Trends, Impacts, and Cost of Catastrophic and Frequent Wildfires in the Sagebrush Biome

ABSTRACT Fire regimes in sagebrush (Artemisia spp.) ecosystems have been greatly altered across the western United States. Broad-scale invasion of non-native annual grasses, climate change, and human activities have accelerated wildfire cycles, increased fire size and severity, and lengthened fire seasons in many sagebrush ecosystems to the point that current wildfire-management practices and postfire restoration efforts cannot keep pace to ameliorate the ecological consequences of sagebrush ecosystem loss. The greatest impact of uncharacteristically frequent fire is the transition from native sagebrush-perennial grass communities to invasive, non-native, annual grasslands that are highly flammable. These community transitions are often permanent, owing to the low probability of reestablishing native perennial plants in non-native annual grass–dominated communities. Moreover, these grasses can form extensive and continuous fine fuel loads that promote more frequent fire and the continued expansion of invasive, non-native annuals. More frequent, larger, and severe wildfires necessitate greater resources for fire-prevention, fire-suppression, and postfire restoration activities, while decreasing critical ecosystem services, economic and recreational opportunities, and cultural traditions. Increased flexibility and better prioritization of management activities based on ecological needs, including commitment to long-term prefire and postfire management, are needed to achieve notable reductions in uncharacteristic wildfire activity and associated negative impacts. Collaboration and partnerships across jurisdictional boundaries, agencies, and disciplines can improve consistency in sagebrush-management approaches and thereby contribute to this effort. Here, we provide a synthesis on sagebrush wildfire trends and the impacts of uncharacteristic fire regimes on sagebrush plant communities, dependent wildlife species, fire-suppression costs, and ecosystem services. We also provide an overview of wildland fire coordination efforts among federal, state, and tribal entities.


Introduction
Historical fire regimes and their impacts on landscape-scale abundance and distribution of sagebrush ( Artemisia spp.) ecosystems are not fully understood, but they were likely highly vari-able over long timeframes and among different sagebrush communities. In recent decades, uncharacteristic fire frequency and behavior caused by the influx of non-native annual grasses (hereafter "invasive annual grasses"; e.g., cheatgrass [Bromus tectorum] ) have become the largest threat to western United States sagebrush landscapes. From 1984 to 2020 alone, wildfires have burned > 9 million ha (22.3 million acres) of sagebrush-dominated lands, primarily in the Great Basin ( Fig. 1 ). Many of these recently burned areas have been affected by altered fire regimes, reflecting more frequent, more severe, and/or larger wildfires than occurred before EuroAmerican settlement ( Balch et al. 2013 ;Bukowski and Baker 2013 ;Brooks et al. 2015 ).  ( Table A.1 ) that burned within the sagebrush biome. Sagebrush dominated areas burned are shown in red and depict where fires have burned in sagebrush dominated communities (light blue). Dark gray represents area burned either outside of the sagebrush biome or area burned within the sagebrush biome that are not sagebrushdominated communities, such as forests, woodlands, and other shrub communities. LANDFIRE Biophysical Settings selected as Sagebrush Dominated ( US Geological Survey 2014 ). Wildfire perimeter information obtained from Welty and Jeffries (2021) .
As altered fire regimes continue to expand across the sagebrush biome, the impacts are often severe for sagebrush ecosystems and sagebrush-dependent wildlife, wildland fire management, and ecosystem services. Uncharacteristically frequent wildfire is considered a major threat to many sagebrush-dependent wildlife species, such as the greater sage-grouse ( Centrocercus urophasianus ), due to loss of habitat and resulting transitions to invasive annual grass −dominated landscapes that contribute to declining wildlife populations ( Knick and Rotenberry 1995 ;Longland and Bateman 2002 ;Brooks et al. 2015 ;Coates et al. 2015Coates et al. , 2016. More fire also results in the need for more resources for fire prevention, firesuppression, and postfire restoration. Although fire-suppression costs are considerable, they only make up a portion (9%) of total wildfire costs ( Barrett 2018 ). Other wildfire costs include emergency evacuation; relief aid, ecological restoration, and rehabilitation; damage to private, commercial, and public-built infrastructure; lost income and tax revenue; housing market impacts; and long-term psychological effects ( Dale 2010 ; Thomas et al. 2017 ; Barrett 2018 ). The impact of wildfire also increases the costs of critical ecosystem services that people rely on for health and survival due to the impacts of smoke, loss of carbon storage, and erosion that affects water quality and availability ( Jones et al. 2017 ;Jones 2018 ;Flerchinger et al. 2020 ). Wildfire also impacts other important ecosystem services such as the loss of recreation opportunities, cultural traditions and sites, and existence values of wildlife and plant communities ( Loomis et al. 2001 ;Hesseln et al. 20 03 , 20 04 ;Rosenberger 2016 ).
Addressing the causes and impacts of altered wildfire regimes on sagebrush landscapes is challenging but essential to long-term conservation and retention of the multiple resources provided by these landscapes. To set the stage for this special issue on rangeland wildfire, we provide a synthesis on historical and contemporary wildfire trends in sagebrush-dominated landscapes and review the impacts of uncharacteristic fire regimes on sagebrush ecosystems and dependent wildlife, fire-suppression efforts (via direct and indirect costs), and ecosystem services. We also provide an overview of national-level strategies used to address the challenges associated with fire coordination effort s among federal, st ate, and tribal entities. Because this synthesis covers a range of topics associated with wildfire impacts to natural and human communities in the sagebrush biome, it will be useful for informing land management planning and decision-making processes.

