Next Article in Journal
How Does Fire Suppression Alter the Wildfire Regime? A Systematic Review
Previous Article in Journal
Experimental Study on Pedestrian Behaviors during Fire Emergency Conditions with Minecraft: Case Studies in a Classroom
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Fire
Volume 6
Issue 11
10.3390/fire6110423
fire-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Fire Regimes of Utah: The Past as Prologue

by
Joseph D. Birch
and
James A. Lutz
*
Department of Wildland Resources, Utah State University, Logan, UT 84322-5230, USA
*
Author to whom correspondence should be addressed.
Fire 2023, 6(11), 423; https://doi.org/10.3390/fire6110423
Submission received: 10 October 2023 / Revised: 25 October 2023 / Accepted: 30 October 2023 / Published: 6 November 2023
Browse Figures
Versions Notes

Abstract

:
(1) Background: Satellite monitoring of fire effects is widespread, but often satellite-derived values are considered without respect to the characteristic severity of fires in different vegetation types or fire areas. Particularly in regions with discontinuous vegetation or narrowly distributed vegetation types, such as the state of Utah, USA, specific characterization of satellite-derived fire sensitivity by vegetation and fire size may improve both pre-fire and post-fire management activities. (2) Methods: We analyzed the 775 medium-sized (40 ha ≤ area < 400 ha) and 697 large (≥400 ha) wildfires that occurred in Utah from 1984 to 2022 and assessed burn severity for all vegetation types using the differenced Normalized Burn Ratio. (3) Results: Between 1984–2021, Utah annually experienced an average of 38 fires ≥ 40 ha that burned an annual average of 58,242 ha with a median dNBR of 165. Fire was heavily influenced by sagebrush and shrubland vegetation types, as these constituted 50.2% (17% SD) of area burned, a proportion which was relatively consistent (18% to 79% yr−1). Medium-sized fires had higher mean severity than large fires in non-forested vegetation types, but forested vegetation types showed the reverse. Between 1985 and 2021, the total area burned in fires ≥ 40 ha in Utah became more concentrated in a smaller number of large fires. (4) Conclusions: In Utah, characteristic fire severity differs both among vegetation types and fire sizes. Fire activity in the recent past may serve as an informative baseline for future fire, although the long period of fire suppression in the 20th century suggests that future fire may be more active. Fire managers planning prescribed fires < 400 ha in forests may find the data from medium-sized fires more indicative of expected behavior than statewide averages or vegetation type averages, both of which are weighted to large fires.

1. Introduction

The long period of fire exclusion and anthropogenic changes to forests in the West has resulted in much less annual fire activity in the twentieth century than during the period of pre-Euro-American settlement [1]. Recent years have seen an increase in fire activity as a result of higher fuel loading and permissive climate and fostered the emergence of ‘megafires’ throughout the West [2], and the prevalence of larger fires is likely to continue [3,4]. Thus, although the recent past may not be an accurate predictor of the full magnitude of future fire activity, particularly because of the compounding effect of decades of fire suppression, it likely forms a baseline which can be useful in projecting a lower bound of future fire activity. Managers are increasingly concerned with understanding the post-fire effects on regeneration in different forest types under changing environmental conditions [5].
Fires and annual fire statistics often focus primarily on the area burned. This characterization is perhaps overly simplistic as it ignores vegetation type, unburned areas within the fire perimeter, and most importantly, the burn severity—the ecological effect of fire and one of the principal determinants of post-fire revegetation. Burn severity, whether measured by satellite or ground data, varies by vegetation type and fire history [6,7,8,9]. Burn severity is frequently measured at landscape scales as the delta normalized burn ratio (dNBR), a measure of the change of surface reflectance, largely from vegetation foliage, between the pre-fire and post-fire condition [10]. However, the ecological effect of remotely-sensed burn severity has different meanings in the context of different vegetation types [11,12]. Satellite-derived spectral data often differs from the actual ecological effect on vegetation (e.g., tree death and soil effects; [11,13,14]).
A very high mean burn severity and its concomitant ecological effects could be characteristic for some vegetation types (e.g., chapparal; [11]) but represent a departure from characteristic fire for others (e.g., ponderosa pine). Therefore, analysis of fire severity and fire severity trends should be presaged on delineation of vegetation types [6,9], a consideration that becomes more important in large, diverse landscapes with a variety of vegetation types that may be present within a large management entity, such as the state of Utah [15] or a large National Park (e.g., Yosemite National Park; [7]). Even at very large scales, the number of fires, area of fires and severity of fires will differ because of climate effects and large-scale topography [16,17] and the interannual variability in fire will be large because of the stochasticity of ignitions and fire weather.
The Monitoring Trends in Burn Severity (MTBS) project provides fire perimeter mapping and severity data for fires ≥ 400 ha in the western United States [18]. However, in many management areas, fires ≥ 400 ha are considered large and potentially unpredictable, with the result that the management decision for fires of this size near values at risk is frequently for full suppression. Medium-sized fires, here defined as those ≥ 40 ha but <400 ha, are more likely to represent fires of similar area to units being considered for prescribed fire activity. Small- and medium-sized fires are more often constrained by topography, ignition sources, seasonal moisture conditions, and fine-scale variation fuels while broad temporal or spatial filters may be responsible for large fires and their behavior [19,20]. Although any fire ignition has the potential to expand to large areas, those above 40 ha (although this is by no means a scientifically delineated size threshold) may have had a sufficient local effect to persist and grow, particularly if the period of fire suppression has led to high fuel loading. Thus, fires that have already reached ≥40 ha are sufficiently large to attract management attention lest they could expand to become large fires if weather and fuel are permissive and suppressive activities are ineffective. Trends in the number of medium-sized or large fires that were successfully managed could therefore be an indicator for increased fire activity in the future when climate and fuel loading make potential management activities less likely to succeed.
High severity fire effects are particularly important for managers as they represent the greatest departure from pre-fire conditions and may exceed ecosystem resistance or resilience [21,22]. Post-fire, the distance to surviving adult plants of reproductive stature may govern the rate of revegetation because seeds must be dispersed longer distances into the burned area. High severity fires may leave large patches of completely burned area that require restoration activity (or long periods of time to become revegetated) and they may kill otherwise fire-resistant large-diameter trees that could be the nucleus of forest revegetation and promote greater biodiversity [23,24]. Thus, the distribution and range of the highest severity fire effects may be more relevant for managers and restoration efforts than mean or median severity values.
Our objectives were to:
  • Establish a 30-year baseline of average fire activity for Utah overall and for each principal vegetation type, considering the number, the area, and the severity of fires ≥ 40 ha.
  • Identify differences in satellite-derived burn severity between medium-sized (40 ha ≤ area < 400 ha) and large (area ≥ 400 ha) fires.

2. Materials and Methods

2.1. Study Area

Utah is an arid state containing 212,761 km2 of land segmented by numerous mountain ranges with topographic relief from 664 m to 4120 m (Figure 1). The discontinuous basin and range landscape of the western half of the state represents the easternmost boundary of the Great Basin [25]. The eastern portion of the state is dominated by the Colorado Plateau and the east-west Uintah Mountains at the northeast boundary of the state. The extreme variety of topography, soils, and local climate results in a wide variety of ecosystems. Northern Utah is characterized as a semi-arid climate zone whereas southern Utah is characterized as a warmer and desert climate. Strong-seasonality and winter-dominated precipitation generally increases with latitude and elevation with shorter fire seasons at high elevations due to a short snow-free season. Much of Utah’s population lives along the foothills of the Wasatch Mountains. The resulting wilderness urban interface (WUI; [26]) occupies a pre-settlement vegetation zone characterized by short-statured trees and shrubs (e.g., Quercus gambelii and Acer grandidentatum) in dissected terrain which has the possibility for extreme fire behavior.
Vegetation varies with topography, fire, and the history of human influences, and generally transitions from low-productivity grassland and sagebrush steppe to pinon-juniper woodlands at intermediate elevations. In the mountains, pinon-juniper, scrub oak and maple woodlands, and riparian hardwood drainages exist along and in the WUI and transition to closed-canopy mixed-conifer forests with increasing elevation.
Common tree species (from lowest to highest elevation) include Juniperus osteosperma (Torrey) Little (Utah juniper), Juniperus scopulorum Sargent (Rocky Mountain juniper), Pinus monophylla Torrey and Fremont (singleleaf pinon), Quercus gambelii Nuttall (Gambel oak), Acer grandidentatum Nuttall (bigtooth maple), Pseudotsuga menziesii var. glauca (Mayr) Franco (interior Douglas-fir), Populus tremuloides Michaux (aspen), Cercocarpus ledifolius Nuttall (curl-leaf mountain mahogany), Pinus contorta Douglas ex Loudon (lodgepole pine), Pinus flexilis E. James (limber pine), Picea pungens Engelmann (blue spruce), Picea engelmannii Engelmann (Engelmann spruce), Abies concolor (Gordon and Glendinning) Hildebrand (white fir), Abies bifolia A. Murray bis (Rocky Mountain subalpine fir), and Pinus longaeva D. K. Bailey (Great Basin bristlecone pine) [30].
Fire behavior in Utah encompasses most of the fire regimes present in the western United States [31,32,33]. Fire return intervals vary from a low of 10 to 20 years in ponderosa pine and mixed-conifer stands to more than 100 years in higher elevation alpine forest types [34,35,36]. The basin and range geography tends to limit contiguous areas of similar vegetation and fuel types, potentially contributing to the relative lack of megafires (so far) compared to other western states (>10,000 ha; [2]).

