2. Fire Management of Woody Encroachment in Canadian and USA National Parks
In cooperation with other federal and local agencies, Parks Canada and the US National Park Service are responsible for brush management and fire management in their respective parks. National parks in Canada and the USA utilize ecological monitoring indicators to assess the integrity of grassland ecosystems
[9][10]. Indicators related to woody encroachment include vegetation composition (e.g., native grass, non-native grasses, shrubs, or trees), vegetation structure (e.g., cover, height, and landscape dynamics), fire disturbance, and the occurrence and habitats of native wildlife affected by woody encroachment.
Divergences of indicator scores from expected values have been assessed as an acceptable variation, potential management concern, or imminent loss based on the resistance, hysteresis, and irreversibility of the indicator
[10]. For example, if returning to the pre-transition fire regime is generally sufficient to maintain the previous grassland state, park managers would identify the grasslands as being within an acceptable range of variation. Otherwise, hysteretic and irreversible responses would raise potential management concerns and indicate a risk of imminent loss.
Prescribed fires and wildfires have played some roles in protecting native grasslands, but in some cases have not been powerful enough to fully reverse shrublands and woodlands in parks
[11][12]. Mechanical, chemical, and biological treatments have also been used in national parks to protect grassland ecosystems from woody encroachment and invasive species
[11][12][13][14].
3. Case Study: Grassland Transitions to Non-Native Prairies and Shrublands
The Grasslands, Badlands, and Elk Island National Parks represent a well-preserved grassland state, a degraded grassland state, and a transition to the shrubland state, respectively (
Table 1). The Grasslands National Park’s iconic mixed-grass prairie is well-preserved, and only slightly threatened by agricultural grass species rather than shrubs or trees
[15]. In Badlands National Park, the non-native Kentucky bluegrass (
Poa pratensis) dominates the non-native prairie–Kentucky bluegrass plant community and only a few native forbs or shrubs are seen
[16]. In Elk Island National Park, shrubs have encroached into the fragmented fescue grasslands
[17].
In practice, wildland fires, including wildfires and prescribed fires, can maintain well-preserved and degraded grassland states. Even under wildfire suppression, prescribed fires may be sufficient to control the non-native species in the Grasslands National Park, and park managers have set the goal to restore the areas disturbed by invasive species and prevent new invasive plants by 2030
[15][18]. However, once non-native species become dominant or shrubland transitions occur, degraded grassland states cannot be reversed by varying the fire frequency alone. For example, in the Badlands National Park, fires only modestly reduced the non-native grass cover in non-native grasslands
[19], and the years since fires have had no significant impact on the richness of native species and relative cover of exotic species
[16]. Generally, high-frequency fires are needed to reverse shrub encroachment
[20], but the fire cycle in Elk Island National Park is not sufficiently restored yet, resulting in the unsuccessful management of shrub and aspen encroachment
[17].
The effectiveness of these management approaches varies due to ecological and social–political constraints. Firstly, fire suppression still occurs and the high-frequency fires needed to halt shrubland encroachment cannot be ensured. Fire suppression was implemented from 2002 to 2022 in Elk Island National Park due to operational and social constraints, such as undesirable weather conditions, inadequate funding, or COVID-19
[21]. Secondly, ecological interactions between fires and grazing, climate, and other disturbances are complex, making the fire thresholds unknown. For example, precipitation has more influence on vegetation cover than fire does in the Northern Great Plains grasslands, and vegetation response to climate is less predictable
[16]. Thirdly, operational constraints also limit the effectiveness of combined management actions on shrub encroachment. Bison grazing and the creation of bison wallows can limit the cover of dominant invasive grass species in mixed-grass prairies
[22]; however, a population of 700 bison in Badlands National Park has had little effect on the grass community due to their relatively low density
[19].
Table 1. Fire management efficiency and recommendations on controlling shrubland transitions in selected grassland national parks.