The Role of Wildfire in Sagebrush Ecosystems
Fire is an important natural disturbance in most terrestrial ecosystems that influences biological diversity, patterns of succession within natural plant communities, and ecological function over time and space. Historical fire regimes in the sagebrush biome likely varied in large part because of the influence of climatic gradients on fuel loads and ignition rates. Sagebrush ecosystems in the eastern part of the sagebrush biome are less prone to fire ignition depending on the timing of summer or monsoonal precipitation, whereas in the western part of the sagebrush biome, most precipitation occurs in the winter and summers are dry. Differing precipitation patterns also occur along elevational and latitudinal gradients, with generally hotter-drier conditions in the south and at lower elevations and cooler-wetter conditions in the north and at higher elevations ( Chambers et al. 2014a ). In response, fuel loads and fire activity vary considerably depending on geography, seasonal precipitation, and ignition patterns. More fire generally occurs where summers are drier than winters, lightning frequency is higher, and fuels are more continuous and less limiting.
As Euro-Americans settled the West, Native American land use practices, such as burning-which is thought to have been relatively common in some landscapes ( Griffin 2002 ;Stewart 2002 ;McAdoo et al. 2013 )-were replaced with new land use practices, such as widespread livestock grazing, mining, and road building. Fire suppression and sagebrush removal were also introduced. After the introduction of extensive livestock grazing in the late 1800s, fine fuels were substantially reduced across many sagebrush landscapes, and fires likely became less frequent and burned with less intensity in many areas ( Miller et al. 2011 ) until the subsequent spread of invasive annual grasses provided contiguous, fine-fuel loads. Currently, these fire-adapted, invasive grasses dominate several million hectares in the Great Basin ( Balch et al. 2013 ;Smith et al. 2022 ), and cheatgrass cover of ≥ 15% is found on more than 21 million ha in the western half of the sagebrush biome . These changes contribute to altered fuel characteristics and ignition patterns that significantly alter sagebrush fire regimes over vast areas and threaten sagebrush ecosystems and associated wildlife species, particularly in the western half of the sagebrush biome ( US Department of the Interior 2015 ).
Cheatgrass and other invasive annual grasses can fill the interspaces between native perennials and facilitate fire spread where it would not otherwise occur, especially in arid regions where native plant productivity is low ( Whisenant 1990 ). Invasive annual grasses also senesce and dry out earlier than most native vegetation, elongating wildfire seasons ( Keane et al. 2008 ;Davies and Nafus 2013 ;Bradley et al. 2018 ). Invasive annual grasses are of particular concern for more arid sagebrush shrublands dominated by Wyoming big sagebrush (Artemesia tridentata wyomingensis) and basin big sagebrush ( A. t. tridentata;Brooks et al. 2016 ). These sagebrush communities are not adapted to frequent fire and have a low resistance to cheatgrass invasion ( Chambers et al. 2014a( Chambers et al. , 2014bBrooks et al. 2016 ;Chambers et al. 2017 ). These conditions can result in greatly reduced fire-free intervals that encourage cheatgrass establishment while preventing reestablishment of the native plant community. This dynamic leads to a self-perpetuating invasive grass-fire cycle ( D'Antonio and Vitousek 1992 ) that favors the dominance and spread of invasive annuals, which in turn facilitates more frequent fire ( Brooks et al. 2004 ;Brooks 2008 ).

Recent Wildfire Trends and Patterns
Commonalities among disparate fire regime studies have emerged, and key fire regime trends analyzed by several of these studies are summarized across three organizational levels: 1) the broader sagebrush biome; 2) among ecoregions or floristic provinces ( Fig. 2 ); and for dominant sagebrush taxa ( Fig. 3 ). Fire regime attributes discussed here include trends in fire area (i.e., area burned), fire intervals (that is, fire rotation and mean fire return intervals), fire size, fire season length, and fire recurrence (reburns).

Fire Area
Across all fire-history studies relevant to sagebrush ecosystems, most have generally concluded that fire area (i.e., area burned) over the past approximately 30 yr has increased in some regions (as reviewed by Crist et al. 2021 ). However, there is mixed agreement regarding landscape trends in area burned owing to different spatial and temporal extents, ecosystem delineations, statistical approaches, and datasets used. Thus, it is important to note that direct comparisons must be considered carefully. For instance, one study found no significant trends in total area burned over a 25yr period (1984-2008) across the sagebrush biome ( Baker 2013 ). However, using different methods and a slightly longer period of record , Brooks et al. (2015) found a potentially significant upward trend in total area burned across the sagebrush biome.
In addition, detection of ecoregional trends in area burned has also varied among studies. Miller et al. (2011) found a weak but significant upward trend in area burned in four of five floristic provinces (Northern Great Basin, Southern Great Basin, Silver Sagebrush, and Wyoming Basin; see Fig. 2 ). Baker (2013) found significant trends in only two of seven provinces (Colorado Plateau and Columbia Basin), although three others (Silver Sagebrush, Snake River Plains, and Southern Great Basin) were nearly significant. In another study, Brooks et al. (2015) found strong evidence of increased fire area in the Wyoming Basin, Snake River Plain, Columbia Basin, and Great Plains (comparable with the Silver Sagebrush Province) but not in the Northern Great Basin, Southern Great Basin, and Colorado Plateau. These discrepancies among studies are not surprising given methodological differences and the different periods of record considered and because of the substantial limitations of statistical trend detection for a short record of time relative to high interannual variability in area burned over the long term.
Despite differences among studies, there are key points of agreement to highlight in area-burned trends and patterns. First, most studies documented general upward trends in annual fire area across the sagebrush biome, even if an increase was not detected as significant or across all floristic provinces. There is also agreement in some ecoregional trends of increasing area burned, especially for the Columbia Basin and somewhat for the Silver Sagebrush (Great Plains) floristic provinces. Second, these studies document that a disproportionately larger area has burned in the western region of the sagebrush biome than in the eastern region. Twenty-one percent of the total wildland area (i.e., landscapes mostly dominated by sagebrush and other semiarid ecosystems) in the western half of the sagebrush biome burned during 1984-2013 (17% when considering repeatedly burned area only once), representing 82% of the total burned area within the entire range of the greater sage-grouse , a sagebrush-obligate bird species of conservation concern Coates et al. 2016 ). In contrast, only 5% of the total wildland area in the eastern half of the sagebrush biome burned during that same period . Third, much of the total area burned over time occurred during the relatively few years with exceptionally large fires. Temporal patterns of large fires within bioclimatic regions suggest the strong influence of interannual climate variability on area burned ( Littell et al. 2009 ;Balch et al. 2013 ;Pilliod et al. 2017 ). Wet years increase fine fuel biomass, and this often leads to increases in the amount of area burned when followed by normal to dry years in sagebrush ecosystems ( Balch et al. 2013 ;Pilliod et al. 2017 ;Smith et al. this issue).