2.2. Classifying Vegetation

We classified vegetation types using the 2018 30 × 30 m LANDFIRE National Vegetation Classification Existing Vegetation Type (EVT) data for Utah, [27,28]. The EVT is a national-level dataset curated by the U.S. Department of Agriculture and the U. S. Department of the Interior that classifies vegetation using a moderated classification and regression tree approach which uses a combination of plot-based data, local climate, topography, LANDSAT imagery, and temporal changes in normalized difference vegetation index (NDVI) to assign classifications to each 30 × 30 m pixel [27]. Multiple levels of classifications and species associations are provided within the EVT with varying levels of details in species associations ranging from the broad physiological categorizations of trees, shrubs, and grasses, to individual species-assemblages on the landscape [27]. We aggregated the 61 EVT ‘group’ classifications into 23 broader classifications of Utah vegetation that are likely to have similar fire behavior (Table 1 and Table S1). We did not analyze areas categorized as “Agriculture”, “Developed”, “Snow”, or “Water”. Although there are some uncertainties with LANDFIRE data, particularly with respect to vegetation conversion (particularly in or near the WUI) or forest successional stage (with composition and structure potentially altering fuel loadings and potential fire effects), our aggregation into broad classifications (and the large number of LANDFIRE pixels; 24,703,822) enables us to characterize fire at the landscape scale.

2.3. Identification of Fire Perimeters

We sourced fire perimeter and ignition data from the Wildland Fire Interagency Geospatial Services (WFIGS) which curates wildland fire incident data within the USA [37]. In cases with medium-sized fires, prescribed fires, and fires < 1990, the WFIGS dataset was occasionally incomplete, and we sourced additional perimeters and fire information directly from the managers of the Dixie, Ashley, Uinta-Wasatch-Cache, and Manti-La Sal National Forests as well as the Bureau of Land Management and Utah state land managers. We selected all fire perimeters ≥ 40 ha between 1984 and 2022, the Landsat Thematic Mapper (and later) period of record and analyzed each fire individually. For all fires that burned across state boundaries, we analyzed data only for the area burned within Utah but classified the fire size according to its full extent. We identified and removed all fire perimeters from our analyses that were identified or suspected to be prescribed fires. We identified prescribed fires using associated fire metadata, naming conventions (e.g., “BrianHeadFireRehabProject” or “DuckCreekFuels1”), or in rare cases from perimeters with right-angles or unusually linear shapes that did not follow visible landscape features. Escaped prescribed burns were analyzed as wildfires.
Table 1. Vegetation categories aggregated from the Existing Vegetation Type of the LANDFIRE National Vegetation Classification [27]. The number of fires burned refers to those fires that burned at least one LANDFIRE pixel of that vegetation type.
Table 1. Vegetation categories aggregated from the Existing Vegetation Type of the LANDFIRE National Vegetation Classification [27]. The number of fires burned refers to those fires that burned at least one LANDFIRE pixel of that vegetation type.
Vegetation CategoryTypical SpeciesArea Burned
1984–2022 (ha)		Area Burned 1984–2022
(%)		Total
Area
(ha)		Total Area
(%)		# of Fires Burned
AlpineHerbs and graminoids13,7790.672,3590.3605 (40%)
AgricultureHerbs and graminoids31,9571.3927,0144.2793 (53%)
Annual grasslandBromus tectorum graminoids205,1658.9389,8161.81254 (84%)
AspenPopulus tremuloides
Abies bifolia		63,9022.8777,0453.5475 (32%)
ChaparralArctostaphylos spp. Ceanothus spp.17,8790.847,9690.2577 (39%)
Developed-22,3550.9410,2621.9765 (51%)
Douglas-firPseudotsuga menziesii
Acer grandidentatum		64,0292.8445,0292.0556 (37%)
Five-needle pinePinus flexilis
Pinus longaeva		20,8550.9143,6490.6353 (23%)
LodgepolePinus contorta
Pseudotsuga menziesii		97910.4126,8030.644 (2%)
Mountain MahoganyCercocarpus ledifolius
Juniperus spp.		14,5490.687,4510.4498 (33%)
Pinon-JuniperPinus monophyla
Juniperus osteosperma		249,14110.93,926,19418.01234 (83%)
Perennial grasslandElymus elymoides
Agropyron cristatum		125,4635.4341,0101.51325 (89%)
Ponderosa PinePinus ponderosa23,1971.0214,7731.0315 (21%)
RiparianJuncus spp.
Salix spp.		49380.295,3690.4498 (33%)
Riparian-hardwoodPopulus trichocarpa
Salix spp.		71020.3146,5110.7681 (46%)
SagebrushArtemisia spp.776,51034.04,318,83219.71422 (96%)
ShrublandSarcobatus spp.
Ericameria nauseosa		371,55616.24,535,39120.71296 (87%)
Snow---680.0-
SparseChenopodiaceae spp.27,9891.23,020,92613.8927 (62%)
Spruce-firAbies bifolia
Picea engelmannii		46,0552.0438,1291.9289 (19%)
Water---635,8712.9-
WUI ShrubPrunus virginiana22,8571.0224,8301.0857 (58%)
WUI WoodlandAcer grandidentatum
Quercus gambelii		149,0826.5599,9302.7900 (60%)
Total 1-2,268,15110021,925,2311001477
1 Area for each vegetation type was calculated with a Transverse Mercator projection (UTM Zone 12). Total area burned includes agricultural and developed areas which were not analyzed.

2.4. Image Acquisition and Calculation of Remotely Sensed Fire Severity

To assess the accuracy of the fire perimeter delineation and assess fire severity we examined each fire perimeter individually with current and historical National Agriculture Imagery Program (NAIP) imagery to assess the pre-fire vegetation type and continuity [29]. We sourced burn severity for most fires ≥ 400 ha from the MTBS database [18] with pre-calculated metric of dNBR following the same equations as Miller and Thode [10]. For all MTBS-derived fire data we adjusted the default dNBR values using the MTBS provided offset, provided in the metadata associated with each fire. The offset adjusts burn severity by subtracting background changes in reflectance due to non-fire related stressors. In some cases, large fires were missing from the MTBS database and we calculated burn severity manually.
For each fire we estimated the percent of forest cover within the fire boundary using pre-fire NAIP imagery. Fires with ≥50% forest cover were analyzed with an extended assessment [38] in which we used post-fire LANDSAT imagery approximately one year after the date of burning. We analyzed fires with <50% forest cover with an initial assessment where post-fire imagery was selected as close to the date of fire extinguishment as possible and within the same year as the pre-fire image [38]. For both initial and extended assessments, we selected LANDSAT scene pairs that minimized smoke, clouds, and particulates around the fire boundary, had similar solar angle, and matched pre-fire and post-fire vegetation phenology [39]. We assessed phenology using non-burned vegetation adjacent to the fire perimeter and high-elevation snowpack extent in the spring and fall. In cases where we could not ascertain the date of fire extinguishment, we used the first post-fire LANDSAT image that did not have signs of fire or smoke within the fire boundary.
After selecting an appropriate pre- and post-fire scene pair, we calculated the normalized burn ratio (NBR), delta NBR (dNBR), and the relative dNBR (RdNBR). We calculated the NBR (Equation (1)) and dNBR (Equation (2)) using the near-infrared (NIR) and shortwave infrared (SWIR) bands [10,38].
NBR = (NIR − SWIR)/(NIR + SWIR) × 1000
dNBR = NBR(Pre-fire) − NBR(Post-fire)
Because changes in vegetation reflectance between images may also be due to drought or annual differences in plant phenology, we selected a neighboring ≥ 90 ha unburned region of comparable vegetation with similar aspect and elevation to control for non-fire-induced changes in vegetation reflectance. We calculated the median dNBR offset over the entire non-burned region and used this value to adjust the dNBR of the burned area. We limited our maximum offset to bounds of −50 to +50 dNBR for phenological change and reassessed any offset selection that produced values outside of these limits. The median offset value was 3. All reported values of dNBR within the manuscript are offset-adjusted values. Preliminary analysis showed that RdNBR values were not markedly different from dNBR values, consistent with results reported by others [40,41].
After calculating the offset-adjusted dNBR, we used a combination of pre-and post-fire scenes and dNBR to assess the accuracy of each fire perimeter and make minor adjustments to perimeters. We were conservative with perimeter adjustment and altered fire perimeters only when clearly burned vegetation and elevated dNBR values were visible outside fire perimeters. We avoided reducing fire perimeter size because remotely sensed imagery may fail to capture low-severity, understory burns that were delineated by ground-crews [42]. We recategorized dNBR outliers of <−300 as ‘−300′ and >1200 as ‘1200′ for all analyses and graphs. We classified dNBR burn severity using the threshold values defined by Miller and Thode [10] which are: ‘Unchanged’ < 41 dNBR, 41 < ‘Low’ ≤ 176, 176 < ‘Moderate’ ≤ 366, and ‘High’ > 366.