Alternative States |
National Parks |
Fire Regimes |
Fire Management Effectiveness |
Management Recommendations |
Grassland states |
Grasslands National Park: invasive grass [15] |
Prescribed fires, a 5-year fire cycle since 2000 [15], Suppressing all wildfires [18] |
Effective to maintain the grasslands [15] |
Reintroduced bison in 2005 and uncleared impacts [15] |
Shrubland transitions |
Elk Island National Park: shrubs and aspen [17][21][23] |
Prescribed fires since 1979 [24]; Fire suppression from 2002 to 2022 [21] |
Not effective, due to fire suppression |
Restore the fire cycle [17] |
Badlands National Park: non-native grass formed a near-monoculture |
Historical fire return 8–25 years [19]; Prescribed fire since the 1980s [25] |
Fire alone cannot reverse non-native grass [16] |
Combinations of native seeding, fire, and/or herbicide [19] |
4. Case Study: Woodland Transitions
Three USA national parks and three Canadian national parks were selected as having representative grasslands that are being encroached by typical woody encroachers, such as ponderosa pine, Ashe juniper, red cedar, mesquite, and aspen (
Table 2). The Lyndon B. Johnson National Historical Park is heavily invaded by exotic grasses, forbs, and the native honey mesquite in the mixed grasslands
[26]. Ashe juniper and red cedar have rapidly spread throughout the tallgrass and mixed grasslands due to fire suppression in the Chickasaw National Recreation Area
[27][28]. The Lake Meredith National Recreation Area also features open stands of mesquite in the prairies
[29]. The Riding Mountain, Prince Albert, and Waterton Lakes National Parks in Canada were selected to represent the fescue grasslands being encroached by sprouting aspen woodlands
[5][30][31].
Despite efforts to reduce woody encroachment by obligate seed species via prescribed burning and wildfires, encroachment remains a persistent challenge in these parks. Social–political constraints may be responsible for the failure of fires to reduce the woody encroachment of non-sprouting species or mature trees of sprouting species. The frequency and intensity of fires have not reached the fire thresholds of these encroachers at the three USA sites. For example, in the Lyndon B. Johnson National Historical Park, prescribed fire was only used in 2015, due to the urban location of the prairie in need of restoration
[26][32]. This social constraint is challenging. Efforts have been made to increase the intensity of prescribed burns, such as by shifting the timing of the burns, to control mesquite in the Lake Meredith National Recreation Area, but this approach faces challenges due to safety considerations and the low postfire recovery of grasslands under extended drought conditions
[29]. In the Southern Great Plains network, intense fires can kill large juniper trees, and subsequent fires within 10 or 15 years have the potential to kill immature junipers
[28]. Intense fires might therefore be effective in reducing the encroachment of non-sprouting species in the Chickasaw National Recreation Area. However, the uncertainty associated with the use of fire and vegetation responses is still being considered by fire managers
[28].
The situation is different for sprouting species in the three Canadian national parks. Even when prescribed fires and high-intensity wildfires occur, sprouting species encroachment cannot be reversed due to their rapid postfire recruitment. This suggests the presence of ecologically irreversible fire thresholds. Frequent prescribed fires in the Waterton Lakes National Park, aiming at restoring the historical fire regime that maintained the fescue grasslands, were insufficient to reverse, or even halt, shrub and aspen encroachment. The areas of aspen canopy (aspen higher than 2.5 m), aspen regeneration, and shrubs did not change significantly before and after the prescribed fires
[11]. Aspen cover still expanded into burned areas in the Prince Albert National Park, indicating that prescribed fires there had limited long-term effectiveness in controlling aspen encroachment
[13]. Similarly, a high-severity wildfire can halt but not necessarily reverse woody encroachment. The Kenow fire caused the mortality of mature aspen and altered the aspen stand structure from a late-seral to an early-seral state one-year postfire; however, the extent of aspen cover remained the same, as the fire stimulated vigorous aspen sprouting
[6]. In the Riding Mountain National Park, fire alone may be insufficient to suppress aspen, and a combination of brush management and prescribed fires has been recommended
[12].