Fire Intervals
The time between fires, or fire interval (often quantified as either a mean fire return interval or mean fire rotation to characterize the fire regime over time for either a given point or for a landscape area, respectively), has great importance for the sustainability of sagebrush ecosystems. This is particularly important if average fire-free intervals are too short for sagebrush plants to regenerate and provide adequate habitat conditions for sagebrushdependent wildlife. Most sagebrush taxa are slow to recover after fire because of limited seed dispersal, inability to resprout, and poor seed viability ( Young and Evans 1989 ;Miller et al. 2011 ). Several studies have documented that sagebrush recovery to near preburn cover after fire can take anywhere from a few decades to more than a century depending on geography, seasonal pre-cipitation, and soil types (e.g., Welch and Criddle 2003 ;Lesica et al. 2007 ;Shinneman and McIlroy 2016 ;Chambers et al. 2017 ). Sagebrush landscapes were characterized by large patches of both dense and scattered sagebrush, as well as large, grass-dominated areas based on historical General Land Office Survey data from the late 1800s to the early 1900s ( Bukowski and Baker 2013 ). Before Euro-American settlement, small fires likely occurred more often and large fires were more infrequent within sagebrush stands. This resulted in dynamic sagebrush landscapes that alternated between periods of ecosystem recovery and more extensive maturity, punctuated by fine-scale disturbance effects ( Bukowski and Baker 2013 ).
A key issue is whether modern fire intervals for sagebrush ecosystems are different from historical intervals and whether differences between the two suggest fire regimes have departed from their historical ranges of variability, thus limiting or prohibiting sagebrush recovery after fire. Modern fire intervals (post-1980) among floristic regions and sagebrush communities have been more accurately assessed than older fire intervals by using consistently and accurately mapped fire perimeter data. Contemporary fire intervals are likely shorter than older historical intervals in many but not all sagebrush ecosystem and regions ( Baker 2013 ;Brooks et al. 2015 ). Modern fire intervals have shortened for some big sagebrush (A. tridentata) communities in the west- ern part of the sagebrush biome compared with historical fire intervals based on land-survey data, particularly for Wyoming big sagebrush, where historical fire rotations likely exceeded 200 yr in most regions ( Bukowski and Baker 2013 ). In cheatgrass-dominated areas of the Great Basin, contemporary fire return intervals averaged just 78 yr ( Balch et al. 2013 ), and areas with ≥ 15% of cheatgrass cover were found to have burned twice as often as areas with less cheatgrass  ).

Fire Size
The number and/or area burned by large fires ( > 405 ha) increased with time throughout most ecoregions of the western United States between 1984 and 2011, and many regions also demonstrated trends toward larger fire sizes ( Dennison et al. 2014 ). Specific to the sagebrush biome, Baker (2013) compared the top fire years in sagebrush types in the western United States based on the total area burned over two consecutive 12-yr periods (1985 −1996 and 1997 −2008) and suggested that fire sizes may be increasing. Brooks et al. (2015) found that larger fires generally occurred in the Northern Great Basin, Snake River Plain, and Southern Great Basin compared with other regions, and notable and significant upward shifts in annual fire-size distributions occurred throughout the western region (but not the eastern region) of the sagebrush biome over a recent 30-yr period (1984-2015). Most (39 of 50) of the largest fires that occurred in the Great Basin during 1980-2008 were associated with cheatgrass, suggesting a significant conversion to a grass-fire cycle in that region ( Balch et al. 2013 ). Increases in fire sizes on the Snake River Plain have also been attributed to extensive cheatgrass invasion ( Knapp 1998 ).

Fire Season
At national and regional scales, studies suggest that the increasing prevalence of human-ignited wildfires and climate change are contributing to longer fire seasons and increased duration of fire-weather conditions ( Abatzoglou and Williams 2016 ;Balch et al. 2017 ;Syphard et al. 2017 ). The median discovery date for human-started fires was more than 2 mo earlier than lightningstarted fires nationwide, and the most common day for humanstarts was July 4 ( Balch et al. 2017 ). Human fire ignitions also had a stronger influence on lengthening fire seasons than climate change ( Syphard et al. 2017 ). For the North American desert region (which includes the Great Basin and much of the sagebrush biome), human-ignited fires expanded the wildfire season length by 230% ( Balch et al. 2017 ).
In the western United States, Dennison et al. (2014) did not find significant trends in large fire start dates between 1984 and 2011 Table 1 Frequency and area burned of sagebrush dominated types within the sagebrush biome from 1984 to 2020. Fire data obtained from Welty and Jeffries (2021) . Sagebrush dominated pixels selected from LANDFIRE Biophysical Settings ( US Geological Survey 2014 , across large ecoregions. Specific to the sagebrush biome, significant increases in fire-season lengths were detected over a recent 30yr period for the Southern Great Basin, Wyoming Basin, and Great Plains (i.e., Silver Sagebrush) floristic provinces .
Increasing fire-season length in the Southern Great Basin may be of particular concern considering the relatively low resilience of sagebrush ecosystems to fire in that region. Fires starting earlier in the season may be driven by invasive annual grasses, which dry out about a month earlier than most native herbaceous species  ).