2.5. Analyses of Differences in Fire Regimes

We conducted analyses in R 4.2.1 [42] using the graphical user interface R Studio 2022.02.3 [43]. We generated maps in ArcMap10.8.1 (ESRI, Redlands, CA, USA) using background data from 2021 0.6 × 0.6 m NAIP imagery [29]. We generated graphs using the ggplot2 3.3.3 and ggpubr 0.4.0 R packages [44,45]. We used the R packages raster 3.6-3 [46], rgdal 1.6-2 [47], and rgeos 0.5-9 [48] to load, analyze, and export raster data. Area calculations were done in ArcMap based on a Transverse Mercator map projection (Zone 12).
We used an analysis of variance (ANOVA) in base R to test significant differences in mean burn severity between medium-sized and large fires, by vegetation type. We analyzed assumptions of normality using Shapiro-Wilk tests, and homoscedasticity using a Bartlett test in base R [42]. To assess differences in severity distributions, we binned dNBR distributions by 10, and used a Kolmogorov-Smirnov (KS) test in base R. To assess variation in burn severity through time we calculated Mahalanobis dissimilarity of the mean of severity quartiles, by year, vegetation type, and between medium-sized and large fires. To test for differences in homogeneity of variance between medium-sized and large fires and between vegetation types we used the ‘betadisper’ test and a permutational analysis of variance with 999 permutations (random seed = 4711) [49]. To measure if the proportion of area burned by the largest fires varied through time, we calculated the Gini coefficient, a measure of inequality among numeric values that is well represented in the natural and social science literature. The Gini is bounded between 0, representing complete equality among values (e.g., all fires burned the same area in a given year), and 1, complete inequality between values (e.g., one fire burned all area in a given year). We calculated the Gini coefficient using the DescTools 0.99.48 R Package [50]. After visualizing the Gini, we detected non-linear trends with time and used the mgcvv 1.8-40 R package [51] to model the relationship with a nonlinear generalized additive model (GAM) with a ‘betar’ distribution. Because we had only partial LANDSAT data for 1984 and 2022, we did not include these years in analyses relating to fire area.

3. Results

3.1. Wildfire Frequency, Area Burned, and Severity

Between 1984–2022, there were 1652 fires ≥ 40 ha, comprising 24,703,822 analyzed Landsat pixels. Of those, 180 were prescribed fires and excluded from our analyses. Within the characteristic vegetation types of Utah (Table 1) there were 775 medium-sized (40 ha ≤ area < 400 ha) and 697 large (≥400 ha) wildfires 1984–2022 (Figure 1, Table 2). From 1985 to 2021 there were an average of 20 medium-sized fires each year which burned an average of 2901 ha and 18 large fires that burned an average of 55,341 ha (Figure 2). Area burned varied widely among years (Figure 2b), partially driven by the differing consumption of vegetation types (Table 3). Annual variation in area burned was 98% of mean annual area burned. Large fires burned more area (mean percent burned 90 ± 10% SD) than medium-sized fires (10 ± 10%), and the inequality in area burned between fires, as measured by the Gini coefficient, increased with time from 1985 to 2021 (R2 = 0.23; p < 0.001; Figure 3), with most of the change occurring from 1985 to 2005 (Figure 3). Large fires burned at least one LANDFIRE pixel of a mean of 13 vegetation types (median = 13) while medium-sized fires burned a mean of 9 of vegetation types (median = 9). However, both large and medium-sized fires had the majority of burn area within a single vegetation type (proportionlarge = 53 ± 17%; proportionmedium = 57 ± 18%), whose type varied depending on the fire. Fire severity varied widely by year and vegetation type (Figure 4). Median severity (−300 ≤ dNBR ≤ 1200) for all burned pixels was 157, although this was highly influenced by the results from sagebrush and shrublands, which were 50.2% of area burned, a proportion which was relatively stable (18% to 79% yr−1, 17% SD). All large fires and 97% of medium-sized fires burned at least one pixel classified as either sagebrush or shrublands.
Fire 06 00423 g002
Figure 2. The (a) number of large (≥400 ha) and medium-sized (40 ha ≤ area < 400 ha) fires that burned in Utah, USA from 1984 to 2021. (b) The total area burned in fires ≥ 40 ha from 1984 to 2022 in Utah, USA.
Figure 2. The (a) number of large (≥400 ha) and medium-sized (40 ha ≤ area < 400 ha) fires that burned in Utah, USA from 1984 to 2021. (b) The total area burned in fires ≥ 40 ha from 1984 to 2022 in Utah, USA.
Fire 06 00423 g002
For forested vegetation types, large fires had a higher top quartile severity than medium-sized fires. In general, medium-sized fires had greater burn severity at lower quartiles while large fires had 7.5% greater 4th quantile burn severity relative to medium-sized fires (Figure 5). Kolmogorov-Smirnov tests indicate that 13 of 19 vegetation types differed in their burn severity distribution (Table S2) with aspen, Douglas-fir, ponderosa pine, riparian-hardwood, spare, and spruce-fir having similar burn severity distributions (Figure 4). Notably, annual and perennial grasslands, chaparral, and sparse vegetation had greater burn severity in medium-sized fires (+7.9% Q4) relative to large fires (Figure 6). Large fires had greater interannual variation in burn severity (Figure S1) than medium-sized fires (F1139 = 30.43, p < 0.001) but this differed by vegetation type (Figure S2) with forested vegetation types having greater interannual variation in burn severity than non-forested vegetation types (Figure 6, Figure 7 and Figure S2). Sagebrush, shrubland, and annual grassland had the lowest interannual variation in burn severity (Figure S2) among vegetation categories.
Table 2. The number of wildfires ≥ 40 ha, the area burned of wildfires ≥ 40 ha, and the severity of wildfires ≥ 40 ha from 1984 to 2021 in Utah, USA.
Table 2. The number of wildfires ≥ 40 ha, the area burned of wildfires ≥ 40 ha, and the severity of wildfires ≥ 40 ha from 1984 to 2021 in Utah, USA.
Year40 ha ≤ Area < 400 haArea ≥ 400 haTotal ≥ 40 ha
# of FiresArea Burned (ha)# of FiresArea Burned (ha)# of FiresArea Burned (ha)
198436341395644590
1985813081112,0131913,321
19861115882147,0443248,632
198777921423,6692124,461
1988911821424,3982325,580
198959191319,8051820,724
19907114868483139631
199123133216852481
1992101485894501810,935
199318141577008258423
19942435043484,9345888,438
19951828942771,8604574,754
199627422138193,61265197,833
199713198276322208304
199879631742,1992443,162
19992330153051,7985354,813
20002743363298,36459102,700
20012337092346,9574650,666
200219298227109,04546112,027
20032231521440,4123643,564
20041929211433,6533336,574
20054061383199,17671105,314
200662890647125,040109133,946
200736452735234,20771238,734
2008315276766233811,899
20091827051641,1853443,890
2010111582621,2631722,845
20112426631118,1793520,842
201210141237163,08747164,499
20131624131133,5982736,011
2014172334880192510,353
201515186821680173548
20162734332437,4505140,883
20174666762489,9467096,622
201836498423132,56759137,551
20191824562337,8104140,266
20203956422394,3266299,968
2021202777821,6522824,429
202276385 19493 11210,131 1
Total775110,8937032,122,45114772,223,344
1 Includes wildfires that were not analyzed for extended burn severity.
Fire 06 00423 g003
Figure 3. The trend in the Gini coefficient for the area burned in fires ≥ 40 ha from 1985 to 2021 in Utah, USA. An increasing Gini coefficient shows that the annual area burned in Utah is becoming more concentrated in fewer, larger fires. Dashed lines mark the 95th percentile confidence interval.
Figure 3. The trend in the Gini coefficient for the area burned in fires ≥ 40 ha from 1985 to 2021 in Utah, USA. An increasing Gini coefficient shows that the annual area burned in Utah is becoming more concentrated in fewer, larger fires. Dashed lines mark the 95th percentile confidence interval.
Fire 06 00423 g003
Fire 06 00423 g004
Figure 4. Cumulative severity distribution for fires ≥ 40 ha in Utah for (a) all fires, (b) by non-forested vegetation, and (c) forested vegetation types. Curves further to the right (left) represent more (less) sever fire years and vegetation types.
Figure 4. Cumulative severity distribution for fires ≥ 40 ha in Utah for (a) all fires, (b) by non-forested vegetation, and (c) forested vegetation types. Curves further to the right (left) represent more (less) sever fire years and vegetation types.
Fire 06 00423 g004