In places, such as in the wildland–urban interface or where irreversible ecological fire thresholds exist, a comprehensive approach that combines fire management with other brush management methods is necessary. This approach should take into account the unique challenges posed by these environments and strive to minimize the risk of catastrophic fire events while balancing ecological concerns and social needs. For example, in the Southern Plains network, fire managers are monitoring the effects of wildfire, prescribed fire, and mechanical treatments on the ecosystems
[32]. In Waterton Lakes National Park, high-severity prescribed fires (e.g., applying prescribed fires in late season to utilize sufficient fuel to generate high-severity fire) and the use of Traditional Ecological Knowledge (TEK) (e.g., bison dynamics) have been recommended if management is to better conserve the grasslands
[6]. However, many uncertainties and knowledge gaps exist regarding fire management and grassland conservation in these national parks. The detailed thresholds of fire frequency and severity/intensity, the interactions of fires with other disturbances, and the effectiveness of these brush management methods on woody encroachment are unknown, and long-term monitoring is needed in practice.
Table 2. Fire management efficiency and recommendations on controlling woodland transitions in selected grassland national parks.
National Parks |
Fire Regimes |
Fire Management Effectiveness |
Management Recommendations |
Riding Mountain National Park: Aspen [5] |
Prescribed fires (1996–2010) |
Fire alone may be insufficient to suppress aspen |
Combination of brush management [12] |
Prince Albert National Park: Aspen and shrubs [30] |
Prescribed fires since 1975, a 5-year fire returns 2018 Rabbit Creek Wildfire |
Insufficient, limited long-term effectiveness [13] |
Interactions between fire and grazing [13] |
Waterton Lakes National Park: Aspen [33] |
Prescribed fires since 1989, a 5-year fire cycle; 2017 Kenow Wildfire [31] |
Prescribed fires have no impacts [11]; Wildfire was insufficient due to vigorous sprouting [6] |
Improving fire severity in late season, bison [6] |
Lyndon B. Johnson National Historical Park: Native honey mesquite [26] |
Prescribed fire in 2015 [26][32] |
Unknown |
Restoring fire cycles, high-intensity fires |
Chickasaw National Recreation Area: Ashe juniper, and red cedar [27][28] |
Prescribed fire since 1998; several incidents of wildfires |
Insufficient, due to resprouting |
Mechanical removal of junipers in combination with fire treatments |
Lake Meredith National Recreation Area: mesquite [29] |
Prescribed fires since 1998, several incidents of wildfires |
Insufficient, reduced invasive grass, but not woody encroachment |
High-intensity, shifting the timing of the prescribed burns |
5. Social–Political Constraints of Resilience-Based Fire Management on Woody Encroachment in National Parks
Fire experiments conducted in the Great Plains region and North America have demonstrated the resistance, hysteresis, and irreversibility attributes of grasslands when reversing woody encroachment by fires. Grasslands in some national parks demonstrated their resistance and irreversibility, but in other parks, the hysteretic shrubland and woodland transitions were irreversible using fire (Table 6). In the early stages of woody encroachment, intact grasslands exhibit resistance to shrub encroachment, and the prescribed fires that restore historical fire frequencies may help to maintain the grasslands within the acceptable range of variability. Once the shrublands or non-sprouting woodlands are established, the grasslands may exhibit hysteresis, creating management concerns and challenges. High-intensity fires that exceed the fire-mortality thresholds are necessary to exclude encroachment, but social–political constraints often hinder fire management efforts. In cases where resprouting species have encroached on grasslands, the grasslands may become irreversibly altered, signaling imminent loss. While fire can halt encroachment, it cannot reverse the transitions caused by resprouting species. Other postfire programs, such as encouraging elk browse on aspen regeneration or bison reintroductions, may be necessary for brush management.