Fire Recurrence-Reburns
Since 1984, within the sagebrush biome, approximately 2.4 million ha have experienced multiple wildfires with some areas reburning 2 −3 × and a few areas up to 8 × during the 37-yr time frame ( Table 1 , Fig. 3 ). As changing fuels, ignition rates, and climate conditions promote greater annual and cumulative area burned and shorter fire intervals, the probability of specific parts of the landscape burning repeatedly also increases. As fire recurrence over a given time-period increases, conditions become more suitable for the persistence of annual plants, such as cheatgrass, and less suitable for the persistence of woody perennials, such as sagebrush, resulting in a high probability of transitioning to a grass-fire cycle ( Whisenant 1990 ;D'Antonio and Vitousek 1992 ).
A previous study by Brooks et al. (2015) showed that of the 1.4 million ha of recurrent fire area reported for greater-sage grouse management zones during 1984-2013, roughly two-thirds of that area occurred in the Snake River Plain, constituting approximately 25% of that ecoregion's total fire area and approximately 8% of its big sagebrush area. Most of that recurrent fire area burned twice (71%), resulting in an average fire return interval of 15 yr for those areas, and the remainder (29%) burned ≥ 3 × for an average fire return interval of 7.5 yr or less. However, the region with the highest percentage (34%) of its fire area classified as recurrent was the Columbia Basin, potentially indicating a similar risk of conversion to a grass-fire cycle as the Snake River Plain. These two provinces have some of the highest proportions of landscapes with low resilience to fire and low resistance to cheatgrass invasion, especially in Wyoming big sagebrush and other low-productivity sagebrush communities ( Chambers et al. 2014a( Chambers et al. , 2014b( Chambers et al. , 2017.

Human-Caused Wildfires
Although the ratio of human ignitions to lightning ignitions are found to be lower across the sagebrush biome in comparison with most other regions of the United States ( Balch et al. 2017 ), the proportion large fires caused by humans across the sagebrush biome can still be substantial ( < 10%) ( Nagy et al. 2018 ). Human ignitions typically occur near wildland-urban interfaces and require a substantial fire suppression response for community protection. There are many causes of human ignitions, such as campfires left unattended, target shooting, powerlines, fireworks, debris burning, intentional arson, and heat and sparks from vehicles and equipment. Although each ecoregion of the sagebrush biome likely has its own unique set of common human ignitions, two common causes are powerline failures in areas with improper clearance and roadside ignitions along highways and major roads bordered by hot, dry, invasive grass fuels in the late spring and summer season. Human ig-nitions could be reduced with targeted fire-awareness and prevention programs focused on the causes of human ignitions in sagebrush ecosystems, particularly if delivered to local communities surrounding public lands that have higher fire risk ( McCaffrey et al. 2012 ;Hamilton et al. 2018 ;Butry and Prestemon 2019 ;Meldrum et al. 2019 ).

Postfire Recovery of Sagebrush Ecosystems
Historical fire frequencies are reasonably presumed to have been generally compatible with postfire succession rates of corresponding sagebrush ecosystems. Specifically, minimum or average fire-free intervals can be roughly estimated on the basis of the time required for sagebrush community recovery, which can range from 25 to well over 100 yr depending on precipitation, edaphic conditions, and other environmental factors ( Baker 2006( Baker , 2011Nelson et al. 2014 ;Shinneman and McIlroy 2016 ). However, there is relatively limited information on successional processes and composition dynamics for intact sagebrush ecosystems as they recover after fire over time.
Big sagebrush is intolerant of fire and has a limited seedbank life. Often, it must recover from seed sources on the perimeter of the fire ( Young and Evans 1989 ;Applestein et al. 2022 ). Lowerelevation sagebrush ecosystems are hotter and drier than higher elevation ecosystems, and recovery from a fire is expected to be exceedingly slow in comparison ( Winward and Tisdale 1977 ;West et al. 1978 ;Winward 1980 ). What information is available is relatively short term compared with how long it may take for sagebrush recovery in these hotter-drier sites. After 23 yr, sagebrush recovery was only 2% in Wyoming big sagebrush communities in Montana ( Lesica et al. 2007 ). Thus, when full recovery will occur is generally unknown and likely to vary by a suite of factors. For example, recruitment (germination and survival of seedlings) of sagebrush at lower elevations is greater with above-average coolseason precipitation ( Maier et al. 2001 ).
Recovery of other shrub species immediately after fire is variable, but most recover more rapidly than sagebrush due to their sprouting ability. For example, if green or rubber rabbitbrush ( Ericameria teretifolia and E. nauseosa, respectively) were present in the prefire community, they often increased in abundance and cover after fire ( Beck et al. 2009 ;Davies et al. 2009 ). Over time, as sagebrush eventually seeds in from adjacent unburned areas and redominates the plant community, shrubs like rabbitbrush are often outcompeted and reduced ( Young and Evans 1974 ). Other resprouters also increase after fire in sagebrush communities such as prickly phlox (Linanthus pungens) ( Young and Evans 1974 ). Antelope bitterbrush (Purshia tridentata) can sprout after fire or experience significant mortality depending on fire intensity, season, and other site-specific factors ( Clark et al. 1982 ). The response of herbaceous understory vegetation to wildfire varies with differences in species composition, preburn site condition, fire intensity, and prefire and postfire patterns of precipitation ( Miller et al. 2013 ).