3.2. Fire Severity and Area Burned by Vegetation Type

Burn severity varied substantially among vegetation types, and between medium-sized and large fires (Figure 5). Medium-sized fires generally had significantly lower Q4 severity than large fires, except for annual grassland, perennial grasslands, sparse, and chaparral which had higher Q4 severity (p < 0.05). Medium-sized and large fires had similar (p > 0.05) Q4 severity in the WUI woodland. The annual area burned varied widely by vegetation type (Table 3, Tables S5, S6 and S7) with consistently high burn areas in non-forested vegetation of annual grassland, perennial grassland, shrubland, and sagebrush steppe which cumulatively burned an average of 72% of the total area (annual range 26–98%). In contrast, Douglas-fir, aspen, and spruce-fir (Table 4, Tables S4, S6 and S8) were the forested vegetation types which contributed the greatest proportions to the annual area burned (mean: 8%, annual range 0–39%) and burned at the highest severities (Figure 6). The WUI woodland had an average of 4 fewer fires per year than WUI shrubland, but burned 6.5 times more area (Table 3).
Fire 06 00423 g005
Figure 5. Delta normalized burn ratio (dNBR) averaged across quartiles, by vegetation type, for medium-sized (40 ≤ area < 400 ha) and large (≥400 ha) fires in Utah, USA between 1984 and 2021.
Figure 5. Delta normalized burn ratio (dNBR) averaged across quartiles, by vegetation type, for medium-sized (40 ≤ area < 400 ha) and large (≥400 ha) fires in Utah, USA between 1984 and 2021.
Fire 06 00423 g005
Fire 06 00423 g006
Figure 6. The distribution of satellite-derived fire severity (dNBR) for medium-sized (40 ≤ area < 400 ha; a,c,e,g,i,k,m,o,q,s) and large (≥400 ha; b,d,f,h,j,l,n,p,r,t) fires across primarily unforested vegetation types in Utah, USA from 1984 to 2021. Colors indicate classified fire severities using delineations from Miller and Thode 2007 (red—high severity, yellow—moderate severity, cyan—low severity, and dark green—no change detected by satellite).
Figure 6. The distribution of satellite-derived fire severity (dNBR) for medium-sized (40 ≤ area < 400 ha; a,c,e,g,i,k,m,o,q,s) and large (≥400 ha; b,d,f,h,j,l,n,p,r,t) fires across primarily unforested vegetation types in Utah, USA from 1984 to 2021. Colors indicate classified fire severities using delineations from Miller and Thode 2007 (red—high severity, yellow—moderate severity, cyan—low severity, and dark green—no change detected by satellite).
Fire 06 00423 g006
Fire 06 00423 g007
Figure 7. The distribution of satellite-derived fire severity (dNBR) for medium-sized (40 ≤ area < 400 ha; a,c,e,g,i,k,m,o,q) and large (≥400 ha; b,d,f,h,j,l,n,p,r) fires across primarily forested vegetation types in Utah, USA from 1984 to 2021. Colors indicate classified fire severities using delineations from Miller and Thode 2007 (red—high severity, yellow—moderate severity, cyan—low severity, and dark green—no change detected by satellite).
Figure 7. The distribution of satellite-derived fire severity (dNBR) for medium-sized (40 ≤ area < 400 ha; a,c,e,g,i,k,m,o,q) and large (≥400 ha; b,d,f,h,j,l,n,p,r) fires across primarily forested vegetation types in Utah, USA from 1984 to 2021. Colors indicate classified fire severities using delineations from Miller and Thode 2007 (red—high severity, yellow—moderate severity, cyan—low severity, and dark green—no change detected by satellite).
Fire 06 00423 g007
Table 3. The area burned (ha) in wildfires ≥ 40 ha, by year and non-forested vegetation type, in Utah from 1984 to 2022.
Table 3. The area burned (ha) in wildfires ≥ 40 ha, by year and non-forested vegetation type, in Utah from 1984 to 2022.
YearAnnual GrasslandPerennial GrasslandSagebrushShrublandSparseRiparianRiparian HardwoodChaparralWUI ShrubWUI Woodland
1984138438311091593601135
1985335664533745143292548253874
198610,412222616,99913,263826446725173933
198778181862593863556751113611575
19883216130910,75826321053530155293120
1989698591406169915543344265071850
19901192293331127389340141662327
19915318514603519001198116
199243872949971431476251211091379
19934122423716246786001292154
199415,143731036,36015,53747511615718215793628
19959545583930,62421,55844013627092621872
199616,43612,63010,422621,2091172115253685156012,477
19973042102445117511910266112611062
199814,97987610,76014,096139507264111280
19993538259331,6316901170351773054012728
20008914657653,56413,001986491774507946135
20014957596716,5576765368551661204395216
20021575326226,82829793514385792108275512,462
2003101716689500994770133613417475158429
200412186965884388214642842617893109423
200510,744268924,72947,60935613312430453951244
20069608370040,59339,6941852113287374913856006
200728,94311,11199,92854,6552942167695457238512,512
2008168355280084115077129262381278
20091713187816,771479019822901515084347
20102559544834423215835222392321
20113116159110,26244817931780116108
2012964610,67475,09520,6051535952701653299612,886
20131330366420,267159416835552288921216
20149778764640146861753151681023
2015115766538383915160467
2016545815,5946644201529910454703171493
201712,735467227,01415,9595046815641413761313
20182042582618,78431812140171637405127021,860
20192570151316,44139013064722521842561938
20209134294316,75217,37752235754825153212567
2021105061360671688538450149585331511
202230162291273914821743
Total205,166125,463776,510371,55727,9924939710318,07522,855149,078
Table 4. The area burned (ha) in wildfires ≥ 40 ha, by year and forested vegetation type, in Utah from 1984 to 2022.
Table 4. The area burned (ha) in wildfires ≥ 40 ha, by year and forested vegetation type, in Utah from 1984 to 2022.
YearPinon-JuniperPonderosaDouglas-FirAspenMountain MahoganyLodgepoleSpruce-FirFive-Needle PineAlpine
19840300000000
198541140104101430
198638101399175123839
19872520285424012524
1988174196262455114209817499
19893047341450802126224896304180
19903817625187620223890
19915205130051
199229516724815908295228
199336731392241601263625
1994506928712109173769108506416
19956113154526076121
1996358705923126810590282358336
1997183441632380121821
19980100700076
1999189458225231744117189127
2000180398595749442070180402
2001304583108494540128104304104
200246222609903058321102696337846222089
200347563592174312070212475200
20045326073061311807853166
200525331729960113162630125371
20061515656562543771278151181
2007622147413322482102875010316221153
2008136166813834059307213689
2009317200417678745840217317346
201017126473896153695008511712367
2011044506200011
2012134480239995151180313121013441621
20134472585832084755514471753
201431241391101231189
2015167531395852081687
2016383169929199917224213813831509
201717732234759710,211797142821773240
20183064164017,59818,257205387884213064762
201954217792405208120596160654273
2020225544224584301218599819,8892255601
2021121624161020901480123212154
2022200010020
Total20,85423,20064,01463,90214,548979146,05420,85413,781