Grassland resilience to woody encroachment is lower than ecologists expected in some parks due to social–political constraints (Table 3). Fire suppression, COVID-19, inadequate funding, and difficulties in implementing prescribed burning (e.g., fire danger near urban locations, and undesirable weather conditions) have all hindered efforts to reverse shrublands using fire. Furthermore, the uncertainty associated with interactions between fire and drought under climate change has further complicated the fire management efforts. For irreversible woodland transitions with resprouting species, these social–political constraints for fire management still exist, especially the unknown effects of combination management, including unknown ecological interactions and a lack of long-term monitoring. Table 3 compares the theoretical grassland resilience to woody encroachment observed in the Great Plains region to the operational effectiveness of fire management in the selected grassland national parks in Canada and the USA.
Table 3. Theoretical grassland resilience to woody encroachment and the operational effectiveness of fire management in selected grassland national park.
Vegetation States |
Intact Grasslands |
Shrubland Transitions |
Woodland Transitions: Non-Sprouting |
Woodland Transitions: Resprouting Species |
Grass resilience |
Resistance |
Hysteresis |
Hysteresis |
Irreversibly |
Monitoring ranking in national parks |
Acceptable change in variation |
Fire management concerns |
Fire management concerns |
Imminent loss |
Management effectiveness |
Success |
Limited effects |
Limited effects |
Limited effects |
Grass resilience in practice in national parks |
Resistance |
Hysteresis and irreversibility |
Irreversibly |
Irreversibly |
Social–political facilitations/constraints in national parks |
- (a)
-
Restored fire frequency
- (b)
-
Implementation of fire and bison
|
- (a)
-
Fire suppression
- (b)
-
The uncertainty of fire interactions with other disturbances and climate change
- (c)
-
COVID-19
- (d)
-
Inadequate funding
- (e)
-
Implementation difficulties of frequent or intensive prescribed burnings, such as undesired weather
|
- (a)
-
Fire danger near urban locations,
- (b)
-
Unexpected drought conditions increasing fire danger
- (c)
-
Uncertainty associated with the use of fire and vegetation responses.
|
- (a)
-
Ecological irreversibility due to postfire resprouting
- (b)
-
Social–political constraints that limited reversing shrublands
- (c)
-
Challenges of combination management: unknown ecological interactions, lack of long-term monitoring, and unclear management effectiveness
|
6. Resilience-Based Management Framework for Woody Encroachment
The National Park systems in Canada and the USA have established monitoring indicators for grassland ecosystem integrity and ranked each indicator as within an acceptable range of variation, a management concern, or an imminent loss. Here focused specifically on woody encroachment and introduced a fire management framework that is grounded in the resilience theory. Initially, the alternative states of grassland subject to woody encroachment would be identified according to the shrub and tree species. Next, the park’s monitored grassland responses to woody encroachment via fire management reveal resilience characteristics, including resistance, hysteresis, and irreversibility under each potential state. If no management intervention or reverting to the original fire regime is sufficient to retain the previous grassland state, it suggests resistance. On the other hand, the existence of hysteresis is indicated when the vegetation transition pathway from an altered state back to grassland is different from the pathway of grassland transitioning to the altered state. When irreversible fire thresholds segregate the potential grassland states, this signals an imminent risk to the ecosystem, pointing toward management failure.
Identifying key indicators that adequately represent the grassland responses and resilience characteristics is challenging. Vegetation composition and structure serve as direct indicators, while wildlife populations and disturbances can be seen as indirect indicators of woody encroachment within parks. Further, the status of each indicator is classified as either an acceptable change, potential management concern, or imminent loss, based on the resilience attributes of resistance, hysteresis, and irreversibility. To manage resistant, hysteretic, and irreversible grasslands, the restoration of the fire regime is required. This may entail frequent controlled burns, intensive fires, or a combination of brush management strategies.