Impacts of More Frequent Wildfire on Sagebrush Ecosystems and Postfire Recovery
Lower-elevation sagebrush ecosystems have a greater risk of postfire invasive annual grass invasion and dominance than higher-elevation big sagebrush ecosystems ( Chambers et al. 2014a ). This risk is significantly greater if native perennial grasses have been reduced ( Chambers et al. 2007 ). Thus, prefire composition of sagebrush plant communities is an important factor for determining postfire recovery. If native perennial grasses and forbs dominate the community before fire, they are likely to dominate the community after fire ( Bunting 1985 ;Rhodes et al. 2010 ;Bates et al. 2013 ). If native perennial grass and forb cover were low and invasive annual grasses already existed in the community before the fire, invasive annual grasses are likely to dominate the postfire community ( Young and Evans 1978 ;Hosten and West 1994 ;Chambers et al. 2007).
Fuel loading (the amount of fuel available to burn) can influence fire severity and postfire recovery in sagebrush ecosystems. In Wyoming big sagebrush communities in Oregon, the accumulation of invasive fine fuels increased fire-induced mortality of native perennial bunchgrasses and led to a substantial postfire invasive annual grass invasion ( Davies et al. 2009( Davies et al. , 2016. Invasive annual grass dominance of lower-elevation sagebrush ecosystems likely indicates a permanent shift in the plant community without restoration interventions ( D'Antonio and Meyerson 2002 ;Bagchi et al. 2013 ). Substantial invasive annual grass invasion prevents sagebrush reestablishment because it can increase fire frequency to the point that sagebrush cannot reach maturity (i.e., produce seed) before the next fire occurs. As a result, the sagebrush seedbank is depleted ( D'Antonio and Vitousek 1992 ; Rossiter et al. 2003 ). Invasive annual grass competition for soil moisture can also prevent sagebrush establishment ( Booth et al. 2003 ).
Increased fire frequency favors invasive annual grasses and is detrimental to native floras that are not adapted to frequent fire ( D'Antonio and Vitousek 1992 ). This creates a positive feedback cycle between fire and continued invasive annual grass dominance (grass-fire cycle) of the community ( D'Antonio and Vitousek 1992 ; Rossiter et al. 2003 ;Shinneman et al. 2021 ). Thus, the effects of increased fire frequency cannot be separated from the effects of invasive annual grass invasion. Invasive annual grass invasion exponentially decreases plant community biodiversity and native perennial species abundance ( Davies and Svejcar 2008 ;Davies 2011 ). Invasive annual grasses use soil water earlier than native plants ( Melgoza et al. 1990 ), resulting in vegetation drying out as much as a month earlier than it would have if invasive grasses were not present ( Davies and Nafus 2013 ). This allows earlier-season wildfires to occur ( Davies and Nafus 2013 ) at a time when native bunchgrasses are more susceptible to fire ( Wright and Klemmedson 1965 ;Britton et al. 1990 ;Davies and Bates 2008 ). Frequent fire in lower-elevation sagebrush ecosystems results in a threshold being crossed to an annual grass −dominated state that has proven to be exceedingly difficult and expensive to reverse at a meaningful scale for conservation and land management Miller et al. 2011 ). Therefore, there is a substantial risk that lower-elevation sagebrush ecosystems will not recover from fire.
Increased incidences of large and more severe fires (no patches of unburned areas within fire perimeter; Adams 2013 ) potentially pose an additional challenge to timely natural recovery of sagebrush ecosystems. Sagebrush seeds only disperse a few meters from the parent plant ( Young and Evans 1989 ), and a sagebrush seed source may be many kilometers away from the interior of large wildfires. Therefore, if sagebrush does not establish from seed in the first yr or 2 post fire, the sagebrush seed bank will be depleted ( Young and Evans 1989 ;Wijayratne and Pyke 2009 ) and sagebrush will need to disperse from the exterior of these large fires. Sagebrush establishment from the seedbank after wildfire seems moderate to exceedingly unlikely following the environmental gradient from cool and wet to hot and dry sagebrush ecosystems ( Baker 2006 ;Lesica et al. 2007 ;Ziegenhagen and Miller 2009 ;Nelson et al. 2014 ). How long it takes the sagebrush seedbank to disperse and establish into the interiors of these large fires is unknown. However, it is likely to significantly lengthen the time for sagebrush recovery.