4. Discussion

The discontinuous, variable landscape of Utah has experienced a tremendous range of wildfire behavior over the past 38 years (Figure 2 and Figure 4). This behavior has varied with fire size, vegetation type, and the legacy of decades of fire suppression. Medium-sized fires, often overlooked, comprised 5% of all burned land and had greater mean severity and lower interannual variance across most vegetation types. However, medium-sized fires had a lower severity for the highest severity quartile, relative to large fires. For vegetation types that burned at overall higher severity (i.e., predominantly forested vegetation types; Figure 4 and Figure 5) medium-sized fires burned at lower severity than large fires.
Although remote sensing of fire is the only practical method to analyze landscapes, the variation in landforms, vegetation, and fuel loading will always introduce considerable uncertainty. In particular, Landsat-based analyses of fire severity have high uncertainties at moderate levels of severity—tree death is often poorly correlated with changes in reflectance [13] and actual surface fuel combustion can also likewise be poorly correlated with satellite-derived severity [52]. The correlation between satellite-derived fire severity and ground-based surveys tends to increase at very high or very low levels of severity [13]. However, the delineation between truly unburned and very lightly burned remains unclear [41,52,53] and our severity assessments likely encompass small- and large-scale unburned refugia located in the interior of fire perimeters. Development of new remote-sensing technologies such as hyper-spectral imaging and post-fire LiDAR as well as building region-specific relationships between burn severity and ecological effects [54] may improve our ability to delineate fire perimeters, changes in fuels, and burn severity [55].
Our results (Figure 6 and Figure 7) demonstrate the differing effects of fire on diverse vegetation types (see also Thode et al. [6]). Although dNBR, RdNBR, and other satellite-derived metrics of fire severity may be broadly comparable when vegetation and fuel loading are similar (e.g., between Douglas-fir and spruce-fir forests), comparisons will be less meaningful when vegetation and surface fuel loading differ (especially if a defined vegetation type has experienced a long period of fire suppression). The issue of differential fuel loading is perhaps more important in Utah spruce-fir forests because of the large quantity of heavy fuels created by the Dendroctonus rufipennis outbreak in the 1990s and the consequent effects on surface fuel evolution [35] as well as the effects of fuel evolution after fires burn in forests where fire has been long excluded [35,56]. Importantly, dNBR and RdNBR can struggle to adequately characterize the ecological effects of fire in non-forest systems [56,57] owing to the potentially rapid sprouting after fire, or inherent differences in fire effects between ecosystems dominated by annual or perennial life. Furthermore, the effects of fire are not limited to mortality and consumption of vegetation. The differential effects of fire on soil between different landcover types makes direct comparison difficult [11].
Although LANDFIRE classification has known accuracy issues with narrow classifications of vegetation [58,59], we aggregated categories of similar vegetation (Table S1) which are more likely to agree with reference plots [26]. However, no single remote sensing classification of vegetation will be completely accurate and will include misclassification errors due to changes in vegetation and land use through time. The summary statistics presented here represent the average of 19 vegetation categories split unevenly across 24,703,822 analyzed pixels, and may not adequately represent fire effects in fringe-case communities and those with underrepresented successional stages or with uncharacteristic fuel loadings.

4.1. Large Fires Have More Variable Burn Severities Than Medium-Sized Fires

Larger fires exhibited lower mean severity, but this was due to a relatively high area in low and unchanged severities. This satellite interpretation of lower severity could in turn be due to large patches of truly unburned vegetation [60] or to fast regrowth of herbaceous vegetation [41,61]. For some vegetation types, the area of the fire had little impact on either mean severity or the form of the cumulative distribution of severity. For example, sparse and riparian vegetation had the lowest mean quartile of burn severity, suggesting that underlying fuel structure and conditioning may not promote flame propagation and results in unburned refugia inside of fire perimeters.
Importantly, those ecosystems with the greatest Q4 severity, such as spruce-fir or Douglas-fir, also had the greatest interannual variation, highlighting difficulty in generalizing fire effects across time. Other systems, such as sagebrush or shrubland, had low interannual variation in burn severity, indicating that if conditions are suitable for burning (e.g., sufficiently low fuel moisture and suitable weather) that these ecosystems may generally have similar fire effects, regardless of the year. Though, the confluence of fire and encroachment or invasion of species that alter fuel structure, such as Bromus tectorum (cheatgrass), may see altered fire behavior and regimes that exceed the resilience of even fire-adapted systems [62,63].
Although medium-sized fires had higher mean severity, large fires may have more negative ecosystem impacts, depending on the vegetation [64], because large fires include larger patches that burn at high severity which may cause delayed vegetation recovery. As well, dispersal and germination of seeds may be hampered over large distances, such as those present in large fire scars [65,66]. Notably, we detected large proportions of ‘unchanged’ severity in large fires, potentially indicating the presence of unburned refugia in the interior of many of fires. These unburned refugia may act as important sources for the dispersal of seeds into the interior of large, high-severity fire footprints on the landscape [60].
Large fires dominated the area burned in Utah and, have greater upper limits of burn severity that may surpass the natural range of variability, enable establishment by problematic species, and exceed ecosystem resilience [63,67]. Additionally, because large fires are often controlled by top-down climatic influences, the confluence of drought and large fires may contribute to long-term mortality of surviving woody vegetation [68] and more high-severity fires [69].
Predicting the post-fire effects of large fires may be more difficult than medium-sized fires due to the higher interannual variability of burn severity and more extreme values of the least severely burned quartile (of area) and the most severely burned quartile. Importantly, local controls on fire behavior from topography, fuel conditioning, and loadings will act as proximate controls on fire effects [40,70] and average burn severities and area may not be applicable for fringe-case fire weather, communities and landscape positions.

4.2. Variation in Area Burned and Severity across Vegetation Types

Forested vegetation types and large fires had high interannual variation in burn area and severity, likely due to the regional climatic influences on fuel accumulation and drying over multi-year and seasonal scales [70,71]. As well, much of the area burned in Douglas-fir, aspen, and spruce-fir has been within the last decade (Table 4), and may represent abnormally high burn area or severity owing to fuel buildups as a result of fire suppression policies and criminalization of cultural burning over the preceding century [72,73]. While herbaceous fuels can also exhibit multi-year responses to climate [74], it is possible that other confounding factors, such as less topographic variation, fewer physical barriers to fire spread, lower variability in fuel sizes, or uniformly low fire season fuel moistures [75] in non-forested systems may be responsible for the lower interannual variation in burn area and severity. Brown et al. [31] found that regional fire years in Utah occurred approximately every 8-yrs from 1630 to 1900, and were associated with drought and La Niña conditions during the year of fire, though this pattern varied with latitude in Utah, with northern sites having less forcing from the El Niño Southern Oscillation. Altered climates have already shifted western North American fire seasons [76] and further warming and more variable climates may see greater shifts in the timing, frequency, and size of Utah’s fires.
Most of the area burned in Utah is driven by fires in lower-elevation, arid vegetation types such as annual and perennial grasslands and the widespread sagebrush steppe and shrublands. However, the more productive vegetation types of Douglas-fir, aspen, and spruce-fir were those with the greatest burn severities and likely those with the most extreme fire intensities owing to their perennial lifecycles, greater fuel loadings, and upper slope positions likely to burn under intense heading fire [40]. Most fires were not confined to a single vegetation type or land cover.

4.3. Wildfires in the WUI

Large fires pose considerable risks to people, developed infrastructure, and ecosystems, although they can lead to some desirable post-fire conditions [77,78]. From 1990 to 2010 the US has seen a 33% increase in the area, and a 44% increase in the number homes within the WUI [79], which poses considerable fire dangers and complications in successfully managing fire on the landscape. Nationally, 32% of all wildfires originate in the WUI and are overwhelmingly human ignitions which are responsible for the majority of threatened and damaged structures [80]. Forested WUI in Utah burned 6.5-fold more area than herbaceous and shrub-dominated WUI, and at higher severity. This suggests that the WUI most at-risk are those occupying the productive and upper-elevation vegetation types in Utah which are capable of supporting trees.

5. Conclusions

Continued monitoring of fires in Utah will further refine the variability in fire regimes and better delineate the scope and severity of wildfire within Utah. Examining the relationships between vegetation and fires in the recent past, we may be able to better predict and mange future fire under increasingly variable climates and the continued expansion of the WUI into wildland systems. Fires of any size can have considerable ecosystem benefits including reducing the fuel loads that can lead to extreme fire behavior and reducing forest density—both of which may become even more important in droughtier conditions. Prescribed fires that are as large as practically manageable can also provide these benefits, and we suggest that a tractable size for prescribed burns- and the characteristic results—may be exemplified by the data on medium-sized fires in each vegetation type.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/fire6110423/s1, Figure S1: Distance to centroid for dNBR quartile values for medium-sized and large fires from 1991 to 2020, Utah, USA; Figure S2: Distance to centroid for dNBR quartile values for medium-sized and large fires, by vegetation type, from 1991 to 2020, Utah, USA. Table S1: Vegetation categories aggregated from LANDIFRE Existing Vegetation Type categories (LANDFIRE 2022); Table S2: Kolmogorov-Smirnov tests on the distribution of burn severity between medium-sized (40 ha ≤ area < 400 ha) and large (≥400 ha) wildfires in Utah, USA from 1984 to 2022; Table S3: The number of medium-sized (40 ha ≤ area < 400 ha) and large (≥400 ha) wildfires in predominantly non-forested vegetation types in Utah, USA from 1984 to 2022; Table S4: The number of medium-sized (40 ha ≤ area < 400 ha) and large (≥400 ha) wildfires in predominantly forested vegetation types in Utah, USA from 1984 to 2022; Table S5: The area burned in medium-sized (40 ha ≤ area < 400 ha) and large (≥400 ha) wildfires in predominantly non-forested vegetation types in Utah, USA from 1984 to 2022; Table S6: The area burned in medium-sized (40 ha ≤ area < 400 ha) and large (≥400 ha) wildfires in predominantly forested vegetation types in Utah, USA from 1984 to 2022; Table S7: The median differenced Normalized Burn Ratio (dNBR) in medium-sized (40 ha ≤ area < 400 ha) and large (≥400 ha) wildfires in predominantly non-forested vegetation types in Utah, USA from 1984 to 2022; Table S8: The median differenced Normalized Burn Ratio (dNBR) in medium-sized (40 ha ≤ area < 400 ha) and large (≥400 ha) wildfires in predominantly forested vegetation types in Utah, USA from 1984 to 2022.

