accessed January 11, 2026, https://www.mdpi.com/1999-4907/13/3/371

7. accessed January 11, 2026, https://www.frontiersin.org/journals/ecology-and-evolution/articles/10.3389/fevo.2021.653647/full#:~:text=Compound%20disturbances%20are%20those%20affecting,develop%20a%20canopy%20seed%20bank.

8. Understanding and Modeling Forest Disturbance Interactions at the Landscape Level, accessed January 11, 2026, https://www.frontiersin.org/journals/ecology-and-evolution/articles/10.3389/fevo.2021.653647/full

9. Mechanisms of forest resilience - the NOAA Institutional Repository, accessed January 11, 2026, https://repository.library.noaa.gov/view/noaa/66351/noaa_66351_DS1.pdf

10. Review of Conceptual Advances in Disturbance Interactions - Canadian Council of Forest Ministers, accessed January 11, 2026, https://www.ccfm.org/wp-content/uploads/2024/07/Review-of-Conceptual-Advances-in-Disturbance-Interactions.pdf

11. COOPERATIVE MONITORING, EVALUATION AND REASEACH 02-205 A Review and Synthesis of Available Information on Riparian Disturbance R - DNR.wa.gov, accessed January 11, 2026, https://dnr.wa.gov/sites/default/files/2025-05/fp_cmer_02_205.pdf

12. Impact of duration of prior interspecific disturbance on the likelihood... - ResearchGate, accessed January 11, 2026, https://www.researchgate.net/figure/Impact-of-duration-of-prior-interspecific-disturbance-on-the-likelihood-of-observing_fig5_384148469

13. Vulnerable regions and tipping reason. (A) The likelihood of tipping... - ResearchGate, accessed January 11, 2026, https://www.researchgate.net/figure/ulnerable-regions-and-tipping-reason-A-The-likelihood-of-tipping-vulnerability-as-an_fig2_362439603

14. Understanding and Modeling Forest Disturbance Interactions at the Landscape Level, accessed January 11, 2026, https://www.researchgate.net/publication/355045806_Understanding_and_Modeling_Forest_Disturbance_Interactions_at_the_Landscape_Level

15. (PDF) Climate Change Effects on Interacting Disturbances in Forest Ecosystems, accessed January 11, 2026, https://www.researchgate.net/publication/395207140_Climate_Change_Effects_on_Interacting_Disturbances_in_Forest_Ecosystems

16. A tropical forest network of interactions and feedbacks. Solid lines... - ResearchGate, accessed January 11, 2026, https://www.researchgate.net/figure/A-tropical-forest-network-of-interactions-and-feedbacks-Solid-lines-indicate-feedbacks_fig2_361668697

17. The Ecology of Disturbance Interactions - Oxford Academic, accessed January 11, 2026, https://academic.oup.com/bioscience/article-pdf/70/10/854/33894673/biaa088.pdf

18. Modelling the multi-scaled nature of pest outbreaks, accessed January 11, 2026, https://ideas.repec.org/a/eee/ecomod/v409y2019ic8.html

19. Plant hydraulic modelling of leaf and canopy fuel moisture content reveals increasing vulnerability of a - USDA Forest Service, accessed January 11, 2026, https://www.fs.usda.gov/rm/pubs_journals/2023/rmrs_2023_ruffault_j001.pdf

20. The Effect of Ecophysiological Traits on Live Fuel Moisture Content - MDPI, accessed January 11, 2026, https://www.mdpi.com/2571-6255/2/2/28

21. Model response curves predicting defense resistance in sugar pine.... - ResearchGate, accessed January 11, 2026, https://www.researchgate.net/figure/Model-response-curves-predicting-defense-resistance-in-sugar-pine-While-other-defense_fig4_366924716

22. Outbreaks of Douglas-Fir Beetle Follow Western Spruce Budworm Defoliation in the Southern Rocky Mountains, USA - NSF Public Access Repository, accessed January 11, 2026, https://par.nsf.gov/servlets/purl/10341600

23. Tree resistance to drought and bark beetle-associated mortality following thinning and prescribed fire treatments - ResearchGate, accessed January 11, 2026, https://www.researchgate.net/publication/366924716_Tree_resistance_to_drought_and_bark_beetle-associated_mortality_following_thinning_and_prescribed_fire_treatments