Impacts of More Frequent Wildfires on Wildlife
Uncharacteristic frequent fire has implications for many wildlife species that are dependent on sagebrush for their survival because the resulting landscape mosaic of burned and unburned areas affects wildlife habitat availability and connectivity until native plants recover. Owing to the delay in sagebrush recovery in some regions, large and frequent fires that lead to extensive loss of sagebrush cover will likely have negative effects on wildlife populations over longer periods of time ( Longland and Bateman 2002 ;Coates et al. 2015 ). In addition, remaining unburned areas may be too small to support the habitat requirements of some sagebrushdependent wildlife (e.g., Castrale 1982 ;Kerley and Anderson 1995 ;Knick and Rotenberry 1995 ). Despite multiple studies being available on the impact of wildfire on many sagebrush-obligate species, they are often limited in scope and results vary over different spatial and temporal scales. Some studies have identified direct relationships between sagebrush obligates and wildfire, but findings are often limited in their scope to local sites ( Connelly et al. 20 0 0 ), movements and habitat associations ( Fischer et al. 1996( Fischer et al. , 1997Nelle et al. 20  Many studies suggest that large-scale changes in low-elevation sagebrush habitat associated with fire have had a negative influence on sagebrush-obligate species. Although the majority of studies addressing the effects of fire on sagebrush bird communities have been short term ( < 5 yr; Knick et al. 2005 ), most studies have found negative effects of fire on population trends and abundance for sagebrush-obligate and/or dependent avifauna, including the sagebrush sparrow ( Artemisiospiza nevadensis ;Welch 2002 ;Reinkensmeyer et al. 2007 ;Earnst et al. 2009 ( Kerley and Anderson 1995 ;Knick and Rotenberry 1995 ) that are dependent on infrequent fire regimes. The long-term effects of uncharacteristic frequent fire on many of these species is relatively unknown. However, Holmes and Robinson (2013) found that the impact of fire on bird abundance in mountain big sagebrush communities persisted for at least 2 decades.
Wildfires can impact pygmy rabbits (Brachylagus idahoensis) directly through mortality and indirectly through habitat modification by depletion of concealment cover and food resources, fragmentation of sagebrush habitat, and facilitating the invasion of in-   ( Fig. 4 ). During that same period, 1.6 million ha ( ≈4 million acres), or 11% of areas modeled as highly suitable habitat for pygmy rabbits , have burned ( Fig. 5 ). Almost all this impact occurred within the Great Basin (see Fig. 4 ).
While the specific effects of wildfire on other sagebrush wildlife species (e.g., big game species, other small mammals, sagebrushdependent amphibians and reptiles) remains unclear, the response of sagebrush-obligates to other forms of habitat disturbance may give some indication of how these species will respond to wildfirecaused habitat disturbance. Widespread sagebrush removal treatments (including fire, mechanical, or chemical treatments) that reduce shrub dominance or reduce fine fuels in sagebrush ecosystems can result in significant declines in sagebrush-obligate bird species ( Magee et al. 2011 ) and may be detrimental to pygmy rabbits owing to their reliance on sagebrush ( Wilson et al. 2011 ;Woods et al. 2013 ). Additionally, many native small mammals (e.g., dark kangaroo mouse [Microdipodops megacephalus] ) may be at risk of extirpation owing to fragmentation of sagebrush habitats ( Hanser and Huntly 2006 ; Hafner and Upham 2011 ).
Wildfire is considered the largest threat across the western part of the sage-grouse range, particularly in the Great Basin Coates et al. 2015 ), contributing to declining sagegrouse populations ( Coates et al. 2016 ). Fire occurring within the range of sage-grouse can cause direct loss of habitat, resulting in negative effects to breeding, nest site selection and nest success, brood survival, feeding, and sheltering opportunities for the species ( Call and Maser 1985 ;O'Neil et al. 2020 ;Anthony et al. 2021, Brussee et al. 2022, Dudley et al. 2022. Large-scale wildfires have also been shown to negatively impact survival ( Lockyer et al. 2015 ;Foster et al. 2019 ), lek attendance ( Steenvoorden et al. 2019 ), and population growth ( Dudley et al. 2021 ). In addition to the direct habitat loss, fire can also create a functional barrier to sage-grouse movements and dispersal that compounds the influence wildfire can have on populations and population dynamics ( Fischer et al. 1997 ;Crist et al. 2017but see O'Neil et al. 2020Brussee et al. 2022 ). In some cases, fire can isolate sage-grouse populations, thereby increasing their risk of extirpation Wisdom et al. 2011 ). The frequency, size, and severity of fires is increasing, and fire has cumulatively removed a significant and growing amount of sage-grouse habitat ( Fig. 6 ), particularly in the Great Basin. Since 1984, nearly 9 million ha ( ≈22 million acres) within the greater sage-grouse range have burned and 74% of the area burned occurred within the Great Basin (see Figs. 5 and 6 ). A fire-threats assessment indicates that threats of too much fire are higher in the western portion of the sagebrush biome than in the eastern portion of the range , raising concern in the western region of sage-grouse range. Overall, these findings corroborate models that projected approximately one-half of the current pop-ulation of sage-grouse will remain in the Great Basin by the mid-2040s if current fire trends continue unabated .

Impacts of More Frequent Wildfires on Ecosystem Services
Beyond the direct costs of fire suppression, more frequent wildfires can affect people in three main ways: 1) impacts to sagebrush Figure 6. Wildfires from 1984 to 2020 that occurred within sage-grouse ( Centrocercus spp.) range and greater sage-grouse (C. urophasianus) priority habitat management areas. Wildfire information from Welty and Jeffries (2021) . Greater sage-grouse data obtained from US Fish and Wildlife Service (2014) . The Gunnison sage-grouse (C. minimus) range used data from Schroeder et al. (2004) . Priority habitat management area data obtained from the Bureau of Land Management (2019) . DPS indicates distinct population segment.