Author Contributions

Conceptualization, J.A.L.; methodology, J.A.L. and J.D.B.; software, J.D.B.; validation, J.A.L. and J.D.B.; formal analysis, J.A.L. and J.D.B.; resources, J.A.L.; data curation, J.A.L. and J.D.B.; writing—original draft preparation, J.A.L. and J.D.B.; writing—review and editing, J.A.L. and J.D.B.; visualization, J.D.B.; project administration, J.A.L.; funding acquisition, J.A.L. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Utah Legislature through the Public Land Initiative, Utah State University, and the Utah Agricultural Extension Station which has designated this as Journal Paper #9649. The funders had no role in the study design, analysis, interpretation, or writing of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Acknowledgments

We thank the Utah State University technicians for fire severity analysis. We thank the Utah Legislature for supporting this work. We greatly appreciate the cooperation we received from the state and federal land, fire, and GIS managers who provided fire perimeters and fire metadata. We thank the anonymous reviewers for the helpful feedback.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Stephens, S.L.; Martin, R.E.; Clinton, N.E. Prehistoric fire area and emissions from California’s forests, woodlands, scrublands, and grasslands. For. Ecol. Manag. 2007, 251, 205–216. [Google Scholar] [CrossRef]
  2. Stephens, S.L.; Burrows, N.; Buyantuyev, A.; Gray, R.W.; Keane, R.E.; Kubian, R.; Liu, S.; Seijo, F.; Tolhurst, K.G.; van Wagtendonk, J.W. Temperate and boreal forest mega-fires: Characteristics and challenges. Front. Ecol. Environ. 2014, 12, 115–122. [Google Scholar] [CrossRef]
  3. Abatzoglou, J.T.; Williams, A.P. Impact of anthropogenic climate change on wildfire across western US forests. Proc. Natl. Acad. Sci. USA 2016, 113, 11770–11775. [Google Scholar] [CrossRef] [PubMed]
  4. Bowman, D.M.; Kolden, C.A.; Abatzoglou, J.T.; Johnston, F.H.; van der Werf, G.R.; Flannigan, M. Vegetation fires in the Anthropocene. Nat. Rev. Earth Environ. 2020, 1, 500–515. [Google Scholar] [CrossRef]
  5. Larson, A.J.; Jeronimo, S.M.A.; Hessburg, P.F.; Povak, N.A.; Cansler, C.A.; Kane, V.R.; Churchill, D.J. Tamm Review: Ecological principles to guide post-fire forest landscape management in the Inland Pacific and Northern Rocky Mountain Regions. For. Ecol. Manag. 2022, 504, 119680. [Google Scholar] [CrossRef]
  6. Thode, A.E.; van Wagtendonk, J.W.; Miller, J.D.; Quinn, J.F. Quantifying the fire regime distributions for severity in Yosemite National Park, California, USA. Int. J. Wildland Fire 2011, 20, 223–239. [Google Scholar] [CrossRef]
  7. Lutz, J.A.; Key, C.H.; Kolden, C.A.; Kane, J.T.; van Wagtendonk, J.W. Fire frequency, area burned, and severity: A quantitative approach to defining a normal fire year. Fire Ecol. 2011, 7, 51–65. [Google Scholar] [CrossRef]
  8. van Wagtendonk, J.W.; Moore, P.E.; Yee, J.L.; Lutz, J.A. The distribution of woody species in relation to climate and fire in Yosemite National Park, California, USA. Fire Ecol. 2020, 16, 22. [Google Scholar] [CrossRef]
  9. Picotte, J.J.; Cansler, C.A.; Kolden, C.A.; Lutz, J.A.; Key, C.; Benson, N.C.; Robertson, K.M. Determination of burn severity models ranging from regional to continental scales for the conterminous United States. Remote Sens. Environ. 2021, 263, 112569. [Google Scholar] [CrossRef]
  10. Miller, J.D.; Thode, A.E. Quantifying burn severity in a heterogeneous landscape with a relative version of the delta Normalized Burn Ratio (dNBR). Remote Sens. Environ. 2007, 109, 66–80. [Google Scholar] [CrossRef]
  11. Keeley, J.E. Fire intensity, fire severity and burn severity: A brief review and suggested usage. Int. J. Wildland Fire 2009, 18, 116–126. [Google Scholar] [CrossRef]
  12. Smith, A.M.S.; Kolden, C.A.; Paveglio, T.B.; Cochrane, M.A.; Bowman, D.; Moritz, M.A.; Kliskey, A.D.; Alessa, L.; Hudak, A.T.; Hoffman, C.M.; et al. The science of firescapes: Achieving fire resilient communities. BioScience 2016, 66, 130–146. [Google Scholar] [CrossRef] [PubMed]
  13. Furniss, T.J.; Kane, V.R.; Larson, A.J.; Lutz, J.A. Detecting actual tree mortality with satellite-derived spectral indices and estimating landscape-level uncertainty. Remote Sens. Environ. 2020, 237, 111497. [Google Scholar] [CrossRef]
  14. Furniss, T.J.; Larson, A.J.; Kane, V.R.; Lutz, J.A. Wildfire and drought moderate the spatial elements of tree mortality. Ecosphere 2020, 11, 03214. [Google Scholar] [CrossRef]
  15. Bradley, A.F.; Noste, N.V.; Fischer, W.C. Fire Ecology of Forests and Woodlands in Utah; General Technical Report INT-GTR-287; US Department of Agriculture, Forest Service, Intermountain Research Station: Ogden, UT, USA, 1992.
  16. van Wagtendonk, J.W.; Lutz, J.A. Fire regime attributes of wildland fires in Yosemite National Park, USA. Fire Ecol. 2007, 3, 34–52. [Google Scholar] [CrossRef]
  17. Abatzoglou, J.T.; Kolden, C.A.; Williams, A.P.; Lutz, J.A.; Smith, A.M.S. Climatic influences on interannual variability in regional burn severity across western US forests. Int. J. Wildland Fire 2017, 26, 269–275. [Google Scholar] [CrossRef]
  18. Eidenshink, J.; Schwind, B.; Brewer, K.; Zhu, Z.-L.; Quayle, B.; Howard, S. A project for Monitoring Trends in Burn Severity. Fire Ecol. 2007, 3, 3–21. [Google Scholar] [CrossRef]
  19. Lutz, J.A.; van Wagtendonk, J.W.; Thode, A.E.; Miller, J.D.; Franklin, J.F. Climate, lightning ignitions, and fire severity in Yosemite National Park, California, USA. Int. J. Wildland Fire 2009, 18, 765–774. [Google Scholar] [CrossRef]
  20. Falk, D.A.; Heyerdahl, E.K.; Brown, P.M.; Farris, C.; Fulé, P.Z.; McKenzie, D.; Swetnam, T.W.; Taylor, A.H.; Van Horne, M.L. Multi-scale controls of historical forest-fire regimes: New insights from fire-scar networks. Front. Ecol. Environ. 2011, 9, 446–454. [Google Scholar] [CrossRef]
  21. Larson, A.J.; Belote, R.T.; Cansler, C.A.; Parks, S.A.; Dietz, M.S. Latent resilience in ponderosa pine forest: Effects of resumed frequent fire. Ecol. Appl. 2013, 23, 1243–1249. [Google Scholar] [CrossRef]
  22. Churchill, D.; Larson, A.J.; Dahlgreen, M.C.; Franklin, J.F.; Hessburg, P.F.; Lutz, J.A. Restoring forest resilience: From reference spatial patterns to silvicultural prescriptions and monitoring. For. Ecol. Manag. 2013, 291, 442–457. [Google Scholar] [CrossRef]
  23. Lutz, J.A.; Furniss, T.J.; Johnson, D.J.; Davies, S.J.; Allen, D.; Alonso, A.; Anderson-Teixeira, K.; Andrade, A.; Baltzer, J.; Becker, K.M.L.; et al. Global importance of large-diameter trees. Glob. Ecol. Biogeogr. 2018, 27, 849–864. [Google Scholar] [CrossRef]
  24. Birch, J.D.; Lutz, J.A.; Struckman, S.; Miesel, J.R.; Karst, J. Large-diameter trees and deadwood correspond with belowground ectomycorrhizal fungal richness. Ecol. Process. 2023, 12, 3. [Google Scholar] [CrossRef]
  25. Parry, W. Geology of Utah’s Mountains, Peaks, and Plateaus: Including Descriptions of Cliffs, Valleys, and Climate History, 1st ed.; Friesen Press: Victoria, BC, Canada, 2016. [Google Scholar]