24. A review of the relationships between drought and forest fire in the United States, accessed January 11, 2026, https://www.fs.usda.gov/rm/pubs_journals/2016/rmrs_2016_littell_j001.pdf

25. thesis interactions and impacts of multiple bark beetle outbreaks in, accessed January 11, 2026, https://api.mountainscholar.org/server/api/core/bitstreams/2cca3f03-45e4-4729-bc6d-d7705ae166ea/content

26. Analysis of the Amazonian tipping element in the 21st century within the local safe operating space framework - DiVA portal, accessed January 11, 2026, http://www.diva-portal.org/smash/get/diva2:1698637/FULLTEXT01.pdf

27. How Amazon deforestation could trigger a climate tipping point - ThinkLandscape, accessed January 11, 2026, https://thinklandscape.globallandscapesforum.org/54916/the-amazon-rainforest-is-nearing-its-tipping-point-but-what-does-that-mean/

28. Collision Course: Development Pushes Amazonia Toward Its Tipping Point, accessed January 11, 2026, https://www.tandfonline.com/doi/full/10.1080/00139157.2021.1842711

29. The Susceptibility of Southeastern Amazon Forests to Fire: Insights from a Large-Scale Burn Experiment - ResearchGate, accessed January 11, 2026, https://www.researchgate.net/publication/281444102_The_Susceptibility_of_Southeastern_Amazon_Forests_to_Fire_Insights_from_a_Large-Scale_Burn_Experiment

30. Climate change, ecosystems and abrupt change: science priorities | Philosophical Transactions of the Royal Society B, accessed January 11, 2026, https://royalsocietypublishing.org/rstb/article/375/1794/20190105/30570/Climate-change-ecosystems-and-abrupt-change

31. Climate Change and Sustainability in Amazonia - Oxford Research Encyclopedias, accessed January 11, 2026, https://oxfordre.com/climatescience/display/10.1093/acrefore/9780190228620.001.0001/acrefore-9780190228620-e-978?p=emailAgIbX.POo.B7s&d=/10.1093/acrefore/9780190228620.001.0001/acrefore-9780190228620-e-978

32. (PDF) Future transition from forests to shrublands and grasslands in the western United States is expected to reduce carbon storage - ResearchGate, accessed January 11, 2026, https://www.researchgate.net/publication/378150911_Future_transition_from_forests_to_shrublands_and_grasslands_in_the_western_United_States_is_expected_to_reduce_carbon_storage

33. Impacts of Changing Disturbance Regimes on Serotinous Plant Populations and Communities | BioScience | Oxford Academic, accessed January 11, 2026, https://academic.oup.com/bioscience/article/63/11/866/2389884

34. Understanding and Modeling Forest Disturbance Interactions at the Landscape Level, accessed January 11, 2026, https://www.fs.usda.gov/nrs/pubs/jrnl/2021/nrs_2021_sturtevant_002.pdf

35. Changing disturbance regimes, ecological memory, and forest resilience, accessed January 11, 2026, https://turnerlab.ibio.wisc.edu/wp-content/uploads/sites/43/2021/12/Johnstone_etal2016Frontiers_Resilience.pdf

36. Ecology of Disturbance Interactions | BioScience - Oxford Academic, accessed January 11, 2026, https://academic.oup.com/bioscience/article/70/10/854/5901988

37. A proposal for capturing interaction and effect modification using DAGs - PMC - NIH, accessed January 11, 2026, https://pmc.ncbi.nlm.nih.gov/articles/PMC9365632/

38. Reflection on modern methods: constructing directed acyclic graphs (DAGs) with domain experts for health services research - Oxford Academic, accessed January 11, 2026, https://academic.oup.com/ije/article/51/4/1339/6609840

39. (PDF) A directed acyclic graph for interactions - ResearchGate, accessed January 11, 2026, https://www.researchgate.net/publication/347389116_A_directed_acyclic_graph_for_interactions

40. Exploring Interactions Among Multiple Disturbance Agents in Forest Landscapes: Simulating Effects of Fire, Beetles, and Disease - University of Montana, accessed January 11, 2026, https://www.umt.edu/media/leopold/pubs/842.pdf

41. 
    

42. Climate change, forests, fire, water, and fish: Building resilient landscapes, streams, and managers, accessed January 11, 2026, https://www.fs.usda.gov/rm/pubs/rmrs_gtr290.pdf