Table 2
Summary of main potential effects of wildfire on the nonmarket goods and services provided by sagebrush ( Artemisia spp.) ecosystems (Modified from Venn and Calkin, 2011 ,  ecosystems and wildlife; 2) changes in critical services that people rely on for health and survival; and 3) impacts to recreational and cultural resources ( Table 2 ). However, few studies quantify the effects of wildfires on ecosystem services (the way natural systems provide benefits to people, such as forage for livestock, water filtering, and so on [ Boyd and Banzhaf 2007 ;Brown et al. 2007 ;Calkin 2008 , 2011 ;Milne et al. 2014 ;Pereira et al. 2021 ]), and to the authors' knowledge, no published studies summarize the effects of wildfires on ecosystem services across the sagebrush biome. Impacts on wildlife populations and plant communities affect people if society uses-or values the existence of-the species or communities affected. Fire can either increase or decrease the availability of forage for livestock. Ranchers tend to face higher costs if they must adjust forage following a fire, and the likelihood of ranchers going out of business goes up with increasing fire frequency ( Brunson and Tanaka 2011 ). These costs can be significant; one case study of 647 520 ha (1.6 million acres) burned in northern Nevada in 1999 estimated $12.8 million in losses from lost livestock output, livestock deaths, and damage to fences ( Riggs et al. 2001 ). Economists also recognize nonuse existence values, which refer to the benefits society derives from the survival of a species or plant community, independent of any associated active uses (e.g., Freeman 2003 ;Segerson 2017 ). Although no published study estimates the existence value of sagebrush-obligate species ( Eiswerth and van Kooten 2009 ), studies have estimated the nonmarket value of protecting habitat for other charismatic bird species as approximately $15 to $60 per household, depending on the species ( Richardson and Loomis 2009 ).
Wildland fires can impact critical services that people rely on for health and survival. Fires release smoke and carbon into the atmosphere. Wildfire smoke can affect human health and welfare directly through induced illness and through costs associated with behaviors taken to avoid that exposure, such as deferred recreation or exercise, and increased air conditioning usage ( Richardson et al. 2012 ;Jones 2018 ). The long-term net effect of fires on total storage of carbon in sagebrush ecosystems is not well understood, especially regarding the long-term response to multiple fires or ecosystem transition to other states ( Miller et al. 2013 ;Fellows et al. 2018 ;Flerchinger et al. 2020 ). Between 7 and 97 metric tons of carbon can be stored annually per acre of shrub/scrub ecosystems ( Lacelle 1997 ;Wilson 2010 ), but some sagebrush ecosystems have been found to be carbon neutral in a year with a strong postrain summer respiration pulse ( Flerchinger et al. 2020 ), and plant communities dominated by annual grasses (cheatgrass and mustard) can turn into net carbon sources during times of severe summer drought ( Prater et al. 2006 ). After a wildfire in sagebrush ecosystems, runoff may increase by 3 −7 × over small-to large-plot scales, respectively, and erosion may increase by 40 −125 × over small-to large-plot scales, respectively ( Miller et al. 2013 ). This in turn can lead to sedimentation of water resources, debris flows, and chemical water quality changes and can potentially result in reservoir dredging costs, infrastructure damage, losses to fishery habitats, and increased water treatment costs ( Haas et al. 2016 ;Jones et al. 2017 ).
Finally, altered fire regimes can affect cultural and recreational resources valued by society. Changes to cultural traditions and practices, or damage to culturally important artifacts and sites, could amount to substantial losses, particularly if irreplaceable resources are threatened. However, such impacts to cultural heritage are not particularly amenable to valuation and even less so to generalization. Few studies attempt to value cultural assets, and most that do focus on historical buildings, monuments, and artifacts ( Venn and Calkin 2011 ). Wildland fires can affect recreation through closures of trails and recreation areas, changes to opportunities for hunting and wildlife viewing, and changes to the aesthetics of recreation areas. Due to varied contexts and preferences, wildfires have been found to either increase or decrease people's usage of the burned areas ( Englin et al. 1996 ;Loomis et al. 2001 ;Hesseln et al. 20 03 , 20 04 ). While estimating related damages from wildfires would require accounting for substitution across sites and activities in response to site closures and changing site characteristics, estimates for time spent hunting range from about $40/d for small game species to hundreds of dollars per day for big game species ( Huber et al. 2018 ) and estimates for time spent wildlife viewing, hiking, and off-highway vehicle use similarly are about $60, $78, and $76 daily, respectively ( Rosenberger 2016 ).

Fire-Suppression Costs
Nationwide, fires burned 4.1 million ha (10.1 million acres) of land in 2020, which is over 1 million ha (2.6 million acres) more than the 10-yr average. Within the Great Basin States of Idaho, Nevada, Utah, and extreme western Wyoming, 679 896 ha (1.68 million acres) or 80% of the 849 870 ha (2.1 million acres) burned in 2018 were in areas identified as sage-grouse habitat, meaning high-quality sagebrush ( Figs. 5 and 6 ). As fires become more severe and consume more area, fire-suppression costs rise. In 2018, firesuppression costs were the highest on record for both the USDA Forest Service and DOI agencies, exceeding $3.1 billion ( National Interagency Fire Center 2021 ), more than 5 × the amount spent in 1985 when adjusted for inflation.
Several factors lead to the rising costs of combating wildfires, including climate change and an increasing prevalence of humanignited wildfires contributing to longer fire seasons as noted earlier ( Abatzoglou and Williams 2016 ;Balch et al. 2017 ;Syphard et al. 2017 ). For the sagebrush biome, however, one important factor increasing the cost of fire suppression is the proliferation of invasive annual grasses, which allows fire to spread rapidly and resist suppression effort s. Fine fuel loadings may increase as much as 20 0-30 0%, as was the case in 2018 in the northern Great Basin ( Newmerzhycky and Law 2018 ). Combined with substantial carryover of fine fuels as litter from the year prior, hot temperatures, low relative humidity, and windy conditions, fire behavior becomes explosive. Extreme fire behavior has been observed on recent fires in these sagebrush and invasive annual grass fuel types with several fires growing more than 8 0 0 0 ha (20 0 0 0 acres) in a 24-h burn period. Rapid rates of spread and high flame lengths during windy conditions often prevent a direct-attack strategy and greatly reduce the effectiveness of fire lines and fuel breaks because conditions are too dangerous to place resources.
In the remote and rugged topography of the West, wildfire can expand beyond conditions for initial attack because of the considerable amount of time required for resources to reach a reported fire. This can lead to the need for significant additional aerial and ground resources. For example, in July 2018, the Martin Fire, the largest single fire in Nevada's history, burned more than 176 0 0 0 ha (435 0 0 0 acres) within a 5-dI period. Other large sagebrush fires have also burned thousands of acres and cost millions of dollars for fire suppression.
The United States Congress funds the annual fire-suppression accounts based on a rolling 10-yr average, but with the consistently rising costs of suppression, Congressional funding is often not enough during the fire season. Congress recently approved a measure in 2020 and raised the funding cap on suppression, allowing federal agencies to tap into Federal Emergency Management Agency funds to fight catastrophic fires.