  26. Schug, F.; Bar-Massada, A.; Carlson, A.R.; Cox, H.; Hawbaker, T.J.; Helmers, D.; Hostert, P.; Kaim, D.; Kasraee, N.K.; Martinuzzi, S.; et al. The global wildland–urban interface. Nature 2023, 621, 94–99. [Google Scholar] [CrossRef]
  27. LANDFIRE. Existing Vegetation Type Layer, LANDFIRE 2.0.0, U.S; Department of the Interior, Geological Survey, and U.S. Department of Agriculture: Washington, DC, USA, 2018. Available online: http://www.landfire/viewer (accessed on 28 October 2022).
  28. LANDFIRE. LANDFIRE Remap Vegetation Type Agreement Assessment Summary. Dataset; 2019. Available online: https://landfire.gov (accessed on 16 March 2023).
  29. National Agriculture Imager Program (NAIP). NAIP Imagery 0.6 × 0.6 m Resolution of Utah; United States Department of Agriculture: Washington, DC, USA, 2023.
  30. Flora of North America Editorial Committee. Flora of North America Editorial Committee, eds. 1993+. Flora of North America North of Mexico [Online]. 22+ vols. New York and Oxford. Vol. 1, 1993; vol. 2, 1993; vol. 3, 1997; vol. 4, 2003; vol. 5, 2005; vol. 7, 2010; vol. 8, 2009; vol. 19, 2006; vol. 20, 2006; vol. 21, 2006; vol. 22, 2000; vol. 23, 2002; vol. 24, 2007; vol. 25, 2003; vol. 26, 2002; vol. 27, 2007; vol 28, 2014; vol. 9, 2014; vol. 6, 2015; vol. 12, 2016; vol. 17, 2019; vol. 10, 2021. Available online: http://beta.floranorthamerica.org (accessed on 12 January 2022).
  31. Brown, P.M.; Heyerdahl, E.K.; Kitchen, S.G.; Weber, M.H. Climate effects on historical fires (1630–1900) in Utah. Int. J. Wildland Fire 2008, 17, 28–39. [Google Scholar] [CrossRef]
  32. Kitchen, S.G. Historic Fire Regimes of Eastern Great Basin (USA) Mountains Reconstructed from Tree Rings. Ph.D. Thesis, Brigham Young University, Provo, UT, USA, 2010; 166 p. [Google Scholar]
  33. Wadleigh, L.; Jenkins, M.J. Fire frequency and the vegetative mosaic of a spruce-fir forest in northern Utah. Great Basin Nat. 1996, 6, 28–37. [Google Scholar]
  34. Heyerdahl, E.K.; Brown, P.M.; Kitchen, S.G.; Weber, M.H. Multicentury Fire and Forest Histories at 19 Sites in Utah and Eastern Nevada; General Technical Report RMRS-GTR-261WWW; USDA Forest Service: Fort Collins, CO, USA, 2011.
  35. Lutz, J.A.; Struckman, S.; Furniss, T.J.; Birch, J.D.; Yocom, L.L.; McAvoy, D.J. Large-diameter trees, snags, and deadwood in southern Utah, USA. Ecol. Process. 2021, 10, 9. [Google Scholar] [CrossRef]
  36. Comer, P.J.; Hak, J.C.; Dockter, D.; Smith, J. Integration of vegetation classification with land cover mapping: Lessons from regional mapping efforts in the Americas. Veg. Classif. Surv. 2022, 3, 29–43. [Google Scholar] [CrossRef]
  37. WFIGS. Wildland Fire Interagency Geospatial Services (WFIGS) Interagency Fire Perimeters; National Interagency Fire Center: Boise, ID, USA, 2022.
  38. Key, C.H.; Benson, N.C. Landscape Assessment (LA). In FIREMON: Fire Effects Monitoring and Inventory System; General Technical Report RMRS-GTR-164-CD; Lutes, D.C., Keane, R.E., Caratti, J.F., Key, C.H., Benson, N.C., Sutherland, S., Gangi, L.J., Eds.; Department of Agriculture, Forest Service, Rocky Mountain Research Station: Fort Collins, CO, USA, 2006; p. LA-1-55. [Google Scholar]
  39. Key, C.H. Ecological and sampling constraints on defining landscape fire severity. Fire Ecol. 2006, 2, 34–59. [Google Scholar] [CrossRef]
  40. Birch, J.D.; Dickinson, M.B.; Knapp, E.; Lutz, J.A.; Miesel, J.R. Heading and backing fire behavior mediates the influence of fuels on wildfire energy. Int. J. Wildland Fire. 2023, 32, 1244–1261. [Google Scholar] [CrossRef]
  41. Kolden, C.A.; Lutz, J.A.; Key, C.H.; Kane, J.T.; van Wagtendonk, J.W. Mapped versus actual burned area within wildfire perimeters: Characterizing the unburned. For. Ecol. Manag. 2012, 286, 38–47. [Google Scholar] [CrossRef]
  42. R Core Team. R: A Language and Environment for Statistical Computing; Version 4.2.1; R Foundation for Statistical Computing: Vienna, Austria, 2023. Available online: https://www.r-project.org/ (accessed on 21 September 2022).
  43. R Studio Team. R Studio: Integrated Deveplment for R. Version 2022.02.3. 2022. Available online: https://posit.co/downloads/RCoreTeam (accessed on 3 February 2022).
  44. Wickham, H. ggplot2: Elegant Graphics for Data Analysis; Springer: Berlin/Heidelberg, Germany, 2016. [Google Scholar]
  45. Kassambara, A. ggpubr: ‘ggplot2’ Based Publication Ready Plots. R Package Version 0.4.0. The Comprehensive R Archive Network (CRAN). 2020. Available online: https://CRAN.R-project.org/package=ggpubr (accessed on 10 February 2023).
  46. Hijmans, R. Raster: Geographic Data Analysis and Modeling. R Package Version 3.6-3. 2022. Available online: https://CRAN.R-project.org/package=raster (accessed on 10 October 2023).
  47. Bivand, R.; Keitt, T.; Rowlingson, B. Rgdal: Bindings for the ‘Geospatial’ Data Abstraction Library. R Package Version 1.6-2. 2022. Available online: https://CRAN.R-project.org/package=rgdal (accessed on 10 October 2023).
  48. Bivand, R.; Rundel, C. Rgeos: Interface to Geometry Engine—Open Source (‘GEOS’). R Package Version 0.5-9. 2021. Available online: https://CRAN.R-project.org/package=rgeos (accessed on 10 October 2023).
  49. Oksanen, J.; Simpson, G.; Blanchet, F.; Kindt, R.; Legendre, P.; Minchin, P.; O’Hara, R.; Solymos, P.; Stevens, M.; Szoecs, E.; et al. Vegan: Community Ecology Package. R Package Version 2.6-4. 2022. Available online: https://CRAN.R-project.org/package=vegan (accessed on 10 October 2023).
  50. Signorell, A. DescTools: Tools for Descriptive Statistics. R Package Version 0.99.48. 2023. Available online: https://CRAN.R-project.org/package=DescTools (accessed on 6 September 2023).
  51. Wood, S.N. Fast stable restricted maximum likelihood and marginal likelihood estimation of semiparametric generalized linear models. J. R. Stat. Soc. B 2011, 73, 3–36. [Google Scholar] [CrossRef]
  52. Blomdahl, E.M.; Kolden, C.A.; Meddens, A.J.H.; Lutz, J.A. The importance of small fire refugia in the central Sierra Nevada, California, USA. For. Ecol. Manag. 2019, 432, 1041–1052. [Google Scholar] [CrossRef]
  53. Kolden, C.A.; Abatzoglou, J.T.; Lutz, J.A.; Cansler, C.A.; Kane, J.T.; van Wagtendonk, J.W.; Key, C.H. Climate contributors to forest mosaics: Ecological persistence following wildfire. Northwest Sci. 2015, 89, 219–238. [Google Scholar] [CrossRef]
  54. Reiner, A.L.; Baker, C.; Wahlberg, M.; Rau, B.M.; Birch, J.D. Region-specific remote-sensing models for predicting burn severity, basal area change, and canopy cover change following fire in the southwestern United States. Fire 2022, 5, 137. [Google Scholar] [CrossRef]
  55. Szpakowski, D.M.; Jensen, J.L. A review of the applications of remote sensing in fire ecology. Remote Sens. 2019, 11, 2638. [Google Scholar] [CrossRef]