Other Costs Associated with Wildfire
In addition to fire-suppression costs and changes in ecosystem services, wildfires can also induce numerous other costs largely beyond the scope of this assessment, including restoration and rehabilitation costs; damage to private, commercial, and public-built infrastructure; and indirect costs such as lost income, lost tax revenues, housing market impacts, and long-term psychological effects ( Dale 2010 ;Thomas et al. 2017 ;Barrett 2018 ). However, available information is sparse. For example, a recent analysis estimates the total annualized cost of wildfires in the United States at anywhere from $71.1 billion to $347.8 billion USD in 2016 but also notes substantial gaps in underlying data ( Thomas et al. 2017 ). Suppression costs paid by state and federal agencies, although substantial, typically make up only a small portion of total wildfire costs. One recent review ( Barrett 2018 ) attributes only 9% of total wildfire costs to suppression, whereas approximately 35% of costs are attributed to short-term expenses such as relief aid, evacuation services, and home and property loss, and 65% of total costs are attributed to long-term damages to ecosystems and people.

Current Coordination Effort s Among Federal, State, and Tribal Entities to Address Fire
The challenge of managing wildland fire in the United States is increasing in complexity and magnitude, and not one agency has the resources to address the growing issues and concerns associated with wildland fire. For all agencies, cooperation and coordination is accomplished through cooperative interagency agreements at all levels, from the national, state, and tribal levels down to the local-community level. The Federal Wildland Fire Management Policy ( US Department of the Interior and US Department of Agriculture 1995 ) states that fire management planning, preparedness, prevention, suppression, restoration and rehabilitation, monitoring, research, and education will be conducted on an interagency basis with the involvement of many cooperators and partners.
The National Cohesive Wildland Fire Management Strategy ( US Department of the Interior and US Department of Agriculture 2014 ) outlined new approaches to coordinate and integrate efforts to restore and maintain resilient landscapes, prepare communities for wildland fire, and better address the Nation's wildland fire threats. The Integrated Rangeland Fire Management Strategy ( US Department of the Interior 2015 ) was developed specifically to address the issues of uncharacteristic sagebrush fire and invasive annual grasses through broader wildland fire prevention, suppression, and restoration efforts and to ensure improved coordination with local, state, tribal, and regional efforts. Though cooperative agreements vary by state in terms of their policies and guidelines, state resources are often used to augment and support suppression efforts on federal lands; similarly, federal resources augment and support suppression efforts on state and private lands. While these effort s demonstrate a commitment to address sagebrush wildfires in a coordinated manner, large wildfires will continue to be a management challenge as human populations and cities continue to grow, invasive annual grasses expand, and climate change continues to increase conditions that favor more wildfire.

Conclusion
Western sagebrush ecosystems continue to be threatened by larger and more frequent wildland fires that often result in the loss of large swathes of sagebrush and facilitate invasion by invasive annual grasses. Natural sagebrush recovery times cannot keep up with the expanding invasive annual grass/fire cycle, and some areas may have crossed thresholds of no return. In response, sagebrush-dependent species that serve as indicators of ecosystem conditions are declining throughout the sagebrush biome. Uncharacteristically frequent fire in sagebrush ecosystems has a large impact on ecosystem services, including human well-being, and cultural and recreational resources. Moreover, fire-suppression costs and other costs associated with wildfire impacts will likely continue to increase. Figure 7 demonstrates how increases in wildfire frequency and extent can have negative effects, including loss of ecosystem services, a need for more fire suppression activities and resources, and other associated costs. Where feasible, enhancing efforts to manage invasive annual grasses and wildland fire may help to break the invasive grass/wildfire cycle and sustain Figure 7. A conceptual figure demonstrating the net economic benefit (positive) or cost (negative) to society based on contemporary wildfire trends. Economic costs from negative effects of altered wildfire regimes include fire-suppression costs, other direct costs, and loss of ecosystem services. Economic benefits include positive effects of wildfire regimes on these same factors. The domains (ovals) indicate the potential amount of overall negative (orange), neutral (gray) and positive effects (green) of wildfire trends on ecosystem services, based on whether wildfire frequency is uncharacteristically increasing or decreasing, or if it remains relatively compatible with historical conditions (neutral). Thus, the position of each oval indicates relatively higher or lower fire-suppression costs and other economic costs relative to altered fire regimes. fire regimes compatible with maintaining key biological resources, lowering fire suppression costs, and providing positive impacts on ecosystem services. Conversely, where fire frequency and extent are lower than historically occurred, negative effects on sagebrush ecosystems and associated ecosystem services may also occur due to increased fuel loads and altered habitat conditions for sagebrush dependent wildlife populations. This paper covered a broad range of wildland fire topics for sagebrush landscapes, discussing what is known and not well known among ecological, economic, and human dimensions. Synthesizing these diverse topics should help natural resource and wildland fire managers to address more effectively: 1) the historical and ecological roles of wildfire in relation to current wildfire trends and the negative impacts caused by uncharacteristic wildfire and 2) the challenges and opportunities for wildland fire management in the face of altered wildfire regimes, including the effects on wildlife habitat, fire suppression costs, and loss of ecosystem services. This synthesis also emphasizes the need for prioritizing management aimed at addressing interactions between uncharacteristically frequent fire and invasive grass expansion in sagebrush ecosystems, and it suggests areas where more research may be needed by land managers to enhance and better inform their decision-making processes.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.