  56. Lutz, J.A.; Larson, A.J.; Swanson, M.E. Advancing fire science with large forest plots and a long-term multidisciplinary approach. Fire 2018, 1, 5. [Google Scholar] [CrossRef]
  57. Stambaugh, M.C.; Hammer, L.D.; Godfrey, R. Performance of burn-severity metrics and classification in oak woodlands and grasslands. Remote Sens. 2015, 7, 10501–10522. [Google Scholar] [CrossRef]
  58. Picotte, J.J.; Dockter, D.; Long, J.; Tolk, B.; Davidson, A.; Peterson, B. LANDFIRE remap prototype mapping effort: Developing a new framework for mapping vegetation classification, change, and structure. Fire 2019, 2, 35. [Google Scholar] [CrossRef]
  59. Fremgen-Tarantino, M.R.; Olsoy, P.J.; Frye, G.G.; Connelly, J.W.; Krakauer, A.H.; Patricelli, G.L.; Forbey, J.S. Assessing accuracy of GAP and LANDFIRE land cover datasets in winter habitats used by greater sage-grouse in Idaho and Wyoming, USA. J. Environ. Manag. 2021, 280, 111720. [Google Scholar] [CrossRef]
  60. Meddens, A.J.H.; Kolden, C.A.; Lutz, J.A.; Abatzoglou, J.T.; Hudak, A.T. Spatiotemporal patterns of unburned areas within fire perimeters in the northwestern United States from 1984 to 2014. Ecosphere 2018, 9, e02029. [Google Scholar] [CrossRef]
  61. Meddens, A.J.H.; Kolden, C.A.; Lutz, J.A. Detecting unburned islands within fire perimeters using Landsat and ancillary data across the northwestern United States. Remote Sens. Environ. 2016, 186, 275–285. [Google Scholar] [CrossRef]
  62. Balch, J.K.; Bradley, B.A.; D’Antonio, C.M.; Gómez-Dans, J. Introduced annual grass increases regional fire activity across the arid western USA (1980–2009). Glob. Chang. Biol. 2013, 19, 173–183. [Google Scholar] [CrossRef] [PubMed]
  63. Kerns, B.K.; Tortorelli, C.; Day, M.A.; Nietupski, T.; Barros, A.M.; Kim, J.B.; Krawchuk, M.A. Invasive grasses: A new perfect storm for forested ecosystems? For. Ecol. Manag. 2020, 463, 117985. [Google Scholar] [CrossRef]
  64. Keane, R.E.; Agee, J.K.; Fule, P.; Keeley, J.E.; Key, C.; Kitchen, S.G.; Miller, R.; Schulte, L.A. Ecological effects of large fires on US landscapes: Benefit or catastrophe? Int. J. Wildland Fire 2008, 17, 696–712. [Google Scholar] [CrossRef]
  65. Kreider, M.R.; Yocom, L.L. Low-density aspen seedling establishment is widespread following recent wildfires in the western United States. Ecology 2021, 102, e03436. [Google Scholar] [CrossRef]
  66. Donato, D.C.; Fontaine, J.B.; Campbell, J.L.; Robinson, W.D.; Kauffman, J.B.; Law, B.E. Conifer regeneration in stand-replacement portions of a large mixed-severity wildfire in the Klamath–Siskiyou Mountains. Can. J. For. Res. 2009, 39, 823–838. [Google Scholar] [CrossRef]
  67. Abella, S.R.; Gentilcore, D.M.; Chiquoine, L.P. Resilience and alternative stable states after desert wildfires. Ecol. Monogr. 2021, 91, e01432. [Google Scholar] [CrossRef]
  68. Furniss, T.J.; Das, A.J.; van Mantgem, P.J.; Stephenson, N.L.; Lutz, J.A. Crowding, climate, and the case for social distancing among trees. Ecol. Appl. 2022, 32, e2507. [Google Scholar] [CrossRef]
  69. Keyser, A.; Westerling, A.L. Climate drives inter-annual variability in probability of high severity fire occurrence in the western United States. Environ. Res. Lett. 2017, 12, 065003. [Google Scholar] [CrossRef]
  70. Johnson, E.A. Fire and Vegetation Dynamics: Studies from the North American Boreal Forest; Cambridge University Press: Cambridge, UK, 1996. [Google Scholar]
  71. Littell, J.S.; Peterson, D.L.; Riley, K.L.; Liu, Y.; Luce, C.H. A review of the relationships between drought and forest fire in the United States. Glob. Chang. Biol. 2016, 22, 2353–2369. [Google Scholar] [CrossRef] [PubMed]
  72. Stephens, S.L.; Ruth, L.W. Federal forest-fire policy in the United States. Ecol. Appl. 2005, 15, 532–542. [Google Scholar] [CrossRef]
  73. Steel, Z.L.; Safford, H.D.; Viers, J.H. The fire frequency-severity relationship and the legacy of fire suppression in California forests. Ecosphere 2015, 6, 1–23. [Google Scholar] [CrossRef]
  74. Pilliod, D.S.; Welty, J.L.; Arkle, R.S. Refining the cheatgrass–fire cycle in the Great Basin: Precipitation timing and fine fuel composition predict wildfire trends. Ecol. Evol. 2017, 7, 8126–8151. [Google Scholar] [CrossRef] [PubMed]
  75. Burgan, R.E. Estimating Live Fuel Moisture for the 1978 National Fire Danger Rating System; Intermountain Forest and Range Experiment Station, Forest Service, US Department of Agriculture: Ogden, UT, USA, 1979; Volume 226.
  76. Westerling, A.L.; Hidalgo, H.G.; Cayan, D.R.; Swetnam, T.W. Warming and earlier spring increase western US forest wildfire activity. Science 2006, 313, 940–943. [Google Scholar] [CrossRef]
  77. Cansler, C.A.; Kane, V.R.; Hessburg, P.F.; Kane, J.T.; Jeronimo, S.M.A.; Lutz, J.A.; Povak, N.A.; Churchill, D.J.; Larson, A.J. Previous wildfires and management treatments moderate subsequent fire severity. For. Ecol. Manag. 2022, 504, 119764. [Google Scholar] [CrossRef]
  78. Churchill, D.J.; Jeronimo, S.M.A.; Hessburg, P.F.; Cansler, C.A.; Povak, N.A.; Kane, V.R.; Lutz, J.A.; Larson, A.J. Post-fire landscape evaluations in eastern Washington, USA: Assessing the work of contemporary wildfires. For. Ecol. Manag. 2022, 504, 119796. [Google Scholar] [CrossRef]
  79. Radeloff, V.C.; Helmers, D.P.; Kramer, H.A.; Mockrin, M.H.; Alexandre, P.M.; Bar-Massada, A.; Butsic, V.; Hawbaker, T.J.; Martinuzzi, S.; Syphard, A.D.; et al. Rapid growth of the US wildland-urban interface raises wildfire risk. Proc. Natl. Acad. Sci. USA 2018, 115, 3314–3319. [Google Scholar] [CrossRef]
  80. Mietkiewicz, N.; Balch, J.K.; Schoennagel, T.; Leyk, S.; St Denis, L.A.; Bradley, B.A. In the line of fire: Consequences of human-ignited wildfires to homes in the US (1992–2015). Fire 2020, 3, 50. [Google Scholar] [CrossRef]
Fire 06 00423 g001
Figure 1. Wildfire boundaries for fires ≥ 40 ha that burned from 1984 to 2021 (a) and vegetation categories aggregated from existing vegetation type vegetation classifications [27,28] (b) in Utah, USA. Background imagery (a) sourced from 0.6 × 0.6 m 2021 NAIP imagery [29].
Figure 1. Wildfire boundaries for fires ≥ 40 ha that burned from 1984 to 2021 (a) and vegetation categories aggregated from existing vegetation type vegetation classifications [27,28] (b) in Utah, USA. Background imagery (a) sourced from 0.6 × 0.6 m 2021 NAIP imagery [29].
Fire 06 00423 g001
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Birch, J.D.; Lutz, J.A. Fire Regimes of Utah: The Past as Prologue. Fire 2023, 6, 423. https://doi.org/10.3390/fire6110423

AMA Style

Birch JD, Lutz JA. Fire Regimes of Utah: The Past as Prologue. Fire. 2023; 6(11):423. https://doi.org/10.3390/fire6110423

Chicago/Turabian Style

Birch, Joseph D., and James A. Lutz. 2023. "Fire Regimes of Utah: The Past as Prologue" Fire 6, no. 11: 423. https://doi.org/10.3390/fire6110423

APA Style

Birch, J. D., & Lutz, J. A. (2023). Fire Regimes of Utah: The Past as Prologue. Fire, 6(11), 423. https://doi.org/10.3390/fire6110423

Article Metrics

Back to TopTop